Environmental Economics, Experimental Methods

The 1970s and 1980s saw environmental economists increasingly turn to experimental methods in an attempt to discover new ways of protecting people and nature without wasting scarce resources in the process. Today the experimental method is commonly applied to environmental economic questions; this book brings together 63 leading researchers in the area and their latest work exploring the behavioral underpinnings of experimental environmental economics. The chapters in this volume will be illuminating for both researchers and practitioners, specifically in relation to questions of environmental policy and how a proposed change in incentives or benefits might affect behavior and, consequently, the likely success of a policy. This book argues that the experimental evidence complements theoretic insights, field data and simulating models to improve our understanding of the underlying assumptions and incentives that drive behavioral responses to policy. This volume covers topical areas of interest such as tradable permit markets, common property and public goods, regulation and compliance and valuation and preferences. Its critical advantage is that each part concludes with discussion points written by environmental economists who do not use experimental methods. This book will interest students and researchers engaged with environmental economics, both experimental and non-experimental and offer a unique in-road into this field of study. Environmental policy makers will also gain insight into behavior and decision making under alternative institutional and policy designs. Todd L. Cherry is the Harlan E. Boyles Professor in the Department of Economics at Appalachian State University, where he also is a research fellow at the Appalachian Energy Center. Stephan Kroll is Professor at the Department of Economics at California State University in Sacramento. Jason F. Shogren is the Stroock Distinguished Professor of Natural Resource Conservation and Management, and Professor of Economics at the University of Wyoming.

Routledge explorations in environmental economics Edited by Nick Hanley
University of Stirling, UK

1

Greenhouse Economics Value and ethics Clive L. Spash Oil Wealth and the Fate of Tropical Rainforests Sven Wunder The Economics of Climate Change Edited by Anthony D. Owen and Nick Hanley Alternatives for Environmental Valuation Edited by Michael Getzner, Clive Spash and Sigrid Stagl Environmental Sustainability A consumption approach Raghbendra Jha and K.V. Bhanu Murthy Cost-Effective Control of Urban Smog The significance of the Chicago cap-and-trade approach Richard F. Kosobud, Houston H. Stokes, Carol D. Tallarico and Brian L. Scott Ecological Economics and Industrial Ecology Jakub Kronenberg Environmental Economics, Experimental Methods Edited by Todd L. Cherry, Stephan Kroll and Jason F. Shogren

2

3

4

5

6

7

8

Environmental Economics, Experimental Methods

Edited by Todd L. Cherry, Stephan Kroll, and Jason F. Shogren

First published 2008 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016 This edition published in the Taylor & Francis e-Library, 2007. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” Routledge is an imprint of the Taylor & Francis Group, an informa business © 2008 Selection and editorial matter, Todd L. Cherry, Stephan Kroll and Jason F. Shogren; individual chapters, the contributors All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloguing in Publication Data A catalog record for this book has been requested ISBN 0-203-93536-5 Master e-book ISBN ISBN10: 0-415-77072-6 (hbk) ISBN10: 0-203-93536-5 (ebk) ISBN13: 978-0-415-77072-9 (hbk) ISBN13: 978-0-203-93536-1 (ebk)

Contents

List of figures List of tables List of contributors Foreword
VERNON L. SMITH

ix xii xvi xix

Introduction
TODD L. CHERRY, STEPHAN KROLL, AND JASON F. SHOGREN

1

PART I

Tradable permit markets
1 Baseline-and-credit emission permit trading: experimental evidence under variable output capacity
NEIL J. BUCKLEY, STUART MESTELMAN, AND R. ANDREW MULLER

7

9

2 A laboratory analysis of industry consolidation and diffusion under tradable fishing allowance management
CHRISTOPHER M. ANDERSON, MATTHEW A. FREEMAN, AND JON G. SUTINEN

29

3 Caveat emptor Kyoto: comparing buyer and seller liability in carbon emission trading
ROBERT GODBY AND JASON F. SHOGREN

47

4 A test bed experiment for water and salinity rights trading in irrigation regions of the Murray Darling Basin, Australia
CHARLOTTE DUKE, LATA GANGADHARAN, AND TIMOTHY N. CASON

77

vi Contents 5 Aligning policy and real world settings: an experimental economics approach to designing and testing a cap-and-trade salinity credit policy
JOHN R. WARD, JEFFERY CONNOR, AND JOHN TISDELL

100 131

6 Discussion: tradable permit markets
DALLAS BURTRAW AND DAN SHAWHAN

PART II

Common property and public goods
7 Communication and the extraction of natural renewable resources with threshold externalities
C. MÓNICA CAPRA AND TOMOMI TANAKA

141

143 157 184

8 Unilateral emissions abatement: an experiment
BODO STURM AND JOACHIM WEIMANN

9 Voluntary contributions with multiple public goods
TODD L. CHERRY AND DAVID L. DICKINSON

10 Can public goods experiments inform policy? Interpreting results in the presence of confused subjects
STEPHEN J. COTTEN, PAUL J. FERRARO, AND CHRISTIAN A. VOSSLER

194

11 Spies and swords: behavior in environments with costly monitoring and sanctioning
ROB MOIR

212 234

12 Discussion: common property and public goods
CATHERINE L. KLING

PART III

Regulation and compliance
13 Managerial incentives for compliance with environmental information disclosure programs
MARY F. EVANS, SCOTT M. GILPATRIC, MICHAEL MCKEE, AND CHRISTIAN A. VOSSLER

241

243

14 An investigation of voluntary discovery and disclosure of environmental violations using laboratory experiments
JAMES J. MURPHY AND JOHN K. STRANLUND

261

Contents vii 15 Congestion pricing and welfare: an entry experiment
LISA R. ANDERSON, CHARLES A. HOLT, AND DAVID REILEY

280

16 Social preferences in the face of regulatory change
J. GREGORY GEORGE, LAURIE T. JOHNSON, AND E. ELISABET RUTSTRÖM

293

17 The effects of recommended play on compliance with ambient pollution instruments
ROBERT J. OXOBY AND JOHN SPRAGGON

307 324

18 Discussion: regulation and compliance
KATHLEEN SEGERSON

PART IV

Valuation and preferences
19 Preference reversal asymmetries in a static choice setting
TIMOTHY HAAB AND BRIAN ROE

329
331 344

20 Measuring preferences for genetically modified food products
CHARLES NOUSSAIR, STEPHANE ROBIN, AND BERNARD RUFFIEUX

21 An experimental investigation of choice under “hard” uncertainty
CALVIN BLACKWELL, THERESE GRIJALVA, AND ROBERT P. BERRENS

366

22 Rationality spillovers in Yellowstone
CHAD SETTLE, TODD L. CHERRY, AND JASON F. SHOGREN

383

23 Wind hazard risk perception: an experimental test
BRADLEY T. EWING, JAMIE B. KRUSE, AND MARK A. THOMPSON

395

24 Consequentiality and demand revelation in double referenda
KATHERINE S. CARSON, SUSAN M. CHILTON, AND W. GEORGE HUTCHINSON

407

viii Contents 25 Investigating the characteristics of stated preferences for reducing the impacts of air pollution: a contingent valuation experiment
IAN J. BATEMAN, MICHAEL P. CAMERON, AND ANTREAS TSOUMAS

424

26 Forecasting hypothetical bias: a tale of two calibrations
F. BAILEY NORWOOD, JAYSON L. LUSK, AND TRACY BOYER

447

27 Discussion: valuation and preferences
JOHN C. WHITEHEAD

466

Index

476

Figures

1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 2.1 2.2 2.3 2.4 2.5 2.6 3.1 3.2 3.3 3.4 3.5 3.6 3.7 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9

Firm cost curves Sequence of events in a typical period Cap-and-trade equilibrium Baseline-and-credit equilibrium Capacity Output volume Aggregate emissions Efficiency Permit trading prices Aggregate permit inventory Profit functions for operators Diffusion treatment prices with an initial lease period Consolidation treatment prices with an initial lease period Average efficiency in the diffusion treatment Average efficiency in the consolidation treatment Percentage of market shares held by large and medium–large operators Session procedure Efficiency and emission outcomes by treatment Mean aggregate buyer production by treatment Mean aggregate seller production by treatment Mean permit prices by treatment Mean trades by treatment Efficiency and emission outcomes by treatment The Sunraysia irrigation districts and salinity impact zones Market demand and supply for water Transaction price in the water market, treatment 1 Transaction price in the water market, treatment 2 Transaction price in the water market, treatment 3 Transaction quantity in the water market, treatment 1 Transaction quantity in the water market, treatment 2 Transaction quantity in the water market, treatment 3 Transaction price in the salt market, treatment 2

12 13 18 19 22 22 23 23 25 26 35 37 37 39 40 41 56 59 60 61 61 68 72 79 83 87 88 88 89 89 90 91

including gains from trade.3 11.1 5. treatment 3 Transaction quantity in the salt market.2 5.1 8.1 11. mean contributions Aggregate CPR appropriations by treatment Gross efficiency gain over Nash by treatment Net efficiency gain over Nash by treatment Group monitoring levels by treatment Benefits and costs of the risky route An entry game session with 60 rounds (me070804) Predicted and observed distributions of entry outcomes for a session with 60 rounds (me070804) 91 92 92 103 117 119 119 121 123 147–8 149 151 152 152 166–7 167 170–1 172 172 174–5 177 189 189 190 191 203 206 220 221 222 225 283 285 286 4.4 7.4 8.11 4.5 8.7 9.x Figures Transaction price in the salt market.1 10.2 9.3 9.3 5.5 5.12 5. treatment 2 Transaction quantity in the salt market.4 15. comparison of all-human and virtual-player contributions Ferraro and Vossler (2005) experiment.6 7. uniform price.3 8.3 .1 15. treatment 3 Schematic of the hydrogeology of irrigation water quality affected by variable upper catchment salt loads Observed aggregate recharge in the discriminant and uniform price tender treatments Observed and predicted bids for the uniform tender treatment Observed and predicted bids for the discriminant tender treatment Aggregate farm income.3 7. observed in the open and closed call market treatments Recharge units traded in the 70 percent reduction.5 8.4 5.2 8.1 9.1 7.10 4.2 11.6 8.2 15.4 10.2 7. social payment and communication treatments Production functions Average resource amount Resource stock and number of subjects with relevant messages in each period and horizon Resource stock and number of subjects with relevant messages in each period and horizon Resource stock and number of subjects with relevant messages in each period and horizon Abatement per period Scatterplots for seq-treatments Abatement over periods Alpha Individual behavior of country 1 and j Profit per period Classification of 36 groups in the sequential treatments MPCR by group account for multiple heterogeneous treatment Total contributions to group accounts by treatment Group contributions across homogeneous competing group accounts Group contribution across heterogeneous competing group accounts GHL application.2 11.

3 17.1 23.7 19. by treatment.4 15.Figures xi 15.1 17.1 20.1 17. l 21. tax/subsidy Distributions of individual decisions.1 23. tax instrument Mean decision by subject type and period. tax/subsidy instrument Mean group totals by treatment and period. experiment 2 Basic decision tree Histogram of participant criteria selection for scenarios 1–3 Preference reversal rates Subject predictions on failure by shingle loss and building breach (modular and manufactured test specimens) Certainty equivalents for incorrect and correct answers Schematic of the design of the double referendum experiments 286 288 289 301 313 313 314 314 317 318 319–20 337 355 358 370 375 390 401 404 410 . tax/subsidy instrument Distributions of individual decisions.6 16.l 20. tax Between-subject comparison: experiment 1 post-treatment preferences for tasks (by treatment) Average bids for the four biscuits in each period of GMO phase.2 17.4 17.2 24.6 17.2 22.1 Entry game with $2 entry fee (me062904) Entry game with information about other entrants (me071504) Entry game with voting on entry fee (me063004) Distributions of bids for the alternative solutions Mean group totals by treatment and period.5 15. by treatment. experiment 1 Average bids for the two identical chocolate bars in periods 1–3.5 17. tax/subsidy instrument Mean efficiency by treatment and period. tax instrument Mean efficiency by treatment and period.2 21.

6 3.3 3.4 7.3 4. uniform price. social payment and communication tender treatments Parameters of the model 18 18 20 21 24 34 38 43 52 55 63 64 65 66 67 69 71 82 82 86 86 94 111 118 121 123 145 .1 3.Tables 1.5 5.3 3.2 4. crop mix and recharge of two of 12 experimental farms ANOVA of discriminant and uniform price tender auction treatments ANOVA of closed and open call market treatments ANOVA of 70 percent reduction.2 1.9 4.5 2.3 5.1 1. farm income.4 4.8 3.1 4.4 3.5 3.3 1.4 1.2 5.1 2.2 2.1 Cost parameters Variable capacity predictions Equilibrium surplus efficiency Mean values over periods 6 to 9 by treatment Mean efficiency over periods 6 to 9 Market share holdings for operators First-order autoregressive model of average prices in asset rounds 5 and 6 Two-way random effects regression of market share consolidation Experimental design Experiment conditions and producer costs Experiment results by treatment Buyer mean production choices (periods 6–8) Seller mean production choices (periods 6–8) Session mean price results (periods 6–8) Estimated regression results (p-values in parentheses) Experiment results by treatment: additional sessions Overproduction-to-total sales ratio Parameter ranges from Sunraysia Private redemption and cost values Experimental design Model equilibrium predictions Random effects estimation models for average transaction price in the water market Decision set.1 5.2 3.7 3.

3 Within treatment comparisons 11.4 Probit model results 14.4 Across treatment comparisons 13.3 Summary of parameters. abatements. under the tax/subsidy by treatment 17. and payments 8. all-human treatment results 10. by view treatment 15.1 GHL application.1 Experimental design 9.5 Efficiency index 9.a.3b Characteristics of subject pool from each experiment: demographic questionnaire responses 16.1 Sessions.1 Per period activity table for stage one 16. profits. mean contributions 11. expected numbers of violations.2 Design parameters by treatment 13.4 Regression analysis 8.2 Post-treatment preferences by treatment 19.3 Probit results on within-subject preference reversals.3 Mean aggregate decision numbers.2 Number of periods per horizon 7.3a Characteristics of subject pool from each experiment: treatment characterization (N = 168).3 Observed cheating in experiments 13. under the tax by treatment 19.1 Cost and benefit functions 8.3 Mean violation probabilities.1 Experimental design 14.1 Mean aggregate decision numbers by treatment 17.2 Sequence of treatments using a Latin Square 14. treatments. stage 1 and stage 2 results 16.3 GHL application. and expected numbers of enforcement actions 15.4 Regression results 17.1 Group contribution predictions 11.2 Experimental treatments 8. and data averages 16. virtual-player treatment results 10. (2003) VCM experiment.1 Payoff for the risky route 15.2 Variances of entry rates in the first ten rounds.2 GHL application.3 Random effects GLS estimation of the effects of communication on the levels of resource stock 8.l Anchor treatments for experiment 1 19.4 Ferraro et al. experiment 1 146 154 161 161 162 168 175 187 190 201 202 203 206 219 221 223 224 248 253 254 255 268 270 271 282 288 291 296 297 298 299 300 302 312 315 316 335 336 338 .Tables xiii 7.2b Selected distributions for one-stage experiment 16.2 Summary data 11.2a Typical distribution options in stage two 16. estimated logit equilibrium models 10.1 Potential cheating and non-compliance cases 13.2 Individual contributions: two-way fixed effect estimates 10.2 Mean aggregate decision numbers.

1 Sequence of events in GMO phase of an experimental session.3 Outcomes for treatment 2 21.6 Chi-squared p-values for tests of differences between vote distributions.5 Chi-squared p-values for tests of differences between rates of non-demand revelation in the inconsequential double referendum and consequential double referendum treatments 24.2 Participants’ perceptions of the seriousness of the lake trout introduction 22.4 Probit results on within-subject preference reversals.4 Participants’ selections for scenarios 1–3 21.1 Descriptive statistics of experiment 340 353 359 372 372 373 374 376 377 377 385 386 386 387 392 403 411 416 417 417 417 418 432 433 435 436 438 440 441 456 .xiv Tables 19.3 Descriptive WTP statistics by subsample and scheme 25.5 Variable definitions and descriptive statistics 21.1 Regret analysis 21.5 Mean and median WTP (£) for three air pollution impact reduction schemes.2 Outcomes for treatment 1 21.1 Predicted and observed vote distributions – inconsequential double referendum 24.7 Multinomial logit model 2 22.2 Demand revealing predictions and observed vote distributions – consequential double referendum treatments 24.6 Treatment effects 25.1 Experimental design and subsample structure 25. conditioned on first vote outcome 25.6 Multinomial logit model 1 21. induced value phase of both experiments 21.5 Logit regression results for preference reversals 23.4 Levels and rates of non-demand revealing voting 24.7 Comparing stated WTP for Scheme A with values calculated from stepwise first responses for Scheme H and Scheme P 26.3 Chi-squared p-values for differences in vote distributions in inconsequential double referendum and consequential double referendum treatments. experiment 2 20.2 Socio-economic and demographic profile of subsamples 25.1 Descriptive statistics certainty equivalents and accuracy 24.2 Deviations of bids from valuation.3 Participants’ preference for fish 22. experiment 1 20.4 Percentage of participants affected by seeing attractions of the park 22.4 Significance of differences in WTP values for schemes 25. by five treatments 25. by subject type 24.1 Familiarity of participants to the lake trout introduction 22.

hypothetical bid residuals.1 Written answers to question 457 460 462–3 .Tables xv 26.3 Stochastic frontier estimation 26.2 Relationship between certainty question. and hypothetical bias 26.a.

University of Rhode Island. Oklahoma. USA. Todd L. Neil J. Texas Tech University. USA. UK. Stephen J. Policy and Economic Research Unit. Evans. C. North Carolina. Cason. Anderson. Indiana. Appalachian State University. Tennessee. University of Tennessee. Cotten. Georgia. CSIRO. . University of Waikato. Australia. Colorado. York University. Mónica Capra. Michael P. Resources for the Future. Emory University. University of New Mexico. Cameron. Anderson. Georgia. Lisa R. USA. USA. Ewing. Paul J. David L. College of Charleston. Chilton. USA. Bateman. Oklahoma State University. University of Newcastle upon Tyne. Jeffery Connor. UK. Ian J. DC. USA. Australia. USA.Contributors Christopher M. Texas. Purdue University. Charlotte Duke. Carson. Bradley T. Appalachian State University. Cherry. New Mexico. University of East Anglia. USA. USA. USA. North Carolina. New Zealand. Susan M. Victoria. USA. Canada. Tennessee. Mary F. University of Melbourne. Dickinson. Robert P. Toronto. USA. USA. Rhode Island. Dallas Burtraw. USA. United States Air Force Academy. Berrens. Hamilton. Calvin Blackwell. University of Tennessee. Katherine S. Washington. Georgia State University. Norwich. College of William and Mary. Buckley. Newcastle upon Tyne. Virginia. USA. Timothy N. South Carolina. Tracy Boyer. Ferraro.

USA. Oklahoma State University. Brian Roe. Gilpatric. UK. Holt. Colorado. Oklahoma. University of Connecticut. Jayson L. Arizona. David Reiley. USA. Kling. Oklahoma. Tilburg. Iowa. Michael McKee. California. Macon State College. Wyoming. Stuart Mestelman. Queen’s University. Gregory George. USA. USA. East Carolina University. University of New Brunswick. Kruse. University of Tulsa. Tennessee.Contributors xvii Matthew A. Alberta. France. Ohio. Ohio State University. Utah. California State University. E. USA. Elisabet Rutström. Freeman. James J. USA. Tilburg University. Canada. Bernard Ruffieux. Florida. Hamilton. Kathleen Segerson. Strasbourg. J. USA. Catherine L. University of Melbourne. Lata Gangadharan. Stephan Kroll. Canada. Andrew Muller. Grenoble. Jamie B. Chad Settle. R. Australia. Anchorage. Belfast. Rob Moir. F. North Carolina. University of Arizona. . Victoria. McMaster University. Ontario. USA. Lusk. Robert Godby. USA. Robert J. Stephane Robin. University of Alaska. Murphy. France. Oxoby. University of Wyoming. North Carolina. Johnson. University of Tennessee. the Netherlands. USA. Rhode Island. Ontario. Hamilton. Connecticut. USA. USA. New Brunswick. USA. USA. Charles Noussair. University of Virginia. Canada. USA. Charles A. University of Central Florida. USA. Timothy Haab. University of Calgary. University of Denver. USA. George Hutchinson. Therese Grijalva. Virginia. Université Louis-Pasteur. Weber State University. Oklahoma. Canada. Ohio State University. Laurie T. Bailey Norwood. Georgia. Institut National de la Recherche Agronomique. USA. University of Rhode Island. Scott M. Appalachian State University. Ohio. W. Iowa State University. USA. McMaster University. USA. Oklahoma State University.

USA. John R. USA. Massachusetts. Jason F. University of Tennessee. Ward. USA. Arizona State University. Mytilene. Germany. Australia. Queensland. Wyoming. CSIRO. Centre for European Economic Research (ZEW). USA. Otto von Guericke University Magdeburg. Tennessee. Cornell University. USA. John K. New York. Arkansas. Griffith University. North Carolina. Thompson. Policy and Economic Research Unit. Christian A. University of Massachusetts. USA. Rhode Island. Vossler. Australia. Little Rock. Kingston. Germany. Whitehead. University of Wyoming. Stranlund. Arizona. Magdeburg. Tomomi Tanaka. USA. University of Massachusetts. John C. Massachusetts. Jon G. Antreas Tsoumas. Shogren. . University of the Aegean. University of Arkansas. Bodo Sturm. USA. Mannheim. Sutinen. University of Rhode Island. Joachim Weimann. Mark A. USA.xviii Contributors Dan Shawhan. Appalachian State University. John Tisdell. John Spraggon. Greece.

This dramatized the concept of market failure. The prevailing wisdom emphasized the impossibility of the private provision of goods whose outcomes were common across all users. Pre-dating Hardin was the less flashy and not well known but classic paper by Gordon (1954). . emitting signals that all ships could observe at zero marginal cost. This insight was reinforced by Acheson’s (1975) study of the Maine lobstermen who created home-grown privately enforced property rights in the open lobster sea beds off the coast of Maine. and of the impossibility “theorem” in the private provision of public goods. however. Indeed. and these dockings provided an effective and practical measure of lighthouse service utilization and value in consumption. also that almost all such resources users were. that was the conceptual key to finding solutions. The canonical example of a pure public good. excludable if you could just find the way.Foreword Public goods theory and its allied “tragedy of the commons” began on a deeply pessimistic note 40–50 years ago with the contributions of Samuelson (1954) and Hardin (1968). let alone developed its tools for a theory of public goods. The theoretical argument that for “efficiency” the so-called “fixed” cost once incurred should not affect the price of lighthouse services was a fallacious non-starter because it omits the inefficiency that results if the lighthouse is not built! Docks. Samuelson and Hardin had swept the field. These “incentive” contracts allowed the capital cost of lighthouses to be prorated among ship dockings. which emphasized that where there was a resource management failure it was useful to think of the problem as one of property rights failure. lighthouses and ships all value lighthouse services and the contracts uncovered by Coase had focused their mutual interest on a solution to this public goods problem. But Coase in effect asked. All ships have to dock somewhere. By the 1970s. use lighthouses on the way. or potentially were. on fisheries. main stream economics thought and taught that public goods could not be produced efficiently by private means. The problem of supplying incentives for private investments and aborting free riders was solved very practically by lighthouse owners who contracted with port authorities to charge docking ships for lighthouse services. was the lighthouse. “I wonder how the early light houses came about and who financed them?” As it turned out lighthouses were privately financed before economics had become a well defined profession. But a key contribution to inducing an about-face in the thinking of economists came from Coase (1976) on lighthouses.

salinity control. The imagination had to range beyond the mathematics of incentive failure. test bed. Equally important. as in the examples by Netting and Coase. I won’t belabor that story or its literature. As Hayek. the solution that Coase found people had used to build lighthouses was actually based on the same principle used in the Swiss Alps. 139). Moreover. puts it. You will find it in this book and my Papers in Experimental Economics (1991). at least since AD 1224 for the high Alpine Swiss cheese makers who each summer pastured their cows on the commons: entry to summer pastures was controlled by a property right rule that “no citizen could send more cows to the alp than he could feed during the winter” (Netting 1976. We now think differently. water markets. more openly and positively on these issues.xx Foreword And the famous “tragedy of the commons” in grazing cattle was not necessarily a tragedy. and in all such cases the question is whether there are feasible ways of limiting use to avoid or internalize external costs. or assuring payments that cover investment cost. perfect and apply human ingenuity to the creation of incentives that make the solutions to these problems possible. but which alone could not facilitate a solution. This exciting collection of research papers provides a compendium of practical earthy rule-governed examples – fisheries management. p. but experienced enough in their behavioral coordination problem to seek solutions that might work. Public goods theory will never be the same.” The rights to a common were tied to a corollary privatized right. one badly needed an empirical testing ground for exposing new models and ways of thinking to tests of effectiveness. Theory had enabled us to see that these were examples of excludable public goods. emission abatement. Smith Arlington. The reader will have noticed by now that 1976 was a good year. VA . These economic design problems were solved by people completely unschooled in free rider theory. Somehow they perfected them by trial and error “natural experiments” over time. namely. and I will make my point by paraphrasing Netting. That early fledgling literature has grown into an imposing contribution collected in this book on how to model. are often ingenious beyond the imagination of the first pencil and paper theories whose primary value was in enabling us to see why there is and were problems that needed solution.” Enjoy! Vernon L. “Rules alone can unite and extend order. Experimental economics responded positively and effectively to this challenge beginning in the 1970s and 1980s when the Samuelson problem of public goods was addressed. congestion control and related measurement and monitoring issues – all in the ancient problem solving spirit of the Alpine Swiss cheese makers and the entrepreneurs who believed lighthouses could be provided privately and did it. “no shipping company could pass more of its ships past the lighthouse than it paid for as part of ship docking charges. along with the Gordon–Hardin problem of managing common property resources. The solutions.

Netting. “The lobster fiefs: economic and ecological effects of territoriality in the Maine lobster industry. Coase. 1976. Cambridge: Cambridge University Press.A. 50. Gordon. 1954.” Human Ecology 4 (2). J. Papers in Experimental Economics. Andrews.” Science 162: 1243–8. 1975. Scott.” Review of Economics and Statistics 36 (4) (November). Samuelson. St. H. Selected Papers No. 135–46.” Human Ecology 3 (3). 1–33). “What Alpine peasants have in common: observations on communal tenure in a Swiss village. Graduate School of Business. 1968. 1991. “The pure theory of public expenditure. Hardin. 1954. Robert. Garrett. 124–42. P. R. 387–9. University of Chicago (Mont Pelerin Society.” Journal of Political Economy 62 (2).H. Smith. Vernon L.M. . 183–207.Foreword xxi References Acheson. “The tragedy of the commons. 1976. Adam Smith’s View of Man. “The economic theory of a common-property resource: the fishery.

.

and Michael McKee turned to laboratory experiments to test the efficacy of alternative mechanisms to provide public goods efficiently and voluntarily. researchers in environmental economics have discovered and created new methods and tools aimed at protecting people and nature without wasting scarce resources in the process. Shogren Greater environmental protection at fewer costs – few people would disagree with the idea that environmental policy should be trying to achieve this wideranging goal. with the input of researchers in other disciplines like biology. Jack Knetsch. Vernon Smith. A reader might be asking him. Over the last five decades. These economists and policy makers. Such pragmatism led to a natural progression for researchers in environmental economics – they quickly adopted the methods of the newly emerging area of experimental economics in the 1970s and 1980s. “some of the earliest work in experimental economics was done by environmental economists. Stephan Kroll and Jason F. Don Coursey.” Economists like Peter Bohm. and to explore how well Pigovian taxes work relative to Coasean bargaining solutions to resolve externality problems. As Cathy Kling points out in Chapter 12 of this volume. have taken a pragmatic approach to their task. Charles Plott. Today. Lay people and policy makers must make decisions within a mix of biotic and abiotic phenomena combined with social institutions like markets and nonmarket allocation systems. and the experimental method is commonly applied to environmental economic questions. William Schulze. we have come full circle. Elizabeth Hoffman. Jeff Bennett. Do the attempts to use the experimental method to understand better the micromotives that underpin the theory of environmental economics have anything to say about the efficiency and fairness of global environmental policy? . to understand how people value gains and losses of a good or service. forestry and hydrology.Introduction Todd L. Ron Cummings. pursuing what works rather than preconceived mindsets. We all know environmental protection is more complex than any laboratory or field study. Cherry.or herself whether such small-scale experiments are the appropriate tool to test large-scale environmental policy. as evidenced by the research in this book and in the general economics literature. How to identify and implement the strategies that can move people toward this goal falls within the purview of environmental economics.

Some of these researchers are environmental economists who occasionally employ experimental methods. to test the robustness of the axioms underlying the theory. experiments might affect how policy is formed and evaluated. researchers use experiments to test the predictive power of a theory. the huge body of experimental research on contributions to public goods. For example. Second.2 T. By supplying information on the behavioral link between incentives. Third. and test bed new institutions designed to protect nature. to identify and measure breakdowns from rationality. values. and simulation models to improve our understanding of the underlying assumptions and incentives that drive behavioral responses to policy. to determine reactions to new information. The lessons learned by experimental economists can help guide environmental policy by providing insights into how a proposed change in incentives or benefits might affect behavior and. to examine how contextual frames affect behavior. Each part concludes with a discussion chapter. We divide the 24 chapters into four topical parts that cover the range of ongoing research today – tradable permit markets. and choice. For example. First.1 Experiments have proven to be a useful tool to stress-test theory. see Shogren 2006). Since the laboratory environment differs from the wilds by necessity. common property and public goods.L. and mechanisms designed to improve resource allocation.g. would be our answer. This volume brings together 63 leading researchers and their latest work exploring the behavioral underpinnings of environmental economics using experimental economic methods. consequently. and to consider how people coordinate actions voluntarily and under duress. Cherry et al. to test the specific boundaries of a general theory. See. and environmental economists unfamiliar with experimental methods will find useful information and ideas for future research (experimental and non-experimental) here. economists use the lab to look for patterns of behavior. others are experimental economists who use their research method in different subfields of economics. look for empirical patterns of behavior. laboratory experiments are used as a test bed for institutional design – the construction of new institutions. regulation and compliance. it does. one of which happens to be environmental economics. Cason and Plott (1996) examined in a laboratory environment the incentives for sellers in new emission trading mechanisms proposed by the US Environmental Protection Agency (see the overview on incentive design in Bohm 2003). The audience for this volume is as diverse as the authors are – we hope both experimental economists who want to conduct policy-relevant work. the likely success of a policy. Experimental evidence complements theoretical insight. and valuation and preferences. Yes. To bridge the gap between experimental methods and non-experimental . field data. experimental data used to back stated positions of policy should be viewed as support for or against a specific case of a more general phenomenon or theory. the long debate on the divergence between willingness to pay and willingness to accept has been rejuvenated by experimental research (e. for example. markets. and to measure the gradient of behavioral change.

and experimental economists who try to reach an audience broader than just the circle of other experimental economists should read these chapters and the advice therein carefully. Eckel et al. and extreme prices. “along the continuum towards the more immediate policy relevance side by adding more relevant context. John Whitehead points out that on the other end of the continuum “contingent valuation economists go overboard on supplying contextual information. p. and Dan Shanan. from Resources for the Future.Introduction 3 research into environmental economics and policy. and an inquiry of market designs for unconventional tradable permits. It is still useful for an experimental economist to be reminded that the use of context-free language is not the only and not always the preferred option. at Cornell University. the discussants wonder if this neutral-language doctrine can sometimes be relaxed to move. The one common thread through all four discussion chapters is the call for more context. bubbles. 2005) and political acceptance (e.” Given their environmental policy backgrounds. we asked four renowned environmental economists to write the discussion chapters and to give the reader outside views on how relevant the chapters in this book and experimental economics in general are for practical work on real-world environmental problems. recognize the fundamental tradeoff. 13). What follows is a brief summary of the insights found in the discussion chapters. The authors of the discussion chapters did a marvelous job in providing critical summaries of the respective papers in each section.” which makes it difficult to extrapolate beyond the current situation. and give us and the readers your view toward how helpful experimental methods are or could be as a tool to help inform your research. studied the five chapters in Part I on “Tradable Permit Markets” and suggest useful research extensions for each one of them. of course. they present some more general topics for experimental research on permit markets: a laboratory comparison of different auction designs under different (and partly competing) government goals.g. . In addition. describe what you think was missing and still needed in particular or in general. This is particularly true in research areas like valuation and bargaining. or (Pigovian) taxation in which just the use of the word “tax” can have significant impact on behavior (e. and “it is an accepted practice in economics experiments to strip away a lot of social context that is not an essential part of the economic theories being tested” (Holt 2006. Cummings and Taylor 1999. react to what you found useful for the general topic.” All discussants. The instructions we gave the discussants were broad: read a set of chapters. Cherry 2001). Kallbekken and Kroll 2007).g. What other themes do the non-experimentalists address? Dallas Burtraw.g. a test of sensitivity of different permit market designs to market power. We purposefully chose researchers who are not experimental economists per se. as Cathy Kling writes. For example. Experimental economists traditionally use “context-free” settings and instructions in their experiments to make the experiment as general and applicable as possible. most public good experiments employ language such as “contributions to a public account” or “investments. in which context matters a lot (e.

Research into less-than-rational behavior of consumers and voters he calls interesting and productive “but not referenda on neoclassical theory. where one’s strength can help to cover the other one’s weaknesses. Interestingly. from Appalachian State University. all experiments. he calls for combining forces – research studies that include the laboratory and surveys simultaneously. including those in environmental economics. She suggests three “inter-related issues that arise in assessing the usefulness of laboratory experiments in understanding environmental policies”: in addition to context.” that laboratory experiments “provide a very useful middle ground [between economic theory and empirical analysis of actual real world policies] for investigating the likely impacts of proposed policies at lower cost. One controls the experimental circumstances to avoid confounding. and makes a case for adding relevancy by doing three things that most experimental economists do not do: describing the good in question in non-generic terms. which makes sufficient incentive structures expensive to implement). or . it seems “imperative” that certain experiments are conducted in a field environment. it is unclear whether unpredicted behavior is due to a poor theory or experimental design. from Iowa State University. While the latter two suggestions might sometimes be at odds with each other (using subjects from directly relevant pools usually means subjects with higher opportunity costs of time. In his witty essay on the contributions in Part IV. points out in her discussion of Part III.” For her too. At the core. Because in the end. states that despite their flaws due to the lack of context. Whitehead. the experimental method is about control.” She focuses on and emphasizes the question of context.4 T.” Our four discussants have provided the ideal foil against which the contributors and editors can rethink how we use and sell the experimental method. two or more elements change.” He sees laboratory experiments and stated preference surveys as complementary approaches. Cathy Kling. however.” John C. increasing the payoffs for subjects. and using subjects from a directly relevant subject pool. the other two are the actual purpose of the experiment and the evaluation of alternative policy approaches.e. Without control. “Valuation and Preferences. “Regulation and Compliance. economic experiments have done a reasonable job in getting contingent valuation economists “out of their orbit around a far off hypothetical planet. from the University of Connecticut. she points out that she is more interested in knowing that a proposed policy mechanism is shown to be effective under laboratory conditions than that it is efficient since she believes that an effective outcome is more likely to be replicated in the real world than an efficient outcome. i.L. discusses the chapters in Part II on “Common Property and Public Goods. In particular. which confounds our understanding of cause-and-effect. reveal the perpetual scientific tension between control and context. experimental economists are well advised to think each time they plan an experiment about whether the benefits of heeding one or more of her suggestions are not large enough to cover the costs of higher inconvenience. This is the classic question of Saliency raised decades ago by Vernon Smith. Cherry et al. Kathleen Segerson.

M. Shogren and Hurley (1999) and Sturm and Weimann (2006). “Unbiased value estimates for environmental goods: a cheap talk design for the contingent valuation method. and Charles R. “EPA’s new emission trading mechanism: a laboratory approach. and the anonymous referees for their hard work and contributions to this volume. Note 1 For exhaustive surveys of the use of experimental methods in environmental economics see. The future of experimental work will be to help design institutions that address the combination of market failure and behavioral anomalies. Context affects participants’ motivation. Ronald G.G. 1543–1560. eds. In contrast. Johnston (2005). “An experimental test of the crowding-out hypothesis. Cummings.Introduction 5 both. K. others argue context is desirable to avoid a setting that is too sterile and too removed from reality for something so real as environmental policy. 649–665.” Journal of Public Economics 89.” Handbook of Environmental Economics.. . Peter J. Lastly. All experiments face this challenge. Games and Strategic Behavior. and R. References Bohm. Taylor (1999). Plott (1996). we wish to thank our friends and families for their support. a clearer and more definitive picture will emerge about how our institutional choices affect the efficiency of environmental policy and how people value these policies. Timothy N. Todd L. In 2005. otherwise we could find environmental economics falling into a new second-best problem: correcting market failure without addressing behavioral biases could reduce overall welfare. Eckel. P.” Journal of Environmental Economics and Management 30. 605–615. We would like to thank the authors and discussants. Therein lies the beauty of the experimental method as applied to human beings rather than terrestrial plants or subatomic partials – one can use one’s imagination to experiment with alternative degrees of control versus context. for example. “Experimental Evaluations of Policy Instruments. 133–160. pp. Experimental methods will likely provide one of the most useful tools researchers can use to understand the behavioral underpinnings of environmental policy. 437–460. (2001). Holt. Catherine. (2003). Boston: Pearson Education. Amsterdam: North Holland. Mäler and J. “Mental accounting and other-regarding behavior: evidence from the lab. Grossman. Markets. Vincent. As experimental evidence continues to accumulate. Cason. and Laura O. We thank Rob Langham and Routledge for their support and patience. Charles A. from which this project originated. (2006). We wish to extend a special thanks to Appalachian State University.” Journal of Economic Psychology 22. volume 1. the Department of Economics at Appalachian State University hosted an Experimental Economics and Public Policy Workshop. Cherry.” American Economic Review 89.

Bodo and Joachim Weimann (2006).C. Handbook of Environmental and Resource Economics. Cheltenham: Elgar.L.” Journal of Economic Surveys 20. K. Sturm. Amsterdam: North Holland.” in J. van den Bergh ed. or do you not understand him? Tax aversion in the lab. Cherry et al. “Experiments in Environmental Economics. “Experimental Methods and Valuation. Jason F. pp. (2006). Shogren. J.6 T. Norway.” working paper. Kallbekken. Oslo. “Experiments in Environmental Economics and Some Close Relatives.G. eds. Mäler and J. Steffen and Stephan Kroll (2007). Vincent. . 419–439. “Do you not like Pigou. Center for International Climate and Environmental Research. 969–1027. Shogren.” Handbook of Environmental Economics... volume 2. and Terry Hurley (1999).

Part I Tradable permit markets .

.

1 Baseline-and-credit emission permit trading Experimental evidence under variable output capacity Neil J. An example is the clean development mechanism proposed under the Kyoto Protocol. Firms then trade the allowances. credits are often computed on a project-by-project basis rather than on the basis of enterprise-wide emissions. In equilibrium. the most frequently discussed plans assume that the permits will be distributed to the regulated firms on some ad hoc basis. For example. Typical baseline-and-credit plans also differ from classic cap-and-trade in a number of institutional details. and R. The classic analysis assumes that the regulatory authority sets an aggregate cap on emissions from a set of sources and then divides the cap into a number of tradable permits (frequently called allowances). The effect is to limit aggregate emissions to an implicit cap equal to the sum of the individual baselines. Many field implementations of emissions trading take a different approach. Buckley. Such plans are generally known as cap-and-trade plans. Although the allowances could be sold at auction to raise revenue. Firms create emission reduction credits by emitting fewer than their baseline emissions. They must be certified and registered before they . thereby minimizing the cost of maintaining the mandated level of emissions. Andrew Muller Introduction Emission trading is now well established as a method for regulating emissions of uniformly mixed pollutants. They redeem allowances equal to the emissions discharged. This baseline may be derived from historical emissions or from a performance standard that specifies the permitted ratio of emissions to output. establishing a market price. Stuart Mestelman. each of which authorizes the discharge of a unit quantity of emissions. If emissions exceed the initial distribution of allowances the firm must purchase allowances to cover the excess. In these baseline-and-credit plans there are no explicit caps on aggregate emissions. each firm has the right to emit a certain baseline level of emissions. selling or banking the remainder. Instead. A good example is the US EPA’s sulfur dioxide auction. individual firms choose emissions such that the marginal cost of abating pollution equals the allowance price. These credits may be banked or sold to firms who exceed their baselines.

Second. Cason and Plott 1996). This introduces two potential inefficiencies. if the performance standard under baseline-and-credit is tightened so as to meet the aggregate emissions specified under cap-and-trade. the implicit cap on aggregate emissions varies with the level of aggregate output. Baseline-and-credit plans are theoretically equivalent to a cap-and-trade plan if the cap implicit in the baseline-and-credit plan is fixed and numerically equal to the fixed cap in a cap-and-trade plan. fully specified markets for emission rights and output. First. 2003) refers to such plans as tradable performance standards. It should be noted that this reasoning presumes that firms are adjusting to pollution regulation on two margins: the emission intensity of output and the level of output itself. Thus far. Fischer 2001. demonstrate the contrast between baseline-and-credit and cap-and-trade to policy makers. the baseline-and-credit plan will exhibit inefficiently high output. emissions and external costs. then industry costs will increase due to unnecessarily tight restrictions on emitting firms (Muller 1999.J. Fischer (2001. can be traded and there are generally restrictions that credits cannot be registered until the emission reductions have actually occurred. We have implemented a computerized laboratory environment with explicit capacity and emission intensity decision. We have demonstrated that predicted results . at the international level there are more active baseline-and-credit greenhouse gas-trading plans than cap-and-trade greenhouse gas-trading plans (Hasselknippe 2003). Laboratory implementation of baseline-andcredit trading would serve several goals: it would verify that market processes are sufficient to drive agents to competitive equilibrium. and a complete accounting framework. and possibly create a vehicle for training policy makers and practitioners in the nature of alternative emission trading plans. Dewees 2001. However. if the performance standard remains the same in both plans. Currently.1 In this case. 2006) we have developed a tractable model with constant returns to scale in production and multiple firm types. the variable baseline acts as a subsidy on output. however. In many cases. experiments have been fruitful in shaping cap-and-trade public policy (Cason 1995. 2003). In previous work (Buckley et al. Buckley et al.2 Specifically. The variable baseline in a baseline-and-credit plan introduces a critical difference in long-run performance compared to cap-and-trade with the same implied performance standard. the predictions on the relative performance of baseline-and-credit versus cap-and-trade have not been tested in the laboratory. Moreover the reasoning is essentially long run in that output is changed by firms investing or divesting themselves of productive capacity and equilibrium is computed by imposing a zero-profit restriction on firms in the market. We have undertaken a long-term research project to compare the properties of baseline-and-credit and cap-and-trade plans in the laboratory.10 N. the baseline is computed by multiplying a measure of firm scale (energy input or product output) by a performance standard specifying a required ratio of emissions to input or output. Firms receiving this subsidy will tend to expand their capacity to produce output. but as of yet no baseline-and-credit laboratory studies have been published.

We have three objectives. Sessions lasted approximately 3 hours. They sought to evaluate the prediction that the outcome of the two approaches would be the same when the output subsidy inherent in the baseline-and-credit plan could not possibly lead to productive expansion. Second. each involving eight subjects. Previous experiments involving human subjects have focused on the intensity decision. do the theoretical competitive equilibria characterize the behavioral outcomes: does the baseline-and-credit policy lead to higher emissions and output than occur under cap-and-trade? Methods We ran six laboratory sessions (three cap-and-trade and three baseline-andcredit). Those studies confirm that the overall predictions on emissions hold. For the first hour and a half. Subjects earned between $18. we hold emission intensity constant at the optimal level for each type of firm and allow firms to increase or reduce their productive capacity each decision period.75 and $53. students received instruction and participated in four training periods using an alternate set of parameters. are there treatment effects that differentiate the two trading policies? Finally.25 with a mean of $38.91. However. (2006) report on six sessions comparing baseline-and-credit with cap-and-trade when firm capacities are fixed and firm adjustment is limited to emission intensity. However there were some deviations from the benchmark values computed under the assumption of perfectly competitive equilibrium. All subjects were undergraduate students at McMaster University who had completed at least an introductory course in economics. Efficiency in the market was improved. including the training fee. examining the emissions intensity margin and the output market margin one at a time. so we have implemented work with human subjects slowly.Baseline-and-credit emission trading 11 hold in simulated markets with robot traders adjusting on both the output and emissions intensity margins. in September and October of 2004. Subjects then took a short break and returned to participate in ten paid rounds using the parameters reported here. Buckley (2005) and Buckley et al. In the present chapter we investigate the complementary problem of adjustment on the capacity margin. . Any deviation from parallel results would be then laid to the institutional differences between the two plans rather than the implied subsidy on output and emissions. That is. The software implementation of the environment detailed below was programmed at McMaster University using Borland’s Delphi programming environment and the MySQL open source database. Emission permit prices were higher under baseline-and-credit trading and inventories of permits were unpredictably high in both treatments. market instability occurs when capacity is freely adjustable. After ten rounds they were informed of their results and paid privately in cash.3 These training periods were rewarded by a flat fee of $10. First we wish to see whether market equilibria emerge in these markets. although only about one-half the available gains from trade were realized.

Dollars per unit output MCi ACi LACi ki Output.J. There were four types of firms distinguished by emission intensity: two. four. There were two subjects of each type.1 Firm cost curves. In the baseline-and-credit treatment we imposed a tradable performance standard of five. We chose not to present the experiment in neutral terms. Each firm created external costs proportional to its emissions. which was chosen to coincide with the social optimum. In the cap-and-trade treatment 160 permits were distributed each period and aggregate production capacity began at 32 units of output. These parameters were chosen to equate the marginal social cost (MSC) of each firm so that all could be present in final equilibrium. We expect the system to remain stable at the equilibrium point with firms trading 32 permits every period. Buckley et al. In this treatment we expect the output and emissions to increase due to the inherent subsidy to output. Each firm was initially given four units of productive capacity. although the instructions did not explicitly inform subjects of this. six and eight emission units per unit of output for firm types A. Output could be produced at zero marginal cost up to the fixed capacity. The marginal damage of emissions (not provided to the subjects) was assumed constant at $16 per unit of emissions.1 illustrates the short. B.4 Figure 1. C and D respectively. The unit cost of capacity varied from $32 per unit for the dirtiest firms (type D) to $128 per unit for the cleanest firms (type A).and long-run cost curves for a typical firm. There were two treatments: cap-and-trade and baseline-and-credit. . because we believed that the explicit emissions trading environment would help subjects understand the nature of the decisions they were making.12 N. k. This implies an average emission intensity of five at the social optimum. Subjects were told that they represented firms that create emissions while producing output and selling it on a simulated market. qi Figure 1. In both treatments subjects were started off at the cap-and-trade equilibrium. equivalent to the average emission intensity in the cap-and-trade treatment.

.Start of period Subjects begin this period with last period’s capacity Is the subject facing a cap-and-trade scheme? Yes Endow subjects with allowances No Permit market: call auction trading allowances or credits Output market: call auction with simulated buyers Is the subject facing a baseline-and-credit scheme? Yes Credits created if emission rate was below the performance standard No Allowances and credits are redeemed Subject chooses to raise or lower capacity by one unit or leave it unchanged End of period Figure 1.2 Sequence of events in a typical period.

the allowance market cleared. they could only trade credits that were produced in previous periods. The first ask was at the lowest price and the subsequent were at higher prices. subjects decided whether to increase or decrease their output capacity by one unit. In the cap-and-trade treatment subjects began with capacity and allowance holdings determined in the previous period. We first introduce some notation and describe the general model that allows adjustment on both the emission intensity and output margins. determining a price of permits and a quantity bought or sold for each subject. They were provided with extensive on-screen help to aid them in this decision. The quantity of credits created in the current period was determined by the firm’s emission intensity and quantity of output sold. This action required subjects to estimate the price they were willing to pay for additional permits and the price at which they were willing to sell their permits. Consequently. Each subject was required to produce and offer for sale as much output as he could. Subjects were also permitted to place up to three different price asks to sell allowances.J. The next highest price asked was for additional allowances.7 Demand for output was represented by an exogenous demand function with known intercept and slope. The sequence of decisions differed slightly between the two treatments. Permits not requiring redemption were banked for use or sale in future periods. and so permits could only be credited after output for the current period was determined. Buckley et al. Output decisions were automatically made by the computer program. price-taking firms.5 Subjects were permitted to enter three different price bids to purchase additional allowances. The second price bid (if any) had to be lower than the first. The output market then cleared. determining a common output price and an individual quantity sold and revenue earned for each subject. . Parameterization and benchmarks In this section we derive benchmark equilibria for the two treatments under the assumption of perfectly competitive. A third.14 N.2.6 Once all bids and asks were submitted. The number of allowances for sale was only restricted by the individual’s inventory of allowances. and the individual could bid to purchase as many allowances at this price as he wished. The individual could bid to purchase as many additional allowances at this lower price as he wished. Their first action was to trade allowances in a multiple-unit uniform-price sealed bid-ask auction (call market). They received an endowment of allowances. or keep capacity unchanged. and finally present the benchmarks. still lower bid. given his capacity and permit holdings. The first bid was the highest price the individual would pay for an allowance. A flowchart is provided as Figure 1. The baseline-and-credit sequence was identical to cap-and-trade except that subjects did not receive any emission permits before the credit market opened. After reviewing their financial report for the period. Second we report on the parameterization of the model for this experiment. could be entered for allowances beyond those bid for at the first two higher prices.

ei ) = qi C i (1. Each firm i [1. Willingness to pay for the output is a weakly concave function of aggregate output. An omnipotent social planner would choose an output and emission rate for each firm such that it would maximize total surplus.. WTP = ∫ P ( z ) z 0 Q where P = P(Q) is an inverse demand curve with positive ordinate (P(0) > 0) and negative slope (P′(Q) < 0). and i unit variable cost wi. Q {ri . which is a constant function of output. Consequently. Note that the marginal cost of output is ci(ri) + wi and the marginal cost of abating pollution is MAC = − ∂C i ∂ei = − ci′ (ri ) . S. N Aggregate emissions are E = ∑i =1 ei = ∑i =1 ri qi . D′(E) > 0 and D′′(E) ≥ 0.Baseline-and-credit emission trading 15 Theory Consider an industry with N firms. Industry output is Q = ∑i =1 qi . . N] produces qi units of output at an emission rate of ri = ei /qi where ei is quantity of emissions. The social planner’s welfare maximization problem is max S = ∫ P( z )dz − ∑ ci (ri )qi − ∑ wi qi − D(∑ ri qi ) . qi } 0 i =1 i =1 i =1 N N N .ri) can be separated into unit capacity cost ci(ri). The private cost of production is a linear homogeneous function of output and emissions: C i = C i (qi . Unit cost Ci(1.. which is a positive and declining function of the emission rate with ci(ri) > 0 and c´(ri) ≤ 0. ri ) . total cost is Ci = ci(ri)qi + wi qi .. − N N Environmental damages are assumed to be a positive and weakly convex function of total emissions: D = D(E).

For any given marginal damage.2 determines only the aggregate level of output. one for each margin of adjustment. Any combination of qi* such that the qi*s sum to Q* and the ri*qi*s sum to E* satisfies the efficient output condition. that is.2) ensures that output is surplus maximizing by requiring that each firm’s marginal social cost (the right hand side of equation 1. {qi } The first order condition for an interior maximum is P(Q c ) = ci (ri* ) + wi + ri* Pc . They are − ci′ = D ′(∑ ri qi* ) ∀i ∈ N . because two firm types can have identical marginal social cost if the reduced social damage generated by the clean firm type is exactly offset by an increase in private cost. This set may contain more than one firm type. If Pc denotes the price of permits under cap-and-trade.8 In the present experiment we suppress adjustment on the emission intensity margin by setting each firm’s emission intensity to its optimal value ri*. The efficient abatement condition (equation 1.16 N. firm i’s profit maximization problem is max π ic = P(Q)qi − ci (ri* )qi − wi qi − Pc (ri* qi − Ai ) . N there will be a set of firm types with least marginal social cost.2. Buckley et al. The social optimum can be supported as a competitive equilibrium under capand-trade regulation. firms cannot independently adjust the marginal social cost of their output.2) where an asterisk denotes optimal values and it is assumed that qi* > 0 for all i. (1. is fixed. The efficient output condition (equation 1.2). The regulator distributes allowances Ai to each firm so that the sum of allowances granted equals the optimal level of emissions. i =1 N N (1. ∑ N i =1 Ai = E *. ri. Condition 1. i =1 * * * * * and P (Q ) = ci (ri ) + wi + ri D ′(∑ ri qi ) ∀i ∈ N . D ′(∑i =1 ri* qi* ) .3) .2) equal the marginal willingness to pay for output (the left hand side of equation 1.1 vanishes and we are left with condition 1. Because the emissions intensity for each type of firm.1) requires that firms choose emission intensities such that the marginal abatement cost – c´ equals the marginal i damage caused by emissions.J. There are two first order conditions.1) (1. Note that condition 1.

Because equation 1.Baseline-and-credit emission trading 17 Equation 1. (1. Under a baseline-and-credit plan. It is useful to illustrate the equilibria diagrammatically. which acts as an output subsidy to the firm.3 illustrates the cap-and-trade equilibrium when only type A and D firms are in the market. Since the equilibrium of both trading plans involve the same average emission ratio.4 is the zero marginal profit condition which determines Q b. we immediately see that the latter differs from the former only in the last term. Firm i’s demand for credits is (ri* – r s)qi . then firm i’s profit maximization problem is max π ib = P (Q)qi − ci (ri* )qi − wi qi − Pb qi (ri* − r s ) . this higher level of output necessarily implies that aggregate emissions will be higher than optimal under baseline-and-credit regulation. The dirty firms have long-run average costs (LAC) of 32 and create damages of rDMD = 8(16) = 128 per unit of output. a solution to the surplus maximization problem is a competitive equilibrium and vice versa. Parameterization Table 1. – r sPb . Marginal social .9 r s = ∑i =1 ri* qi* Q * . Negative values denote a supply of credits. {qi } The first order condition for an interior maximum is P(Q b ) = ci (ri* ) + wi + ri* Pb − r s Pb .3 requires that each firm earn zero marginal profit.2 by replacing D ′(∑i =1 ri* qi* ) N by Pc and Q* by Q c. will be higher than under cap-and-trade. N Comparing the baseline-and-credit condition (equation 1.3). Consequently. marginal private cost to the firm is less than the marginal social cost and the corresponding output. r s. Let us assume that the regulator sets the emission rate standard equal to the average emission rate under the social planner scenario.4) to the cap-and-trade condition (equation 1.2 summarizes the associated equilibrium predictions under the alternative emission trading mechanisms. Figure 1.1 presents firm-specific parameters used in the sessions reported in this chapter. Table 1. If the price of credits is Pb .4) Equation 1.3 can be obtained from equation 1. the regulator sets an industry-wide performance standard. and identifies Qc . Q b.

yielding the same marginal social cost. Firm type A has a higher unit capacity cost at $128 but lower damages of rAMD = 2(16) = 32 per unit of output. and an 320 Demand curve S(E 160) Dollars per unit output 160 rAPC 128 rDPC 32 Qc 32 64 MSC LAC 2 MD LACA. the regulatory authority allocates 160 allowances and the allowance market clears at $16 per permit. Equilibrium at a price of $160 implies output of 32 units.18 N.3 Cap-and-trade equilibrium. B.J. D A. Long-run average cost is now $160 for each firm type. D Baseline-and-credit 16 Cap-and-trade 16 cost is 160. rD 8 Output. or 32/2 = 16 per unit of emissions.1 Cost parameters Firm type A B C D Unit fixed cost. At the optimal output. B. Table 1. Both types of firms are willing to pay $16 per permit. Under capand-trade.2 Variable capacity predictions Trading Institution Price of Output allowances or price credits (permits) 80 160 Aggregate output 48 32 Aggregate emissions 240 160 Active firm types A. Type D firms earn 160 – 32 = 128 in rent per unit output. or 128/8 = 16 per unit of emissions. Buckley et al. type A firms earn 160 – 128 = 32 in rent per unit of output. C. rA LACD. ci(ri*) 128 96 64 32 Fixed emission rate 2 4 6 8 Endowment 20 20 20 20 Performance standard 5 5 5 5 B&C initial credits 12 4 0 0 Table 1. Optimal output Qc* = 32 is determined by the intersection of the demand curve and marginal social cost. . Q Figure 1. C.

In computing the environmental damages we assume constant marginal damages of $16 per unit of emissions. Efficiency We compute the efficiency of baseline-and-credit and cap-and-trade equilibria relative to the maximum surplus available. type D firms must buy rD – r s = 3 credits per unit of output and they are willing to pay (80 – 32)/3 = 16 per credit. Since there is equal capacity of type A and D firms (24 units for each type). They must receive at least (128 – 80)/3 = 16 per credit to earn non-negative profits under baseline-and-credit. Producers’ surplus is (160 – 80)(32) = 2560. Type A firms create r s – rA = 3 credits per unit of output. The performance standard is rs = 5 units of emissions per unit of output. This determines the inefficient equilibrium output of 48 units. the same amount. Q MSC LAC 2 rS 5 MD rS)PB rSPB LACA. rD 8 Output.Baseline-and-credit emission trading 19 320 Demand curve Dollars per unit output Damages exceed increase in surplus by this area 160 128 (rA 80 (rD 32 QC 32 QB 48 64 rS)PB LACD. This equilibrium implies the presence of 16 units of capacity from type A and 16 units of capacity of type D in the market. we see this implies that there must be equal capacity of each firm type. From Figure 1. Figure 1. the supply of credits equals demand for credits at a price of $16.5(320 – 160)(32) = 2560. Restricting attention to type A and D firms.3 it is clear that under cap-and-trade consumers’ surplus in equilibrium is $0.4 Baseline-and-credit equilibrium. rA LACA&D. The social surplus is equal to the sum of consumers’ surplus and producers’ surplus less any environmental damage. Given equal capacity shares. 24 from each firm type. average emission intensity of five. External damages are equal . r Figure 1. The only way to achieve an average emission intensity of five with type A (rA = 2) and D (rD = 8) firms is to have equal output capacity of each firm type. average LAC = (128 + 32)/2 = 80.4 shows the equivalent baseline-and-credit equilibrium. At this point. The effect of baseline-and-credit trading is to equate the LAC of both firm types.

Results Figures 1. the fact that there was no payoff to subjects’ inventories of permits held at the end of the session may have induced an end-game effect in period 10. the cap-and-trade surplus values are optimal. the corresponding consumers’ surplus. Many series show a distinct time trend. producers’ surplus.20 N.3 Equilibrium surplus efficiency Components of efficiency Efficiency = Cap-and-trade equilibrium Surplus Efficiency index Consumers’ Producers’ Environmental surplus surplus damages + + – $2560 100% $5760 225% $2560 100% $0 0% $2560 100% $3840 150% $2560 100% Baseline-and-credit equilibrium Surplus $1920 Efficiency index 75% to total emissions multiplied by the marginal damage. Using Figure 1. Accordingly we drop periods 1 through 5 and 10 in summarizing the results numerically and report mean values for periods 6 through 9 in Table 1. We have three independent series in each treatment. so that total social surplus is equal to the consumers’ surplus of 2560. external damages and total social surplus under baseline-and-credit are 5760. Note that this exactly offsets the producers’ surplus. 3840 and 1920. Table 1. Moreover.3 reports the equilibrium values for total surplus and its components under the two treatments.5 to 1.4. producers’ surplus and external costs. 0. Thus the consumers’ surplus component of the efficiency index is Consumer Surplus Component = Actual Consumer Surplus . Given these definitions we can compute an efficiency index Efficiency = Actual Total Surplus .4. Optimal Total Surplus Table 1.J. Optimal Total Surplus It is convenient to decompose efficiency into components associated with consumers’ surplus. Buckley et al. The figures show the range and mean of observations for each period. respectively.10 provide an overview of the data. Because the emissions cap was set to the socially optimal level of 160 units of emissions. 160(16) = 2560. We .

output. capacity rises steadily to reach the predicted level of 48 by period 7.50 0.75 36.75 31. Nevertheless. Despite the general resemblance to the benchmarks shown in Figures 1.50 69.00 160.17 48.50 10.50 32.75 36.50 215.10 Capacity.00 45.00 32. significantly above the benchmark of 32. both treatments display substantial deviations from their benchmarks.00b 47.25 142.25 47.50 50.00 217.75 7.75 6.5 to 1. the statistical tests reported in Table 1. c The cap-and-trade treatment mean is significantly different from the baseline-and-credit prediction using a t-test at the 5 percent level. test for treatment effects using parametric (F-test) and non-parametric methods.00 33. The treatment effect is strongly significant.00b 46.00c 159.25 32. so caution should be applied when interpreting the statistics.d 14.00c 17. d The baseline-and-credit treatment mean is significantly different from the baseline-and-credit prediction using a t-test at the 5 percent level.Baseline-and-credit emission trading 21 Table 1.4 provide evidence of a treatment effect (cap-and-trade versus baseline-and-credit trading) that is strongly significant. even adopting a critical level of 10 percent.00 122.00 218.33 48. b The baseline-and-credit treatment mean is significantly different from the cap-and-trade prediction using a t-test at the 5 percent level. Emissions (Figure 1.4 Mean values over periods 6 to 9 by treatment Capacitya Output volumea Aggregate emissionsa Permit market Price Cap-and-trade Session 1 Session 2 Session 3 Treatment mean Prediction Baseline-and-credit Session 4 Session 5 Session 6 Treatment mean Prediction 36.00 49.50 34.00 48. Under cap-andtrade.58 32.58 16.00d Permit inventoriesa Notes a Treatment effect is significant using a t-test and a Mann-Whitney U-test at a 5 percent critical level.25 51.00c.25 48.00 11.25 30. output and emissions.00 160.50 47.7) follow the same pattern as output.58 32. .5 shows the evolution of capacity.6 shows a similar pattern for output.50 119. Figure 1.7.50 10.58 0.83 240.00b 106. This suggests pervasive underutilization of capacity due to inability to acquire permits.00 23. emissions and efficiency Consider first the key predictions on capacity.50 34.00 45.00 Volume 30.33 48.50 163.75 45. capacity stabilizes quickly between 36 and 40.00b 212. these observations conform well to the underlying theory.00 16. except that under cap-and-trade output exceeds the benchmark level of 32 by only a small amount. Overall. However these tests have extremely limited power. Under baseline-and-credit trading.25 46.50 38.25 15. Figure 1.50 9.50 157.

5 Capacity.6 Output volume.Cap-and-trade 55 Baseline-and-credit 48 Capacity 45 35 32 25 1 5 10 1 Period Min/max capacity 5 Mean capacity 10 Figure 1. Cap-and-trade 60 Baseline-and-credit Output volume 48 40 32 20 1 5 10 1 Period 5 Mean output volume 10 Min/max output volume Figure 1. .

.5 1 5 Min/max efficiencys 10 1 Period 5 10 Mean aggregate efficiency Figure 1.7 Aggregate emissions. Cap-and-trade 1 Baseline-and-credit Efficiency (1 = 100%) 0.Cap-and-trade 240 220 Baseline-and-credit Aggregate emissions 180 160 140 100 1 5 10 1 Period 5 Mean aggregate emissions 10 Min/max aggregate emissions Figure 1.8 Efficiency.75 0.

however. Table 1. while the third achieved only 70 percent. c The cap-and-trade treatment mean is significantly different from the baseline-and-credit prediction using a t-test at the 5 percent level. Efficiency was highly variable across sessions. Due to the wide variation the difference in means was not significant. b The baseline-and-credit treatment mean is significantly different from the cap-and-trade prediction using a t-test at the 5 percent level.5 reports the numerical results.24 N. Treatment effects were significant for each of the three components of surplus. Two of the cap-and-trade sessions attained close to 100 percent efficiency.5 Mean efficiency over periods 6 to 9 (%) Components of efficiency Efficiency = Cap-and-trade Session 1 Session 2 Session 3 Treatment mean Prediction Baseline-and-credit Session 4 Session 5 Session 6 Treatment mean Prediction Consumers’ surplusa + 109 115 88 104 100b 215 243 199 219 225c Producers’ surplusa + 85 82 83 83 100b. Figure 1. d The baseline-and-credit treatment mean is significantly different from the baseline-and-credit prediction using a t-test at the 5 percent level. Buckley et al. suggesting that costs were not being minimized. Under cap-and-trade consumers’ surplus and damage were close to their benchmark values. Mean efficiency in the three baseline-and-credit sessions was almost exactly the predicted level of 75 percent.e –8 –39 19 –9 0c Environmental damagesa – 99 98 101 100 100b 132 136 135 134 150c. These results imply that the efficiency losses from baseline-and-credit trading will be similar to those predicted by theory.J. although still higher than in the cap-and-trade treatment. Table 1. while producers’ surplus was significantly lower.d 95 99 70 88 100b 75 67 83 75 75 Notes a Treatment effect is significant using a t-test and a Mann-Whitney U-test at a 5 percent critical level. Figure 1.8 reports the evolution of efficiency over the ten periods. Credit and allowance markets The relatively promising results discussed above were obtained despite some rather strange behavior in the markets for credits and allowances. e The cap-and-trade treatment mean is significantly different from the cap-and-trade prediction using a t-test at the 5 percent level. Under baseline-and-credit trading consumers’ and producers’ surpluses were close to the benchmarks while emission damage was less than expected.9 .

Baseline-and-credit emission trading 25 Cap-and-trade 150 Baseline-and-credit Permit price (Eq'bm = $16) 100 50 16 0 1 5 10 1 Period 5 Mean permit price 10 Min/max permit price Figure 1. With cap-andtrade. leading to the supposition that it is an institutional factor associated with baseline-and-credit plans that is driving the differences. A similar pattern was observed in Buckley (2005). The data do not allow us to differentiate among these explanations. for speculation or inadvertent. due to errors in permit trading and capacity choice. Both treatments exhibit significant inventory build-ups. inventories accumulate in both treatments. Risk averse subjects may hold permit inventories between periods but have no incentive to carry permit inventories past period 10 because they have no redemption value at the end of the session. but while the cap-and-trade inventories . permit prices are consistently very close to the benchmark and not significantly different from it. With baseline-and-credit. The high early prices and rapid decline of permit prices under baseline-and-credit are probably due to bidding errors in the early periods of the session. Permit banking might be intentional. there is no clear explanation for subjects to hold permit inventories. Risk neutral subjects should hold no permits in either treatment. However the two series are not significantly different across periods 6 through 9. permit prices start very high. shows dramatic differences in permit prices across treatments.10. then fall rapidly to below equilibrium levels. One such factor is the requirement that credits be generated before being offered for sale. In fact.9 Permit trading prices. In general. The convergence of baseline-and-credit permit prices toward zero in periods 9 and 10 could likely be caused by an increased supply of permits due to increased permit inventory carryover in the credit treatment. as illustrated in Figure 1. To determine this we must also examine aggregate levels of permit inventories by treatment.

26 N. Buckley et al. we have confirmed that price incentives conveyed through markets perform as theory predicts. Despite differences in early permit trading prices under the two plans.10 Aggregate permit inventory. Results from the laboratory sessions reported here support the theory. Using graphical and tabular data. This chapter reports results from real markets in controlled laboratory sessions in an environment involving fixed emission technologies and variable output capacities. however. whether the theoretical predictions regarding the two mechanisms would hold in real markets. The supply of permits is fixed under cap-and-trade regulation. Our conjecture regarding increased permit supplies causing lower prices under baseline-and-credit trading appears to be supported by the inventory data. Discussion and conclusions Theory predicts higher aggregate output and emission under baseline-and-credit than under cap-and-trade when the former imposes a performance standard consistent with the cap under the latter plan. baseline-and-credit inventories climb steadily. stabilize below 50 units. The question remained. . the results strongly support the theoretical predictions.J. The treatment differences are likely driven by the increased supply of credits generated by expanded output under the fixed performance standard of baseline-andcredit regulation. This is because a performance standard acts as a subsidy on output. Cap-and-trade emission and output levels stay close to their predicted equilibrium values and emissions and output soar and converge to their predicted higher levels under baseline-and-credit. Aggregate permit inventory (Eq'bm = 0 permits) Cap-and-trade 250 200 150 100 50 0 1 5 10 1 Period 5 10 Baseline-and-credit Min/max aggregate permit inventory Mean aggregate permit inventory Figure 1.

Acknowledgments We gratefully acknowledge the support of the Social Sciences and Humanities Research Council of Canada.pdf for the laboratory instructions. see Appendix B. the 2005 Southern Ontario Resource and Environmental Economics Workshop and the 2005 Experimental Economics and Public Policy Workshop at Appalachian State University. (1982). available at: socserv. Even though permit inventories are predicted to be zero in the baseline-and-credit equilibrium. 410-00-1314. online. available at: socserv. while traders have incentives to bid below values and ask above costs. Notes 1 This ratio is generally called the emission intensity. As discussed by Smith et al. For screenshots of the computerized environment. 2 A cap-and-trade plan with aggregate cap on emissions may be said to imply a performance standard of rs = E/Q where E and Q are respectively aggregate emissions and output in long-run equilibrium. evidence shows that permits are accumulated in inventory over the entire experiment. future work can now assess the long-run theoretical prediction of higher output and emissions under baseline-and-credit trading in a full model in which firms choose emission rates and output capacities. traders of infra-marginal units near the margin that determines price should fully reveal costs and values to avoid being excluded from the market by extra-marginal units. 4 Marginal social cost equals unit capacity cost plus the external costs created by each unit of output. the 2004 Meetings of the Southern Economics Association. online. 6 In order to aid them in their bid and ask decisions subjects were provided with a planner window in the computer software that displayed the latest output and permit market prices and volumes.mcmaster. In addition. This chapter reports experimental sessions involving fixed emission rate and variable capacity while Buckley (2005) provides results from sessions assuming fixed output capacity and variable emission rates. See Cason and Plott (1996) for a comparison of the uniform price auction with a discriminatory price auction in the context of emission permit trading.mcmaster. . Asha Sadanand and Bart Wilson for helpful comments. 5 A multi-unit uniform price sealed bid–ask auction was chosen because of the relatively quick trading time and high efficiency associated with it. however.ca/econ/mceel/papers/ varcapercappa. With the theoretical framework and corresponding experimental environment in place. An experimental environment has now been designed and tested. This work was presented at the 2004 Canadian Experimental and Behavioural Economics Workshop.Baseline-and-credit emission trading 27 One caveat.ca/econ/mceel/papers/varcapercappb. This behavior might be caused by the relatively more complex framing of the baseline-and-credit institution in addition to the variable permit supply inherent in baseline-and-credit regulation. Grant No. is that it appears that baseline-and-credit regulation is susceptible to higher levels of permit inventories than cap-and-trade. the planner provided a cost calculator that detailed firm costs under various hypothetical capacity and permit value conditions. For our parameters MSC equals 160 for all four firm types. We thank Daniel Rondeau.pdf. 3 See Appendix A.

. and Plott.J. Canada: McMaster University. Hasselknippe.2 would imply a firm specific output level but would result in a more complicated laboratory environment. If this assumption were relaxed.. Climate Policy. S89–S109. PhD Thesis. This environment assures full compliance with emissions standards and determines the number of permits that will be banked each period. Competitive market institutions: double auction versus sealed bid-offer auctions. 3 (Supplement 2). C. 2003. Emissions trading without a quantity constraint. Combining rate-based and cap-and-trade emission policies. In the present environment output is implicitly determined by the subjects’ capacity choices and the outcome of the permit market. Emissions trading: ERCs or allowances? Land Economics. S. Fischer. McMaster University.N. Mestelman. 85 (4). Cason.. R..G.. condition 1. M. American Economic Review. S42–S57. 10 With a critical level of 10 percent there is about a 45 percent chance of detecting a true difference in means of 1. Unit cost. Hamilton.N. We would need a critical level of 25 percent to get a 70 percent chance of detecting this large a difference in means (two-tailed tests. DC. 2003. Dewees. C. 133–160. We could set a stricter standard so that quantities of output and emissions are optimal but then firm costs will be inefficiently high.. 77 (4). 11 (2). T. 9 As mentioned in the Introduction. we will find that setting the performance standard equal to the optimal average emission rate will result in quantities of emissions and output that are inefficiently high. Climate Policy. ci(ri*). common variance).J. A.. 2001. Discussion Paper 01–22. Considering that both methods yield inefficiency we choose to focus on the case comparing cap-and-trade with a baseline-and-credit system with a performance standard equal to the average emission rate from the optimal scenario. 72 (1). Smith. Williams. Department of Economics Working Paper 99–13. and Vannoni. C. 2005. 1995.5 standard deviations. 1999.W. 30 (2).R. Cason. Hamilton. 1996. R.. Implications of Alternative Emission Trading Plans. 8 This feature of the model is a direct result of the constant marginal cost of output assumption. N. Muller.. We determined that this excessively complicated the environment.. V. Systems for carbon trading: an overview.L. 2001. 513–526. 905–922. EPA’s new emission trading mechanism: a laboratory evaluation. Resources for the Future. 2006. Buckley et al. D.. 58–77. is a function of emission rate but not output. Rebating environmental policy revenues: output-based allocation and tradable performance standards. .A. H. 149–166.J. 3S2. American Economic Review. 7 At the pilot stage output decisions were explicitly made by the subjects. Ontario. Journal of Environmental Economics and Management.28 N. Fischer. 1982. and Muller. W. Buckley. An experimental investigation of the seller incentives in the EPA’s emission trading auction. Bratton. N. References Buckley.. Pacific Economic Review. Implications of alternative emission trading plans: experimental evidence..A. T.K. Washington.

in exchange for a payment that exceeds the profits sellers would make from using them. Freeman. Pew Oceans Commission. In experiences in seven major fishing nations. and totaling 10 percent of ocean harvests worldwide. where stakeholders are traditionally so heavily involved in management that major changes are dependent upon stakeholder support. Anderson. 2002). 2003). OECD. The fear of consolidation can often lead stakeholders to oppose implementation of market-based management. Sutinen and Soboil. Sutinen Introduction Tradable allowance systems are being increasingly applied to address numerous environmental and natural resource management problems. allocates allowances to that level among users and permits users to trade their allocations. In fisheries. In theory. including water use. where the affected parties are usually small and family businesses. Squires et al. pollution and overfishing (Tietenberg.2 A laboratory analysis of industry consolidation and diffusion under tradable fishing allowance management Christopher M. 1997. In a typical application of tradable allowances. and Jon G. The effect of fears about consolidation is particularly acute in fisheries. especially in applications such as irrigation rights and fishery management. a management authority sets an allowable level of activity. These experiences have led three independent national panels in the United States to endorse increased use of tradable allowance management for federally regulated fisheries upon expiration of a 6-year federal moratorium on new programs (NRC. Matthew A. individual transferable quota (ITQ) management has proven effective at constraining exploitation within set limits. even in cases where market fundamentals suggest consolidation is unlikely. the ensuing market allows inefficient resource users to transfer their use or pollution rights to efficient users who can earn more profit from them. 1999. 2002. the resulting industry transformation can threaten local economies and traditional ways of life. 2004). US Commission on Oceans Policy. While economically defensible. such management is dependent on both government and stakeholder support to be . it may be difficult to garner the benefits associated with tradable allowance management. 1995. reducing overcapacity and gear conflicts while improving product quality and availability (Arnason. 2003. mitigating the race to fish. Without addressing consolidation concerns..

acceptable and effective. they affirm the theoretical .30 C. and others diffusion. and in the winners and losers that it determines (Matulich et al. will emerge through trading. in the other. 1996). The sort of consolidation feared by fishermen is illustrated by the Dutch demersal North Sea fishery. economists can use equilibrium models to help fishers understand the likely extent of consolidation (or diffusion!) in their fishery. Rather. or market power exercised by large or well-capitalized allowance owners. we consider whether consolidation is a general property of market-based management. This will likely result in a smaller number of operators. However. where the number of ITQ-holders decreased about 29 percent from 1988 to 1997. or if instead some applications will experience consolidation. have a clear understanding of how tradable allowances will affect their way of life. under tradable allowances. who have an active role in management. First. small operators will acquire marginal units from the larger operators. In a study that followed fishermen after they sold. We assess the extent to which market forces and the equilibration process may lead to consolidation using a controlled economic experiment in which human subjects play the role of fishermen managed under a tradable allowance system. In the absence of excess consolidation. However. they will be unable to make informed decisions concerning their stance on tradable allowance implementation. different cost structures among the large and small operators lead to different theoretical predictions about market share under tradable allowances: in one. but until stakeholders. In this chapter. market volatility inherent in the equilibration process plays a significant role in the transition process.. Economic theory suggests that the equilibrium outcomes. Our results show that concentration is not a necessary consequence of tradable allowance management. and the same allocation of allowances under the new tradable allowance system. including the market shares they imply. market fundamentals predict small operators will sell out to larger operators and. However. The economic analysis of consolidation requires distinguishing among two phenomena. so an understanding of whether and how it occurs can indicate when preventive measures are required. Anderson et al. and the number of vessels employed by the industry decreased about 33 percent from 1987 to 1998 (Davidse. efficient scaling of the market will occur. based on observable market fundamentals. Second.M. 2001). helping them identify the management regime best suited to their goals. Ford (2001) found that many of the 17 percent of vessels who sold out from the Tasmanian rock lobster industry expected to move into other fisheries. Both economists and fishers wish to avoid excess consolidation. leading to a less concentrated industry than before the implementation of tradable allowances. and they were unable to sell their vessels because of a general contraction in all fisheries in the region caused by rationalization. We examine market outcomes in two experimental treatments that reflect a fishery with the same initial distribution of fishing effort. This could result from market volatility. rationalization in those fisheries made switching too costly. as inefficient operators sell out at the equilibrium price. consolidation may take place in excess of that predicted by equilibrium.

as a motivating case.104 pounds per year). First. smaller buyers may buy smaller blocks of traps than larger fishers. medium–small (44 licenses averaging 3. Treatments and hypotheses In 2004. such that fisherman who diversified into other fisheries or who took a year off would not be penalized. we estimated a production function of the form Landings = Days Fished × [Constant × (Traps Fished)bj] where bj is the return to average traps used by a fisher in category j.Industry consolidation and diffusion 31 predictions that the level of consolidation observed is explained by underlying market fundamentals. The Lobster Conservation Management Team (LCMT). the Atlantic States Marine Fishery Council (ASMFC). and small (174 licenses averaging 464 pounds). the LCMT’s draft plan is being reviewed by managers. in the form of trap tags that allow a fisherman to put one trap in the water for a season. With appropriate economic analysis.600 pounds). concerns about initial allocations and latent effort from boats licensed to fish for lobster but who have been recently inactive have delayed implementation past the 2005 season. To relate the experimental environment to the lobster fishery. which advises the ASMFC. as it would be applied to the Rhode Island lobster fishery. In this plan. drafted the plan. we conducted a cluster analysis on inputs and landings to decompose the fishery into five “types” of operations: large (18 licenses landing an average of 27.950 pounds). In the two experimental treatments. which is responsible for managing American lobster in the northeast. approved a management plan that includes a provision for transferability of lobster trap allowances. medium–large (33 licenses averaging 14. We evaluate the role of tradable allowance management in consolidation using this plan. . which were used to estimate the production functions of Rhode Island lobster operations. we constructed profit functions for heterogeneous experimental subjects based on 2001 logbook data. we multiplied this by the average annual ex vessel price ($4. provided by the Rhode Island Department of Environmental Management.15). stakeholders can consider marketbased management with a clear idea of the likely structure of their industry following implementation. a cap on the total number of traps has been determined from stock assessment data and the need to rebuild the lobster stock. No limitations are to be set on allowance transfer except that transfers must involve a minimum number of trap tags. However. Profit functions were estimated in three stages. and subjects could vary the number of traps fished by buying and selling trap allowances in the market. This implies that the extent of consolidation is predictable based on the economic characteristics of the small and large operators within the fishery. will be allocated based on the highest annual value of fish caught during the base period. Second.169 pounds). Currently. To calculate revenue. Allowance. medium (18 licenses averaging 8. revenue for subjects is based on average days fished for each type.

which must be the largest component of a cost function that rationalizes production level choices. This restriction is binding on only the largest operators. the experimental test would indicate whether the model’s predictions based on the actual data are meaningful. and encourage them to consider instead whether the production elasticities of their fishery are likely to lead to undesirable levels of consolidation. the experiments provide a key test of the competitive model’s ability to predict the consolidation process in an equilibrating market based on market fundamentals. In one treatment. The profit functions from the two treatments are designed to have ex ante identical levels of fishing effort. we designed an experiment with two treatments. Therefore. if accurate cost data were to be collected in the future. and even if it were. if equilibrium predictions are upheld. prior to the reduction in traps accompanying introduction of tradable allowances. we determined costs for each type of operation. If the model’s predictions are meaningful. or whether the price discovery process facilitates consolidation to a greater degree than fundamentals suggest. Rather than mislead fishers about the likely industry concentrations resulting from a transition to tradable trap certificates. Unfortunately. any lobster license holder may fish up to 800 traps at a time. stakeholders could be provided with the foreknowledge of whether tradable allowances would result in excess consolidation or efficient scaling in their particular fishery. In current regulation. Thus. Finally.32 C.M. and different allocations of allowance among small and large operators. reliable cost survey data is not available for this fishery. Anderson et al. While this design cannot predict the level of consolidation that would be observed if trading were to be implemented in the lobster fishery. the direct cost data typically collected does not reflect opportunity costs. they will address the popular perception that consolidation is always a consequence of tradable rights management. most operators maximize their profit at trap levels below this limit. In the consolidation treatment. although the allowance endowments and pre-cap profit maximizing trap levels are equivalent. Thus. 2005) predicts that market share will move toward large operators (consolidation). the experiments can contribute to policy analysis in two ways. the model predicts that market share will shift toward smaller operators (diffusion). the two markets will yield different prices. the initial endowments of allowance are also identical within each type of operator between the two treatments. When the . the profit functions are chosen so that optimal number of traps for each type of operation is the same across treatments. Second. Thus. and in a second treatment. This may address stakeholders’ concerns about market-based management. Because the trap certificates are most likely to be allocated proportional to historical landings. costs are chosen so the competitive general equilibrium model (see Anderson and Sutinen. we elect to address the concerns that market-based management inevitably leads to consolidation with experiments that demonstrate the range of outcomes for different plausible cost functions. What differ between treatments are the elasticities of the derived demand curves of each type of operator. First. the large operators have a more inelastic profit function than the small operators.

subjects may exchange allowance for the right to earn profit from fishing in a single . subjects trade in a market for allowance. cost functions consistent with the diffusion treatment may also occur in the field. Each round is then divided into periods. or simply a failure of the market to identify a price within the time it is observed. and subjects are able to trade allowance with each other. Once the trading market closes. Since consolidation has been observed in some fisheries implementing tradable allowance systems. At any available market price for allowance. In the diffusion treatment. In this environment. and subjects earn profit from fishing their allowances. the market opens. The cost functions used in the two treatments are primarily demonstrative. During the trading phase. a controlled laboratory experiment is used in which subjects play the role of fishers in a tradable allowance market. As in a naturally occurring fishery with an allowance market. subjects are given an initial allocation of allowance and cash. Experimental procedure In order to test whether the general equilibrium predictions emerge under each set of trading rules. these two treatments will demonstrate that market fundamentals dictate the level of consolidation that can be expected following implementation of market-based management. 2005.. Each experimental session is comprised of six rounds. If so. the fishing phase commences. 2005).Industry consolidation and diffusion 33 total number of traps is cut back below historical levels. The first four rounds are each one period. additional traps are worth more to larger operators than to smaller operators. 2000.. However. each subject needs to decide what action earns her the most profit: buying some allowance and fishing more. or fishing years. This is tested against the hypothesis that equilibrium allocations do not obtain. selling some allowance and fishing less or fishing her current holdings of allowance. Smith et al. and individual profit from fishing is determined by the quantity of allowance a subject holds. Larkin and Milon. Therein.. Fishing profits are determined from a table based on the amount of allowance the subject holds after trading. At the beginning of each round. especially when small operators particularly value fishing. In these rounds. the hypothesis to be tested is that the allocations of allowance predicted by a general equilibrium model obtain. This could be caused by price volatility (Newell et al.g. small operators have a more inelastic profit function. and so they buy from the large operators with lower marginal values. Each period is divided into two parts: a trading phase and a fishing phase. 1988). speculation in the asset market (e. subjects who better balance fishing and trading allowance to maximize their total profit from both activities earn more profit in the experiment and are paid more money for participating. it is reasonable to assume that the cost functions similar to that of the consolidation treatment occur in the field. Anderson and Sutinen. Insufficient data exist to determine where the Rhode Island lobster fishery falls within that range.

on which profit is based.2) Initial allocation (endowment)* 77 (32.8) 85 (29.6) 70 (29. The first trading period of each round is 5 minutes long and. A centralized price board displays current buy and sell offers where subjects may buy (sell) an allowance unit by accepting the lowest sell (highest buy) price advertised. In this experiment. Subjects’ profit functions were derived from logbook data as described in the previous section. corresponding to an initial lease period.1) 34 (28. as can be seen in Figure 2. the predicted equilibrium price of allowance decreases from period to period by the amount of profit the inframarginal demander earns from holding the allowance in one period. The endowment range for the six small operators reflects wide heterogeneity of fishing histories within that category. Should current prices not be appealing to a subject. while allowance purchased in the last period provides profit only in the final period. the second period is 4 minutes long and the third and fourth are 3 minutes long. Therefore.1) 45 (30. allowance purchased in the first of four periods provides profit in each of the four periods. for each of the four operator types.1) 4 (5. In these rounds.4) 41 (34. subjects may make and respond to both buy and sell offers in a double auction market. trade immediately takes place. when applicable. two playing with profit functions of large operators. four medium–small operators and six small operators.6) Consolidation equilibrium* 85 (35. This trading structure was chosen as it is like field allowance markets in that trading is continuous and asynchronous and that trading can occur at different prices. of subjects 2 2 4 6 Profit function optimum* 90 (30. which allows for trades to occur at different prices throughout a period. Table 2.2) 31 (26.5) 6 (7. The equilibria for market share holdings differ for the two treatments as a result of profit functions varying for the two treatments.4) 74 (31. Upon acceptance of an offered price. two with the profit functions of medium–large operators.1 Rounds five and six are each four periods. As such. Each experimental session was designed for up to 14 participants. Profit. as a function of the number of experimental Table 2.M. trades of allowance carry over from one period to the next – an asset structure – and intertemporal speculation is possible.1.7) 79 (33. .0) Diffusion equilibrium* 68 (28.1 displays the market share holdings.0) Large Medium–large Medium–small Small Note * Total market share (percentage) of subjects are in parentheses.8) 9 (9. she can advertise her own price at which to buy or sell and the quantity of allowance at that price.6) 2–9 (8. The allowance itself is structured as an asset that provides the opportunity to earn profit in each period until the end of the round. year. Anderson et al.1 Market share holdings for operators Operator No.34 C.

Although few managed fisheries exist in a steady state. it is unlikely that equilibrium would obtain in a fishery with a dynamic or stochastic stock. For an experimental session. if equilibrium for market shares in the tradable allowance model fails to obtain in this simple environment. If there were extra subjects (we designed the diffusion experiment to accommodate 13 or 14 subjects and the consolidation experiment to accommodate 12 or 14 subjects). then randomly selected subjects were paid their participation fee and dismissed. subjects were shown into the laboratory and seated at individual computer terminals. although the peaks were scaled differently in the two treatments. explained how to . subjects were recruited – by email and through in-class sign-ups – to appear at an appointed time at the Policy Simulation Laboratory. and because the functions do not change between rounds. the upper bound of the equilibrium price in the diffusion treatment is 17. located at the University of Rhode Island. with barriers to discourage talking and impair visibility of others’ terminals. Each operator type’s profit function peaks at the same number of permits.1 Profit functions for operators. The upper bound of the equilibrium price for which trades would occur in the consolidation treatment is 20. profit functions for the consolidation treatment are noted with a solid line.Industry consolidation and diffusion 35 600 Profit ($) M–S operator Large operator M–L operator Small operator 0 0 10 20 30 40 50 60 Permits 70 80 90 100 110 120 Figure 2. Unchanging profit functions also simplify the subjects’ task.8. The experimenter then read aloud the instructions (available from the authors) as subjects followed along on their computer screens. we are able to monitor the effects of experience and learning. for the diffusion treatment are noted with a dashed line.2 Subjects’ profit functions do not change during the experiment.3 After reading consent forms. They were told they would receive a $5 participation fee and would have the “opportunity to earn considerably more” during the experiment. “permits” (corresponding to roughly eight trap tags). and reflect the assumption that the fishery is in steady state and that the total allowance allocation is such that recruitment exactly offsets harvest and mortality.

earnings averaged $23. In Figure 2.75). obtaining equilibrium prices and highly efficient allocations. there is some dispersion around the equilibrium price. as higher elasticity in the consolidation treatment made deviations more costly for buyers. Average prices are broadly consistent with equilibrium predictions. The market outcomes broadly support the levels of consolidation predicted by the general equilibrium model in both treatments. For the consolidation treatment. but do not display volatility.36 C. and they were paid privately as they left the laboratory. indicating that market fundamentals are driving trades. The data indicate that the market performed well. the observed market shares are examined. respectively.75). After answering any questions. subjects’ earnings were converted to US dollars. or away from. In the third section. as all four sessions have observed prices very close to equilibrium in all periods. Following the experiment.38 with a standard deviation of $3. Sessions lasted approximately 1 hour and 45 minutes each. The first and second sections examine. Anderson et al. respectively. and the convergence towards.57 (range of $12. subjects were provided an allowance and profit table with which to trade with the computer. Next. However. in all rounds there is movement toward equilibrium predictions. use the experimental software and led subjects through a practice round.2. some sessions are slightly higher than equilibrium.75 to $31.3 show the average prices in each period of the diffusion and consolidation treatments. prices in rounds one through four.75 with a standard deviation of $3. For the diffusion treatment.M.75 to $29. In Figure 2. Once the asset market is introduced. the first four rounds of the experiment were run.3. Results The findings of this experiment are separated into three sections. the experimenter read instructions explaining the asset structure for the four-period rounds as subjects followed along on their computer screens. prices and the resulting efficiency observed under the two market environments and show that equilibrium behavior obtains on average.61 (range of $8. there is little such variance. The difference in price variation between treatments may be a result of the difference in elasticities of the derived demand curves of each type of operator.2 and 2. and subjects participated in another practice round before beginning rounds five and six of the experiment. respectively. . In the practice round. By periods three and four of each round. average prices are very close to equilibrium. The two practice rounds were identical to the subsequent rounds. consolidation among large operators is discussed for the consolidation and diffusion treatments. but without financial implications. earnings averaged $24. The two treatments each included four sessions for a total 108 participants. Trade Prices Figures 2.

2 Diffusion treatment prices with an initial lease period.3 Consolidation treatment prices with an initial lease period.Diffusion treatment prices Lease Rounds 1–4 140 120 A B C D Equilibrium Round 5 Round 6 Price (experimental dollars) 100 80 60 40 20 0 Period Figure 2. Consolidation treatment prices Lease Rounds 1–4 140 120 E F G H Equilibrium Round 5 Round 6 Price (experimental dollars) 100 80 60 40 20 0 Period Figure 2. .

and thus should be equal to the predicted single-period equilibrium price. we model observed prices as a function of the number of periods remaining in the round.4 and 2. we cannot reject the hypothesis that the estimated value of 19. For the diffusion treatment. using the average period price from particular sessions as the dependent variable.337) 20. the coefficient on periods left in round is the only significant determinant of price in both treatments.5.596 (2. shown with a dashed line. respectively. the price interval’s upper band. In several cases. The one case where efficiency drops to 57 percent is a result of the outlier trading of two subjects. Consistent with equilibrium predictions. Efficiency levels in both treatments show a tendency to increase into the 90 to 100 percent range. we conclude that prices are consistent with equilibrium predictions. Table 2. The coefficients represent the amount the price changes between periods.918 (1. efficiency rises as high as 100 percent.818 (5. In the diffusion treatment. Table 2.308) 19.2 shows the results of the autoregressive heteroskedastic panel regression of prices in rounds five and six of the diffusion and consolidation treatments.185).57 is different from the predicted price of 20 (p = 0.076 (4. Hence.600) 2. with an indicator variable for the last period in the round to test for any bubble-crash behavior (Anderson.M. the efficiency in the consolidation treatment shows more variation.133 (1. in Figures 2.2 First-order autoregressive model of average prices in asset rounds 5 and 6 (N = 32) Diffusion* Constant Periods left in round Last period in round Total R-square Note * Standard errors are in parentheses.809 1. For the consolidation treatment.990) 0. efficiency shows improvement during repetition in both the initial lease periods and the asset rounds. 2004). In contrast.80 (p = 0.692) –0. Anderson et al.571 (0. The efficiency of the identical endowments. is 76 percent of the potential maximum profit for the diffusion treatment and 80 percent for the consolidation treatment. the price interval’s upper band. we also cannot reject the hypothesis that the estimated value of 20. Efficiency The allocative efficiencies that were realized in the diffusion and the consolidation treatments are displayed.806 To confirm statistically the general impression from the graphs that the market prices are consistent with equilibrium. These same two subjects were earning negative profits in period four of round five and in periods three and four of round . Even the minimum efficiency of 85 percent steadily improves. We obtain results using session-specific heteroskedastic and autoregressive processes with a common autoregression parameter.92 is different from the predicted value of 17. Consolidation* 0.330) 0.38 C.410).

As a result of the high efficiencies occurring in both treatments. the behavior of those two subjects balanced out with regards to average prices. we can expect the market allocations to be similar to the equilibrium predictions made about consolidation for both treatments. These profits were due to one subject selling her allowances for less than they would have earned her in fishing profits and the second subject also selling her allowances for a low price and then buying allowances for more than they were worth to her in fishing profits.3. where efficiency dropped below starting efficiency. whereas a gradual upwards trend is seen in the diffusion treatment. six. Market shares Even with transactions occurring efficiently and at equilibrium prices. In the consolidation treatment. Furthermore.4 Average efficiency in the diffusion treatment. the remaining subjects in that session and subjects in the other sessions behaved as predicted in terms of efficiency.Industry consolidation and diffusion 39 Efficiency of diffusion treatment market Lease Rounds 1–4 Round 5 Round 6 100 Profit from fishing/maximum profit from fishing 95 90 85 80 75 70 65 60 55 50 A B C D Endowment Period Figure 2. which is why no abnormalities were seen in Figure 2. the elasticity of the derived demand curve has the same effect on efficiency as it had on average price relative to the diffusion treatment: an immediate jump in efficiency occurs with the consolidation treatment. the experimental parameters support one equilibrium allocation for the diffusion . Despite these two outliers. we expect the resulting distribution of market shares to be different for the two treatments.

From the initial endowment.6. while in the consolidation treatment. Anderson et al. as seen in Figure 2. Of interest is the fact that in the consolidation treatment the consolidation of market shares sometimes overshot the predicted equilibria. the market shares of large and medium–large operators should converge towards the predicted equilibrium for the respective treatment. a gradual convergence towards the predicted equilibrium occurs in the diffusion treatment whereas convergence occurs more rapidly in the consolidation treatment. in which subjects respond to prices that others submit to the market by trying to overbid buy offers and underbid sell offers.689 for the consolidation treatment and 0. the predicted distribution of market shares should be 0.579 for the diffusion treatment. Since the subjects lack information about each other’s profit tables. treatment and a different equilibrium allocation for the consolidation treatment. The gradual convergence in the diffusion treatment is due to a price discovery process in the market. they require some time to observe price submissions and be able to determine what prices constitute a good deal. it converges to its predicted state of greater concentration.6. Efficiency of consolidation treatment market Lease Rounds 1–4 Round 5 Round 6 Profit from fishing/maximum profit from fishing (%) 100 95 90 85 80 75 70 65 60 55 50 E F G H Endowment Period Figure 2.5 Average efficiency in the consolidation treatment.4 Still. For this experiment. the market seems to converge in the diffusion treatment to its predicted state of low concentration. so subjects trade less than the optimal number of allowances until they attain enough price information.40 C. The market share distribution among operators is displayed in Figure 2.M. However. resulting . the change in market share holdings does not occur immediately.

’s (1995) panel model of price convergence: 1 t −1 Sjt = ∑ β Startj[ Dj ] + β Asymptote + µ jt t t j . market share holdings rarely diffused more than theory predicted. We use a two-way random effects model to control for serial correlation as well as for correlation among errors in sessions. the likely structure of the industry following implementation of a tradable allowance system can be predicted under the proposed cutback level. we model the data from both treatments to test that market shares are converging to their predicted shares.6. 1. given the different profit and cost structures among large and small operators in a fishery.Industry consolidation and diffusion 41 Lease Rounds 1–4 Round 5 Round 6 Large operators’ percentage of market shares (%) 75 73 71 69 67 65 63 61 59 57 55 Period A B C D E F G H Diffusion equilibrium Consolidation equilibrium Endowment Figure 2. whereas in the diffusion treatment. in the consolidation treatment.09 more permits are owned on average by large and medium–large operators than at the predicted equilibrium.6 Percentage of market shares held by large and medium–large operators. we adapt Noussair et al. Thus. the results point toward the consolidation of market shares in one treatment and the diffusion of market shares in the other. However. and that these shares are statistically distinct from one another. To address the effect that time has on convergence. To confirm this impression statistically. From Figure 2. in an even greater concentration of market shares amongst a few firms. systematic excess consolidation does not occur.

The asymptote for the diffusion treatment is not statistically different from the predicted value of 0. proving that the market shares are converging towards points that are distinct from one another. It is the value of the asymptotes for each treatment that is important. The estimated limits (Consolidation and Diffusion asymptotes) are reported in the first section of the table. not less than and not greater than the asymptotes. and convergence towards the asymptotes occurs in both treatments. or 1. As a result of the recent expiration of a 6-year federal moratorium on new programs. where µjt = υj + et + εjt. These asymptotes are statistically distinct from one another (p < 10–16). an obstacle to implementation that remains in many fisheries is stakeholder concern that market-based management will lead to industry consolidation. Anderson et al. the asymptote is only 8.707. and a number of fisheries are considering this newly available option. The asymptote for the consolidation treatment. For purposes of this estimation. The use of tradable allowance systems has been a topic of discussion and debate in many other countries for a number of years.8 percent greater. and a common asymptote to which all rounds are converging. respectively. Of 476 total permits in the treatment market. and all other periods’ indicies are incremented by one such that five periods exist. This model estimates a starting point of the convergence process for each round. this system may now be implemented in US fishery management.3 shows the results of the convergence regression model of the market shares. 0.42 C. The subscripts j and t correspond to round and period.689 (p = 0. As the period number gets larger.025). government and nongovernmental panels have recommended that US policymakers give greater consideration to using tradable allowance systems in fishery management. demonstrated the feared case. The p-values in the right-hand column of the two lower sections of the table test the hypothesis that the starting points are. respectively. One treatment. where larger operators . a period one index corresponds to the initial endowment.568). Sjt is the market share observed in period t of round j and Dj is an indicator variable that takes on the value one in round j and zero otherwise. All are rejected. However. indicating that market shares are moving during trade. is only borderline statistically distinct from the predicted value of 0. the consolidation treatment. and the starting points for each asset round of the consolidation and diffusion treatments are listed in the lower two sections. The asymptotes are statistically close to their respective predicted equilibrium. Table 2. for that indicates the equilibrated behavior in each treatment where market share convergence is heading.5 Discussion Based on their record of effectively reducing harvest below established limits and improving stock health and profitability in other fisheries.568 more permits than the predicted value. predictive weight shifts from the 1/t of the round-specific starting point towards the (t 1)/t of the common asymptote. and that they are moving toward distinct asymptotes as predicted by the general equilibrium model.579 (p = 0.M. we designed an experiment based on market fundamentals from an actual fishery. To address this fear.

633 0.634 0.627 0.025 0.568 p value (1-tail) Consol start # Consol asymptote Consolidation start 1 Consolidation start 2 Consolidation start 3 Consolidation start 4 Consolidation start 5 Consolidation start 6 Consolidation start 7 Consolidation start 8 0. Concentration is not. an automatic result of market-based management. There are a couple of important limitations of this experimental design that . As such.Industry consolidation and diffusion 43 Table 2. the general equilibrium model is useful in identifying when concentration and diffusion are likely to occur.636 0. because observed market shares corresponded to equilibrium predictions in both treatments. While large operators’ share increased. Second.643 0.652 0.3 Two-way random effects regression of market share consolidation Variable Consolidation asymptote Diffusion asymptote Parameter estimate 0.628 0.640 <10–5 <10–5 <10–6 <10–5 <10–4 <10–4 <10–5 <10–5 Note * All Consolidation start and Diffusion start variables have a 0. both Asymptote variables have a 0. market-based management can lead to market diffusion.0126 standard error. excess consolidation – beyond the level predicted by equilibrium – was not characteristic of the treatment on average. These results have two significant implications.638 0.00758 standard error.645 0. governments may not need to put caps on ownerships of market share percentages. A second treatment demonstrated the opposite: smaller operators purchased some of the market share of larger operators after allowance trading was permitted.642 0.641 0.640 0. First. particularly in apprehension of excess consolidation. as many people fear.642 0.623 <10–5 <10–5 <10–6 <10–5 <10–4 <10–4 <10–7 <10–7 p value (1-tail) Diff Start # Diff Asymptote Diffusion start 1 Diffusion start2 Diffusion start 3 Diffusion start 4 Diffusion start 5 Diffusion start 6 Diffusion start 7 Diffusion start 8 0.654 0.575 p value (2-tail) asymptote = predicted 0. based on the profit functions actually present in the fishery.707 0. buy the market share of smaller operators.

Acknowledgments We are grateful to Meifeng Luo for engineering the experiment software.M. This might happen if. the smaller family operations place lot of value on the fishing lifestyle. If larger operators experience higher marginal profits. producing widely varying outcomes. the shift may well go the other way. 2 The upper bound is used in analysis since prices are empirically closest to the upper bound.44 C. Anderson et al. or better diversify the market risk associated with holding a fluctuating asset. (2005) identify price dispersion. if smaller operators obtain higher marginal profits. Notes 1 An initial lease period is part of the design as Anderson and Sutinen (2006) have demonstrated it to dramatically reduce price volatility and improve market performance. should be recognized in applying these results. overcapitalized operators could make use of more traps and would purchase permits from operators who have a lower marginal profit. to Rhode Island Sea Grant (NOAA Award NA16RG1057) for financial support of the experiments and to the Rhode Island Agricultural Experiment Station for support of the Policy Simulation Laboratory. neither of the two treatments allows for a firm’s shutdown. Larkin and Milon (2000) find price ranges spanning from one to four times the mean price in the first 5 years of the Florida spiny lobster trap certificate program. Anderson and Sutinen (2005) argue high levels of volatility support speculation in new markets. it does suggest industry characteristics to consider. which could advantage larger firms that can afford to invest more speculatively in permit purchases and facilitate consolidation. For instance. While this result does not actually allow us to predict whether there will be concentration in the Rhode Island lobster fishery if and when tradable allowances are implemented. as much as 30 percent of mean price. evaluated at the level of trap certificates reflecting the initial cutbacks. can be highly volatile. shutdown is very likely for some firms. Field and laboratory analysis suggests that new asset markets that do include initial lease periods. This volatility may lead to disequilibrium market shares. In a naturally occurring industry. such as those designed for market-based management. as people familiar with family-scale commercial fishing often claim. First. Newell et al. then some shift in market share toward large operators is likely. On the other hand. While these limitations do need to be considered and may be addressed in future experiments. Second. as a prominent feature in the first 4 years of prices in a 30-species study of the New Zealand quota system. This empirical regularity has not been systematically studied and may possibly . In a laboratory analysis of tradable fishing allowances. only marginal scale changes are permitted. they do not alter our conclusion that both consolidation and diffusion may result from tradable allowance management. both treatments include initial lease periods in their design. long-run dispersion is closer to 10 percent.

Reberte. W. Anderson.. Ford. DC. C. The Effect of Initial Lease Periods on Price Discovery in Laboratory Tradable Fishing Allowance Markets. Arlington. 33 (2). 112–128. while the changes taking place are not enough to alter the market’s classification under the Index. Thus. Riezman. . 61 (2). Sharing the Fish: Toward a National Policy on Individual Fishing Quotas. 54–70. Richard. Agricultural and Resource Economics Review. Tech. Econometrica. 3 The treatment endowments are designed such that a certain number of subjects acting as small operators could be removed without affecting the predicted equilibrium. Rome. Noussair. 2002. Journal of Economic Behavior and Organization. Newell. Matulich. A report to the Department for the Environment. 2001. Paris.P. National Research Council. Hatcher. National Academy Press. 1119–1152. PA: Pew Oceans Commission.. FAO. Davidse. Williams. W. The Effects of the Introduction of Individual Transferable Quotas in the Tasmanian Rock Lobster Fishery. Pap. V. C. 2005. using a standard measure of market share. Organization for Economic Cooperation and Development. pp. Mittelhammer and C. 5 Analysis of the Herfindahl–Hirschman Index measure of market concentration shows that. FAO. Toward a More Complete Model of Individual Transferable Fishing Quotas: Implications of Incorporating the Processing Sector. and J. In A. Banks and R.. Plott and R. 462–491. 31 (1). 2000. where operators with large endowments withhold supply and drive prices to the upper end of the range. S. 49 (1). R. America’s Living Oceans: Charting a Course for Sea Change. Sanchirico and Suzi Kerr. References Anderson. 58. S. 412. 437–462. 164–180. 1995. Pascoe. FAO Fish. 193–208. R. we arrive at the same conclusion as looking directly at allocations. Crashes and Endogenous Expectations in Experimental Asset Markets. UK. Marine Resource Economics. The Effects of Transferable Property Rights on the Fleet Capacity and Ownership of Harvesting Rights in the Dutch Demersal North Sea Fisheries. Sutinen. Fishing Quota Markets. OECD. ——. 56 (5). Tradable Effort Permits: A Case Study of the Florida Spiny Lobster Trap Certificate Program. 2006. 20 (1). Pages 139–191. Rome. 1997. How Institutions Affect Outcomes in Laboratory Tradable Fishing Allowance Systems. 4 The length of trading periods was based on pilot experiments such that all trading that would occur did. James N. Washington. 2001. Journal of Environmental Economics and Management.. we do observe a strong bifurcation from the initial market concentration toward the predicted equilibrium points. Suchanek and A.W. Larkin. 2005. R.Industry consolidation and diffusion 45 be explained by an endowment effect. No. Towards Sustainable Fisheries: Economic Aspects of the Management of Living Marine Resources. Pew Oceans Commission.. An Experimental Investigation of the Patterns of International Trade. 412. G. Future Options For Fish Quota Management. Journal of Environmental Economics and Management. Smith. 268 pp. American Economic Review. 1999. NRC. University of Florida Working Paper. 2003.. Bubbles. 85 (3). ITQs in Practice: An International Review. 2004. eds. C. C. A Laboratory Assessment of Tradable Fishing Allowances. 1988. Milon. Arnason. Food and Rural Affairs. Report No. Centre for the Economics and Management of Aquatic Resources University of Portsmouth. and J. 1996.. 1–23. S. FAO Fisheries Technical Paper No. Arnason.

The Performance of Fisheries Management Systems and the Ecosystem Challenge. DC: US Commission on Ocean Policy. Soboil.. The Drama of the Commons. Washington.46 C. An Ocean Blueprint for the 21st Century. 197–232. Reykjavik. pp. United States Commission on Ocean Policy. 1995.A. Final Report. 2004. Tisdell.G.. Ostrom. 2003.M. Tietenberg. 2 (2). Kirkley and C. ed. Sutinen. J. Individual Transferable Quota as a Fisheries Management Tool. A paper presented to the Reykjavik Conference on Responsible Fisheries in the Marine Ecosystem. DC: National Academy Press. The Tradable Permit Approach to Protecting the Commons: What Have We Learned? In E. 141–169. October 2001. T. 2002. J. Reviews in Fisheries Science. Squires D. Iceland. and M. Washington. . Anderson et al.

2003). Emission trade has also occurred at the international level.g. CEA. Dales. Shogren. 1999. A prominent example of such global emission trading systems is in the 1997 Kyoto Protocol (UNFCCC 1999).3 Caveat emptor Kyoto Comparing buyer and seller liability in carbon emission trading Robert Godby and Jason F. 1999. however. it has been noted that “sanctioning authority is rarely .g. Global emission trading is central to cost effective climate policy because by some estimates it could halve compliance costs (e. Kyoto allows for some flexibility in GHG mitigation. and was used successfully. Barrett. 1966. 1998). Demands posed by national sovereignty concerns. contracting other emitters whose abatement costs are less than their own to make reductions for them (see Crocker. Emission markets have several appealing properties over traditional “command and control” regulation. Recall the Kyoto Protocol required the leading industrialized countries to reduce their greenhouse gas (GHG) emissions by an average of 5 percent below 1990 levels by 2008–12. In international treaties. 2000. Stavins. Emission trading allows regulated emitters to buy emission reduction efforts from other emitters – in effect. particularly with respect to Article 18 (see Rabkin. Shogren Introduction Tradable emission markets continue to be promoted as a cost effective tool to manage global public goods such as climate protection. Domestic trading programs in the United States indicate that realized cost savings can be substantial. though in a limited way to administer reductions in the use of ozone depleting substances demanded by the Montreal Protocol (see Solomon. see Nordhaus and Boyer. 1968).1 These reductions are severe relative to current GHG emission rates and will likely be costly to achieve (e. Article 18 in Kyoto).g. 1998. 1998). Schmalensee et al. Domestic trading programs in the United States and elsewhere have relied on strong enforcement and sanctioning frameworks to ensure market compliance. make the usual organization of emission permit markets problematic in an international setting. To minimize control costs. and chief amongst them is the fact that market outcomes can theoretically result in emission reductions occurring at least cost to society. The effectiveness of such global trading depends on the rules of enforcement and sanctions for nations that shirk on their emission commitments (e. including the international trading of GHG emission quotas (Article 17). 2004).

Proceeds from international GHG trading could be significant – the former Soviet Union countries projected emission trading revenues under Kyoto may be as high as $17 billion (Edmonds et al. But this problem should be viewed from the other direction. .. Here caveat emptor rules.g. Overselling eliminates the benefits to emission trading in reducing GHG emissions. While opponents contend that buyer liability will undermine efficient emission trading by depressing market prices and trade volume due to its increased risk.g. Costa Rica). 1999). see the Third Assessment Report of the IPCC. 1999). or equal to the value of all lending between the US and Russia between 1990 and 1996. The critical issue is who should be held liable for overselling permits beyond quotas – the seller or buyer country? Relatively weak under-compliance penalties and ineffective monitoring methods create the incentive for selling nations to oversell permits and shirk on their emission responsibilities due to the magnitude of the potential permit revenues. buyer liability is the choice given their objective – to make sure that the relatively rich buyer nations (e. Cason (2003) notes that this value. Shogren granted by treaty. measured in 1992 dollars.F. Japan) are liable for any shortfalls in emission reductions made by the relatively poorer sellers (e. buyer liability avoids the difficult issues arising from the negotiation of tough enforcement measures while creating incentives for efficient market outcomes (e. Russia. 2001). is equivalent to three-quarters of Russia’s trade surplus in 1997. and it needs to be addressed directly”. 2001). In domestic non-climate emission trading programs.48 R. and likely to be ineffective when used” (Chayes and Chayes. To most economists. International enforcement of emission reductions inside a sovereign nation is unlikely to be authoritative if based strictly on moral suasion or threats of potential trade sanctions (see Rabkin. This was something that the negotiators thought they could add later. 323) the Kyoto treaty “ignored enforcement. . see for instance Baron’s (1999) description of liability rules that could be used to support international GHG trade. Kyoto and climate change provide our motivating example. p. As noted by Barrett (2003. and this commitment creates an incentive for buyers to seek out sellers likely to comply with their emissions obligations thus increasing overall market compliance (see Victor. however. Godby and J. The argument for buyer liability arises from a belief that buying implies a commitment by the purchaser to ensure emission compliance.g. Permits sold by non-compliant nations would be returned to the sellers. US. seller liability is the obvious choice if the goal is to minimize transaction costs because all permits have the identical value to buyers. . Conveniently. Enforcement is the main challenge. proponents counter that fears of thin markets due to high transaction costs are exaggerated. To others. sellers are held liable for any permit shortfalls. 32). and the buyers would face the consequences of any resulting permit shortfalls. rarely used when granted. 1998. and only one market is needed to transact all permits. p. but the problem of relatively weak enforcement holds for any proposed emission trading system that crosses the borders of sovereign nations. .

buyer liability is preferable to seller liability. While not directly comparing seller liability and buyer liability. while less optimistic about the cost savings. Cason (2003) finds that in experimental buyer liability markets where sellers can make a commitment to honor their obligations. For example. whereas buyer liability increases transaction costs but leaves the efficiency untouched. Differences in the perceived likelihood of sellers being in compliance would result in buyers having different valuations for permits based on their origin. the Environmental Defense Fund (1998) said: in view of the limited set of enforcement tools available to an international regime. and the need for separate markets for permits offered by each seller. 2000). 1998). nations should adopt a limited but effective form of “buyer liability” to create incentives in favor of compliance and to ensure that the environment is made whole. Using a stylized Kyoto emission market double auction experiment in which liability rules are the treatment. since it would strengthen compliance incentives by discouraging buyers to purchase emission reductions from (entities in) countries which appear to be heading towards non-compliance. as markets form to capture the gains from trade and reputations work to police market behavior (e. This chapter explores the relative efficiency of a caveat emptor carbon trading program. Other observers. To our knowledge.g. Such concerns have been downplayed by some groups lobbying for particular types of liability rules in global emission trade negotiations. buyer liability is still preferable to a quantitative restriction on trade. market performance is enhanced relative to when such commitments cannot be made or . an outcome that helps the climate. The idea is that buyer liability results in buyers purchasing from those sellers they believe will honor commitments. no empirical studies have attempted to compare buyer and seller liability outcomes in markets.Caveat emptor Kyoto 49 Buyer liability rules increase the complexity of global emission markets. European Business Council for a Sustainable Energy Future. The working thesis is that buyer liability leads to greater climate protection. Both from an economic and environmental perspective. Also see Woerdman (2001) who argues: from the perspective of environmental effectiveness. note “buyer liability increases the environmental performance of the system by reducing the probability of invalid permits circulating” (Kopp and Toman. because such a restriction crudely reduces the overall cost saving potential. we show that buyer liability under relatively weak international enforcement leads to the worst possible outcome – less climate protection at greater costs. Differences in permit prices across sellers and incomplete information regarding the likelihood of seller compliance could both decrease the economic efficiency of using emission markets to achieve emission goals and increase the economic and environmental risks faced by buyers.

Examples include the EPA Open Market Rule (since withdrawn). We consider relatively weak enforcement levels to reflect the view held by many observers. find that introduction of supplementary rules such as reserve accounts and minimum eligibility requirements for market participation can be sufficient to induce market formation and increase market efficiency in international emission markets. We test the robustness of this result by creating a setting more likely to produce spontaneous “good seller” reputations in the permit market such that buyer liability would be more likely to succeed. Experimental environment. Schelling.F. Our findings indicate buyer liability lowers economic efficiency. Adoption of buyer liability with relatively weak enforcement renders emission trading inadequate by subverting the goals of Kyoto – creating less environmental protection at greater economic cost. 1983). Plott. although the results were not inconsistent with a weak reputation effect. Haites and Missfeldt (2001). But contrary to prediction. in which potential sanctions on sovereign nations over international emissions trading are likely to be meager (see for example Cooper. permit trade occurred under buyer liability as some sellers reduced production to support trade. We increased the monitoring probability and introduced a random ending point in the markets. we still observed more permit trading and production than expected. We work with climate but several environmental policy initiatives have considered buyer liability rules. 1997). 2000). however. Shogren are unobservable. They may not. Higher enforcement reduces output and increases emission compliance under seller liability. distorts permit prices and market production patterns. eliminate all losses such rules impose on the market. Overall. no studies exist to determine how the choice of liability rules affects market outcomes. we find that our main finding was relatively robust. 2001.g. Under buyer liability. PERT. A random ending point led to outcomes similar to those observed with a fixed end point. See for instance Muller and Mestelman (1998) for an overview of emission permit . We find such rules also significantly worsen environmental performance through greater noncompliance. Herein the choice of liability rule is a treatment variable to investigate the impact of the choice of liability rules. 1998. however.50 R. Godby and J. To date. in which the general implications apply to a wide variety of settings. design and hypotheses The laboratory is an effective means to test economic institutions. Our laboratory design is a test of concept experiment (see e. several proposed effluent markets in the US. and has been used extensively in the development of emission trading programs. as predicted. the PERT emission trading market in Canada and in a proposed emission market for SO2/NOx in southern Ontario (Ontario Government Ministry of the Environment. We do not calibrate the experimental parameters to exact features one might expect in a climate treaty. rather we test the concept of buyer liability within a global emission market. using a simulation study.

considers how buyer liability rules affect market outcomes when firms commit to technologies that ensure emission compliance to reduce uncertainty about their ability to sell permits. The laboratory environment in each session was designed to replicate the market incentives that would be present in possible emission trading markets without exactly replicating the decisions required in any particular market setting that might occur. Expenditures for permit purchases are deposited in these accounts and not released to sellers until it had been verified they had not overproduced. Bohm and Carlen (1999) and Carlen (1999) report experiments testing institutional issues other than liability in international GHG trade (also see Soberg. each was aware that their actual production level could be either one more. each subject would choose a level of production. All costs were subjects’ private information and production profits earned “lab-dollars”. Experimental environment. 1999). For buyer liability. for example. treatments and procedures The 12 experimental sessions reported were conducted. Since an international trading framework remains unspecified. while sellers forfeit any permit revenues. We include a refund treatment to test this hypothesis explicitly. In each period. Production costs varied across subjects. Each treatment “round” consisted of eight market periods. it was possible to produce without a permit – but each faced a fixed and known probability of . (ii) buyer liability – seller non-performance inflicts sanctions on buyers only. though production capacity did not. one seller and two buyer liability treatments were considered using an ABA crossover design. one less or the exact level they chose with an equal probability.2 Our design extends this work by comparing how buyer and seller liability affects market outcomes. Subjects were further informed that while production of each unit of output was expected to occur only if they held a permit to do so. each using a different set of student subject volunteers. We compare the performance of three liability rules: (i) seller liability – seller non-performance inflicts sanctions on the seller only. requiring 24 market periods to be conducted in each experimental session. Cason (2003). 2000). our laboratory environment purposefully focuses on the essential features of liability choice in a market under relatively weak international enforcement of emission shirking by sovereign nations. and (iii) buyer liability with refund – seller non-performance inflicts sanctions on buyers only. Subjects recorded production choices on their own computer terminal.Caveat emptor Kyoto 51 experiments conducted to evaluate American and Canadian program proposals. it has been suggested the incentive for sellers to oversell permits could be eliminated with the creation of “escrow accounts” (see Baron. To investigate the effect of alternative liability rules on market outcomes. Each session conducted three treatment rounds as described in Table 3. however.1. causing trades to be canceled and forfeiture of buyer’s permit costs. Each session involved eight experimental subjects who were informed they represented a firm producing and selling a product at an announced market price.

1 Experimental design Session 1 Treatment Seller liability Buyer liability Seller liability Buyer liability w/refund Buyer liability Buyer liability w/refund Buyer liability Seller liability Buyer liability w/refund Buyer liability w/refund Seller liability Buyer liability Treatment Treatment Seller liability Buyer liability Seller liability 2 3 Symbol Round 0 Treatment Practice Practice Practice Practice Buyer liability w/refund Buyer liability Buyer liability w/refund Practice Practice Practice Practice 011003 011004 011017 011018 011010 011011 011012 011015 011022 011023 011025 011026 020307 020315 020415 020418 020604 020607 020615 Seller liability High enforcement Buyer liability High enforcement Seller Liability Buyer liability High enforcement Seller liability High enforcement Buyer liability-Random end point SBS-1 SBS-2 BSB-1 BSB-2 SRS-1 SRS-2 RSR-1 RSR-2 BRB-1 BRB-2 RBR-1 RBR-2 SBS-HM-1 SBS-HM-2 BSB-HM-1 BSB-HM-2 SB-RE-1 SB-RE-2 SB-RE-3 Practice Seller liability High enforcement Buyer liability High enforcement N/A .Table 3.

while sellers faced no such fine. Using a whiteboard at the front of the laboratory. After production decisions were made. any subject found to have produced without enough permits was fined for each unit produced without a permit. the same rules applied as those just described. If found to have produced without enough permits to cover production.70 and standard deviation of $7. which opened for a fixed period of time. subjects were paid privately in cash an amount based on their accumulated labdollar earnings. 87 were undergraduates. Instead. Of these subjects. One session included three graduate . Under both buyer liability treatments. We imposed three treatments on the market. At the start of each period in the session.5–3 hours. permits could not be resold to other subjects and could not be carried over to future periods. After permit sales were canceled. whether they were monitored and any monitoring fines incurred and their earnings in the period. These markets were open for 150 seconds in rounds 1 and 2. individual subjects faced a fixed fine for every unit produced in excess of their permit inventory. converted to US dollars at a fixed and announced exchange rate. any buyers forced to return permits to a seller were informed of this outcome on their computer. with the exception that permit expenditures for canceled trades were refunded to buyers.Caveat emptor Kyoto 53 being monitored individually to determine if they held enough permits to cover their production. two in each experimental session (see Table 3. each lasting 2. Sellers also kept any permit sales revenues. while nine were first or second year graduate students in Finance or Economics. any seller who had sales canceled was identified by their ID number. sellers were not fined even if their production exceeded their permits. if sellers were monitored and found to have overproduced. and then a new period began. any permit sales made in the period were canceled in reverse order until they either had enough permits to cover their production or until all sales had been canceled. with a mean payment of $45. After this market closed. production profits earned. Under buyer liability with refund (BLR) treatments. In seller liability (SL) treatments. At the end of the experiment. Each subject was endowed with an initial lab-dollar balance at the beginning of the experimental session. some subjects were also endowed with permits while others were not. primarily from business disciplines. subjects were informed of their actual production levels.02. Subjects’ permit endowments and auction roles each period remained constant throughout the experiment. Buyer liability (BL) treatments imposed per unit fines on any buyer found to have produced without permits.1). even if permits were returned. Buyers who had to return permits were then monitored and fined for each unit produced without a permit. and 120 seconds in round 3 of each session. their permit revenues or expenditures. Individual subject earnings ranged from $27 to $73. A total of 96 subjects were recruited to participate in one of the 12 experimental sessions. the period the cancellation occurred and the number of sales canceled. Once purchased. subjects with permits (referred to as sellers) had the opportunity to sell permits to those who were not endowed any (buyers) using a computer-mediated double auction.

In buyer liability treatments. random production outcomes were generated using a random number generator drawing over the unit interval. buyers should be indifferent to the origin of any permits purchased so only one auction market opened.2 reports the actual subject production costs. The monitoring probability of one in . since it is a familiar setting to subjects. Buyer’s valuations for permits could differ across sellers thus we accommodated these concerns by simultaneously opening four permit auctions in which each seller could sell permits only in the market corresponding to their own their own ID number. An instruction phase followed that lasted approximately 45 minutes and included two unpaid practice periods. Figure 3. To determine which subjects were monitored in any period. Experiment instructions included details regarding the number of periods in each round and the number of rounds in the session. Following standard experimental protocol. Under seller liability conditions. each folder contained worksheets and experimental instructions particular to that session describing how to use the computer software developed for the experiment and subjects’ experimental tasks. We never mentioned pollution or production emissions during the experiment and we used context-neutral terms in the instructions to avoid potential framing effects. while six other sessions included one graduate student. redemption values (production price) of output. A monitoring probability of one in six was maintained for each subject in every period. Table 3. Since none had previous experience with the market environment or decisions they faced. Seller liability was used in practice sessions. probability of being monitored and fine per unit of excess production used in the experiment. initial permit endowments. All sessions used the same set of random monitoring and stochastic production outcomes in each period to ensure session outcomes were comparable. Subjects were asked to follow along as the instructions were read aloud. Prior to the experiment. with the number rolled on the dice determining the subject monitored in the period. two different subjects were excluded from each period’s dice roll thus every three periods another dice roll was necessary to determine which of six excluded subjects would be also be monitored to maintain the stated monitoring probability. For purposes of analysis. experimental sessions began with each subject being given a folder and randomly assigned to an experimental role and ID number. Since there were eight subjects in each period. To aid subjects in their production choices. Shogren students. a dice roll was used.54 R. All instructions are available by request. a profit calculator was provided on their computer to explore their potential profits for various production and permit holding levels if they were or were not monitored. we expected buyers to be concerned with the source of the permit and how many permits any seller had sold since buyers might expect different sellers to have different probabilities of overproducing. Godby and J.F. we assume that one unit of emissions was created for each unit of output produced.1 shows the general experimental procedure in each session. one using the training parameters and one unpaid and unreported practice session using the actual experiment parameters.

The analogy to dice rolls could then be used to ensure the random nature of monitoring and the concept of statistical independence was more likely to be understood by all subjects.70 115 136 151 161 185 315 3 500 0.00 25 139 148 164 235 265 0 3000 1.00 25 136 151 161 225 275 0 3000 1. Subjects were also told the approximate odds of being monitored every period of a round and of not being monitored in any period of a round.167) six was used intentionally.70 105 139 148 164 195 305 3 500 0.00 Production price: 170 Monitoring fine/unit: 240 Monitoring probability: 1/6 (0. under the presumption that subjects would have experience with rolling dice.00 25 142 145 167 245 255 0 3000 1. .Caveat emptor Kyoto 55 Table 3.70 25 133 154 157 215 285 0 3000 1.70 105 142 145 167 205 295 3 500 0.2 Experiment conditions and producer costs Marginal costs of production subject SL treatments Production unit 1 2 3 4 5 6 BL treatments Production unit 1 2 3 4 5 6 BLR treatments Production unit 1 2 3 4 5 6 Initial permits Initial balance Exchange rate Seller1 Seller2 Seller3 Seller4 Buyer5 Buyer6 Buyer7 Buyer8 128 133 154 157 175 325 125 136 151 161 185 315 60 139 148 164 195 305 60 142 145 167 205 295 8 133 154 157 205 285 5 136 151 161 225 275 2 139 148 164 235 265 2 142 145 167 245 255 115 133 154 157 175 325 115 136 151 161 185 315 105 139 148 164 195 305 105 142 145 167 205 295 25 133 154 157 215 285 25 136 151 161 225 275 25 139 148 164 235 265 25 142 145 167 245 255 115 133 154 157 175 325 3 500 0.

sales cancelled in reverse order – audit all buyers whose sales cancelled and assess fine if they are non-compliant.Sign-in Instructions/risk aversion test/DA practice New market: initial assignment of • cash balances • production costs • auction roles (buyer/seller) • auction rules Start period: initial period permit inventory assigned to each subject Production decision Permit market opens Permit market closes • updated cash balance • final permit inventory for period “Stochastic” production outcome assigned. .1 Session procedure. If buyer audited – assess fine if noncompliant. Compute period profits – update cash balance If period If round period 8 of round round 3 If period = period 8 of round If round = round 3 End session Figure 3. actual profits computed Monitoring – 1/6 probability of monitoring = yes • audit if monitoring = yes • seller liability – assess fine if non-compliant • buyer liability – if seller audited found non-compliant. Refund to buyers if treatment = BLR only.

and also on the number of permits these sellers have already sold.Caveat emptor Kyoto 57 After eight periods in each treatment round were completed. permit price predictions under seller liability (between 33 and 40) exceed those under the buyer liability with refund (between zero and three) and buyer liability (between zero and two) treatments in Table 3. which reintroduced the liability and cost conditions used in the first round.3. A buyer’s valuation now depends on the risk of overproduction they perceive of each seller. the costs subjects 3 and 4 faced in the first round were switched with those of subjects 1 and 2. For these reasons. Subsidies were paid to all subjects holding permits at the end of a period regardless of whether their permits covered production. To ensure that subjects did not focus on previous market outcomes. or on the marginal profit generated by the unit of output the permit is to be applied to. all agents value permits for this reason. Refunds raise buyers’ expected profits. cost conditions were similar across treatments. Under seller liability. Firms that produce without a permit face a potential fine. and only told their costs had changed. production costs were changed in each round without affecting aggregate production cost conditions. with only the first unit costs being changed to equalize predicted earnings differences across treatments. which reduces total permit market demand. Experimental predictions In our laboratory environment. but do not remove the discount buyer liability creates. whichever is less. Subjects were unaware that the overall market cost structure was the same as that used in the first round. while the costs of subjects 5 and 6 were switched with subjects 7 and 8. Earnings made in the previous round carried forward.1. In contrast. Buyers face higher risk and lower expected profits under these rules since their probability of being monitored increases when they buy permits from a seller that overproduces. Their valuation of each permit depends on the expected market fine if they are non-compliant. Prices under buyer liability will be higher if refunds are possible than when they are not. thus the following permit price predictions arise: PSL ≥ PBLR ≥ PBL. In the third round of each session. buyer liability removes the need for sellers to value permits. accumulative earnings could affect risk taking behavior if people have decreasing or increasing absolute risk aversion over wealth. predictions of market outcomes under the alternative liability rules considered can be identified using expected profit maximization under the assumption of risk neutrality. Table 3. subjects opened a new set of instructions that described the next liability treatment. and this may depend on their priors.3 summarizes our predicted market outcomes:3 Prediction 1: Permit prices will be lower under buyer liability than under seller liability. Induced permit demand and supply curves were also shifted by paying different fixed “permit subsidies” in each round. . As shown in Table 3. but less than those under seller liability.

Given our assumption of risk neutrality. predicted trading volume increases under the assumption of risk neutrality. the resulting flow of total emissions increases and the stock of GHGs accumulates faster when buyer liability is used. Risk aversion among subjects would lower the predicted trading volume. creating a vertical supply curve at the total permit endowment level in the market. In our environment. Prediction 3: Buyer liability in permit markets distorts production outcomes. In general. thus the following permit trade volume predictions arise: QBL = QBLR > QSL . quantity traded equals the number of permits endowed to sellers assuming buyer demand equals or exceeds supply. Predicted aggregate market output increases by eight units under both buyer liability conditions. Sellers no longer value permits. Given the assumption of emissions proportionate to output. seller subjects increase production to their unconstrained profit maximizing levels. Although the increased risk of default could decrease buyer demand to the point where expected trading volume is less than that under seller liability. however. however. increasing their total production by seven units. In this experiment. Without an effective emission constraint due to the liability rules imposed here. however. demand for permits can still exist despite the fact that all buyers may expect all sellers to shirk. even though under buyer liability permit demand functions rotate downward due to increased risk (the horizontal intercept remains fixed). Quantity traded will be unaffected by the presence of permit refunds. Probabilistic monitoring and higher predicted permit inventories.58 R. . whether the imposition of buyer liability results in buyer’s production increasing or decreasing depends on the probability of monitoring. increasing production and emissions on the seller side of the market without necessarily reducing buyer’s output and emissions. Godby and J. one might expect that a “lemons-market” outcome could occur here – that trade in permits would disappear since all buyers should expect all sellers to overproduce.F. although exact predictions depend on the subject’s degree of risk aversion. given the probabilistic monitoring. Given the fact that the length of the round is finite and known. and using the permit and the cost parameters used in this experiment. Shogren Prediction 2: Quantity of permits traded will be greater under buyer liability than under seller liability. increase the buyer’s predicted production levels relative to those under seller liability by one unit in total. Seller output. thus the following predictions regarding emissions arise: EBL = EBLR > ESL . which reduces the effectiveness of emission trading to achieve cost effective climate protection. the market conditions created here result in the predicted volumes under seller liability (six) or both buyer liability treatments (12) when risk neutrality is assumed. always increases to unrestricted levels. Buyer liability effectively relaxes the emission constraint faced by permit sellers and transfers it to buyers.

worse outcomes – higher emissions and lower efficiency – are in the upper right. we summarize our key finding over these criteria. Buyer liability leads to less economic efficiency and lower environmental protection. Better outcomes – higher efficiency and lower emissions – occur in the lower left corner. and the monitoring and production choices of others over the last three periods of each round by treatment. . The measure defined here indicates the potential efficiency given subject choices (not realized). and then discuss the specific market factors that created this outcome. while in seller liability treatments they improved. the inefficiency and emissions levels worsened under buyer liability.Caveat emptor Kyoto 59 Results We first present the key result. Result 1: Instituting buyer liability in the laboratory markets decreased market efficiency and increased carbon emissions relative to levels observed under seller liability.2 Efficiency and emission outcomes by treatment. Efficiency here is the ex ante expected profit a subject would realize given their period production and permit holding choices relative to the expected profit possible had they made the predicted decisions. As subjects gained experience. production and monitoring outcomes.2 captures this finding by describing the average efficiency and emissions-via-production by treatment and round. The realized ex ante efficiency loss is computed as 1 – ((predicted Eπ – actual Eπ)/predicted Eπ) given subject’s decisions recorded in 250 BLR–1 BL–2 BL–3 BLR–3 Over-emission percentage 200 SL–1 SL–3 150 BL–1 BLR–2 SL–2 100 50 0 0 5 10 15 20 25 30 35 Unrealized efficiency percentage Figure 3. Since liability rules in climate negotiations are debated on two main criteria – the cost effectiveness (or market efficiency) and carbon emissions outcomes. Figure 3.

00 and p = 0. Efficiency increased with experience in the seller liability treatments. round 1.3 Mean aggregate buyer production by treatment. Using ANOVA analysis for descriptive purposes. but worsened under buyer liability. Round 3.2 there was little difference in average emissions levels across the buyer liability treatments.3 reports that differences in mean production outcomes by round are stable over time.6 percent and 73.F.9 percent across all buyer liability without and with refund treatments. while average seller liability emissions levels were always lower than those under either buyer liability treatment regardless of experience. round 3. Godby and J. round 2.00 in both cases) and their interaction (p = 0.60 R. Highest average efficiency and low emissions occurred under seller liability observed in round three (SL-3) of the experimental sessions. experience (p = 0. BPC T3–1 = buyer production choice buyer liability w/refund (BLR). BPC T2–2 = buyer production choice buyer liability (BL). Figure 3. BPC T3–3 = Buyer Production Choice Buyer Liability w/Refund (BLR). Emission levels are the percentage deviation from the emission quota defined by the total number of permits available. round 1. and reflect the qualitative differences predicted between treatments.3 presents a time 15 SL BPC T1–1 BPC T1–2 BPC T1–3 BPC T2–1 10 BPC T2–2 BPC T2–3 BPC T3–1 BL 5 0 8 Period 16 24 BPC T3–2 BPC T3–3 Figure 3. BPC T2–3 = buyer production choice buyer liability (BL). Table 3. BPC T1–3 = buyer production choice seller liability (SL).2 shows that the highest efficiency outcomes were always observed under seller liability: average efficiency in the last three periods of all seller liability treatments was 82. Efficiency under buyer liability was significantly lower – 71.00 in both cases).03).4 Figure 3. round 2. Quantity chosen . Notes BPC T1–1 = buyer production choice seller liability (SL).1 percent. Emissions levels directly reflect production levels in our environment. while lowest average efficiency and high emissions were observed in third round observations of the buyer liability treatments (BLR-3 and BL-3). mean production choices by sellers and buyers appear to be influenced by treatment (p = 0. These results strongly support Prediction 3 – looking at Figure 3. BPC T3–2 = Buyer Production Choice Buyer Liability w/Refund (BLR). BPC T1–2 = buyer production choice seller liability (SL). Shogren the last three periods of each round. Round 2. round 1. BPC T2–1 = buyer production choice buyer liability (BL). round 3.

4 Mean aggregate seller production by treatment. round 3. round 2. SPC T2–1 = seller production choice buyer liability (BL). SPC T3–2 = seller production choice buyer liability w/refund (BLR). round 1. round 3. Notes SPC T1–1 = seller production choice seller liability (SL). Normalized price . SPC T3–3 = seller production choice buyer liability w/refund (BLR). Np23 = normalized price buyer liability (BL). SPC T1–2 = seller production choice seller liability (SL). round 1. SPC T2–3 = seller production choice buyer liability (BL). round 2. SPC T1–3 = seller production choice seller liability (SL). round 2. round 3. round 1. Np22 = normalized price buyer liability (BL). round 1. Notes Np11 = normalized price seller liability (SL). round 2. round 3. round 3. SPC T2–2 = seller production choice buyer liability (BL). round 2. Np12 = normalized price seller liability (SL). Np33 = normalized price buyer liability w/refund (BLR). Np32 = normalized price buyer liability w/refund (BLR). 50 SL 40 np11 np12 np13 30 20 10 0 0 BL 8 Period 16 24 np21 np22 BLR np23 np31 np32 np33 Figure 3. round 1. SPC T3–1 = seller production choice buyer liability w/refund (BLR). round 2. round 3. Np31 = normalized price buyer liability w/refund (BLR). Np21 = normalized price buyer liability (BL).5 Mean permit prices by treatment. round 1.20 BL SPC T1–1 SPC T1–2 SPC T1–3 Quantity chosen 15 SPC T2–1 SPC T2–2 10 SL 5 0 8 Period 16 24 SPC T2–3 SPC T3–1 SPC T3–2 SPC T3–3 Figure 3. Np13 = normalized price seller liability (SL).

6 support Prediction 1 – we see lower permit prices under buyer liability. Table 3.6 We assume the random disturbance eit exhibits first order autocorrelation with previous disturbances.4).5 reports average permit prices by treatment observed over time. such that eit = ρieit–1 + εit . where ρi is the autocorrelation coefficient in session i. Prices under the buyer liability treatments are also shown to differ significantly using a Mann-Whitney U-test across all three rounds. On average. the results support our benchmark predictions regarding permit prices and volume.3 and 3. transaction and production volumes observed by round and treatment. In rounds in which subjects were less experienced.F.4).4. the last two or over the third round only (p = 0. Considering seller’s production (emission) choices (see Tables 3. Shogren series of mean buyer production choices by treatment and round and indicates treatment effects occurred early in rounds. Result 2: Average permit prices deviate by treatment as predicted – prices were greater under seller liability relative to buyer liability. mean production (emission) choices by sellers increased by approximately four units under buyer liability relative to choices under seller liability conditions. which presents a time series of mean seller production choices by treatment and round. and buyer liability prices were only slightly lower than those observed under buyer liability with refund. . pricing behavior appears essentially similar.00).3 and 3. We estimate: Pit = a + b1BL + b2BLR + b3Rnd2 + b4Rnd3 + b5BL-Rnd2 + b6BL-Rnd3 + b7BLR-Rnd2 + b8BLR-Rnd3. average production (emission) levels chosen across all buyers converged toward the predicted levels (see Table 3. A KruskalWallis test of mean prices by treatment indicates a significant effect due to treatment (p = 0. The results in Figure 3. Direct comparison of seller and buyer liability (with and without refund) price outcomes indicate mean prices are always higher under seller liability conditions in later periods. But as they gained experience both within and across rounds. buyers often overproduced.62 R. We confirm these unconditioned tests with a random effects regression: Pit = α + ′xit + ui + eit where Pit is the price observation at time t of session i in the experiment. shows a contrasting outcome. Godby and J.5 Figure 3. α is an intercept. Overall. and εit is a classical disturbance term.5. while comparison of the two buyer liability treatments both within and across sessions indicates that although variation in observed prices occurs. is a vector of coefficients on the liability rule and order effects as defined in the matrix xit .5 and Tables 3. the effect of treatment seems apparent. which we summarize below. especially after subjects gain experience with the markets. Figure 3.00). and Figure 3.3 summarizes the results on permit price.

96) 7.22) 14.09 (2.07) 5.97 (1.41 (2.84 (2.84 (2.39) 5.84 (1.47) 24.07 (33.65) 24.34) 24.88 (1.58) 4.97) Mean permit price Treatment: SL Prediction: Mean obs. dev.88) 19. (std.57) 7.39) 6.75 (3.03 (1.69 (1.28 (1.25 (1.25 (3.20) 12 15 13.31 (2.36) 9 9.60) 4.07) 6.) Round 1 Round 2 Round 3 33–40 42.72 (1.3 Experiment results by treatment Aggregate permit holdings Sellers 6 6.95) 9 11.32) 15. dev.06 (2.35) 40.53) 24.91 (1.98 (8.33) 24 24.93) 6.89) 9.60) 13.48) 6 8 8 11.17) Buyers Sellers Buyers Aggregate production choice Total 16 21.92) 10.71) 9.95) Note Predictions assume risk-neutral traders.72 (8.20) 0 5.12) 24 23.82 (18.93) 9.71) 0 5.42) 9.76 (7.34 (1. that the buyers assume a probability of p = 1 that sellers will overproduce.26) 5.34 (2.26 (4.84 (9.) Round 1 Round 2 Round 3 0–3 21.69 (1.50 (2.19 (1.97 (1.44 (1.88) 15 13.59 (2.33) 5.34 (2.97 (1.84 (1.61) 9.38 (2.33) 14. dev.06 (1.16 (1. (std.63 (4.75) 9. .69) 22.94 (1.88 (1.69) 8.10) 8.93) 12 6.59 (2.Table 3.47 (1.38 (1. and that in the case of both buyer liability treatments.) Round 1 Round 2 Round 3 0–2 5.05) Treatment: BL Prediction: Mean obs.39 (15.33) 6.69 (2.95) 9.50 (0. (std.64) 7.42) Treatment: BLR Prediction: Mean obs.60) 7.03 (1.38 (2.75 (1.98) 19.00 (1.16 (2.57) 4.50) 9.03 (1.96) 4.41 (2.

08 0.F. Frequency Mean price Std.29 3 2. dev. dev.14 3 2.58 0.14 3 2.333333 0.14 3 1.381881 3 2.75 0.14 3 2.144338 3 2.67 0.4 Buyer mean production choices (periods 6–8) Round Session 11003 SL-BL-SL 11004 SL-BL-SL 11010 SL-BLR-SL 11011 SL-BLR-SL 11012 BLR-SL-BLR 11015 BLR-SL-BLR 11017 BL-SL-BL 11018 BL-SL-BL 11022 BLR-BL-BLR 11023 BLR-BL-BLR 11025 BL-BLR-BL 11026 BL-BLR-BL Total Statistic Mean price Std.5 0.5 0.333333 0. dev. Frequency Mean price Std.34 0. Frequency Mean price Std. Frequency 1 3.08 0.92 0. dev. dev.29 3 2.92 0.14 3 2.083333 0. dev.25 3 2. Frequency Mean price Std.14 3 3 0 3 2. dev. Shogren Table 3.67 0.14 3 2. dev.144338 3 2.25 3 2. Frequency Mean price Std.14 3 1.83 0. Frequency Mean price Std.52 3 2.64 R.25 0 3 2.39 36 2 2.83 0. Frequency Mean price Std.38 3 2. Frequency Mean price Std.41 36 3 2.25 3 2.67 0.5 3 2. dev.14 3 1.25 0. .33 0.14 3 2 0 3 3. dev.33 0.381881 3 1. Frequency Mean price Std.25 0 3 2. b3 and b4 are estimated order effects and indicate the average deviation in prices observed in rounds 2 or 3.08 0.14 3 2.29 3 2.33 0.25 3 2 0 3 2.14 3 2.29 3 1. dev. Frequency Mean price Std. dev.08 0.67 0.67 0.144338 3 2.54 0. Godby and J.416667 0.42 0.33 0. Frequency Mean price Std.92 0.14 3 2.75 0.583333 0. dev.25 3 2. the coefficients b1 and b2 are the estimated average price deviations from those observed under seller liability in round 1 induced when buyer liability and buyer liability with refund rules are used.75 0. Frequency Mean price Std.45963 36 Here a is the estimated average price in markets using seller liability rules in round 1.75 0 3 2.

dev.29 3 2. dev.25 3 3.00 3 3.00 3 4.29 3 3. Frequency Mean price Std. dev.00 3 3.42 0.00 0. and that these differences do not require previous experience in markets to be realized. dev.25 3 2.25 0.29 3 3. Frequency 1 2.38 3 3. Frequency Mean price Std.75 0. Frequency Mean price Std.00 3 3. On average.92 0.92 0. Frequency Mean price Std. Frequency Mean price Std.25 0.50 0. Frequency Mean price Std.00 3 4.75 0.17 0.83 0.14 3 3.08 0.00 3 3.00 3 2 3.00 0. dev.92 0.00 0.00 3 3. while prices under buyer liability with refund fall by between 20 and 24 lab-dollars.7 reports the regression results.00 0. no significant difference emerges in how experience . dev.29 3 2. and indicates liability treatment has a significant impact on closing prices.42 0. dev.83 0. Frequency Mean price Std. Further.00 3 3 2.00 0.17 0.08 0.38 3 3.75 0.92 0.29 3 2.00 3 3.14 3 4.5 Seller mean production choices (periods 6–8) Round Session 11003 SL-BL-SL 11004 SL-BL-SL 11010 SL-BLR-SL 11011 SL-BLR-SL 11012 BLR-SL-BLR 11015 BLR-SL-BLR 11017 BL-SL-BL 11018 BL-SL-BL 11022 BLR-BL-BLR 11023 BLR-BL-BLR 11025 BL-BLR-BL 11026 BL-BLR-BL Statistic Mean price Std.00 3 and the estimated coefficients b5–b8 indicate the average interaction effects of buyer liability and market order.00 0.25 3 2. dev.25 0. Table 3. dev.80 3 1.00 3 2.14 3 4.00 0.33 0. dev. Frequency Mean price Std.14 3 2.50 0.14 3 3.00 0.14 3 3. Frequency Mean price Std.83 0. Frequency Mean price Std. dev.38 3 4.14 3 2. Frequency Mean price Std.14 3 3.00 3 4.Caveat emptor Kyoto 65 Table 3.83 0.43 3 2.58 0.33 0.58 0.00 3 3. prices under buyer liability fall by about 30 lab-dollars relative to prices observed under seller liability.75 0. dev.29 3 4.00 0.

87 2. dev.53 2.73 0.91 3 2.70 3 2 18.66 R. Frequency Mean price Std.65 5. dev. however.35 0. .30 3 36.79 3 0.33 3 43.38 3 1. however.36 0.54 0. Frequency Mean price Std. Shogren Table 3.91 3 15. Frequency Mean price Std. Frequency Mean price Std.14 7. Frequency Mean price Std.95 3 6.63 3.F. are not observed in all sessions.11 3 3 57.46 2.94 0.06 0.91 3 1.61 3 15. dev.94 3.91 3 2. dev.57937 1.58 3 51.30 3 34.82 3 32.87 1. dev.13 3 12.57 3 2.17 0.44 3 3.09 0. Result 3: Permit trading volume increases under buyer liability.36 2.19 3 6.41 2. as supported by the time paths of prices in periods 9–16 (see Figure 3.42 0. Predicted differences in volume between liability treatments.57 3 3.63 0. dev.38 3 5. that early negative price and risk experiences under buyer liability initially depress closing prices under seller liability.33 1. Frequency Mean price Std. dev.5).40 3 29.30 3. dev.89 0.6 Session mean price results (periods 6–8) Round Session 11003 SL-BL-SL 11004 SL-BL-SL 11010 SL-BLR-SL 11011 SL-BLR-SL 11012 BLR-SL-BLR 11015 BLR-SL-BLR 11017 BL-SL-BL 11018 BL-SL-BL 11022 BLR-BL-BLR 11023 BLR-BL-BLR 11025 BL-BLR-BL 11026 BL-BLR-BL Statistic Mean price Std.59 3 7.75 2 influences outcomes under alternative liability treatments.45 1.62 3 2.7 Our results do suggest.07 0.73 3 –2.30 1.92 0.19 3 2.07 2.81 3 11. Godby and J.52 3 18.43 1.69 0.26 2.13 0. Frequency 1 45. dev. Frequency Mean price Std. Frequency Mean price Std.13 3 8.92 3 35. dev.33 0. dev.09 0. Frequency Mean price Std.33 3. dev. Frequency Mean price Std.44 3 37.05 3 21.54 3 42.73 3 7. Frequency Mean price Std.29 3 2.

volume converges toward predicted levels. whereas in the other half. trade could collapse to zero.629 (0. In about half the sessions. the average buyer holds two more permits relative to holdings under seller liability.799) 6.918 (0.678 (0. With such risk.813) –7. supporting our benchmark prediction. and volume differences under buyer liability increased with experience. markets were thin.221) 1. A Mann-Whitney U-test also verifies that no significant difference exists in buyers’ permit inventories after trade between buyer liability treatments (p = 0. Figure 3.7 Estimated regression results (p-values in parentheses) Estimated coefficient Dependent variable (Pit ) Closing prices all periods1 Constant (a) BL (b1) BLR (b2) Rnd2 (b3) Rnd3 (b4) BL-Rnd2 (b5) BL-Rnd3 (b6) BLR-Rnd2 (b7) BLR-Rnd3 (b8) Number of observations 38.609 (0.Caveat emptor Kyoto 67 Table 3.639) 5.000) –9. allowing for heteroskedasticity between sessions (cross sections) and assuming an AR1 process present in the data.640 (0. Table 3.000) –29. volume increased but remained below predicted volumes.3 shows differences in permit trading occurred across treatments. this is not a predicted outcome.695) 1. What is unclear in the figure is that significant deviation occurs in trading volume under buyer liability across sessions.794) 1. While these results indicate trading volume (and therefore buyer inventories) increases under buyer liability.187) –1.025 (0. A zero trading outcome would be consistent with a “lemons-market”: buyer liability creates an incentive for sellers to cheat and if they are monitored permit purchases could be lost. 1 Estimates for all periods are the GLS coefficients.195 (0.168 (0.000) –30.000) –1. 2 Estimates for all periods are the OLS coefficients.681 (0.000) –24.515 (0. and this divergence increases with experience.000) –19. 1995). buyer liability outcomes diverge toward a volume appreciably greater than seller liability.047 (0.887 (0. Although in the early periods of a round.6 shows the average permit trading volumes by round and treatment. the increase is less than expected.47).8 Overall. By the third round. Using these coefficients is recommended when the number of time periods is small relative to the number of panels (see Beck and Katz.049 (0. the causes of the inferior emission and efficiency outcomes under buyer liability seem apparent – they arise due to the predicted distortions in observed permit .912 (0.815 (0. A Kruskal-Wallis test confirms a significant difference between average buyer’s permit holdings over the last three periods of the each round (p = 0. allowing for heteroskedasticity between sessions (cross sections) and assuming an AR1 process present in the data. But since risk of monitoring is low. trade rarely failed to occur.825 (0.001) –4.124) –0.739) 179 Notes p-values in parentheses – bolded values indicate significance at the 5 percent level. In later rounds.088) 287 Closing prices periods 4–82 39. trading is thin. By periods 4–8.00) by treatment.780 (0.000) –11.814 (0.

68 R. Godby and J.F. Shogren
12 10 8 BL ph11 ph12 ph13

Trades

ph21 ph22 ph23 ph31 SL ph32 ph33 0 8 Period 16 24

6 4 2 0

Figure 3.6 Mean trades by treatment.
Notes ph11 = Permit holding seller liability (SL), round 1. ph12 = Permit holding seller liability (SL), round 2. ph13 = Permit holding seller liability (SL), round 3. ph21 = Permit holding buyer liability (BL), round 1. ph22 = Permit holding buyer liability (BL), round 2. ph23 = Permit holding buyer liability (BL), round 3. ph31 = Permit holding buyer liability w/refund (BLR), round 1. ph32 = Permit holding buyer liability w/refund (BLR), round 2. ph33 = Permit holding buyer liability w/refund (BLR), round 3.

prices and trading volumes. These distortions result in allocative inefficiencies and production pattern distortions that reduce permit market cost effectiveness, and increase emissions.

Robustness tests – high enforcement and random endings
Less climate protection at greater cost under buyer liability is a strong result. Consider the robustness of this outcome given two changes in the permit market: an increase in the monitoring probability to two-thirds (0.667) from one-sixth (0.167), and removal of fixed-ending point effects caused by using a known number of periods within each round by introducing a random ending point (i.e. an infinite horizon) after the eighth round. These conditions create an environment more likely to produce “good seller” reputations in the permit market so buyer liability stands a better chance of success. Increased monitoring probability Table 3.8 summarizes the revised predictions and results in the high monitoring environment. Under seller liability with a higher enforcement, eight permit trades are predicted with permit market prices predicted to lie within an interval of 27–71. This wide interval occurs due to the wide discrepancy between buyer and seller valuations at this quantity due to the characteristics of the expected

Table 3.8 Experiment results by treatment: additional sessions Mean permit price Aggregate permit holdings Sellers Buyers Sellers Buyers Total Aggregate production choice

High enforcement, fixed length sessions Treatment: SL Prediction: Mean obs. (std. dev.) Round 1 Round 2 Round 3 27–71 81.57 (16.81) 61.59 (8.64) 69.22 (7.85) 0 50.51 (17.68) 38.94 (16.57) 38.43 (8.96) 8.13 (2.06) 4.75 (1.12) 7.50 (1.15) 3.88 (2.06) 7.25 (1.12) 4.50 (1.15) 12 0 15 9.25 (3.09) 8.44 (2.16) 10.63 (3.01) 6.50 (1.32) 8.25 (1.65) 5.81 (1.11) 5.50 (1.32) 3.75 (1.65) 6.19 (1.11) 6.00 (2.16) 7.19 (2.46) 5.94 (1.06) 4 8 4 8 7.06 (1.24) 3.88 (1.54) 5.94 (0.93) 0 3.94 (1.44) 6.31 (1.54) 2.81 (1.05)

12 13.06 (2.98) 11.06 (2.88) 11.875 (1.31) 15 13.19 (3.31) 14.75 (2.38) 13.44 (3.03)

Treatment: BL Prediction: Mean obs. (std. dev.) Round 1 Round 2 Round 3

Low enforcement, random length sessions Treatment: SL Prediction:* Mean obs. (std. dev.) Round 1 33–40 28.48 (8.61) 0–3 9.73 (6.20) 0 6.80 (2.14) 7.42 (2.86) 6 6 12

8 4.58 (2.86) 12.25 (1.83) 15 5.20 (2.14) 12.92 (1.35)

8 9.21 (1.22) 9 8.16 (1.93)

16 21.46 (2.19) 24 21.08 (2.71)

Treatment: BL Prediction:* Mean obs. (std. dev.) Round 2

Note * Predictions in random length sessions are those for fixed length sessions, and not necessarily the only possible prediction.

70 R. Godby and J.F. Shogren profit maximization. Although this interval allows for prices that are lower than those predicted with lower monitoring odds, one could expect that since the probability of being monitored and caught producing without a permit has increased, permits would be more valuable and sell at a higher price than observed when monitoring occurred only one-sixth of the time. The center of this interval occurs at 49, approximately 10 lab-dollars higher than expected previously. Relative to the lower enforcement setting, increasing the monitoring probability increases permit prices and permit trades, while reducing market output. In seller liability sessions, no subjects are predicted to produce without a permit; sellers are predicted to reduce output to a level commensurate with their predicted permit holdings. Under buyer liability, increased monitoring probability should significantly affect predicted market outcomes. Buyers should presume that sellers will always overproduce given they face no sanctions. Buyers now should not value permits, and one should observe a complete market failure – valueless permits should not be exchanged. Facing no risk of a fine, we predict sellers will produce at their unconstrained profit maximizing levels, increasing total market output 25 percent beyond the emission cap imposed. Buyers are predicted to produce nothing. Result 4: Higher enforcement reduces output and increases emission compliance under seller liability, as predicted. Contrary to prediction, however, permit trade occurred under buyer liability as sellers reduced their production to support trade. Buyer liability with high enforcement made the problems observed in low enforcement sessions more costly due to naive and strategic behavior – “good sellers” exploit their reputations, while still promoting overemissions, albeit at reduced levels. Tables 3.8 and 3.9 present the summary statistics on how high enforcement altered market performance as measured by permit price, permit holdings, production, and the overproduction-to-total sales ratio.9 As predicted, seller liability resulted in almost perfect compliance with the emission target. For seller liability, Table 3.8 shows that trading led to higher mean permit prices relative to the low enforcement sessions, as predicted. The price increase is more extreme than expected – mean price (81.57) is 10 lab-dollars greater than predicted (although in the later rounds, mean prices fall to within the predicted range). We observe little change in permit trade volume relative to the low-enforcement sessions. Aggregate output decisions are not substantially different from aggregate permit holdings, supporting the prediction that higher enforcement reduces output and increases emission compliance. High monitoring reduced mean production outcomes by about eight units (nearly 40 percent) from low monitoring. As predicted, aggregate output choices are, by the second round, compliant with the emission target in the market. Contrary to predictions, however, the permit market did not completely collapse under buyer liability. Permit trades still occur at prices similar to the low

Caveat emptor Kyoto 71
Table 3.9 Overproduction-to-total sales ratio Treatment Round 1 Low monitoring probability fixed length sessions Seller liability 0.62 (0.53) Buyer liability 1.27 (0.21) Buyer liability w/refund 1.21 (0.23) High monitoring probability sessions Seller liability –1.12 (0.40) Buyer liability 0.33 (0.83) Random end-point sessions Seller liability Buyer liability 1.04 (0.53) N/A 2 0.76 (0.38) 1.38 (0.39) 1.29 (0.65) –0.31 (0.54) 0.49 (0.35) N/A 1.13 (0.27) 3 0.51 (0.26) 1.64 (0.54) 1.34 (0.16) –0.003 (0.22) 0.76 (0.64) N/A N/A

enforcement case. The extent of market failure is shown in Figure 3.7, which redraws Figure 3.2 adding the additional sessions. Trades occurred because sellers reduced their output to cover permit sales and because not all permits sold were “bad” (i.e. buyers were not fined) – seller output averaged 40 percent less than predicted. Market efficiency now decreased because production was lower and some permit trades went “bad” and buyers were fined. Market inefficiency resulted from myopic or naive market decisions, especially by buyers who had greater costs through realized fines and lower subject earnings. Increased enforcement made the market more unforgiving – subjects endured high penalties for poor production decisions and sudden strategic switches in seller behavior. This strategic effect held when “otherwise-good” sellers turned on their buyers and began overproducing in the later periods of the rounds. Since buyers did not refrain from trading under these circumstances, efficiency losses arose from the penalties they incurred. In buyer liability sessions under high enforcement, seller overproduction occurred in all rounds, and increased over time (see Table 3.9). In the first round, these values indicate on average two-thirds of permit sales did not risk being returned if a seller was monitored. Increasing enforcement resulted in overproduction ratios 75 percent lower than those in the low enforcement sessions. Similarly, in the last round, although overproduction/total sales values and the risk of returning permits rose over time, on average, almost 25 percent of the permits sold were backed by output reductions. Compared to the low monitoring sessions, this overproduction ratio was less than 50 percent of the value observed under buyer liability (0.76 versus 1.64), and 44 percent lower than observed under buyer liability with refund (0.76 versus 1.34). Buyer liability in a market with high enforcement made the problems observed in low enforcement sessions more costly, while still promoting over-emissions, albeit at reduced levels.

72 R. Godby and J.F. Shogren
BL–2 BLR–1 BL–3 BLR–3 BLR–2

200

Emissions – percentage of cap

BL–1 SL–1 SL–3

BL–2RE SL–1RE SL–2

150

Buyer liability with refund Random end SL-high monitoring Seller liability Buyer liability BL-high monitoring
BL–3HM BL–2HM

100
SL–3HM

SL–1HM

BL–1HM

SL–2HM

50 0 20 40 60 80 100 120 140 Unrealized efficiency percentage 160 180

Figure 3.7 Efficiency and emission outcomes by treatment.

Random ending points Our markets were hostile to reputation effects since our rounds were short and of known length. This design reflects current climate policy – the original Kyoto timetable from 2008–2012 is fixed and short. Compliance periods beyond 2008–2012 have yet to be negotiated. This contrasts with other emission markets where buyer liability has been promoted that do not end at a fixed date. It is assumed participants in these markets would assume market conditions are permanent or at least that the end of the market is unpredictable. To capture this belief within our experiment and to avoid the end-period strategic effects that arose in the high enforcement sessions, we replaced the fixed and known end point to each round used previously with a random end point. This allowed us to explore whether “good seller” reputations might emerge in the labortory, and whether such outcomes could alter our key results. We focus on the low monitoring probability treatments. All conditions of the original sessions were maintained with two exceptions: (1) in buyer liability, we used a random ending point; in seller liability, we did not tell subjects when the period would end; and (2) we used an AB experiment design. Seller liability always occurred for the first eight periods (A), with buyer liability rules for the rest of the sessions (B) (see Table 3.1 for sessions 020604–020615). The AB design ensured sessions would not be too long, since the random ending point could cause the buyer liability rounds to go on for some time. Although using an AB design theoretically confounds the effect of experience and treatment on interpretation of observed results, we expected the effects of time and treatment would be easily identified by comparison to previous ABA sessions. If they

Caveat emptor Kyoto 73 were not, it was expected that the effect of changing the fixed to random ending point would have significant impacts on market outcomes if the new sessions had the predicted effect. After the eighth period in buyer liability markets, we used a random number generator to draw a number between one and 100 at the end of each period.10 Subjects knew the period ended if the number drawn was 20 or less – a one-fifth chance – otherwise another period would take place. We use the same predictions as in the low enforcement probability sessions under seller liability as a benchmark. Under buyer liability, our predictions are those computed for the fixed length rounds. If a reputation effect emerged, market outcomes will differ from predictions, suggesting the random end point changed behavior and market outcomes. Result 5: A random ending point had little effect in seller liability sessions – outcomes were similar to those observed with a fixed end point. Under buyer liability, the results of with random ending point are not inconsistent with a weak reputation effect – we observed fewer permit trades and lower production levels. Overall, the random ending point sessions were broadly consistent with the levels of over-emission and efficiency loss by treatment observed earlier. Again Tables 3.8 and 3.9 summarize the results. Although we see minor differences across the random and fixed ending treatments, our basic results still hold. Under seller liability, an uncertain ending point had an almost identical outcome to that observed when end points were known. Under buyer liability, emissions increased and market efficiency fell, as reported in Result 1. Buyer liability also lowers permit prices relative to seller liability, supporting Result 2. Output patterns reflect earlier behavior – sellers produced the highest proportion of total output. Total output level observed, however, is approximately three units less than previous sessions and total seller output levels are as much as two units less on average. The patterns of permit trade and output decreases observed are broadly consistent with the hypothesis that the inclusion of a random ending point in buyer liability rounds reduced the incidence of “cheating”, at least in later periods of a round; lowering the production levels in the market. Table 3.9 shows the overproduction: total sales ratios observed over all periods reinforce this conclusion, as they are 20 percent lower than those observed in the second round of previous buyer liability sessions.

Conclusions
Successful global climate protection policy depends in part on the cost effectiveness of greenhouse gas emission trading. We present evidence that emission trading under a buyer liability rule could lead to less climate protection at greater costs given relatively weak international enforcement for environmental shirking. Promoting a caveat emptor liability rule backfired in our experiment on both economic and environmental criteria. Holding the subjects that represented highemission buyer nations responsible for climate shirking rather holding the

74 R. Godby and J.F. Shogren relatively poorer low-emission seller nation subjects responsible resulted in average emission levels exceeding those observed under seller liability by nearly 34 to 40 percent. The imposition of an escrow-like refund system did not alter this result; and neither did the introduction of tighter enforcement or conditions that could create stronger seller reputations. Our findings support the notion that buyer liability in global emission trading might lead to less climate protection at greater cost.

Acknowledgments
We thank the College of Business and the Stroock Professorship for financial support. Thanks to Tim Cason, Kate Carson, Chuck Mason and participants at University of Victoria, the Economic Science Association Meeting in Tucson, AZ, and American Economic Association Meetings in Atlanta, GA, for helpful comments. This chapter is dedicated to Peter Bohm.

Notes
1 Specific emission targets for developing countries were not set in the protocol. 2 Also see Cason’s (1995) test bed experiments on efficiency of the acid deposition auctions run by the US Environmental Protection Agency. 3 In policy documents, proponents of buyer liability often suggest reputation effects will result in market outcomes equivalent to or approaching those expected under seller liability. In general, however, multiple equilibria exist in games with reputation, and which of these are reasonable outcomes to expect will be critically dependent on the characteristics of the market setting and the prior distribution of player types assumed. Results of reputation games in a number of settings are reviewed in Fudenberg and Tirole (1991). In the predictions presented here, since each round consisted of eight periods, backward induction suggests that in buyer liability rounds, buyers should expect sellers to overproduce since permits have no other use value to sellers. Using this assumption and the conditions described in Table 3.2, we find the values reported in Table 3.3. We do not attempt to develop alternative reputation–effect predictions that might be possible in this environment. Instead, we suggest that outcomes varying from the risk neutral predictions presented could indicate shortcomings in the method of computing predictions we use. 4 This test is performed on mean outcomes by round, but is inappropriate if one expects that the independence assumption required is violated across rounds. 5 Experimental predictions are computed assuming equilibrium market outcomes, but early periods appear to have convergence processes at work thus resultant mean values across periods vary in the predicted direction by treatment, if not by the predicted magnitude. This occurs in the summary data (see Table 3.3), and is present to a lesser degree when permit price outcomes are averaged over the last three periods of each round to allow for convergence (see Table 3.6). Treatment outcomes across rounds differ in the directions predicted, and Table 3.6 shows mean session outcomes are closest to the values predicted. 6 Since experimental sessions used separate subject groups, a random-effects model is employed in which the fixed and constant subject or session effect on price observations is described by ui , while eit is a random disturbance term. 7 There are no significant interaction effects of order of presentation and liability treatment as estimated by coefficients b5–b8. The decreased significance of b8 across the last five periods relative to all periods where it is significant at the 10 percent level

Caveat emptor Kyoto 75
may suggest that initially refunds may cause subjects to increase their permit valuations in the first round; however experience appears to result in closing price outcomes that are not significantly different across rounds in this treatment. 8 Volume differences are consistent with subject-pool differences in risk aversion, which could explain the lower than predicted average trading volumes observed under buyer liability treatments in some sessions. It could suggest the assumption of risk neutrality to generate the experimental predictions may be an excessive abstraction in this environment. 9 Overproduction is the difference between a firm’s output and its permits. For the overproduction-to-total sales ratios, a positive value means sellers overproduced relative to predictions; a negative value means they underproduced; a value over unity says sellers did not back permits sold with an output reduction. 10 We used an Excel spreadsheet to generate random numbers between one and 100.

References
Baron, R., 1999. An Assessment of Liability Rules for International GHG Trading. International Energy Agency Information Paper, October. Barrett, S. 2003. Creating Incentives for Cooperation: Strategic Choices. Providing Global Public Goods. I. Kaul, P. Concei, K. Goulven and R. Mendoza, eds. New York: Oxford University Press, 308–328. Beck, N. and Katz, J., 1995. What To Do (and Not To Do) with Time-Series CrossSection Data, American Political Science Review 89: 634–647. Bohm, P. and Carlen, B., 1999. Emission Quota Trade Among the Few: Laboratory Evidence of Joint Implementation Among Committed Countries, Resource and Energy Economics 21: 43–66. Carlen, B., 1999. Large-Country Effects in International Emissions Trading, Department of Economics, Stockholm University, mimeo. Cason, T., 1995. An Experimental Investigation of the Seller Incentives in EPA’s Emission Trading Auction, American Economic Review 85: 905–922. Cason, T., 2003. Buyer Liability and Voluntary Inspections in International Greenhouse Gas Emissions Trading: A Laboratory Study, Environmental and Resource Economics 25: 101–127. Chayes, A. and Chayes, A., 1998. The New Sovereignty: Compliance with International Regulatory Agreements. Cambridge, MA: Harvard University Press. Cooper, R., 1998. Toward a Real Global Warming Treaty, Foreign Affairs, March/April: 66–79. Crocker, T., 1966. The Structuring of Atmospheric Pollution Control Systems. The Economics of Air Pollution. H. Wolozin, ed. New York: Norton, 61–86. Council of Economic Advisers (CEA) 1998. The Economics of the Kyoto Protocol, J. Yellen’s Statement before the Committee on Agriculture, Nutrition, and Forestry, US Senate, 5 March. Dales, J.H., 1968. Pollution, Property and Price. Toronto: Toronto University Press. Edmonds, J., Scoot, M., Roop, J. and MacCracken, C., 1999. International Emissions Trading and Global Climate Change: Impacts on the Costs of Greenhouse Gas Mitigation. Report prepared for the Pew Center on Global Climate Change, Washington, DC. Environmental Defense Fund, 1998. Cooperative Mechanisms Under The Kyoto Protocol: The Path Forward. Online, available at: environmental defense.org/ documents/747_PathForward.pdf (accessed September 2007).

76 R. Godby and J.F. Shogren
European Business Council for a Sustainable Energy Future, 2000. Position for the COP 6, November 11, Online, available at: e5.org/pages/st-08e.htm (accessed March 2001). Fudenberg, D. and Tirole, J., 1991. Game Theory. Cambridge, MA: MIT Press. Haites, E. and Missfeldt, F., 2001. Liability Rules for International Trading of Greenhouse Gas Emissions Quotas, Climate Policy 1: 85–108. Intergovernmental Panel for Climate Change (IPCC) 2001. Climate Change 2001: Synthesis Report. A Contribution of Working Groups I, II, and III to the Third Assessment of the Intergovernmental Panel on Climate Change. R. Watson, eds. Cambridge and New York: Cambridge University Press. Jacoby, H., Prinn, R. and Schmalensee, R., 1998. Kyoto’s Unfinished Business, Foreign Affairs July/August: 54–66. Kopp, R. and Toman, M., 1998. International Emissions Trading: A Primer. Resources for the Future. Online, available at: weathervane.rff.org/features/feature049.html Muller, R.A. and Mestelman, S., 1998. What Have We Learned from Emissions Trading Experiments? Managerial and Decision Economics 19: 225–238. Nordhaus, W. and Boyer, J., 2000. Warming the World: Economic Models of Global Warming. Cambridge, MA: MIT Press. Ontario Government Ministry of the Environment, 2001. Emissions Reduction Trading System for Ontario: A Discussion Paper. Toronto, March. PERT, 2000. Clean Air Mechanisms and the PERT Project: A Five Year Report. Pilot Emissions Reduction Trading, Toronto, June. Plott, C., 1983. Externalities and Corrective Policies in Experimental Markets, Economic Journal 93: 106–127. Rabkin, J., 1999. Why Sovereignty Matters. Washington, DC: AEI Press. Schelling, T., 1997. The Costs of Combating Global Warming, Foreign Affairs November/December: 8–14 Schmalensee, R., Joskow, P., Ellerman, D., Montero, J. and Bailey, E., 1998. An Interim Evaluation of Sulfur Dioxides Emission Trading, Journal of Economic Perspectives 12: 53–68. Shogren, J., 2004. Kyoto Protocol: Past, Present, and Future, GeoHorizons 88: 1221–1226. Solomon, B., 1999. New Directions in Emissions Trading: The Potential Contribution of New Institutional Economics, Ecological Economics 30: 371–387. Stavins, R., 1998. What Can We Learn From the Grand Policy Experiment? Lessons from SO2 Allowance Trading, Journal of Economic Perspectives 12: 69–88. Soberg, M., 2000. Price Expectations and International Quota Trading: an Experimental Evaluation, Environmental and Resource Economics 17: 259–277. UNFCCC (United Nations Framework Convention on Climate Change), 1999. The Kyoto Protocol to the Convention on Climate Change. UNEP/IUC/99/10. France: published by the Climate Change Secretariat with the support of UNEP’s Information Unit for Conventions (IUC). Victor, D., 2001. The Collapse of the Kyoto Protocol and the Struggle to Slow Global Warming. Princeton: Princeton University Press. Woerdman, E., 2001. Limiting Emissions Trading: The EU Proposal on Supplementarity. Department of Economics and Public Finance ECOF, Faculty of Law, University of Groningen, Groningen (the Netherlands), ECOF Research Memorandum No. 29, p. 63.

4

A test bed experiment for water and salinity rights trading in irrigation regions of the Murray Darling Basin, Australia
Charlotte Duke, Lata Gangadharan, and Timothy N. Cason

Introduction
The Murray Darling Basin covers more than 1,000,000 square kilometers, or one-seventh of the Australian continent, and is the catchment for the Murray and Darling rivers. The river system extends from Queensland, and includes threequarters of New South Wales and half of Victoria, through South Australia and drains out of a single exit of Lake Alexandriana to the sea. The Basin supports almost three-quarters of Australia’s irrigated agricultural production, an important export for Australia. While naturally saline, human activity has significantly changed the Basin’s natural water balance systems. The removal of (deeprooted) native vegetation and the replacement with (shallow rooted) European style annual crops and pastures has resulted in excess rain and irrigation water entering the groundwater systems. As groundwaters rise, the gradient between these saline groundwater mounds and the rivers decrease. In some regions groundwater salinity can reach 50,000 EC, i.e. electroconductivity, which is a measure of salt concentration: 600 EC is thought to be suitable for potable water.1 Saline water contained in the soils then percolates into the rivers, increasing water salinity. While estimates vary, costs reported by the Murray Darling Basin Commission place annual costs due to human-induced salt impacts at $130 million in agricultural costs, $100 million in infrastructure and $40 million in environmental costs across the Basin.2 In response to increasing salinity costs, and informed by improved scientific information on salinity impacts, in November 2000 the Commonwealth of Australian Governments, agreed to the National Action Plan for Salinity and Water Quality (the NAP).3 This plan is the Australian government’s vehicle to deliver on-ground change, across this complex Basin, which includes the coordination of institutions and resources across four states and one territory. The government agreements supporting the NAP specifically include the facilitation of new market-based incentives, including trading mechanisms

2001). In the Murray. Figure 4.78 C. This chapter reports a part of one of these pilots. there is a Salinity Target at the town of Morgan in South Australia. experiments have been used in the design of the Victorian Bush Tender Program. funded under the NAP.7 These zones allow regulators to attribute salt concentrations to individual irrigators based on their zone and monitored water use. We focus on the design of experimental simultaneous double auction markets for water and salt.4 This program allocated $5 million across 11 pilots to test MBIs. farm extension staff and farmers in this region are innovative and they recognize the need to carry some costs for salinity externalities in private production costs. and how can private salinity management efforts be recognized in policy? Tradable emission permits use market incentives to allocate pollution control responsibility at least cost between regulated firms.8 This means a diffuse source of pollution can be managed as a point source. a discriminative sealed bid offer auction to secure land management change on private land for environmental outcomes (Cason et al. there exists mature scientific knowledge about the non-standard cause and effect between irrigation practices and salinity concentrations in the river for this region.6 Second. tradable permits do not require the regulator to estimate the demand and supply response of firms to different prices for pollution. the regional natural resource management plan had zoned the irrigation regions into salinity impact zones. the Market Based Instruments (MBI) Pilots Program was announced as a process. Duke et al. and perhaps most importantly when attempting to influence policy direction. Legislation requires the salinity concentration level at Morgan to remain under the Target of 800 EC for 95 percent of the time (MDBMC. We selected this region for a number of important reasons. We use the Sunraysia region in northern Victoria as the pilot region. 2001a and 2001b).1 shows the region and salinity impact zones.. The salinity impact zoning in Sunraysia allows regulators to attribute irrigation impacts to EC at the Target. and in one treatment we allow subjects to make a technology choice that changes parameters in both markets. regional government officers. Experimental economics is increasingly being used to provide solutions to complex real world allocation problems. to irrigators based on their zone and water use. 2003). for building Australia’s capacity to use MBIs for natural resource management.5 The objective of this chapter is to report preliminary economic experiments that pilot test bed a tradable permit market for irrigation induced salinity in the Murray. Within natural resource management. Compared to price-based MBIs such as taxes.9 Third. Tradable permits differ from the more traditional command and control mechanisms because they equalize marginal control costs between sources. (Commonwealth Government of Australia. First. In April 2003. Knowing the position and the elasticity of aggregate marginal control costs is important when there is some safe minimum standard above which pollution cannot rise.10 . The two questions raised most frequently in our (two) stakeholder workshops held in Mildura (the largest regional centre in the Sunraysia) during 2003 and 2004 were: how can salinity be cost effectively managed.

Second. Further. resource intensive process with uncertain outcomes.1 The Sunraysia irrigation districts and salinity impact zones. Creating legal property rights in a complex field environment before we understand how the markets work in a simple controlled environment is potentially very costly. The experiments allow us to study how the two markets transfer this information. Experimental economics allows us to test the main design features of a tradable permit market for salinity in the Murray at low financial and political cost. We chose to use experiments to test bed a tradable permit market for salinity in the Murray for two reasons. 2002. and are particularly useful in the area of environmental economics for communicating how complex incentive systems operate to multidisciplinary groups (Roth. Experiments were also used to evaluate the United States EPA allocation auction for the Sulfur Dioxide Trading Program (Cason and Plott. 2003. 1998. we are applying a point-source policy to a non-point-source problem. without having to wait for more complete information to become available. and without changing legislation. The methodology is also proving . We are relying on the accuracy of the hydrological models used to design the salinity impact zones. First. and alternative proposals for tradable permit markets for air pollution control (see Muller and Mestelman. for a survey of the experimental literature on tradable emission permits). and this can be a time-consuming. and the resulting change in salt and water allocations. legislative change would be needed to pilot test the mechanism in the field.Water and salinity rights trading 79 Figure 4. and Bohm. 1996).11 Experiments use special cases to inform more complex field situations (Plott. Cummings et al. 1994). we currently have incomplete knowledge about private salt abatement options and their effectiveness over long time horizons. to allocate emission rights.. 2001).

The first reason is communication. In an oral auction. The time taken to run oral auctions limits the number of periods that can be implemented in a session. the value of irrigation water to a farmer will also depend on the cost of purchasing salt licenses in the salt market. For this reason. as was the case in this pilot test bed.14 We chose to calibrate the experiments to the field for several reasons. (2004) use experiments to investigate the operation of sequential markets in which subjects first trade in a sealed bid uniform price multiple unit market for emissions rights. is the system doing what it is designed to do? This means. We use a double auction institution as it most closely approximates the emerging water trading markets in the Basin. The parameters used in the experiment approximate the production conditions and environment characteristics of Sunraysia in Northern Victoria. When the experimental design is unclear. We pilot test simultaneous double auctions for water and salinity rights in the Murray River. are resources being allocated to least opportunity cost players. First.13 This is because if bidders must make bids for related goods in independent markets. the policy should generate the observed outcomes (allocations) for reasons we understand.15 Second but . and Plott (1983) uses sequential double auction markets to investigate the operation of a tradable permit market and a primary output market. quantities and efficiencies moving towards the predicted theoretical equilibrium given the environment? Second. The methodology allows us to ask the following questions as posed by Plott (1994). and these reasons should be supported by economic theory. as compared to computerized experiments. In an oral auction the experimenter can learn more about how the markets transfer private information. The value of each good to them depends on the final bundle of goods they are successful in winning (Milgrom. then they do not have all the information they need to make a profitable bid on the bundle. and are prices. is the system working for understandable reasons? In other words. Buckley et al. which act as inputs in the production process. and how transaction prices are formed. We chose to implement the markets simultaneously as water and salt are complements in production and therefore must be held as a bundle if the farmer wants to irrigate. we have implemented the markets simultaneously in this experiment. Duke et al.12 We want to observe if the institution can efficiently transfer private information between subjects as market complexity is increased across treatments.80 C. it is easier to explain the institution to stakeholders if it is designed using parameters with which they are familiar. In computerized experiments the experimenter is an observer. an oral auction allows mistakes or misunderstandings to be identified. In our pilot. and then sell output in the output market. recorded and (where possible) even corrected within the session. the experimenter is (more of) a participant in the experiment who controls the trading floor. very useful for communication with policy makers unfamiliar with economic decision making. Goodfellow and Plott (1990) use experiments to investigate simultaneous input and output markets organized as double auctions. Failure to account for interdependencies between goods can lead to costly policy design mistakes. The pilot experiments are implemented as oral auctions. 2004).

The pilot experimental results are then presented. (1999) investigate investment decisions in experiments designed to test features of the tradable emissions permit program under the US Clean Air Act. Subjects in our experiment can choose to change production technology. By using parameters calibrated to the field. the experiment employs parameters that correspond to field conditions in the Sunraysia region in Northern Victoria. where subjects must choose emissions rates for their output between the closing and opening of their sequential markets. Subjects in the experiment were allocated to one of two salinity impact zones. at a preannounced point in a session. We introduce endogenous technology change in one treatment to represent private abatement choice. Gangadharan et al. Our pilot test bed experiment reported in this chapter is.1. followed by a conclusion with a discussion and some design lessons for future treatments. a lack of parallelism with the real world is sometimes a criticism of experiments. it is possible to improve parallelism without introducing all the complexity of the field. Experimental design Environment and procedures The aim of this research is to investigate if the two markets organized as double auctions and implemented simultaneously can efficiently allocate water and salt units between subjects in our three treatments. Table 4. The first part of the chapter describes the experimental design. Ben-David et al.16 Third. They find that subjects often invest more than optimal. (2004) introduce technology choice in their experiments. such that in equilibrium only a subset of firms would choose to invest. The rest of this chapter is organized as follows.1 shows the main regional industries. we think. a high impact zone (HIZ) or a low impact zone (LIZ). For each unit of water used in the LIZ the external cost is $40. (2005) allow subjects to choose to invest in abatement technology at a cost. one of the first to test simultaneous double auctions with endogenous technology choice using variables that approximate field conditions. or to remain with their starting technology.Water and salinity rights trading 81 related. These external costs represent the cost to government of intercepting the salt .17 For each unit of water (megaliter) used in the HIZ the external cost is assumed to be $80 in this experiment. In these pilot experiments. The slope of these curves was discussed with regional government officers and crop specialists to ensure the parameters approximate field conditions.2 shows subjects’ private redemption and cost values used in the experiment. Table 4. The private values and costs were drawn randomly from uniform distributions with the ranges shown in Table 4. Buckley et al. Cost and value ranges remained private information throughout all sessions. their production technologies and value ranges for water. the joint relationship between salt and water was important in our choice of simultaneous markets as opposed to a sequential market design. As noted above. the institutional design can depend upon field specific characteristics. leading to cleaner production but reduced efficiency.

depending on their impact zone. This implies that the price of a permit in the salt market should be 20 experimental dollars. If the price .2 Private redemption and cost values Unit 1 Redemption value of units (buyers) Buyer 1 1167 Buyer 2 1131 Buyer 3 1142 Buyer 4 1109 Cost of units (sellers) Seller 21 Seller 22 Seller 23 Seller 24 908 858 900 860 Unit 2 1077 1031 1062 1049 968 948 1000 940 Unit 3 987 931 982 989 1038 1038 1100 1020 Unit 4 897 831 902 929 1088 1128 1200 1100 using engineering solutions.2). Production technology for the industries was provided by SunRise 21.82 C.1 Parameter ranges from Sunraysia Industry Wine grapes Irrigation technology Drip irrigation Overhead sprinkler Pressurized mix Flood (2 subjects) Overhead sprinkler Flood Overhead sprinkler Value ranges for water (per megaliter = 1000 liters)* 1100–1500 900–1300 700–1100 400–800 1000–1400 600–1000 900–1300 Orchard and citrus fruit Dried grapes Note * Value ranges were provided by the NSW and Victorian Department of Primary Industries. Duke et al. and therefore Buyer 1 must purchase four salt permits for each unit of water he buys in the water market. The transfer coefficient is therefore two between irrigators located in the HIZ and the LIZ for salt concentration impact at the Target: irrigators located in the HIZ must hold twice the number of salt licenses for each unit of water as compared to irrigators located in the LIZ. and discussed with regional government officers and field extension staff. We implement this in the laboratory by requiring subjects in the HIZ to hold four salt licenses for each unit of water. and in the LIZ they are required to hold two licenses for each unit of water. 2004. Buyer 1’s private value for his first unit of water is 1167 and the value for his second unit is 1077 (see Table 4. Table 4. For example. The salinity permit internalizes the salt externality. Table 4. when they must pay for salt permits. Buyer 1 is located in the HIZ. Buyers’ private values for water will decrease by 80 or 40 experimental dollars. and approximate the field parameters for two salinity impact zones in the region (MDBC 2003).

For example. The technology change is assumed to reduce salt impact for each unit of water used in production. if Buyer 1 chooses to change technology then his private values for water will increase by $100. Buyer 1 will also incur a technology cost equal to $15 and this will decrease his marginal values by 15. Sellers’ opportunity costs for water also decrease by $40 or $80 as sellers can sell salt permits if they sell some water rights. In this pilot experiment this technology investment changes all values and costs by $100 for each unit of water. then the opportunity cost for her first unit of water is 860. and the equilibrium price and quantity for water will decrease as shown in Figure 4. The introduction of the salt market therefore shifts the water market demand and supply curves down. 1000–1020. Subjects also incur a cost for this technology change. technology 1. . equal to $15 per unit of water. ie. In one of our treatments (Treatment 3) subjects have the choice to change their production technology. technology 1 Treatment 3.Water and salinity rights trading 83 for a salt permit is $20. For example. technology 2 Treatment 1 Treatment 2 and 3.18 If Seller 23 sells her first unit of water then she can also sell two units of salt. i. then Buyer 1’s demand curve for water will shift down by $80 for each unit of water (1167 − (20)*4 = 1087 and 1077 − (20)*4 = 997).e. Seller 23 is located in the LIZ. i. 7 units Treatment 1. 1000 − (20)*2.e. technology 2. 900 − (20)*2 and the opportunity cost for her second unit is 960.2. Seller 23’s private cost for her first unit of water is 900 and the cost for her second unit is 1000 (see Table 4. 11–12 units 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Units of water Treatment 1 Treatment 2 and 3.2 Market demand and supply for water. 1300 Treatment 2 and 3. technology 2 Figure 4. technology 1 Treatment 3. 900–997. 963–1032. 8–9units 1200 Price of water 1100 1000 900 800 700 Treatment 3. If buyers choose to change technology then their license requirements are halved for each unit of water purchased. If the price of salt permits is $20. and therefore holds two salt permits for each unit of water.2).19 If subjects choose to change technology then they receive a (private) benefit and their value (cost) for water units increase (decrease).

i. buyers can either make a price and quantity buy offer or accept a seller’s sell offer. Buyers are better off with the new technology because their (private) marginal values increase and the number of permits they require for each unit of water purchased is reduced. The design of the salt license market is still in development. i. The effect of using this fine to define the opportunity cost of a salt permit. Seller 23 also has fewer salt licenses to sell. Duke et al. and is equal to $20 in this pilot test bed. This means Buyer 1 must now purchase two as compared to four salt permits for each unit of water. Buyer opportunity cost for each license is therefore $20. Seller 23 had two licenses that could be sold in the salt market for each unit of water she sold in the water market. We have chosen not to use standard seller costs and buyer values for salt licenses. The water market operates as a standard multiple unit double auction. and this is equal to the external cost of another salt concentration unit at the Target. If Seller 23 chooses to change technology then he receives a $100 benefit and incurs a $15 cost. At any time while the trading period is open. and therefore sellers should learn that buyers’ maximum willingness to pay for each license is $20. Sellers are better off with the new technology because the net change in their (private) marginal costs for water (equal to 85 experimental dollars) is greater than the reduction in the number of licenses they can sell in the salt market multiplied by the price at which they can sell licenses in the salt market (equal to $20 to $40 on each unit of water). Seller 23’s private cost for his first unit of water is 900. Buyers should be willing to buy a license from the salt market if license price is less than $20. and sellers’ earnings on a unit are equal to the price at which they sell water minus the cost of water. is that the efficient market price of salt is not determined by private marginal salt abatement costs but . With her new technology Seller 23 has only one license to sell for each unit of water.e.84 C. Buyers’ earnings on a unit of water are equal to the redemption value for water minus the price paid for water. If sellers choose to change technology then the number of licenses available to them to sell in the license market is also halved. and sellers can either make a price and quantity sell offer or accept a buyer’s buy offer. The fine is equivalent to the buyer buying a salt license from the regulator as described above. In these pilot experiments we want to focus on learning how to implement the simultaneous oral markets and to test if they work as theory predicts. Each buyer and seller has up to four units of water they can buy or sell in a trading period.e. We instead introduce a uniform cost that buyers must pay for each required license they fail to buy. 1167 + 100 − 15 − (20)*2 = 1212. this means she loses $20 in potential license sales on each unit of water. So all traders should adopt the new technology given the parameters we have implemented. higher buy offers or lower sell offers) are recognized by the market.e. We call this cost a fine. For example. The technology change is salt saving and therefore Buyer 1 must now hold only half the number of salt permits per unit of water as compared to his starting technology. The fine is public information. 900 − 100 + 15 = 815. as opposed to standard demand and supply curves for salt usage. With her starting technology. Only offers that improve the proposed terms of trade (i. 1167 + 100 − 15 = 1252.

20 Subjects participated in either one or two (simultaneous) oral double auction(s). All treatment 1 sessions included ten trading periods of 5 minutes each. depending on the treatment. Future experiments can also employ private values for water that change by non-uniform amounts due to adoption of the technology. presents the market demand and supply for water in each treatment. in which six periods were run. sessions included eight trading periods of 8 minutes each. Sessions lasted between 1. Subjects recorded their transaction prices on these excel sheets.21 In treatments 2 and 3. we do not use the words water and salt in the experiment in order to avoid uncontrollable bias due to private opinions about salinity and water. Subjects were read aloud the instructions reproduced in the appendix. randomly assigned as four buyers and four sellers. which influences parameters in both markets. We report two treatment 1 sessions in which buyers and sellers trade units of water in a multiple unit oral double auction.5 and 2 hours. Experimental subjects were undergraduate students from the University of Melbourne. as explained above. and their license requirements per unit of water were displayed on excel sheets for each subject.2. Table 4. at a pre-announced point in the experiment. and this increases the cost of water to buyers and decreases market demand as compared to treatment 1. Subjects received their earnings in cash at the end of the session. In treatment 2 we introduce the salt license market. which is more realistic for the field. Table 4. Treatments In this chapter we report six pilot experimental sessions.3 summarizes the experimental design. In treatment 3 we report two sessions in which buyers and sellers can choose to change production technology. Subjects’ (private) values and costs for water. session 2. Eight subjects participated in each session. and Figure 4. In treatment 3 we allow subjects to choose if they want to change production technology at the end .Water and salinity rights trading 85 is instead determined exogenously and is set equal to the government salt interception cost. and the computer automatically calculated profits and license sales and purchases for the subject (example excel sheets are reproduced in the appendix.4 presents the predicted equilibrium outcomes for the three treatments. which are high profile environmental issues in Australia. We report two sessions for treatment 2 where buyers and sellers also traded licenses for salt in a multiple unit double auction implemented simultaneously with the water market. The equilibrium quantity of water traded should decrease in treatment 2 by one to two units and salt impact should decrease by four to six licenses (or EC units) as compared to treatment 1. except for treatment 2.) Buyers and sellers traded units and licenses. In future experiments we plan to include individual demand and supply curves for salt based on private abatement. and earnings ranged between A$20 and $30. All offers to buy and sell and transaction prices were recorded on a white board at the front of the room. The salt market only allocates the salt rights and should determine the efficient quantity for each farmer given this common salt interception cost and private values and costs for water.

The expected EC units reported in this table for treatment 1. periods 1 to 4. and salt is reduced by four licenses (EC units) as compared to technology 1. there is no salt market. In periods 5 to 8. the net change in subjects’ marginal values and costs is 85 experimental dollars plus the avoided cost for buyers and the lost revenue for sellers in the salt license market. The new expected equilibrium price range for water is 963–1032. Given the costs and values used in these pilot sessions all subjects should choose to change. and 8 trading periods session 2. Duke et al. of trading period 4 and before trading period 5 begins. is the magnitude of the salt externality (equivalent to the number of licenses traded in treatments 2 and 3) when there is no policy to manage salt operating.3 Experimental design Treatment Treatment 1 – Water market (called the primary market in the experiment instructions) Treatment 2 – Water market plus salt market (called the license market in experiment instructions) Treatment 3 – private salinity abatement Features Multiple unit oral double auction Simultaneous multiple unit double auctions Sessions 2 inexperienced* 10 trading periods in each session 2 inexperienced 6 trading periods session 1.86 C. Table 4. but they had not participated in an experiment with this environment and institution. ** In treatment 3. Technology 2 reduces salt impact for each unit of water used in production. from 11–12 units as compared to seven. Table 4.4 Model equilibrium predictions Treatment 1 Transaction price for water Quantity of water traded Transaction price for salt licenses Quantity of salt licenses traded 1000–1020 8–9 units 24–26 EC units* Treatment 2 900–997 7 units 20 20 licenses Treatment 3** 900–997 963–1032 7 units 11–12 units 20 20 licenses 16 licenses Notes * In treatment 1. so that more water can be traded efficiently. . 2 inexperienced 8 trading periods in each session Simultaneous multiple unit double auctions with endogenous technology change between trading periods 4 and 5 Note * Subjects may have participated in previous experiments. the first values presented in this table refer to the expected equilibrium outcome with technology 1. the expected market outcomes are the same as those for treatment 2. the second values refer to technology 2.22 In treatment 3.

When technology change occurs. average transaction price falls by 6 experimental dollars. The variance in prices (not shown) appears to decline in this session as periods increase. In session 2. In session 2 transaction price adjusts from below. We expect the price of water to increase when new technology is adopted because irrigators receive a productivity improvement and the cost of salt licenses decreases. Average transaction price continues to fall in period 6 before beginning to adjust upward in periods 7 and 8. session 1 Predicted lower eq. In treatment 3 (Figure 4. treatment 1.3. In treatment 1 (Figure 4.5 present average period transaction price in the water market for the three treatments. session 2 Figure 4. treatment 1 Predicted upper eq. both sessions exhibit convergence towards the lower end point of the predicted equilibrium price range. 4. Both sessions appear to be converging as variance in price between periods decreases and price is moving closer to the midpoint of the equilibrium price interval. In session 1 of treatment 3 average transaction price is within the predicted equilibrium range in all periods. After 1100 1050 Price (per megaliter) 1000 950 900 850 800 1 2 3 4 5 6 7 8 9 10 Trading period Treatment 1. .3). In session 1.Water and salinity rights trading 87 Results Figures 4. In session 2 transaction price for this treatment is more erratic in the early periods but then stabilizes near the midpoint of the predicted range between periods 5 and 8. subjects can make a choice to change production technology before period 5.5). and is within the treatment 2 predicted range in all periods.3 Transaction price in the water market. transaction price for water adjusts from above to within the treatment prediction for periods 1 to 4. This technology change is salt saving and therefore reduces the cost of salt to the irrigator. average transaction price is lower compared to treatment 1.4) when tradable salt licenses are introduced we expect the price of water to be lower since water buyers must now pay for salt licenses.4 and 4. and in period 10 of this session average transaction price is 992. price treatment 1 Treatment 1. In treatment 2 (Figure 4.

session 2 Figure 4. 1100 1050 Price (per megaliter) 1000 950 900 850 800 1 2 3 4 5 6 7 8 Trading period Treatment 3.8 present average transaction quantity per period in the water market for the three treatments.88 C. session 1 Predicted lower eq.6. This is consistent with observed transaction price for this session in . Duke et al. treatment 3.5 Transaction price in the water market. treatment 2 Predicted upper eq. price treatment 3 Treatment 3. In treatment 1. treatment 3 Predicted upper eq. treatment 2. although it remains too low in session 2. 1100 1050 Price (per megaliter) 1000 950 900 850 800 1 2 3 4 5 6 7 8 Trading period Treatment 2. the technology change transaction price continues to fall before adjusting towards the lower end point of the new prediction in periods 6 through 8. session 2 Figure 4.7 and 4. Figures 4. 4. price treatment 2 Treatment 2. session 1 Predicted lower eq.4 Transaction price in the water market. average transaction quantity adjusts to the treatment prediction from below.

shown in Figure 4.7. transaction quantity in session 1 is equal to the treatment prediction in periods 3 to 8. Figure 4. In both sessions.3.Water and salinity rights trading 89 Quanitity of water traded (megaliters) 12 10 8 6 4 2 0 1 2 3 4 5 6 7 8 9 10 Trading period Treatment 1. session 2 Predicted eq. when salt licenses are introduced.7 Transaction quantity in the water market. qty treatment 2 Figure 4. session 1 Figure 4. session 1 Treatment 2. and quickly . Both price and quantity are below the equilibrium prediction. session 2 Predicted eq. treatment 2.6 Transaction quantity in the water market. 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 Trading period 6 7 8 Quantity of water traded (megaliters) Treatment 2. in this session subjects are failing to realize all gains from trade in the water market. treatment 1. as predicted the transaction quantity is lower than in treatment 1. qty treatment 1 Treatment 1. the quantity of water traded should fall. In treatment 2. In treatment 3.

although in session 2 transaction price and quantity are too low but exhibit an upward trend towards their predicted equilibrium. prices and quantities in the water market do adjust towards the predictions for the treatments. Treatment 3.6.4 and 4. qty treatment 3 Figure 4. session 2 Predicted eq. some of our subjects in treatment 2 apparently did not understand how the salt market operated. but both sessions show signs of convergence to the treatment predictions. Duke et al.12 present market outcomes for the salt permit market. adjusts when technology change occurs between periods 4 and 5.5 and 4. treatment 3. is particularly interesting as the market is clearly responding to the technology change. Figures 4.9 shows that the average period transaction price in the salt market for treatment 2 differs substantially from the equilibrium price of 20 experimental . and transaction quantity increases in period 5. information about the new (private) costs and values is quickly reflected in the market. Figure 4. session 1 Treatment 3.90 C. The poorer performance of treatment 2 is probably influenced by some initial confusion and learning by the subjects. Then in periods 6 to 8 prices begin to rise towards the new equilibrium. 12 Quantity traded (megaliters) 10 8 6 4 2 0 1 2 3 4 5 6 7 8 Trading period Treatment 3. In treatment 1. shown in Figures 4.8. price and quantity in session 1 stabilize at their predicted equilibrium early in the session. shown in Figures 4. Figures 4. Subjects first realize they can profitably trade more units of water.7.3 to 4. average transaction price and transaction quantity are initially low.8 indicate that in most of our sessions.8 Transaction quantity in the water market. In treatment 2. perhaps because subjects are unsure about the impact of the change on market equilibrium. shown in Figures 4.9 to 4. As we discuss later when we explore prices in the salt market. Transaction quantity in session 2 takes longer to respond to the new technology. Although in period 5 average prices in both sessions fall.3 and 4.

treatment 2. In session 1.10). dollars. In session 2. session 1 Treatment 2. . and rapidly fall across periods to reach $39 in period 8. permit prices are very low.9 Transaction price in the salt market. session 1 Treatment 3. 25 Price of salt license (1 license = 1/1000 EC unit) 20 15 10 5 0 1 2 3 4 5 6 7 8 Trading period Treatment 3.Water and salinity rights trading 91 80 Price of salt license (1 license = 1/1000 EC unit) 70 60 50 40 30 20 10 0 1 2 3 4 5 6 7 8 Trading period Treatment 2.10 Transaction price in the salt market. session 2 Predicted eq. treatment 3. In Treatment 3 (Figure 4. transaction price for salt adjusts from below to within a few experimental dollars of the treatment prediction. much lower than the prediction. price salt permits Figure 4. price salt permits Figure 4. session 2 Predicted eq. salt prices begin very high (equal to 204 experimental dollars).

qty treatment 2 Figure 4. We . session 1 Treatment 2. session 2 Predicted eq. Duke et al.12 Transaction quantity in the salt market. qty treatment 3 Figure 4. In these pilot experiments the uniform equilibrium price for salt permits is determined by the $20 fine that buyers incur for each salt permit they failed to buy in a period. treatment 3.92 C. 25 Quantity of salt licenses traded (1 license = 1/1000 EC unit) 20 15 10 5 0 1 2 3 4 5 6 7 8 Trading period Treatment 2. treatment 2. 25 Quantity of salt licenses traded (1 license = 1/1000 EC unit) 20 15 10 5 0 1 2 3 4 5 6 7 8 Trading period Treatment 3. session 2 Predicted eq. In treatment 2 we believe some subjects did not understand that the opportunity cost of salt permits was $20 and this is why transaction price for salt permits is either too high (session 1) or too low (session 2). session 1 Treatment 3.11 Transaction quantity in the salt market.

and they instead used the unintended suggestions given by the experimenter in the instructions in forming beliefs about the reasonable price of salt permits. The observations used in these regressions are transaction prices averaged by period across sessions. Figures 4. we used an example for the salt market with transaction prices between $1 and $4. D2. The equation we estimate is the following: Pit = constant + β11/t + β2D2 + β3D3Tech1 + β4D3Tech2 + eit where P indicates the average transaction price. some subjects chose to incur the $20 fine for salt permits and to trade only in the water market in later periods when permit prices are near 20. in particular. While these graphical impressions provide useful informal information about the operation of the markets. (These transaction costs could include the mental effort and the need to trade and focus on two markets. Figure 4. In session 2. In treatment 2. Transaction quantity then shifts down in periods 5 and 6. the session represents the random effect to account for the correlation of market outcomes within a session. In treatment 3 transaction quantity for salt increases to the technology one prediction of 20 permits. and in the instructions for treatment 2. The inverse specification 1/t allows for the impact of time on prices to be non-linear. If they had not then we clearly told them they must therefore pay a cost of $20 for each license they required but did not buy in the market.11 and 4. transaction quantity for salt in both sessions adjusts from below towards the treatment prediction of 20 permits. In the random effects models. as predicted. The coefficient β1 captures the impact of time on average prices. We also checked buyers’ record sheets at the end of the first few trading periods in treatment 3 to determine if they had bought enough permits. The observed prices suggest that some subjects did not understand the relationship between the salt permits and the enforcement fine. we had adjusted the instructions to more clearly illustrate that 20 is the opportunity cost to buyers. i denotes the session and t represents time as measured by the number of market periods in the treatment.5 presents random effects regressions for average price in the water market. D3Tech1 and D3Tech2 are dummy variables for treatments 2 and 3 respectively. session 1. but it appears to shift down too much. with larger price changes in early periods. (D2 takes the value of 1 when the treatment . Table 4.Water and salinity rights trading 93 used random transaction prices for salt in the instruction examples.) In treatment 3. We think that subjects’ understanding of how the permit market operates in treatment 3 was significantly better than in treatment 2. we used an example with transaction prices between $60 and $80.11. In Treatment 3. we also present some simple regressions to explore the statistical significance of treatment differences in these preliminary sessions. some buyers choose to pay the fine and to avoid the subjective transaction costs of trading in the salt market.12 present transaction quantity in the salt market for treatments 2 and 3. We conjecture that when permit prices are within 1 to 2 experimental dollars of 20.

The results indicate that transaction prices are significantly lower in treatment 2 as compared to treatment 1. The β1 coefficient on 1/period is not significantly different from zero. We do not present regression results for average price in the salt market for treatments 2 and 3. of observations Wald Chi-squared Significance level R-squared (overall) 987.24 (18. This is consistent with the equilibrium prediction: in treatment 1 there is no salinity policy and therefore private values and costs. relative to the baseline treatment 1. D3Tech1 takes the value of 1 for data in periods 1–4 of treatment 3 and D3Tech2 takes the value of 1 for data in periods 5–8 of treatment 3.) The coefficients on the dummy variables D2 and D3.85 0.5 Random effects estimation models for average transaction price in the water market Random effects model Constant 1/period (t) Treatment 2 Treatment 3. Hence the coefficient for the treatment dummy could be capturing the noise in the data and not providing any important information.45 Notes The first number in brackets is the standard error.49) –28.08) –15. .61* (12. periods 1–4 (tech 1) Treatment 3.25) (–1.66) (–2. periods 5–8 (tech 2) No.5 assume that the price of salt licenses does not affect water prices and vice versa.45 (13.55* (17. capture the impact of treatment 2 and the impact of the change in production technology in treatment 3 respectively. Transaction prices in treatment 3 are not significantly different from treatment 1 for either production technology. * significant at 1 percent.9).002 0. as there is substantial variation between sessions in treatment 2 (see Figure 4. being considered is treatment 2 and takes the value of 0 otherwise. and the resulting market equilibrium.17) –46. do not include the external cost of salt.68) (0.52) (–1.54) 50 16.57 (18.64) 9. Models that account for the endogenous Table 4. The results presented in Table 4. indicating that no systematic trend in transaction prices exists across periods. the second number in brackets is the z-statistic.81) (77. Duke et al.94 C.

would release both salt and water permits for sale. Discussion and additional design lessons In this chapter we report preliminary results from experiments that test the implementation of simultaneous markets for water and salinity rights in the Murray River in Australia. Alternative technologies. Changing irrigation technology from. and we held a number of workshops in the region to determine accurately costs and values for water and salt. the kind of technology that is useful to manage the salinity problem can be different across treatments to reflect different technologies available in the field. we use parameters specific to the field. We find that subjects in the experiment respond to the technology change and prices and quantities of water and salt begin to move toward their new equilibrium predictions. In treatment 3 all subjects had an incentive to adopt the new technology. which operates simultaneously with a water permit market. Subjects in treatment 3 could be performing better in the last few periods because of learning effects or due to the impact of adopting new technology.Water and salinity rights trading 95 relationship between water and salt prices in these simultaneous markets would be useful in understanding the interconnectedness of these two markets. In addition. These pilot experiments form a part of the Commonwealth of Australian Governments’ efforts to facilitate market-based incentives for the environment across the Murray Darling Basin. This pilot experiment identifies some useful design lessons for additional sessions. groundwater pumping and improved drainage . Subjects take some time to learn in these markets and it would be useful to design sessions with more periods to enable us to examine their behavior over time. however. We hope to extend the analysis in this direction in future work that is based on additional experimental sessions. so that in equilibrium not all subjects should choose to change. Additional period observations could also help in separating out the impact of learning and technology adoption. such as replanting native vegetation. flood to drip. Given the difficulty in conducting more than eight periods in a multimarket oral double auction lasting no more than 2 hours in total. it is difficult to obtain identifying variables to estimate this endogenous relationship. These preliminary experiments illustrate how the introduction of a salt license market. In one treatment we introduce endogenous technology change to represent private salt abatement by irrigators. In future experimental treatments we plan to explore how many subjects adopt new technology when the incentives to adopt are asymmetric and different across participants. To examine this issue we could analyze the later periods’ data from treatments 2 and 3 to isolate the effect of technology adoption. Given our limited data. for example. As these pilots are designed ultimately to inform government policy. in future experiments we plan to computerize the trading institution to allow us to conduct more trading periods within a session. can reduce water price and quantity traded in the direction of the social optimum. Our pilot region is the Sunraysia located in Northern Victoria.

engineering solutions such as pumping and the purchase of water permits from the water market for environmental flows can also reduce salt concentrations. corrosion of gas.96 C. Subjects in this experiment represent irrigators along the Murray River. available at: mdbc. Duke et al.23 Government also invests to reduce salinity. This is particularly important for illustrating to government how environmental impacts from multiple sources can be managed by markets. The market price for a salt permit would be $20 if government is the lowest cost abater. however in future experiments we would examine the impact of different technologies on market and environmental performance. available at: mdbc.gov. 2 Costs to agriculture include lost production due to saline affected soil and the application of salty water to crops.gov.htm (accessed 30 August 2005). Most of the infrastructure costs fall on local government and residents in regional urban areas. destruction and damage to wetlands and the dependant breeding patterns of migratory birds.htm (accessed 30 August 2005). See surface and groundwater salinity. See the costs of land and water salinity. which corresponds to the fine level for a failure to purchase a salt permit. Environmental costs include impacts on aquatic and terrestrial biodiversity. amenity values and the costs of remedial actions to repair irrigation and drainage infrastructure and farm buildings. killing of trees. and would be less than $20 if private salt abatement is less expensive. The complexity of introducing heterogeneous technology choice in these oral pilots was significant and we wanted to learn how to run the markets before introducing additional complexity. would only release for sale salt licenses and not water permits. willing to supply salt permits at prices equal to its marginal control costs ($20 in the experiment). Private salt harvesters. such as Pyramid Salt. We plan to investigate increasing the number and types of participants in these markets to include these agents. grass and other vegetation. The price for salt permits in our experiment is determined by the government’s cost to build engineering works to reduce salt. reduction in land values.au/education/encyclopedia/naturalresources/env_ issues/water_and_land_salinity. Notes 1 Online. Irrigators would compete with government and would be willing to supply additional salt permits if the market price is greater than their marginal abatement cost. Representatives at our workshops were cautious about publicly revealing the currently available (rough) estimates of benefits for private abatement. . but other agents also impact upon salinity. Government could be a market participant.au/education/encyclopedia/naturalresources/env_issues/water_and_land_ salinity. water and sewer pipes. reduce river salt concentrations and sell the harvested salt as a final product. Online. and include damage to roads and buildings. We have not included this endogeneity in our pilot experiments. In more realistic settings the price for salt permits would also depend on irrigators’ heterogeneous marginal cost for salt reduction (abatement) technology. In the pilot experiments presented in this chapter we assume that the technology improvements only release salt licenses.

available at: napswq. Tradable permits allow firms to trade allocations of pollution (property rights). Behavior may also be induced by incentives not controlled for in the laboratory. a Laboratory Experiment Market. 2002) and the allocation of American medical doctors to their first jobs (Roth. In future experiments. Our objective is to understand how the simultaneous multiple unit double auctions operate. an extension of the Bushtender See online. 5 Online. 19 Many technologies will also save water. 18 Sellers are allocated their equilibrium quantities of licenses free of charge at the beginning of each period.gov. Using two impact zones instead of five provides some understanding without introducing additional complexity. 1999). 9 Without this information the regulator could not define property rights for salt. having subjects trade units labelled water and salt as compared to units 1.au (accessed 30 August 2005). could induce private preferences that cannot be controlled.au/mbi/index.html (accessed 30 August 2005). 14 Land use and production technology distributions across the Sunraysia were provided by SunRise 21. Or. 6 See Sinnclair Knight Merz (2001). 16 Introducing too much complexity from the field can reduce control over incentives in the experiment. See Cap and Trade for Salinity: Property Rights and Private Abatement Activities. we plan to increase the number of subjects and impact zones to model the field more closely. and for information on the new Ecotender program.au/dse/nrence.vic. available at: dpi. 17 In Sunraysia there are five salinity impact zones (see Figure 4. for example. available at: napswq.nsf/LinkView/F18669E8E2A4C02FCA256FDB00031592 E2A52365F438A1EFCA256FFF00236D8E (both accessed 30 August 2005).au/mbi/projects. and therefore if a subject adopts improved technology then they will have additional water to sell that is no longer required in . to name but a few applications.gov. the British Telecommunications Spectrum Auction (Binmore and Klemperer. 11 Outside the environmental domain experimental economics has been successfully used in the design of the Federal Communication Commission’s Spectrum Auction (Milgrom. 2002). 15 Our stakeholders include irrigators and rural water authorities from the region. It is very important to them that any policy experiments. 7 See.1). (2003).com. See also online. For general public information on the Bushtender program. 2004). the Commonwealth Government of Australia released an audit of salinity across the Basin (MDBC. for information provided to irrigators in the region about the salinity impact zones. 2004 and McMillan.au (accessed 30 August 2005). 2 and 3. subjects may choose a trading strategy that minimizes effort and difficulty giving a “safe” average payoff as opposed to trading to maximize earnings. in more complicated experiments. available at: watermove. Victoria. 8 Each irrigator is located within one zone only for the purpose of policy implementation (SRWA.vic. In this pilot test bed experiment we have chosen to simplify the environment.nsf/FID/ -15F9D8C40FE51BE64A256A72007E12DC?OpenDocument. It would be too costly or infeasible to create property rights for sources that cannot be measured. which may in the future influence production in the region. for a technical discussion of the salinity impact estimates. For example. available at: napswq.gov.gov. this can be mitigated through training periods. Subjects may find the experiment too difficult to understand. accurately reflect the field conditions.Water and salinity rights trading 97 3 In October 1999. These groups are familiar with local production costs and values. Mildura.gov. 12 Online. 13 See. SRWA (2002). 10 Also see Stoneham.au/DSE/nrence. et al. See online. available at: dpi.html (accessed 30 August 2005). 4 Online. 2002). the New Zealand Spectrum Auction conducted in 1990 (Milgrom. This can arise for a number of reasons. 1994).

K. New South Wales. 46(3). and Plott. Economic Journal. The Value of Water – A Guide to Water Trading in Victoria. Vol. Mimeo. or native trees and shrubs. Duke et al. Cason.com. 446–471. and Klemperer. eds. 2001. A laboratory study of auctions for reducing non-point source pollution.. 2004. Brookshire. In session 2 of treatment 2. Bohm. and Vincent.. Department of Economics Working Paper Series. R.au/ (accessed 30 August 2005).. University of Melbourne. Burness.. Frontiers of Environmental Economics. To whisper in the ears of princes: laboratory economic experiments and environmental policy. 2001b. N. and Muller. Commonwealth Government of Australia. 1996. Catchment and Water Resources. ACT. drainage schemes that capture excess water and drain away from rivers. McKee.. irreversible production choices. Binmore. Cummings. C. A.. Gerking. An Agreement Between the Commonwealth of Australia and the State of Victoria for the Implementation of the Intergovernment Agreement on a National Action Plan for Salinity and Water Quality. and Gangadharan.98 C. H. Canberra. an experimental investigation of the Sunraysia Salinity Levy. L. available at: pyramidsalt. M. C... . Buckley. An announcement was made in all treatments informing subjects there was 1-minute until market close. and Taylor. Salinity in water markets. S. Department of Natural Resources and Environment. Tasmania. T.. an agreement between the Commonwealth of Australia..G. Online. It is possible to design experiments in which subjects change from buyer to seller within a period depending on market price. 1. June. 20 21 22 23 References Ben-David. 112(478).. 2003. Experimental evaluations of policy instruments. T. Journal of Environmental Economics and Management. and efficiency in emission permit markets.N. Heterogeneity. Journal of Environmental Economics and Management. Handbook of Environmental Economics. Victoria. L. As our focus is to study how these markets operate it was not necessary to include this additional feature at this stage. Melbourne. D.. Journal of Environmental Economics and Management. Gangadharan. Commonwealth Government of Australia. Victoria.. 2002.. Implications of alternative emission trading plans: experimental evidence.. 74–96. Mäler. of Handbooks in Economics. S.... Vic. and Schmidt.. We assume in these pilot experiments that technology only reduces salt impact. Gabel. M. 2001. Examples of technologies that reduce salt are groundwater pumps.R. South Australia. subjects could expand production for a given level of water input. 133–160. S. Queensland. Folmer. C.A. October. 1999. 2003. Mestelman. P. production. McKee. eds. R. K. Duke. P. McMaster University. and Duke. L. Cason. S. ACT. This introduces additional complexity particularly in oral auctions. London: Elgar Publishers. C. Canberra. Alternatively. The biggest auction ever: the sale of the British 3G telecom licences. Intergovernment Agreement on a National Action Plan for Salinity and Water Quality. and Rose.. 2004. the Northern Territory and the Australian Capital Territory. only Buyer 2 did not choose to change even though it was profitable for her to do so. 2001a. EPA’s new emissions trading mechanism: a laboratory evaluation.N. Amsterdam: Elsevier. 30(2). Western Australia. J. G. 38(2). 176–194. and revegetation with native crops.

Mapping and Information Services. 9(4–5)... Murray Darling Basin Ministerial Council. Southern Economic Journal. 3–10. A. Canberra. 2005. L. Stakeholder workshop. Victoria. 1994. .R. ACT. C. Goodfellow. L.A.. Journal of Economic Perspectives. August. Sinnclair Knight Merz. May 2004. Melbourne.. 477–500. McMillan. University of Melbourne Working Paper # 942. Irrigator.. ACT. 56(4). 2003. 1983. June 2002. 1998. Econometrica.. Milgrom. Murray Darling Basin Commission. Chaudhri. Plott. C. J. A 100 Year Perspective. Muller. J. and Croson. Mallee Region Salt Impacts of Water Trade: A Proposed Method for Accounting for the EC Impacts of Water Trade in the Victorian Mallee. and Mestelman. 93(369). Market architectures. 2001. 2002.Water and salinity rights trading 99 Gangadharan.R. 2001. Selling spectrum rights... SunRise21. Externalities and corrective policies in experimental markets. Economic Journal. C. October. 8(3). 1994. 2002. Putting Auction Theory to Work. and Plott. 106–127.. a report prepared for the Department of Natural Resources and Environment.. S. 969–983. 1999. 225–238. 1(16). 2003. Mildusa. Roth. institutional landscapes and testbed experiments. P. Investment decisions and emissions reductions: results from experiments in emissions trading.E. Economic Theory.R. Australian Journal of Agricultural and Resource Economics. 2004. A. Farrell.. Murray Darling Basin Commission. Basin Salinity Management Strategy 2001–2015. 145–162. experimental economics and computation as tools of design economics. An experimental examination of the simultaneous determination of input and output prices. The Salinity Audit of the Murray Darling Basin. R. V. G.. Refinement of the river salinity impact zoning system. 70(4). 4(1). The economist as engineer: game theory. 1341–1378. Stoneham. and Strappazzon. R. Plott. Auctions for conservation contracts: an empirical examination of Victoria’s Bushtender Trial. A. Ha. Cambridge: Cambridge University Press. 47(4). 6–17. Sunraysia Rural Water Authority. Canberra. What have we learned from emissions trading experiments? Managerial and Decision Economics. 1990.

Jeffery Connor. Ward. Despite regional . Recharge rates and associated rates of salt mobilization in the areas depend on geomorphology with localized fractured rock conducive to high groundwater recharge rates. grazing and management decisions. and John Tisdell Introduction This chapter describes context specific lessons about the application of experimental methodologies to aid in the design and implementation of cost effective water quality and land management policy.5 Aligning policy and real world settings An experimental economics approach to designing and testing a cap-andtrade salinity credit policy John R. Australia. The increased recharge leads to episodes of increased mobilized salt loads. Alternate institutional structures were tested in an experimental economics laboratory including an array of potential market settings to inform an on-ground trial of a recharge cap-andtrade scheme. connected groundwater aquifers. The Bet Bet catchment contributes the majority of the more than 40. The case study focus is the Bet Bet dryland farming community in a catchment located in north central Victoria.000 tonnes of salt annually entering the Boort irrigation area from the Loddon dryland catchment areas. legal and statutory remedies and a scheme providing scheduled payments for the re-establishment and management of deep rooted perennials. The additional salt is exported into connected river systems where it jeopardizes the long-term viability of downstream irrigated horticultural and agricultural crops. In addition the rate of recharge and thus external salinity impacts depend on the type of vegetative ground cover and farm specific cropping. Extensive replacement of deep rooted woody perennials and perennial pasture with shallow rooted annual pastures has been identified as a key factor in increased rainwater soil percolation and subsequent groundwater recharge. and threatens the functional organization of downstream riparian ecosystems. Levels of river salinity (measured as units of electrical conductivity or EC) are partially a consequence of the volume of rainwater recharging highly saline. Government sponsored efforts to motivate changes in management regimes on privately tenured land have relied on traditional farm extension processes.

The goal of the experimental economics reported on here was intended to test the capacity of alternative policy designs to: • • • • motivate persistent land use change appropriate to the specified groundwater condition. In addition. Shortle and Horan. as opposed to the explicit directives for environmental management associated with regulatory and centralized planning measures (Stavins 2003). . achieve targets at the lowest cost to society. testing and implementation (see for example. The primary motivation of market based instrument approaches is that if environmentally appropriate behavior can be made more rewarding to land managers. mobilize high levels of participation in strategically beneficial localities and. The local agencies responsible for the management of groundwater recharge and salinity have endorsed market based instruments as an alternative policy approach. Market based instruments (MBI) rely on regulation or laws that encourage behavioral change through the price signals of markets. have not been at a scale sufficient to comply with prescribed salinity targets (Connor et al. Appraisals of their relative importance in Australian policy portfolios have also been informal. The project that this research focused on was in response to a specific call for proposals to test a cap-and-trade approach to motivate revegetation efforts.Aligning policy and real world settings 101 promotion. simple rules and evaluation protocols to identify a priori the relative advantages over other instruments to resolve specific environmental problems have not yet emerged. then private choice will better correspond to the best social and environmental choice. This chapter describes the design and the results of a series of economic experiments to test the reliability and performance of potential policy instruments and provide robust ex ante guidance in the design features for a trial in the market exchange of recharge credits to reduce river salinity. the level of established revegetation and consequent groundwater recharge and river salinity. The primary objective of the experiments was to empirically inform the contractual design of an on-ground trial of a recharge cap-and-trade scheme in the Bet Bet dryland farming community. The advocacy of MBI as policy tools for managing both diffuse and point source environmental problems has a recent heritage. 2004). The objective was to develop and test the capacity of a recharge credit scheme to provide flexibile incentives sufficient to motivate extensive revegetation efforts and as a consequence ameliorate high volume reductions in groundwater recharge. limiting opportunities for policy makers to gain experience and expertise in their design. limited and ad hoc. mobilized salt loads and eventual levels of river salinity. encourage change at scales that contribute substantially to salinity targets. a discussion is provided of how economic experiments informed actual groundwater recharge policy design. Despite improvements in the analysis of market based instrument performance. 2001). There are several market based approaches that are widely applied including environmental charges and auctions.

Ward et al. groundwater levels drop and this imposes group shared costs on the common pool community such as higher pumping costs. Groundwater in the Bet Bet (and many other Australian settings) is also characterized as a common pool resource.g. controls. which also outlines an argument for expanding the potential function of experimental economics from primarily theory testing to wider use in the institutional design process. Costly exclusion implies it is not possible to compel those who enjoy the additional benefit of appropriations to compensate those who incur costs. where typically increased recharge ameliorates the social allocation dilemma facing abstractors. player payments and protocols. The first section of the chapter describes the hydrological process of saline groundwater recharge that characterizes many Australian aquifers. The figure shows that groundwater recharge increases as an inverse function of the level of deep rooted perennial vegetation (illustrated in panel B).102 J. Figure 5. individual appropriations occur to a level where resource overuse arises. A description of a context rich simulation setting is the focus in the third part of the chapter followed by details of the experimental setting. most of North America). however rising rather than falling levels of groundwater constitutes the common pool externality. The results are then set out and are discussed further in the penultimate part of the chapter. arising as a result of the costly exclusion of beneficiaries and rival resource utilization (Ostrom 1998). declining groundwater levels constitute a common pool externality. The increased hydraulic pressure in the mound above the saline aquifer causes a subsequent rise in both the water table and the level of salt intrusion in the river system. Without coordinating institutions. including the Bet Bet catchment. Such a model is a necessary antecedent to the market exchange of recharge units. The development of a proxy monitoring and auditing model is also described. which quantifies variation in groundwater recharge rates at the farm scale as a function of land management action and landscape position. In typical settings. In the Bet Bet the majority of salinity impacts are exported to downstream river districts. An evaluation of potential market impediments and the experimental rationale and research hypotheses to test solutions are described in the next part of the chapter.R. incentives exist for individuals to act opportunistically.1 is a schematic representation of land management effects of rising groundwater recharge levels. Conclusions drawn from the experiments in developing implemented policy for the market exchange of recharge credits in the Bet Bet catchment complete the chapter. The chapter is set out as follows. When joint outcomes depend on multiple actors contributing inputs or actions that are costly and difficult to quantify and there is lack of institutional protocols to restrict usage. where the costs of salinization are incurred primarily by . Modeling interdependent land management and hydrological processes The effects of recharge are different in the Bet Bet catchment than in fresh groundwater aquifers in many other parts of the world (e.

1 Schematic of the hydrogeology of irrigation water quality affected by variable upper catchment salt loads. land managers need to be able to evaluate the implications of their management decisions prior to implementation and to monitor progress against their targets or commitments. The increased salt load confers increased irrigator costs brought about by increased soil salinity and correlated reductions in crop yields. downstream irrigators. rising water tables can create external costs for dryland farmers in the form of land with shallow rooting depth to saline water leading to crop yield losses or reduced rotational options. Since groundwater recharge and salinity are not readily measured directly. a prerequisite to implementing a cap-and-trade is the development of a reliable and transparent . accelerated infrastructure degradation and reduced ecosystem function. administrators of the scheme must also have the capacity to monitor and audit the outcomes of changes in land use or management practice and to attribute change in recharge to either landholder action or climate.Aligning policy and real world settings 103 Figure 5. In addition. To participate effectively in the market based instrument scheme. Similarly.

Gi. k = 1. RAi. RAi is annual rainfall. The estimated costs of groundwater recharge for land management activity at farm i. Aij is area of crop type. Lk) and Cij is crop type and management. Gi is soil type and geomorphology and Lk is landscape position. Thus a first step in this project was the development of a robust and community validated biophysical and hydrological modeling to provide information about groundwater recharge rates as a function of variable vegetation cover and land management at the farm scale. recharge Ri is such that Ri Lk MP < Ri Lk MG. subject to land management regime MP (Panel A) or MG (Panel B). Lk represent exogenous variables in a farm decision set.R. RL1 < RL2 < RL3.104 J. Cij. for crop j. for landscape position Lk. Ri represents the recharge rate for farm i. and RAi. Ward et al. where: Ri = α (Cij.1: 1 Ceteris paribus. CMG > CMP – external costs of recharge imposed on downstream irrigation are greater for upstream annual grazing compared to perennials. differentiated according to landscape position (Clifton 2004. WTMG > WT MP – the water table level is higher for annual grazing than perennials. In the model. Increased deep rooted perennial vegetation reduces groundwater recharge: namely. 2 or 3. 2004). which is less than recharge from upper slopes (L3). where 1 represents lower slope. surrogate metric to assist all participants in evaluating recharge and salinity impacts of land management actions. at landscape position Lk is such that: RMG > RMP – recharge from annual grazing is greater than recharge from perennial grazing or forestry. Namely. 2 represents break of slope and 3 represents ridge and upper slope. 2 represents perennial pasture (phalaris set grazing). j – 1–5. Connor et al. Shortle and Horan (2001) and Schary (2003) argue that developing policy capable of realizing savings by focusing on performance coupled with compliance flexibility when monitoring actual outcomes are technically infeasible or 2 3 . Aij. The model developed accounts for three key biophysical determinants of recharge differences across locations and actions shown in Figure 5. managing crop j. 4 represents native tree vegetation and 5 represents farm forestry (less than 10 years old). Gi. SLMG > SLMP – alt load is greater for annual grazing than perennials. recharge from lower slope (L1) is less than the recharge from break of slope (L2). Aij represent endogenous variables in a farm decision set. 3 represents perennial pasture (phalaris rotational grazing). where: 1 represents annual grazing.

Several commentators note that successful MBI schemes have not necessarily substituted for regulatory approaches but are more generally complementary (Stavins 2003. 177) articulate the economic case for and potential advantage of tradable credit systems applied to environmental policy. 2004. A challenge for policy is to create the opportunity for a “frictionless” market setting expediting rapid learning and understanding of the advantages of trade with low learning and exchange costs relative to trade benefits. In many cases MBI may be advantageous. 1996). Tietenburg (1998) notes that global experience indicates that many past attempts to implement well intentioned tradable permit schemes for both diffuse (e. relative to approaches that allow less flexibility such as uniform standards. and others will choose to use less (being rewarded with credits). MBI have been increasingly endorsed across an array of agency jurisdictions as cost effective policy instruments to address environmental targets and protection. in a well functioning market the opportunity to trade has the potential to compensate that loss or reduce the cost burden. but in others the relative advantages over other instruments may be limited. advocating that environmental goals can be achieved at lower cost to society than alternative policy approaches. However. Young et al. poorly defined. Since the mid 1990s (Tietenburg and Johnstone 2004). The recharge accounting model described above was designed to act as a reliable means of auditing and monitoring land management actions to facilitate performance based payments. savings to landholders through market exchange between individuals with surplus credits and those in deficit may be considerable. but may not perform well in the dimensions of adoption . Tietenburg and Johnstone 2004. ground water recharge) and point source emissions have failed because of inadequate attention to the design and timing of implementation of the market architecture deployed. Subject to controversy and debate ten years ago (Keohane et al. information expressed as coherent price signals from market exchange would reveal any differences in returns to management options that reduce environmental consequences and these would be immediately discovered and exploited. Baumol and Oates (1988. In a frictionless market. Some individuals will choose to use more than their quantum (and incur a debit). state contingent and subject to change through time.g. Tisdell et al. In addition. MBI have evolved to the point of becoming received wisdom in many environmental policy circles (Stavins 2003). 1998). Selecting experiments to address market impediments Whilst the introduction of maximum recharge quanta imposes costs on individuals. p. For example a market based instrument may be cost effective. Environmental policy analysts dating back to Dales (1968) and Montgomery (1972) have advocated tradable credit approaches based on the argument that such approaches can increase economic efficiency.Aligning policy and real world settings 105 very costly is a substantial challenge to effective cap-and-trade schemes.

rates. the likelihood of thin markets and the effect of non-market motivations were identified as likely impediments to an effective recharge market. To improve ex ante predictions of policy outcomes. administrative and transaction costs. 2003. experimental economics was employed to provide empirically based analysis of observed behavioral responses to the implementation of potential economic and community governance instruments to reduce recharge.1 Previous Australian experience (Stoneham et al. administrative alignment. 1995. a feasible surrogate property right is to pay landholders to enter into a temporary contract prescribing recharge obligations. The rationale for the three sets of experiments that were chosen through this process is summarized below. Ward et al. environmentally reliable and politically feasible trial implementation and empirically guide policy design features to mitigate identified impediments. Given that a mandatory approach is not institutionally possible in the short run. Experiments were part of a wider process of policy design involving formal and informal community and expert consultation. formal experiments were applied to pretest feasible on-ground market impediment solutions and longer term policy options that may require changes to current institutional structures. the single model terrain of economic efficiency or cost effectiveness may not be sufficient to inform reliably policy makers of aggregate instrument performance. calibrated to represent the salient economic and biophysical features of the actual catchment (Clifton 2004). (2004) identified market impediments that were qualitatively evaluated and ranked according to the likely significance of the impediment on the catchment.R. In a controlled decision environment. analysis of a sociological survey in the catchment and farmers participating in field demonstrations of an experimental land management and recharge trading simulation. The motivation for experimental work was to help understand and compare key factors that could impede cost effective. A lack of enforceable property right obligations to manage recharge. Latacz-Lohman and Van der Hamsvoort 1997) suggest that a competitive . and by introducing an incentive to maximize participation rates addresses the likelihood of thin markets. The payment scheme also provides a straightforward trial entry.106 J. Kagel. Ward and Connor (2004) and Connor et al. 2004) and theoretical insights (inter alia: Holt 1980. Thus a first step in trialing the market exchange of transferable recharge credits is to establish limits on recharge. concentration of environmental consequences and political feasibility. When these are articulated policy objectives. political feasibility and the likely efficiency gains of proposed impediment solutions. the high transaction costs of complex information processing. Milgrom 2004. Bryan et al. Experiment 1: testing assignment of property right obligations for recharge management with discriminant and uniform price auctions There are currently no obligations for farmers to manage recharge to prescribed limits.

Experiment 2: assessing impacts of information processing cost Social survey work from this research (Thomson 2004) as well as experience with market based policy (Hahn 1989. Milgrom (2004) notes a uniform price. One of the premises of theoretical analysis of perfectly competitive markets is that agents are fully and costlessly informed of the consequences of all potential market actions they can take. Sterman 1987. 62) “Plott characterizes this as a three stage process with a particular trajectory. p. Smith 2002). p. is that “rational economic behavior emerges at different stages in the development of an agent’s decision skills” (see Guala 1999. 2003. autocorrelation or interdependence of decision sets and the existence of public goods. Myerson 1981). is more incentive compatible to the revelation of true costs. Cason and Gangadharan 2004) agencies have relied on maintaining the extant information asymmetry3 to mitigate incentive incompatibilities and maximize cost effectiveness for the agency. Relying on experimental analysis (Cason et al. In implementing novel natural resource policies involving transitions to new institutional arrangements such as recharge credit trade. called the “discovered preference hypothesis” by Plott (1996) and referred to by Binmore (1999) and Baga and Starmer (2005). 190. Cason and Gangadharan (2005) note that violations of the assumptions of the revenue equivalence theorem (Vickery 1961. variable risk aversion. Thus a set of experiments was designed to test how the level of complexity of information processing required of program participants in alternative policy designs influences tradable credit market outcomes. 2004). namely. Randall 2003) suggests that the issue of low participation rates and unreliable recharge reduction could be particularly acute with novel market designs that require more complex information processing. to induce sellers to minimize the level of trade surplus included in their tender bid. 569). precludes coherent theoretical analysis of the tender formats. multiple or endogenous quantities. Thus the first experimental analysis conducted is to provide a formal comparison of the bidding behavior and the magnitude and reliability of recharge reduction with discriminant versus uniform tender formats.Aligning policy and real world settings 107 tender should reveal the true costs of recharge abatement and act as a cost effective way to assign recharge obligations. p. common to many environmental applications. Bryan et al. where an equal payment level is established exogenously from successful bidders. A premise of the iterative process of rational expression of preferences. Current Australian applications of tenders for conservation provision have relied on a sealed bid2 discriminant price tender format (Stoneham et al. Milgrom (2004) summarizes these as budget constraints. decisions . agencies may need to account for behavioral and cognitive limitations of informational processing to understand how to adapt successfully and introduce market based approaches (Simon 1972. both to reveal individual opportunity costs in a competitive environment and to minimize public expenditure by capturing available producer surplus. 2003. According to Baga and Starmer (2005.

especially in the domain of public goods and environmental quality. Experiment 3: testing policies to overcome low participation rates and non-market motivation The final issue investigated is policy designed to manage common pool resources. 1991. Prestige. given the opportunity for repeated market participation and with feedback signals of sufficient clarity and strength. Survey results from this research (Thomson 2004) indicate high levels of social cohesion. Poe et al. Therefore experiments were designed to provide experimental evidence relating levels of information complexity to recharge market outcomes. Ostrom et al. expressed as seemingly random impulsive responses. According to Vatn and Bromley (1995). 12 months). Gintis 2000. 1992. In another Australian catchment. extending to trade in recharge units (Ostrom 1998. public recognition. the expression of increasingly rational behavior can be expected to evolve. Smith (1982. Additionally. 2002) and Vatn and Bromley (1995) argue that the complexity of informational processing necessary to enable coherent choices (extended here to environmental goods). which have remained functionally invisible. given behavior of individuals with non-market and social motivations. According to Plott (1996) and Binmore (1999). 2004. precluding repeated farmer participation. through a more systematic selection of choices that reflect the decision environment and finally to a stage where expressed choices reflect the choices of others. 2004). Tisdell et al. There is currently a paucity of empirical evidence or theoretical insights to help anticipate the time steps and explanatory variables of emergent learning in specific market contexts (Braga and Starmer 2005).R.108 J. progressively exhibit less randomness and greater rationality.” That is agents progress from an initial state of untutored and limited cognitive capacity. it may not just be individual choice but a variable sequence of choices (Kahneman and Tversky 1979) potentially filtered through the focal length of collective choice (Ostrom 1998). reciprocal behavior and concerns about community reputation. group belonging. Lowenstein (1999) and Loomes (1999) the expression of individual preferences in market settings is a complex psychological process where personal welfare may not be sufficient to explain the nature of choice. Ward et al. Additionally. may impede convergence with a theoretical market equilibrium. avoidance of group sanction and desire to contribute to the public good can all represent powerful motivators in some contexts. Failure to account for such motivations in policy design can result in reductions in the willingness to supply environmental quality compared to policy designed to harness such motivations. Past research suggests that behavior may diverge substantially from expected utility theoretical predictions of the outcomes of trade. The recharge trial is time constrained to a single iteration of the clearance market (namely. Marshall (2004) also found the level of preparedness to cooperate and adhere to group compacts for salinity control were more sensitive to perceptions of . This is especially important for an evaluation of the initial recharge clearance market implemented in the field.

modeling recharge and opportunity cost to actual farm activity choices in the trial area.Aligning policy and real world settings 109 community benefits and a belief that others will reciprocate compared to private considerations like business security. These combined findings suggested that a policy approach to reward scheme participation with some form of collective award for reaching an aggregate recharge reduction level may increase the motivation for trial participation. the introduction of a social payment. the additional complexity of available information in an open call market will lead to more volatile market outcomes and impede convergence to predicted equilibrium. there was concern that once the few enthusiastic participants had initially been engaged primarily in pasture improvement actions. Research hypothesis 1: The incentive to reveal the true cost of individual recharge reduction is less for a discriminant price tender than for a uniform price tender and a discriminant price tender will result in more variable recharge reduction and more volatile market outcomes. the violation of theoretical assumptions of auction theory and the public nature of groundwater recharge may mean that theoretical prediction is not sufficient as an analytical tool or a reliable precursor for policy decision making. experiments were chosen to test three research hypotheses. biophysical and economic data. To gain an empirical basis for a priori evaluation of recharge policy. In particular. Consultation with local landholders and agency personnel raised important concerns that a voluntary tendering process may result in insufficient participation levels to achieve the desired level of land management change. Research hypothesis 2: Compared to a closed call market. will result in increased voluntary participation and hence recharge reduction. Research hypothesis 3: Compared to payments for individual recharge remediation. it would be difficult to motivate further land management change to reduce recharge. Synopsis of rationale for chosen experimental treatments The overall conclusion of the review of potential impediments to recharge credit trade policy was that introduction of novel market institutions. Experimental farms are thus heterogeneous and represent the main relationships between landscape positions. The introduction of communication as a coordination mechanism will reinforce voluntary social contracts and result in increased recharge reduction. Context rich economic experiments The experiments were designed with field calibrated hydrological. especially tree planting at the locations where this would contribute most to recharge reduction. given the poorly defined state of information discovery. farm . equally distributed to all players for collective recharge reduction in excess of target levels.

2003. All experiments involve the same simulated experimental catchment comprising a total of 12 heterogeneous farms located in three landscape positions with four farms located in each landscape position.110 J. farm income and groundwater recharge specific to the Bet Bet catchment. There is now a growing body of experiments conducted in context rich environments. (2004). Experimental settings and experimental controls common to all experiments All experimental sessions were held at the Griffith University experimental economics laboratory in November 2004. Poe et al. Gintis 2000. management regimes. 2004). they may not tell us much about how people are likely to react in real world contexts where confounding factors exist. As an important insight from cognitive psychology. experiments may need to be designed to include salient features of the policy setting of interest (Lowenstein 1999. using the MWATER experimental software platform developed and administered by Dr. Contextualization is not usual experimental protocol. Lowenstein (1999) and Loomes (1999) advocate that decision making is highly context dependent. however this research follows the lead of Cardenas (2000). The students represent a number of faculties and the majority do not have an economics background. Loomes 1999.R. Krause et al. Ward et al. Tisdell et al. by informing participants that they are hypothetically farmers located in a closed catchment and produce recharge that in turn potentially affects other farmers. The farms represent . Using the experimental setting. 2003). Adherents conclude that while experiments designed to eliminate any confounding effects are useful for isolating influence of single treatment factors. we sought to elicit and measure behavioral responses to hypothetical decisions simulating policies and diverse institutional structures across an array of market conditions in an analogue specific to the Bet Bet catchment. (2003). While experiments from context rich settings may allow only limited inference about behavior in other contexts. Participants were selected from an existing pool of approximately 200 undergraduate students. they are employed in this work because they represent in the view of some. Results demonstrate that differences in context lead to differences in bargaining behavior. John Tisdell of Griffith University and the Cooperative Research Centre for Catchment Hydrology. risk taking and sharing (inter alia Camerer and Lowenstein 2004. Krause et al. the most appropriate way to draw inferences about behavior that are valid for specific contexts where policy design is being investigated (Lowenstein 1999. This has led some experimentalists to conclude that to inform policy meaningfully. familiar with experimental protocols and procedures. Plott and Porter 1996). Krause et al. (2004) and Tisdell et al. Additional context is introduced into the experimental domain.

1 allocations were distributed according to farm characteristics and landscape position.868 3. characterized by different levels of farm income and recharge rates.484 7. FF = Farm forestry.1 Decision set. For all treatments. The duration of experimental sessions was approximately two hours. .911 7.210 1.948 2. NV = Native vegetation. Throughout the experimental sessions.280 2.1 The decision set provided to participants comprised five discrete farm management options. Ps = Phalaris set stocking.280 3. Farm 2 allocated 20 units. calibrated to simulate the main economic and biophysical decision environment facing farmers in the Bet Bet catchment (Ward and Connor 2004).1 was allocated 45 recharge units. the marginal value of recharge units and the required recharge balance for the selected option. farm income.183 5. Options characterized by higher income levels were associated with higher recharge levels for all farms. Pr = Phalaris rotational grazing. After entering the chosen management option.399 7. each participant was randomly assigned to a single farm. As shown for two farms in Table 5. each option associated with a unique level of farm income. recharge and a marginal value of a recharge unit based on recharge modeling summarized earlier in the chapter. crop mix and recharge of two of 12 experimental farms Landscape position Ridge and upper slope Farm 1 Mid slope and break of slope Farm 2 Decision 1 2 3 4 5 1 2 3 4 5 Crop type (%) 100a An 100 Phs 100 Phr 100 NV 100 FF 50 An + 50 Ps 50 An + 50 Pr 50 An + 50 NV 50 An + 50 FF 50 Pr + 50 FF Farm income (A$) 3.Aligning policy and real world settings 111 a synthesis of existing farm management styles. To create realistic incentives. An = Annual pasture. for example Farm 1 in Table 5. Lu = Lucerne. subjects were paid according to the combined simulated farm and trading income they achieve in experiments. Two sessions were held for each experimental treatment. Player payments and non-compliance penalties in the open and closed call clearance market adhere to induced value theory and were standardized across Table 5. experimental subjects participated in sessions that involved ten independent periods. participant screens were updated with the option specific income. and could select one of the five possible farm management options. when the recharge quantum was set at 50 percent of maximum recharge.923 Recharge (Mls) 91 73 62 22 0 39 33 31 26 7 Notes a % refers to the proportion of 65ha under specified crop management.

The successful completion of the quiz was a necessary prerequisite for participation .112 J. Player payments were calculated as a function of aggregate farm management and trading outcomes. adherence to farm management agreements and trading skill. a player selects the 50 percent recharge reduction income choice). Talking. If you do not reach the target. protocols. The cost you paid for purchasing the surplus units will be deducted from your farm income. Optimal incomes were calculated assuming players act as profit maximizers.R. For example: Say you have an allocation of 2 units. To ensure salience of player behavior and response to income variance in the simulated catchment.50 per period for achieving the derived optimum farm income and $0. rules. your excess farm recharge will affect other farmers. payments and noncompliance penalties specific to each experimental treatment. Ward et al. You purchase 7 units in the market. an income of $1 (and a target of 5 – 2 = 3). In addition to a $10 attendance payment. decision sets. You cannot reach the option 3 target. specific farm (player) payments were calibrated by regressing (OLS) achieved farm income against a payment schedule of $1. To control for variable learning and to ensure consistent understanding. all treatments. was forbidden except to clarify questions from individuals regarding the experimental setting or instructions. participants were required to answer accurately a quiz comprising 10–12 questions specific to the experimental treatment. The penalty for non-compliance in the clearance market treatments reduces farm income to a level where the recharge allocation can be satisfied and deducts the equivalent market value of the shortfall of recharge units necessary to satisfy the recharge target associated with the income level selected.40 for static decisions (namely. where option 3 produces a total of 10 recharge units and an income of $2 (with a target of 10 − 2 = 8) and option 2 has a total of 5 units. You will not be able to trade or be compensated for any surplus units. unless formalized in the treatment. salient for each treatment were provided via individual internet access to a power-point display. You choose option 3. Carefully designed instruction sets. The recharge target is the total recharge produced by the farm option minus your allocation. Your income will be reduced to the management option where the recharge target can be met. acting optimally to available information. the player payments were therefore a scaled representation of the income decisions confronting farmers in the Bet Bet catchment. reflecting player capacity to comply with recharge targets. Supervising staff did not verbally present the instructions to avoid personality or behavioral biases and correct for possible delivery nuances. The noncompliance penalty was explained to experimental participants thus: To receive the income for the farm management option you have chosen you must reach the recharge target. The instructions explain the farm characteristics. You receive the option 2 income of $1 and have 4 surplus units. but can meet the option 2 target.

Players were restricted to selling surplus units in the tender treatments.1: Farm 1 Ra = 45. The market price was announced (in the event that there are no matching buy and sell offers. or choose an option that required less recharge credits than their quota and offer for sale the excess credits. Players could access bidding history at any stage during the experiment. In the closed call. The instruction sets are available from the authors upon request. Market entrants enter bid quantities (based on Rb) and their price (based on the marginal value of recharge). Subjects had the choice of acting in one of three ways in each decision making period. the market price and their total income from the period. Anonymity of bidders was maintained throughout the session and players could access their individual bidding history at any stage of the experiment. and public declaration of all bids in the open call. Marginal value (MV) for each option is calculated as the marginal cost of recharge abatement for the option compared to the recharge allocation. Participants’ screens were then updated to reveal the outcome of their bids. then Rb = 45. For all experiments. Players voluntarily enter the market to meet recharge shortfalls and sell surplus recharge units. but may only enter a single bid in each period. Players were prevented from offering surplus recharge units for sale in excess of their calculated Rb. choose an option with recharge in excess of their quota and make up the deficit by buying credits. individual farms are differentially allocated a recharge entitlement according to farm characteristics. nominally set as 50 percent of the maximum farm recharge rate. 12 experimental participants are provided with details of their farm’s initial recharge allocation (Ra). Participants could either buy or sell (subject to having surplus recharge units).Aligning policy and real world settings 113 in the experiments. Participants only have access to their own farm information. The recharge balance (Rb) is calculated as the initial recharge allocation less the amount needed for the farm management option the participant selects. it was announced that no trades occurred). Roption5 = 0. the value of a recharge unit and the recharge levels for each decision variable or option. requiring purchase in a recharge market. then Rb = −46. the control treatment was the behavioral response that would be predicted if players acted as profit maximizers. For example from Table 5. Using individual computer terminals. Option 1 has a 46 recharge unit shortfall. Players were prevented from offering surplus recharge units for sale in excess of their calculated Rb. Bids were accepted over a 90 second period. They could choose a farm management option that exactly satisfied their recharge quota. facilitating a selling option in the market. farm income. Rb can represent a surplus or deficit of recharge units depending on the farm allocation and management option selected. Roption 1 = 91. In the cap-and-trade treatment. Market entrants entered a sealed bid quantity based on Rb and their market exchange price (based on the marginal value of recharge). Bidding information is restricted to individual bids only in the closed call. and the market clearing price calculated. acting optimally to . successful bids were only revealed to successful individuals. option 5 has a surplus of 45 units.

If participants are assumed to maximize profit. Dependent on the marginal cost of reduction. available information provided by the given market structure treatment instructions. in a discriminant price auction (first price) the purchaser pays a range of prices that match the bids offered by successful sellers. potentially resulting in a less reliable and more costly recharge outcome. Traders strategically seeking an optimal and maximum price may tend to explore the price opportunities in the market for a longer period compared with strategies in a uniform price auction. Design of uniform and discriminant price tender experiment to test initial recharge credit allocation options In the uniform price tender experiments. As a corollary. understand their farm characteristics perfectly and responded optimally to available price information. In all experiments.R. the discriminant tender format is not compatible with true cost revelation (Milgrom 2004) but may be more cost effective for the purchasing agency (Cason and Gangadharan 2005). Alternatively. the recharge volumes traded. That is all sellers receive a market clearing price that both exceeds their original offer and is determined exogenously of individual bids. levels of . Individual payments to sellers are thus inclusive of a trading surplus. behavioral responses to the treatments were measured as the aggregate recharge volume. We experimentally tested a first price discriminant price tender with a first price uniform tender. Experiments designed to compare the results of tendering with context specific upper Bet Bet supply. regardless of their initial bid. a seller faces a tension and costly learning effort in determining the balance between a higher probability of bid acceptance versus a higher trading surplus. The incentive for sellers is to increase the bid value to include a component of variable trading surplus. When payment is determined by an own bid. Sellers therefore face uncertain bid ranking and acceptance although price is assured.114 J. demand and market conditions provided the opportunity to test for differences between participation rates. players were asked to submit bids to the authority to undertake land use change that reduces recharge. In a uniform price auction the purchaser offers a single uniform purchase price that is paid to all successful sellers. without a commensurate raising of the market price. Trading surplus is not determined exogenously and must be included in the bid offer. The bids were ranked according to price per unit recharge reduction and accepted until a targeted level of reduction was achieved (nominally assigned at 367 units or 50 percent of maximum recharge) or bid offers exhausted. Ward et al. The purchase price can be determined as either equal to the lowest rejected bid (second price) or the highest accepted bid (first price). the recharge and price values may be more volatile in the discriminant price auction. the predicted reduction of 367 units will be realized at a tender price of $56 per unit. as increasing their bid reduces the probability of bid rejection. As a consequence sellers have a greater incentive to reveal their true costs of recharge reduction. market price and aggregate farm income.

We sought to establish the tender mechanism associated with a recharge reduction outcome that best approximates a predetermined level of recharge reduction (namely.Aligning policy and real world settings 115 recharge reduction and purchase costs for the uniform and discriminant price tender system. As participants eventually have the opportunity to exchange tradable recharge units. In all the experimental tender treatments the bidding strategy of the purchasing agency is quantity constrained. In contrast. Payment rates per unit of recharge were a function of aggregate recharge reduction. in contrast to a budget constrained strategy. simulates a hypothetical cap of 367 ML). Landscape context in the Bet Bet catchment matters in addressing dryland salinity. The purpose of the experiment was to test prior hypotheses about the initial cost of complex information processing required of untutored program participants in closed and open call market designs and the effect on the observed gains from trade. Design of tendering experiments with social payments The social payment experiments were variants of the uniform price auction. We therefore sought the best approach to maximize participation rates. The authority computes a single equilibrium price at which all trade takes place based on the aggregate supply of and demand for credits. An important characteristic of the closed call auction is the limited disclosure of bidding information. We therefore did not apply a strategy of asymmetric information to mitigate incentive incompatibilities. an open call market publicly declares all individual bidding and volume offers prior to the market clearing. Our motivation for deploying the tender process in the recharge trading trial departs from the objectives of previous experimental analysis of tender schemes for conservation provision. When the price has been computed. implicitly they are fully informed of recharge outcomes associated with their individual land management actions. in this case the experimental recharge management authority. particularly those farms with low marginal costs of recharge reduction. The only information disclosed is market price and volumes traded. The results of experimental treatments were compared with the theoretical predictions of a no-trade regulatory control. There is no public disclosure of individual bidding information or the individual volumes traded. the authority notifies successful traders and announces the market price and informs successful traders of the individual volume traded only. In a closed call market potential buyers submit sealed bids to buy and potential sellers submit sealed offers to sell. . Design of open and closed call market exchange credit trade treatments to test impacts of information processing requirements The recharge credit trade market can be administered as either a closed call or an open call exchange. The market is called and trades are executed by a clearing house. no information on specific transactions is reported.

30 547 − 367 where: Ro equals aggregate recharge reduction by players 1–12. The level of recharge reduction achieved and the level of the actual social payment was provided at the end of each period. Individual adherence with the group crafted agreement was voluntary. The “communication” treatment consisted of the same social payment mechanism in addition to a formal forum for group discussion. The treatment allowed for an initial face to face. The standard uniform auction price was paid to successful bidders if total recharge reduction was less than 50 percent of maximum recharge. non-anonymous communication and discussion forum of 10 minutes whereby players could craft a social compact for recharge management. The 70 percent reduction level corresponds to 547 units. players received a two part payment. Supervising staff were able to facilitate the discussion forum by answering technical questions and calculating aggregate recharge reduction and social payment estimates only.116 J. If aggregate recharge reduction exceeded 50 percent of maximum recharge. Participation in the treatment was voluntary and there is no penalty for non-compliance or non-adherence to group consensus. a social or group payment component was paid equally to all players. The communication treatment consisted of the same payment mechanism as the social payment plus a formal forum for group discussion. Subsequent face to face discussion periods were of 3 minutes’ duration (a similar process to that reported by Tisdell et al. regardless of recharge reduction efforts. In addition. 2004). The experiment involved two treatments of the social payment mechanism. the social or group payment was equally distributed amongst all participants. Ward et al.R. The intended aggregate recharge reduction was disclosed at the beginning of the period. In this treatment players were informed of the level of achieved aggregate reduction at the end of each period for all treatments when the reduction target is 70 percent of maximum recharge. In this treatment in addition to payment for individual performance. The discussion forum was designed to test the effect of reinforcing group crafted social coalitions to improve recharge reduction. One excluded communication among subjects. actual aggregate reduction was disclosed at the end of the period. increasing market participation and improving individual contract adherence. Players cannot reveal their farm characteristics. They did not engage in any strategic discourse with the players. Communication between players was not permitted between the discussion forums. players were publicly informed of the agreed and intended recharge reduction and the magnitude of the social payment at the end of the discussion forum. Based on the marginal cost of recharge . intended or historical market strategies or the value of their recharge units. Consensus is achieved by majority vote. The group payment for aggregate reduction exceeding the 50 percent reduction target of 367 recharge units is calculated as: R0 − 367 × $2. In addition to the standard uniform price payment.

ANOVA assumes independence of responses between periods.2. For both ANOVA (F) and t-tests a * entry in the table indicates significantly different values at α = 0.2. Experimental results Uniform and discriminant price tender treatments Results of aggregate recharge observed in the uniform and discriminant price auction treatments are compared in Figure 5. Therefore Dunnett’s post hoc test was used for pair wise comparison and described by subscript letters across rows.05.Aligning policy and real world settings 117 reduction of the 12 experimental farms. Differences in subscripted letters across rows indicate that values across treatments are significantly different.30 group payment is equivalent to the ratio of the additional surplus of a 70 percent reduction compared to a 50 percent recharge reduction payment of $1. To test for the degree of inter-period independence we ran a linear mixed model imputing periods (1–10) as a covariate to estimate 600 550 Aggregate recharge (ML) 500 450 400 50% recharge reduction 367 MLs 350 300 250 Discriminant price Uniform price Figure 5.05) for all treatments.2 Observed aggregate recharge in the discriminant and uniform price tender treatments.70.50 minus a transaction cost of approximately $0. . A summary of the ANOVA and Student’s t-test statistical analysis is presented in Table 5. Homogeneity of variance was tested by Levene’s statistic and found to be significantly different (p < 0. the $2.

992* 13.2 ANOVA of discriminant and uniform price tender auction treatments Periods 1–10 Statistics1 Control2 Discriminant2 Uniform2 Aggregate recharge (MLs) F(2.61) 203 (103.25 (s.2 summarizes the statistical analysis of the bidding responses to the uniform and discriminant price treatments of periods 1–5. .552 Periods 1–5 Aggregate recharge (MLs) Qty traded (MLs) Player income ($) Trade income ($) Discriminant Uniform F(2.210 (9036) 15.47) = 4.05. p = 0.402* 367 441 (82.47) = 7.26) 15. 2 Treatment means (across rows) with the same letter were not statistically different at α = 0.77) 280 (60. The primary experimental metric is the level of aggregate recharge (or recharge reduced) observed for each treatment.845 (3.20) 294 (52.327 13.78) t (9) = 1.36) Trade income ($) Discriminant t (19) = 3.398 (2.118 J. A total of 367 units at a value of $56 (in the uniform price) are predicted to be realized in the control treatment.22) = 7. = 46.552 Notes 1 Statistical tests * indicates significant difference across treatments at α = 0. The control values for all treatments represent those predicted according to expected utility theory.084 (7.05) 14.22) = 7. The observed bidding behavior of the uniform and discriminant price treatments is illustrated in Figure 5.188* 367 367 18.620.954* F(2.63 (2.42) 16.49 (3.202) 410 (76. random effects (AR-1 heterogeneous) using a restricted maximum likelihood estimator.83) Player income ($) F(2. the aggregate recharge resulting from the behavior of players trading in the uniform price tender is not significantly different from the control prediction (a proxy for a regulatory cap on recharge).89)3 411 (60.756* 20.e.80 (2. the Wald Z coefficient is 0.729* F(2.007) 14. Table 5.47) = 19.R.54) Qty traded (MLs) F(2. 3 Mean (standard deviation). The estimated covariance parameter for periods 1–10 is 23.497. management strategies and payoffs instantaneously. indicating there are no significant unobserved random effects between periods.05.144 t (9) = –5.22) = 6.573* 367 245 (116. Results indicate that: 1 The number of aggregate recharge units observed in the discriminant price tender is significantly more than the number predicted for the control treatment for periods 1–5.897* 18.144 21. In contrast.22) 18.3 and Figure 5.971) Uniform t (19) = –7. representing the expected outcome where participants are assumed to behave as profit maximizers.00 493 (74.82).00 20.40 (2.4 respectively. who completely understand and act upon optimal trade. Table 5. Ward et al.636* 20.

The mean aggregate player income for the ten period session is significantly greater in the discriminant price tender than the value predicted for the control and that observed in the uniform price tender treatment.4 Observed and predicted bids for the discriminant tender treatment. 160 140 120 Price ($) 100 80 60 40 20 0 0 20 40 60 Recharge units 80 100 Experimental bids Predicted bids Figure 5.3 Observed and predicted bids for the uniform tender treatment. despite fewer recharge units traded in the discriminant price tender than in the uniform price tender treatment. = 116. Higher than predicted bids for recharge reduction actions were therefore purchased if the 50 percent recharge reduction stipulated in the .45 for periods 6–10. The experimental design deploys a simulated purchasing strategy of the management agency that prioritizes achieving recharge reduction targets as opposed to being budget constrained.d. the mean value of recharge units traded in the uniform price is greater than the observed mean value traded in the discriminant price tender treatment. Player income increased from a mean value of $18. The variance of the discriminant price tender (s. 2 3 Although not significantly different.62 in periods 1–5 to $22.Aligning policy and real world settings 119 140 120 100 Price ($) 80 60 40 20 0 0 20 40 60 80 100 Recharge units Experimental bids Predicted bids Figure 5.4).203) indicates the tender format is characterized by substantial bidding volatility for the ten period experimental sessions (illustrated in Figure 5.

In a general linear model. In contrast to these experimental results. The uniform price is therefore selected as the appropriate tender mechanism to establish payments for farmers to enter into recharge management obligations.080.055).143: interaction effect of treatment * period: F(2. The independence of periods was tested at two factor levels (dependent variable net farm income). in the discriminant and uniform tender auctions in the context of these experimental treatments. providing the opportunity for strategic.184). Empirical evidence provided by these authors found that offers in the uniform price auction experiment they conducted were within 2 percent of individual abatement costs while offers in the discriminative price auction were at least 8 percent greater than cost. p = 0. Cason and Gangadharan (2005) found the discriminant tender was more cost effective compared to a second price uniform tender.44) = 1. Successful bids in the discriminant tender ranged from $23–120/unit (see Figure 5. The results are in accord with the experimental results reported by Cason and Gangadharan (2005) that compared uniform and discriminant price auctions for land management. The control treatment predicts a market price of $61 and 144 recharge units traded. p < 0.23 in periods 1–5 to $86.111. p = 0. measured as the amount of aggregate recharge reduction. .5. level 2 = periods 6–10.R. The behavior of players in the uniform tender results in significantly increased recharge reduction. The first factor used the ten periods as levels: periods did not significantly affect observed net income (F(9. The control values represent those predicted when participants are assumed to behave as profit maximizers.4) increasing from a mean cost of $79.05) and described by different superscript letters across treatment rows. there is a significant difference. Significant differences (p < 0.3. Successful bids for the uniform price tender ranged from $23–56/unit for periods 1–10 (see Figure 5.120 J. A summary of the ANOVA statistical analysis is presented in Table 5. p = 0. There was no correction on observed treatment means for the random effect of period.3) at a mean cost of $52.49) = 2. The two level factor did not significantly affect observed net income (F(1. the second factor tested the random effect of two period levels: level I = periods 1–5.759.49) = 5. target had not been fulfilled by purchasing sufficient units offered by lower priced bidders. The experimental purchasing strategy is sensitive to the suboptimal decisions of low marginal cost players. Ward et al.57 for all ten periods. less volatile market bidding and recharge is reduced at a lower unit cost.37. The analysis of the experimental results indicate that the research hypothesis 1 should be accepted: that is. Market exchange in open and closed call treatments The results of the gains from trade observed in the open and closed call cap-andtrade experiments are presented in Figure 5.05) are estimated using Dunnett’s T3 post hoc comparison (Levine’s. who completely understand and act upon optimal trade and management strategies and payoffs instantaneously. successful bidding behavior by high marginal cost players.

978 Periods 1–5 367 Aggregate recharge (MLs) F(2.Aligning policy and real world settings 121 70.05.3 ANOVA of closed and open call market treatments Periods 1–10 Statistics1 Control2 Open2 Closed2 Aggregate recharge (MLs) F(2.47) = 27.978 385 (65.505) 59. .054* 144 Market price ($) F(2.18) 60 (42. Results indicate that there is no significant difference in the observed mean market price between treatments.22) = 0.88) 83 (20.22) = 0.05) than the control.60) 56.175 Qty traded (MLs) F(2. 3 Mean (standard deviation).681) 61.150 (4.978.000 Net aggregate farm income ($) 60.22) = 27.950) 353 (83. The mean number of recharge units traded observed in both the open and closed call treatments are significantly fewer (p < 0.182) 68 (26.668* 144 Market price ($) F(2.47) = 1.000 45.142 61 Gains from trade ($) F(2.000 65.825 (4. 2 Treatment means (across rows) with the same letter were not statistically different at α = 0.31) 93 (28.5 Aggregate farm income. There is no significant difference between treatments in Table 5.05) 79 (19.05.53) 61 (38. observed in the open and closed call market treatments. cap with no trade 50.47) = 6.83) 67 (21.338) Notes 1 ANOVA tests: * indicates significant difference across treatments at α = 0.257 367 Qty traded (MLs) F(2.22) = 6.729* 56.51) 371 (71.000 Predicted income $56.47) = 0.000 Open call Closed call Figure 5.000 55. but are not significantly different from each other.4458* 56.762 (3.284 61 Gains from trade ($) F(2.40)3 355 (68. including gains from trade.167) 53.000 40.54) 77 (19.584 (3.

2 The analysis of the experimental results indicate that the research hypothesis 2 should be accepted: that is there is a significant difference in the gains from trade of the closed call market treatment compared to the open call treatment.d.4.05) and described by different superscript letters across treatment rows. The social payment treatment involves payments based on individual performance and an additional social payment (equally distributed to all players regardless of contribution) is paid based on aggregate recharge reduction achieved by the group. the volume traded and market price did not converge to a predicted equilibrium value.05) than the control value of regulated. stable market prices and increased convergence to the predicted equilibrium. = 17.175) compared to periods 6–10 (s. In contrast. The uniform price treatment involves player payments made according to individual land management and trading performance only. The communication treatment involves individual and social payments. there is no social payment related to group aggregate reduction. the variance observed in the closed call decreased from periods 1–5 (s.05) are estimated using Dunnett’s T3 post hoc comparison (Levine’s. = 42.825) observed in the open call market trading for periods 1–10. Results of aggregate recharge observed in the uniform and discriminant price auction treatments are compared in Figure 5. prior to the selection of a farm management option. The closed call market is therefore selected as the appropriate market protocol based on the observed significant gains from trade. a uniform price tender with social payment and a uniform price tender with social payment and communication. In a general .d. The market prices observed in the closed call (s.60) were less variable than the open call market. = 49.d. no trade institution. A summary of the ANOVA statistical analysis is presented in Table 5. The continuing variability of observed volumes traded suggests that participants were continuing to explore for an optimal strategy up to the final periods of the experiments. = 21.6. The result indicates that the volumes traded and the market prices in an open call market are more volatile and unpredictable. = 38. Social payment and communication treatment This experiment involved three treatments: a uniform price tender.d. the net income (including gains from trade) observed in the closed call market are significantly greater (p < 0. Significant differences (p < 0. increasing in periods 6–10 (s.d. Results indicate that: 1 In contrast to the open call. Ward et al. There is substantial variance in the market price (s.122 J. the value of aggregate recharge in the cap-and-trade treatments.d. p < 0.165). In addition it includes a discussion forum to craft a voluntary social compact for all players.90) compared to periods 1–5 (s.R.182). = 26. Despite increased experience and practice in making farm and trading decisions.

the unobserved random effects of periods were tested at two factor levels (dependent variable = aggregate recharge (MLs)). level 2 = periods 6–10.47) = 4.99) 440 (73. uniform price. .47) = 4.23) Notes 1 ANOVA tests: * indicates significant difference across treatments at α = 0.6 Recharge units traded in the 70 percent reduction. The two level factor Table 5.794.60) = 2. The first factor used the ten periods as levels: periods did significantly affect observed total recharge (F(9.231* 528 (MLs) Player F(2.66) 30.47) = 1.66 (2.36) 453 (74.Aligning policy and real world settings 123 550 Predicted quantity traded (528 MI) 500 450 400 350 300 250 200 5 Uniform tender 6 Uniform and social 7 Uniform and comm Figure 5.66) 28.05.67 (10. 2 Treatment means (across rows) with the same letter were not statistically different at α = 0. social payment and communication treatments.05.629 35 income ($) 294 (82.46) 32.017). p = 0. The second factor tested the random effect of two period levels: level I = periods 1–5.912* 206 recharge (MLs) Qty traded F(2. social payment and communication tender treatments Periods 1–10 Statistics1 Control2 Uniform2 Uniform + social2 Uniform + communication2 268 (54.49)1 277 (52. 3 Mean (standard deviation).96 (10.4 ANOVA of 70 percent reduction.61)3 437 (80.62) Aggregate F(2. uniform price. linear model.

Results indicate that: 1 All the treatments are described by significantly fewer (p < 0. the provision of a form of social payment. Due to random effects across the ten periods. but characterized by high levels of variance (s. Poe et al.e.268 and the control corrected to 14. Ward et al. does not reduce farm income nor impede the reliability of recharge outcomes compared to individual payments only. finite period. p = 0.62) = 1. p = 0. The experimental results are consistent with Ostrom et al.075: interaction effect of treatment * period: F(7.292. Summary and conclusions This chapter reports on the use of experimental economics to guide and inform the design and pre-implementation testing of an on-ground trial of a recharge cap-and-trade scheme implemented in 2005.543. Mean player payments observed in the uniform price treatment ($28. (1992). There is no significant difference in the mean aggregate recharge revealed in trading in the three treatments although they are significantly less than the 206 MLs predicted in the control. is to invite and remunerate landholders . There is no significant difference between the three experimental treatments. There are currently no obligations for farmers to manage recharge below a maximum threshold. Incorporating an additional social payment based on collective action into the on-ground trial implementation design does not significantly reduce recharge reduction and may encourage increased trial participation as it is aligned with the high levels of social cohesion observed in the catchment. the standard error of the treatment means was corrected to 10. Thus a first step in trialing the trade in exchangeable recharge credits is to establish limits or a cap on recharge. (2004) and Tisdell et al. did not significantly affect observed net income (F(1. In the n player. A feasible approach to developing effective obligations for the trial. significant free riding and rapid and substantial decay in observed public good contributions (Ledyard 1995).124 J.66) were significantly less than the control prediction of $35.R.285). in addition to revealing accurately individual costs of abatement. (2004). public good game. based on cooperative action to achieve aggregate recharge reduction.62 respectively). Player payments in the social payment and communication treatments were not significantly different from the control or the uniform treatment.23 and 10.69) = 6.521. = 10. A mandatory approach is not institutionally possible at least in the short run. The control represents a 70 percent reduction target of 528 units at a recharge value of $109. 2 3 The analysis of the experimental results indicate that the research hypothesis 3 should be rejected: i. game theory predicts little or no public good contribution. in contrast to game theoretic predictions.05) quantities of recharge units exchanged when compared to the 528 units predicted in the control.d.

Ostrom 1998. In both treatments the purchasing strategy of the recharge authority prioritizes achieving recharge targets rather than being constrained by budget or price limits. Gintis 2000) reports the potential for significant divergence from individualistic profit maximizing behavior in small.g. Sociological survey work conducted for this project indicates levels of social cohesion within the community are very high.Aligning policy and real world settings 125 to enter into a contracted obligation through a competitive tender process. We recognize that the results may well be only relevant to the specific experimental context. increased rent seeking and impeded convergence to a predicted equilibrium (expressed as more volatile bidding behavior over the experimental periods).e. with high levels of voluntary social contract adherence and low levels of observed free riding. cohesive communities. These combined findings suggested that a policy approach to reward participation with some form of collective award for reaching an aggregate recharge reduction level might increase trial participation. To test the potential for social payments to enhance willingness to enter into contracts to reduce recharge. Experimental results indicate a significantly (p < 0. Previous research (e. there was relatively little opportunity to obtain additional gains when social payments and communication treatments were . an incentive compatible uniform price tender may result in higher participation. high levels of free riding and decay of initial public contribution). and increase cost effectiveness. The control treatment was a uniform auction with the same total player payoff but no social payment. a uniform and discriminant price tender. rates of convergence toward theoretical predictions were high. Generally. For both the control and social payment treatments. with over 80 percent of the survey respondents indicating involvement in the local Landcare group (Thompson 2004). outcomes were not statistically different across treatments.05) greater quantity of aggregate recharge for the discriminant price tender compared to the uniform price tender and control treatments. (2004) there was no significant effect on the level of recharge reduction observed for the communication treatment. variants of the uniform price auction treatment were run inclusive of a social payment component. Whilst not predicted by game theory (Ledyard 1995) (i. The experiments involved two social payment treatments: one with communication and one without. (2004) and Tisdell et al. Experimental economics was used to compare the performance of alternative tender processes. The analysis supports the research hypothesis that the increased incentive incompatibility of the discriminant tender results in lower levels of recharge amelioration. We suspect that the particular results were likely at least to some extent an artifact of the particular experimental context. An implication of these results for the Bet Bet catchment recharge credit policy design is that in a context where a new tendering process is being introduced. Because participants were able to extract nearly all possible gains to trade even without social payment and communication. In contrast to Poe et al. the results are in accord with the empirical work summarized by Ostrom (1998) and Gintis (2000) and the level of public good contribution reported in inter alia Smith (1991).

context rich and complex information. This feature addresses thin market challenges by harnessing the power of community cohesion to bring in reluctant or undecided participants with high recharge reduction potential properties. The results indicate that the gains from trade observed for the closed call treatment are significantly greater than the control and open call treatments. Additionally. the volatility of observed market prices and quantity traded (expressed as variance) is substantially higher in the open call market. Binmore (1999) and Plott (1995) predict delays in convergence to a theoretical equilibrium when the decision environment requires the processing of novel. The National .R. the objective of the experimental component of the MBI trial was to provide an empirical basis for selecting appropriate instruments to establish reliable recharge reduction obligations and cost effective recharge trading opportunities in the on-ground trial implementation phase.126 J. a closed call auction format has been selected in preference to an open call market structure. Based on experimental results a competitive uniform price tender was selected to elicit and rank the marginal abatement costs of individual participants and establish initial land management obligations. increased market predictability and reduced market volatility. an open call market should lead to more rapid convergence to a predicted equilibrium as participants receive additional market information (the bidding information is publicly disclosed for all players whereas players in a closed call are only informed about the success of their own bids). Conceptually. The intent of the trial is to allow credit/debit positions to be cleared through trade. To understand the implications of credit trade market structures. Based on the contextualized experimental findings of significantly increased gains from trade. The experimental economics reported on here informed recommendations for contractual obligations in three key areas: 1 Multiple year agreements with landholders that involve providing payment in exchange for land management obligations to provide recharge reduction. Market exchange is facilitated using the closed call structure. introduced. a closed call and a theoretical prediction (regulation in absence of trade) were compared. behavioral responses to an experimental open call. 2 3 Acknowledgments This research was funded by the Commonwealth National Action Plan for Salinity and Water Quality Market-based Instruments Pilot Program. Possibly if experiments had been constructed such that individual gains realized with the baseline treatment were less. although not significant. Ultimately. A group performance component of payment that involves withholding available funds unless landholders cumulatively achieve a designated recharge reduction level (or threshold). Ward et al. increased gains from social payment and communication may have been observed. In contrast.

Gatti. J. G. and Lowenstein. preference elicitation and the discovered preference hypothesis. M. which if withheld can adversely affect the value of the transaction with the agency. To minimize the level of farmer’s trade surplus.C. pp. J. S.) Camerer. and Oates. (2004) Behavioral economics: past present and future. Environmental and Resource Economics: 32(1). Adelaide. (2004) Auction design for voluntary conservation programs. farmers possess information about their opportunity costs of recharge reduction actions. and Rabin. Advances in Behavioral Economics. (2005) Preference anomalies. 55–89. 3 When the quality of a groundwater is not readily observable. Bryan. G. Notes 1 Thin markets and the anticipated low market participation rates in the geographically constrained catchment may be associated with price volatility and distortion (Stavins 1995. American Journal of Agricultural Economics: 86(5). The agency possesses information (biophysical and hydrological) about the marginal benefits of specific land management actions. C. Cardenas. Development and Sustainability: 2(3–4). (2000) How do groups solve local commons dilemma? Lessons from experimental economics in the field. Camerer. C. B. K. F16–F24. W. spatial concentration (Tietenburg 1998). CSIRO Land and Water (Policy and Economic Research Unit). Milestone 13 Report for the Onkaparinga Catchment Water Management Board.E. Cason.. a reduced probability of satisfying market needs associated with increased transaction costs (Stavins 1995) and. Connor. Garrod.. the discriminant price tender reveals farmer costs of reduction plus a variable level of trade surplus. T. L. New Jersey: Prentice Hall. unreliable recharge outcomes (Dinar and Howitt 1997). (1999) Why experiment in economics. 4 More detail on the experimental demand and supply schedules is available on request from the authors.J. as a corollary. 2nd edn. W. (2004) Catchment care – developing a tendering process for biodiversity and water quality gains.Aligning policy and real world settings 127 Action Plan for Salinity and Water Quality is a joint initiative between the State and Commonwealth governments. M. References Baumol. 2 A sealed bid in this case attempts to reduce the possibility of collusion to seek excessive profits when a tendering agency is confronted with a small number of land managers in a spatially constrained catchment. L. but not the actual costs of individual reduction actions. in this case at the farm scale. . Lowenstein. and Scriver. Environment. Kampas and White 2003). J. 3–51. 2000. Economic Journal: 109. Binmore. pp. New Jersey: Russel Sage Foundation and Princeton University Press. pp. 305–322. The Theory of Environmental Policy. and Starmer. In (eds. and Gangadharan. In the tender schemes deployed for conservation provision. The Victorian DPI has provided information input into this report. Two anonymous referees provided insightful and helpful comments. Braga. All errors remain the responsibility of the authors.. 12–17. pp. (1988). C. 2003). pp. but not the accruing benefits (as these primarily occur off farm). the agency has tactically withheld information about the benefits of those actions (Stoneham et al.

Victoria. David S.. Folio no. Journal of Economic Perspectives: 3(3). Gangadharan. (2003) A laboratory study of auctions for reducing non-point source pollution. 433–455. N. Journal of Political Economy: 88(3). A.... Clifton. S/04/1845. Harvard Environmental Law Review: 22(2). D. 510–585.E. 555–573.. number 455. Kagel. and Roth. J.E. (1997) Mechanisms for allocation of environmental control cost: empirical tests of acceptability and stability. B. Cason. Revesz. Cambridge: Cambridge University Press. J. pp. In (eds.H. K. Journal of Regulatory Economics: 23(2). Holt. Ecological Economics: 47(2–3). 263–291.) Kagel. Kahneman. (2004) Design and implementation of a land holder agreement for recharge credit trade in the Upper Bet Bet Creek Catchment. J. D. Ledyard. Thomson. (1995) The Handbook of Experimental Economics. U. American Agricultural Economics Association Annual Meeting. Journal of Economic and Environmental Management: 46(2). New Jersey: Princeton University Press. Withers P. 3–109. pp. Latacz-Lohman. (2003) Pollution Trading to Offset New Pollutant Loadings – a Case Study in the Minnesota River Basin. (1979) Prospect theory. (1995) Public goods: a survey of experimental research. Property and Prices. F. A. pp. and Gangadharan. Chermak. C. (1998) The choice of regulatory instruments in environmental policy. pp.O. .) Kagel. pp. CSIRO Land and Water Report. (1997) Auctioning conservation contracts: a theoretical analysis and application. 167–191. T. The Handbook of Experimental Economics. and Duke. Dwyer. Connor. Land Economics: 81(1). Social Science Information: 38(4). Journal of Environmental Management: 49(2).128 J. H. (2005) A laboratory comparison of uniform and discriminant price auctions for reducing non-point source pollution. Farmer. English nature research reports. J. 95–114. and Clifton. In (eds. (1999) The problem of external validity (or “parallelism”) in experimental economics. J. pp.. R. J. R. (2000) Beyond Homo Economicus: evidence from experimental economics. Guala. A. C. R. and White. C. New Jersey: Princeton University Press. L. and Silcock P. and Stavins. Toronto. 135–147. C. Ecological Economics: 35(2). American Journal of Agricultural Economics: 79(2). pp. and Roth. Ward et al. pp. A. 407–418. pp.E. Baldock. (2003) Selecting permit allocation rules for agricultural pollution control: a bargaining solution. M. pp. Kagel. R.R. C. T. 311–322. Dinar. Hahn. 446–71. Gintis. Krause. J. (2004) Water Salt Balance Model for the Bet Bet Catchment. an analysis of decision under risk. 183–203. Canada 27–30 July 2003.M. (1980) Competitive bidding for contracts under alternative auction procedures. J. (2003) The demand for water: consumer response to scarcity. J. D. pp. R. pp. L. A. (2002) Policy mechanisms for the control of diffuse agricultural pollution with particular reference to grant aid. Montreal. Common.W. Econometrica: 47(2). and Tversky. and Van der Hamsvoort. pp. and Brookshire. (1995) Auctions: a survey of experimental research. (1995) Sustainability and Policy: Limits to Economics. Keohane. J. 55–70.. K. Eaton. and Howitt. pp. A. Bendigo Victoria: SKM. (1968) Pollution.. Dales. F. Cason. Fang. (1989) Economic prescriptions for environmental problems: how the patient followed the doctor’s order. 313–367. Canada: University of Toronto Press. Ward. Kampas. and Easter.

Ostrom. (1996) Rational individual behavior in markets and social choice processes: the discovered preference hypothesis. Social Change in Rural Australia.R. In (eds. pp. (1996). 35–45. Lyons. and Gardner. (2003) Can the acid rain program’s cap and trade approach be applied to water quality trading? EPA Conference.. . S. 225–250. (2004) Farmers cooperating in the commons? A study of collective action in salinity management. 1–22.P. and Schmidt. R. Plott. J. Segerson. Ostrom. 237–272. 404–417.) The Rational Foundations of Economic Behavior. Montgomery. pp. and Radner. Washington. F. E. G. Canberra.. C. pp. (2003) Market based instruments – international patterns of adoption. Poe. Walker. p. Schulze. (1998) A behavioral approach to the rational choice theory of collective action. L. Australia. R. October. pp. Mathematics of Operations Research: 6(1). C. Economics Journal: 109 (February). J. pp. DC 1–2 May 2003.) Radner. (1996) Market architectures and institutional test-bedding: an experiment with space station pricing policies. pp. The Handbook of Experimental Economics. R. C.B. W. R. MA: Harvard University Press. K. pp.E. Market-Based Instrument Workshop. American Journal of Agricultural Economics: 86(5). American Political Science Review: 92(1).) Lawrence. Randall. American Political Science Review: 86(2). pp. C. Cambridge: Cambridge University Press.. Simon. In (eds. Amsterdam: North Holland. Economic Inquiry: 20(2). (1982) Markets as economizers of information: experimental examination of the Hayek Hypothesis. (1981) Optimal auction design. Marshall. (1991) Game Theory: Analysis of Conflict. E. (eds. G. MBI Working Group on Market-based Instruments (2002) Investigating new approaches: a review of natural resource management pilots and programs in Australia that use market-based instruments. 1203–1210.J. National Action Plan for Salinity and Water Quality..L. Lowenstein. 165–179. Smith. J. (2004) Putting Auction Theory to Work. K. 111–194.B. C. Journal of Economic Surveys: 15(3). E. G. D. Journal of Economic Behavior and Organization: 31(2). Economics Journal: 109(February). and Horan. K.E. M. Milgrom. Farming styles amongst grape growers in the Sunraysia district.R. Suter.. Cambridge: Cambridge University Press. and Montaz. (1972) Markets in licenses and efficient pollution control programs. A. (2004) Exploring the performance of ambient based policy instruments when non-point source polluters can cooperate. 255–289. A.Aligning policy and real world settings 129 and Roth. pp. Mesiti. W.D. V. and Porter. 271–286.S. Cambridge. G. (1992) Covenants with and without a sword: self governance is possible. 395–418. pp. remaining challenges and emerging approaches. 64–97. Schary.. C. pp. 275.. P. 58–73. and Vossler. Journal of Economic Theory: 5(3). 25–34. H. Colombatto. (1972) Theories of bounded rationality. Plott. New Jersey: Princeton University Press. R. Loomes. (1999) Some lessons from past experiments and some challenges for the future. Decision and Organization. and Vanclay. Canberra. (1999) Experimental economics from the vantage point of behavioral economics. Rockhampton: Central Queensland University. (2001) The economics of non-point pollution control. Myerson. G. In Arrow. Myerson. pp. Ecological Economics: 51(3–4). pp. Shortle.R. Perlman. pp.

Experimental Economics: 5(2). Resources for the Future Discussion Paper 00–09. W09S03. Stavins.L. Young. p. T.. J. property right and regulatory incentives for the conservation of biodiversity. pp. (2003) Attitudes towards water markets: an Australian case study. 133–148. Tradable Permits: Policy Evaluation. pp. Water Resources Research: 40(9). R. CSIRO Client Report. (2004) Understanding the social drivers and impediments to participation in the upper Bet Bet MBI pilot. J. J.130 J. Paris: OECD. pp. and Strappazzon. CSIRO Land and Water Report. Folio no. Department of the Environment. MA: Blackwell. Australia. Society and Natural Resources: 16(1).. Journal of Environmental Economics and Management: 29(2). T. J. Sport and Territories Biodiversity Unit. (1991) Papers in Experimental Economics. S/04/903. Tietenburg. A. A. J. 190–211. T. auctions and competitive sealed tenders. W.) Bromley. 8–37. 91–110. The Handbook of Environmental Economics. Grabosky. DC. R. price based. pp. Canberra. Stoneham. Smith V. Howard..L. New York: Cambridge University Press. and Bromley. E.. pp. Biodiversity Series Paper 9. (2002) Method in experiment: rhetoric and reality. Cambridge. (1998) Ethical influences on the evolution of the US tradable permit approach to air pollution control. and Capon. (1995) Choices without prices without apologies. V. J. .and Ward. T. and McCrone. and Ward. (2000) Bargaining and Market Behavior: Essays in Experimental Economics. pp. D. J. Ha. and Connor. D. December 2004. Ward. University of California. G. (2004) Evaluation of options for recharge credit trade in the Bet Bet Creek catchment.L. Stavins. Thomson. Elix. Tisdell. (2000) Experience with Market-Based Environmental Policy Instruments. Sterman. 61–75.) Tietenburg. (1996) Reimbursing the future: an evaluation of motivational. Ward et al. N. Cambridge University Press. Design and Reform. D. (2003) Actions for conservation contracts: an empirical examination of Victoria’s BushTender trial. New York. Smith. Chapter 3 in Connor. J. (1987) Systems simulation: expectation formation in behavioral simulation models. Australian Journal of Agricultural and Resource Economics: 47(4). Presented at Twenty years of market based instruments for environmental protection: has the promise been realised: Donald Bren School of Environmental Science and Management. In (ed. Victoria... (Collected works). January 2000. Tietenberg. Smith. Design and implementation of a landholder agreement for recharge credit trade in the upper Bet Bet Creek catchment. P. B. Santa Barbara 23–24 August 2003. (Collected Works). V. V. J. J. Behavioural Science: 32(3). (2004) Ex post evaluation of tradable permits: methodological issues and literature review: In (ed. Ecological Economics: 24(2–3). Vatn. R.D. (2004) Impact of communication and information on a heterogeneous closed water catchment environment. Tisdell. N. Cunningham. 477–500. 241–257. Vickery. Stavins. Ward. (1961) Counterspeculation. voluntary. Lambert.. Victoria. Washington. M. J. Chaudri. L. pp. Resources for the Future.R. Journal of Finance: 16(1). (2003) Market based instrument policies: what can we learn from US experience (and related research). and Johnstone.. (1995) Transaction costs and tradable permits.

and baseline-and-credit plans. each firm earns an amount of tradable emission permits for each unit of product output. product output and emissions are higher under the baseline-and-credit plan because of the implicit subsidy given to increases in product output. “Baseline-and-credit plans are theoretically equivalent to a cap-and-trade plan if the cap implicit in the baseline-and-credit plan is fixed and numerically equal to the fixed cap in a capand-trade plan. The two types of market based instruments are cap-and-trade programs.” A baseline-and-credit plan can accommodate a fixed cap by adjusting the permitted emission rate as aggregate product output varies such that the aggregate emissions are held constant. Mestelman. Mestelman. In their theoretical discussion. Buckley. which do not necessarily have such a cap. and Muller Buckley. such as making precise the design of institutions. However. In Buckley et al. identifying potential multiple equilibria and the guiding influence of psychological or social factors. This creates an incentive to expand production because the permit allocations grow with output and function like a subsidy to output. The assumptions behind the theoretical predictions include perfect competition and risk neutrality. In the predictions and the experiment results.6 Discussion Tradable permit markets Dallas Burtraw and Dan Shawhan Five teams of authors have contributed chapters that illustrate the relevance of experiment methods to the analysis of quantity-based tradable permits for environmental protection. generating new hypotheses and illustrating for policy makers how markets for environmental goods can function. Various ones of these chapters illustrate the various purposes to which experimental methods can be applied. which have a cap on aggregate emissions. and Muller seek to test experimentally theoretical predictions about the performance of two types of tradable emission permit markets when each producer can incrementally vary its output capacity but cannot change its emissions per unit of output. The experiments verify the prediction of theory and show that the baseline-and-credit plan leads to a lower level of economic efficiency than cap-and-trade. as long as a producer .’s version of the latter. the authors assert.

and Sutinen One of the lingering concerns about the introduction of the new institution of market based policies into resource markets is the effect on incumbents and market structure. depending in part on the profit function elasticities of small firms relative to those of large firms. and Sutinen demonstrate that a cap-and-trade program for fishing rights can either increase or decrease market concentration. speculation. The role of the output subsidy in tradable performance standards has been characterized as a theoretical oddity by some analysts. Freeman.1 Hence. 2001. the authors’ measure of market concentration is slightly larger in the experiment than in the equilibrium model. this does not lead to a theoretically equivalent result as cap-and-trade. Using those elasticities. Sometimes laboratory experiments provide evidence suggesting that a design of a new institution such as tradable fishing rights is likely to have one particular . adjusting the permitted emission rate in order to maintain aggregate emissions. Similarly. the authors’ chosen measure of market concentration in the experiment is statistically indistinguishable from the equilibrium model’s prediction. because the incentive to increase output still exists. This chapter is a good illustration of the potential role of experimental methods in demonstrating for policy makers a theoretical result in a practical context. In an experiment modeled after a local fishing industry. Shawhan can earn more credits by increasing product output.132 D. Anderson. the authors might find an even more general result were they to repeat their experiments with the aggregate emission cap fixed under their baseline-and-credit plan. profit function elasticities that increase market concentration in the equilibrium model also do so in the experiment. By demonstrating that a cap-and-trade program for simulated fishing rights may decrease rather than increase market concentration in an experiment. Burtraw and D. However. The authors conjecture that concentration in excess of the predicted amount may occur as a result of price volatility.” The experiment supports the approximate accuracy of the predictions of the general equilibrium model. but this experiment shows it has practical implications. and especially the likelihood that it will lead to market consolidation. Fischer 2001). Freeman. as suggested. market power. or “a failure of the market to identify a price within the time it is observed. The authors report that the experiment results approximately match the predictions of a general equilibrium model described by two of them in a separate paper. Profit function elasticities that decrease market concentration in the equilibrium model do so in the experiment as well. Theory and previous simulation models show that the equilibrium will lead to increased output and lower product prices and also lower welfare than under cap-and-trade (Burtraw et al. the chapter may reduce the reported resistance of small fishers to a cap-and-trade system. especially if further research indicates that the profit function elasticities in the industry are conducive to decreased industry concentration under a cap-and-trade system. Anderson.

“buyers. the experiment in the chapter by Anderson. In another treatment labeled the “buyer liability” scheme. In the experiment. In contrast. A third treatment labeled the “buyer liability with refund” . In the experiment. A concern under the Kyoto Protocol is that nations might sell permits in excess of actual emission reductions (henceforth. sellers are liable if their permits are found invalid. this finding illustrates the complementarity of experimental methods and other research methods. Players run the risk of detection and fine if production exceeds permitted levels. each subject represents a country. but also because some approaches might do a better job of inducing participation of developing countries in the international regime. In one treatment. sellers cannot be fined. The question that is addressed is how performance of the trading system varies with the assignment of liability for non-compliance. As a result. The authors model the program in their chapter after the Kyoto Protocol for reduction of greenhouse gas emissions through emission trading. The question of liability is important not only for ensuring the integrity of environmental agreements. some or all of the players have a one-insix chance of being “monitored” and fined a set amount for each unit produced without a corresponding valid permit. In all three experimental treatments. which is a feature that resembles technological uncertainty. Freeman. a country might initiate a project for biomass sequestration and sell allowances based on expected capture of carbon. Such evidence can provide some confidence about likely outcomes. A key question in the implementation of the Protocol is how to assign liability for the possibility that emission allowances are not backed up by valid emission reductions. players can intentionally produce in excess of permitted levels. Godby and Shogren Godby and Shogren report on an experiment designed to compare the performance of three different enforcement schemes for a multiparty cap-and-trade program in which the only feasible enforcement options may be weak ones.” can acquire permits only by purchase. and Sutinen usefully varies the fundamental parameters of the problem and proves the positive hypothesis that a particular treatment can have either of two qualitatively different impacts. The actual market outcome will depend on the underlying fundamental parameters. “invalid permits”). but buyers can be fined if they buy permits that turn out to be invalid.Discussion 133 impact rather than another. perhaps due to a failure to operate investments that lead to intended emission reductions. while the other participants. In addition. a stochastic element leads actual production to vary slightly from anticipated production. but cannot prove that other outcomes cannot happen. For example. The authors suggest an interpretation that identifies sellers as developing nations and buyers as the industrialized nations. “Sellers” are endowed with tradable production permits (analogous to emission permits) at the beginning of each period. but over time the country might fail to maintain adequately the project and the anticipated sequestration goal may not be realized.

as the identity of a seller who markets invalid permits is revealed. Duke. It could do so by fining a seller. and Cason investigate linked markets for water and salinity and the prospects for trading of salinity . Gangadharan. Burtraw and D. through the Clean Development Mechanism established by the Kyoto Protocol. Unlike the previous chapter that treated costs in a heterogeneous way. Godby and Shogren point out that some previous authors claim that buyer liability could encourage the participation of more countries as venues for greenhouse gas emission reductions. Therefore. “After [invalid] permit sales were cancelled. The authors state this finding strongly: “Buyer liability with relatively weak enforcement renders emission trading inadequate by subverting the goals of Kyoto – creating less environmental protection at greater economic cost. Shawhan scheme is the same as the “buyer liability” scheme except that if a permit is canceled. A central assertion of advocates of buyer liability is that it will influence participation decisions of developing countries in a positive way. and Cason Two chapters address the question of salinity and water use in Australia. In Godby and Shogren’s buyer liability treatment. Gangadharan. the authors’ strongly stated conclusion should be weighed with caution because seller liability may lead to less overall participation by developing countries and. in each buyer liability scheme than in the seller liability scheme. but this assertion is not examined.134 D. there is another version of buyer liability that could be considered. reputation plays a role. emissions are higher. Just as with the underlying theoretical results. less environmental protection overall – a possibility that is precluded in the experiment. in which both sellers and buyers are liable for sellers’ excess emissions. Duke. both of these chapters treat costs as homogeneous. In the authors’ theoretical predictions and in the experiment results. Godby and Shogren’s experiment results support their theoretical prediction that emissions will be higher. As the authors explain. it also provides an example of the need to interpret results with caution. The chapter inspires possible extensions. An alternative to seller liability and buyer liability is joint liability. In each treatment. Increasing the probability of being monitored did not change these relationships. if monitored. neither buyers nor sellers are liable for a seller’s production in excess of its initial permit endowment. and economic efficiency lower. and economic efficiency lower.” An alternative buyer liability treatment could eliminate this exemption. conceivably. for any excess production that remained after cancellation of any invalid permit sales it had made. under buyer liability than under seller liability. the experimental results depend on the limitations imposed for the analysis. sellers were not fined even if their production exceeded their permits.” While this chapter provides a striking example of how experimental methods can be used to verify and illustrate theoretical predictions. In addition. the buyer receives back from the seller the money she paid for that permit. This option could be considered theoretically and experimentally.

The pilot studies used students and did not reveal the commodity or other aspects of the environmental problem in the simulations.Discussion 135 rights in the Murray Darling Basin. some buyers choose to pay the fine and to avoid the subjective transaction costs of trading in the salt market. Even in the context of a quantity instrument.” One of the widely recognized values of experimental methods is the generation of new hypotheses. and this chapter does so. one might be losing the defining features of a problem if one strips away all detail entirely. However. and in so doing place the experimental literature within a broader context in environmental economics. However. The authors find the safety valve affects the behavior of participants in the qualitative direction that was anticipated. the problem structure allows for a kind of safety valve. which they label as a monetary fine. the authors choose to frame their pilot experiments in an entirely abstract manner.” This leads the authors to “conjecture that when permit prices are within one to two experimental dollars of 20. but it appears to affect it “too much. An important part of the work summarized in Duke et al. perhaps should be re-examined in every study. The pilots used an oral auction. is the building of a link between natural science and social science. even though mainstream. Were the environmental costs more continuous. one might be compelled to address a larger set of issues about instrument choice. The authors state that the justification for a quantity based approach to the salinity problem is the presumption of “some safe minimum standard” below which salinity does not impose significant harm. This fine represents the engineering cost of an outside option that the government could institute – an abatement technology that is described as the government’s “salt interception cost. This approach follows the main thread in the academic literature. With too much detail and complexity. the instructions given to participants were complicated. This is one of the only chapters to acknowledge explicitly the justification of allowances (permits) as a policy instrument. wherein mistakes and misunderstandings were identified and even corrected within the session. although the experimental parameters approximate this technical information including the economic factors that affect the industry. To know whether this applies or not in a specific context one might vary the treatment to explore how contextual detail affects the behavior of participants. .” Alternatively. and above which it is very harmful. without a specific context for the problem that is modeled in the experiments. The chapter reports on the six pilot experimental sessions that were an attempt to characterize the relevant markets and institutions. it appears such behavior could be the result of the introduction of an asymmetric payoff function when payoffs in one direction are censored by the safety valve mechanism. Some subjects did not understand the instructions and took the unintended suggestions in the instructions as data in forming beliefs about the reasonable price of salt permits. Nonetheless. The authors begin their study with a series of workshops in the region to gather information about scientific and economic issues that face the agricultural community. The merits of this “context free” setting. one can question what if anything is being learned in an experiment.

One variation allowed for communication among participants. and Tisdell The second chapter on salinity credit trading addresses the Bet Bet dryland farming community in north central Victoria. The results conformed to expectations. This finding is a bit unsubstantiated. compared to the absence of the social payment. The availability of a supplementary social payment could increase voluntary participation in practice if there are important social norms that support group activity. a market clearing price is calculated. The third hypothesis addressed the role of a supplementary social payment to the group for aggregate environmental performance above a threshold level. and the authors endorse the strategy of using social payments. In a closed market bids to buy and sell are sealed. and successful traders are informed. However. with increased environmental performance. In an open market information is provided about the bid structure. in which farm income can be calculated directly. rather than by the desire to test theory per se. Three hypotheses are tested directly. based on the previous literature. although variability of payments increased. it is based on the absence of evidence that the social payment causes a problem and it hinges on a continued conjecture that the social payment may encourage increased participation. Concern about free riding is rejected. Moreover. which is not demonstrated. In contrast to Duke et al. The experiments supported the latter conjecture. this chapter embeds the technical information that characterizes the salinity problem in a context-rich setting for its experiments. Shawhan Ward. This research is motivated by the desire to identify the performance of actual market based instruments in an applied context to guide the design of institutions. Australia. the reader has to work hard to dissect the finding because the . and a more cost effective outcome. but no further information on the bid structure or outcome is revealed. Connor. a less volatile market. A uniform price auction led to a significantly different outcome. A challenge in this project is to balance the amount of information that is needed in the experiment in order to develop the resemblance of an identifiable institution against the potential confusion that can emerge from the introduction of detail that may be irrelevant to a specific research hypothesis. or it could lead to a reduction in participation if individual incentives are scaled down and there is important free riding. One might conjecture that an open market design would allow participants to learn faster and for a more efficient equilibrium to be achieved sooner. other authors have suggested that the additional information can introduce complexity about issues beyond the payoff facing an individual and it can slow learning about the best performance for that individual in the market. The results find that environmental performance and mean payments to participants are unaffected. Burtraw and D. and found in favor of a closed market design. The second hypothesis compared a closed market design with an open design. The authors do so by comparing the behavior of participants in the experiments with a regulatory control. The authors appear to have traversed this dangerous slope successfully.136 D. One dealt with discriminating versus uniform price auctions.

An implication of factors favoring optimal decision making One way in which an experiment can differ from a comparable “real world” situation is the relative extent to which optimality of decisions is favored in the experiment versus in the real world situation. For example. chapter may be one of the gems in this volume – it is strongly motivated and incorporates helpful literature throughout to visit some of the most important conceptual issues in experimental methods. There was a quiz that participants had to pass before participating. rather than effects of tightly constructed treatments. so the conclusions could conceivably be artifacts of confusion or lack of understanding. In fact. If one believes that optimality is more favored in the experiment than in the real world situation. Perhaps a lesson from this reading is that if a complex or non-intuitive setting is chosen. then such a belief may play a useful role in using the experiment to update expectations about what will happen in the real world situation. The chapter’s conclusions hinge on such things. by which they might mean participants with low marginal abatement costs submitted very highly priced offers or something else similarly irrational. but we do not know what was in that quiz or what portion of answers had to be correct. even greater attention should be directed at evaluating the understanding of participants. and Muller. There were no verbal instructions other than answering specific questions. The instructions given participants seem unclear. for . This is a matter of judgment and opinion. The reported results are consistent with these possibilities. and increase recharge compared with what would have happened if the subjects had understood.Discussion 137 statement of the hypothesis ex ante and the statement of the hypothesis in the results appear to contradict each other. Applying this to the experiment by Buckley. or alternatively if one believes that the real world situation does. This would be consistent with confusion. reduce bid and offer volume. the possibility of subject confusion may cast some doubt on the authors’ interpretation of the results. although the issues that are addressed are rich. on the contrary. The authors mention suboptimal bidding by low-marginal-cost players. then one should expect decisions to be less optimal in the real world situation than they were in the experiment. the Ward et al. Confusion could have a large effect on the results. since profits were in fact low and recharge was in fact high compared with the theoretical predictions. the reader has to work harder than one would like to grasp the issues that are presented. then one should expect decisions to be more optimal in the real world situation than in the experiment. it could reduce subject profits. Moreover. one believes that optimality is more favored in the real world situation than in the experiment because of higher stakes and greater experience of real world emission credit traders. perhaps through followup interviews and debriefing. If one believes that the experiment favors optimal decision making more. Mestelman. perhaps because of a simply constructed choice context. If. However.

such an experiment might serve other purposes such as validating underlying predictions of theory.” permit or final-product prices that are inefficiently high. an experiment that plausibly favors optimal decision making more than does some important comparable real world situation may not be able to establish convincingly that outcomes in that real world situation will be within some tolerance of the optimum.138 D.) • • What is the performance of different market designs for unconventional tradable permits. one might conclude that their finding of irrationally low emissions and high permit banking in their experimental tradable performance standards market is not a reliable indicator that the same will occur in a real world tradable performance standard market. minimizing “mistakes” by participants. or volatile? (For a possible empirical example. indicating that outcomes will be far from the theoretical optimum. To various degrees. regulators. testing notions that might go beyond the limits of established theoretical models. see Joskow and Kahn 2002. market “bubbles. Naturally. Shawhan example. We have mentioned a few elsewhere in this commentary. even if the outcomes are well within that tolerance in the experiment. we see these chapters testing theoretical predictions. minimizing cost of participation. Burtraw and D. and demonstrating theoretical findings for policy makers. Conclusion As a set. such as maximizing perceptions of fairness. the chapters in this Part illustrate the various ways that experimental methods can make a unique contribution to economic analysis. especially as such designs more precisely target environmental measures but in doing so introduce increasing levels of complexity? One example is locational ambient pollution permits (see Montgomery 1972). it may be useful to consider the above in choosing or designing an experiment. A few additional experimental research ideas These five chapters sample from a set of topics related to tradable permits that might be researched experimentally. Here are a few more. • What auction designs will best achieve the goals of government officials. However. . For example. generating new hypotheses. identifying unanticipated outcomes or generating new hypotheses. low. and maximizing economic efficiency and government revenue? To what extent are different tradable permit market designs susceptible to the following: • • • market power.

D.” We would characterize this phenomenon differently.. Fischer. Online.mit. Joskow. As Buckley.php?id=548 (accessed 12 September 2006). Lawrence H. 1999). especially variations that may apply outside of the laboratory? This is an issue facing the researcher in any context. and Muller point out. “The Effect of Allowance Allocation on the Cost of Carbon Emission Trading. experiments can also be reused as participatory learning exercises for policy makers and stakeholders.” Mimeo dated 4 February 2002.Discussion 139 and stakeholders. 2001. but experimental methods are especially well suited to test the robustness of theory by varying the margins of the analysis. Paul and Edward Kahn. 1999. Roberton C. Dallas. available at: econ-www. Parry. “The Cost-Effectiveness of Alternative Instruments for Environmental Protection in a Second-Best Setting. 2001. and Anthony Paul. Note 1 The authors interpret the theory to say that if the emission rate is adjusted as output changes so as to maintain a fixed aggregate level of emissions this raises industry costs “due to unnecessarily tight restrictions on emitting firms. Mestelman. but also empower the researcher to speak with confidence about specific aspects of a problem. Ranjit Bharvirkar.” Resources for the Future Discussion Paper 01–22. which is the source of inefficiency in a general equilibrium context (Goulder et al. “Rebating Environmental Policy Revenues: Output-based Allocation and Tradable Performance Standards. and Dallas Burtraw. How robust is each finding? Does it survive variations in the margins of the analysis such as the institutional details of constraints and framings.” Journal of Public Economics 72(3): 329–360. . Williams III.” Resources for the Future Discussion Paper 01–30. Carolyn.edu/faculty/ download_pdf.. The output subsidy implicit in the baseline-and-credit approach leads to more output than is efficient. Montgomery. 1972. The boundaries of the experiment framework necessitate caution in formulating general conclusions about real institutions outside the experimental context. “Markets in Licenses and Efficient Pollution Control Programs.” Journal of Economic Theory 5(3): 395–418. Ian W. 2002. References Burtraw. “A Quantitative Analysis of Pricing Behavior in California’s Wholesale Electricity Market During Summer 2000: The Final Word. W.H. Karen Palmer. Goulder.

.

Part II Common property and public goods .

.

there are two steady states. In this chapter we use a simple neoclassical growth model with multiple equilibria to investigate the mechanism by which non-binding communication results in lower equilibrium resource extraction. in this dynamic setup. 1991. we model the extraction of a natural renewable resource using a coordination game environment rather than the typical prisoners’ dilemma environment. then the regeneration rate is lower. Perhaps communication results in an increased sense of group identity. Ostrom and Walker.1 There are two innovative aspects in this chapter. agents attain a lower overall resource stock and consumption. 1992.. has been associated with the resolution of coordination failures (see Cooper et al. or a necessity to seek or avoid verbal approval or reprimand when promises of cooperation are fulfilled or violated.. 2004). Consider a case in which the available stock of fish determines its regeneration rate. Most experimental papers on communication and coordination have analyzed .7 Communication and the extraction of natural renewable resources with threshold externalities C. Thus. First. communication is expected to reduce the uncertainty of what other players are likely to do and hence facilitate coordination in the better equilibrium. If fishing or extraction is such that the stock does not surpass the threshold level. et al. We use a growth model because it provides an adequate dynamic framework for modeling extraction of a natural resource with threshold externalities. 1988. In contrast. Clark et al. 1994. Second. one of which is optimal. Cardenas et al. In the lower equilibrium. Ostrom. In social dilemma games. In simple coordination games. a higher regeneration rate is achieved. the reasons why communication works are still unclear. which renders a lower equilibrium. Mónica Capra and Tomomi Tanaka Introduction Non-binding communication. an enhancement of normative orientations toward cooperation. In equilibrium. 2001) and social dilemmas in both laboratory and field experiments (see Isaac and Walker. higher resource stock and consumption can be attained.. we use a dynamic setup to see the effects of communication on equilibrium selection. or cheap talk.. when extraction is such that enough fish stock is left to pass the threshold. when the resource surpasses the threshold. One specific example is fisheries.

then the value if A is A .144 C. Theoretical framework In the neoclassical growth model a representative agent maximizes his lifetime utility given by: ∑ (1 + ρ ) t =0 ∞ −t U (C t ) subject to the resource constraint given by the expression Ct + Kt+1 ≤ AF(Kt) + (1 – δ)Kt.25. In contrast. Ct. we are interested in discovering the circumstances under which communication works. By analyzing the content of the messages. Tanaka static games. ρ equal to 0. These authors add institutions – rules that govern subjects’ choices – to the basic Ramsey–Cass–Koopmans (RCK) neoclassical growth model with a resource externality. In addition. With such a perishable resource. we study aspects of the content of messages that have not been included in previous works. K would represent a renewable resource such as fish.M. In our resource extraction environment. This is implemented by letting the parameter A take different values depending on the amount of the resource that is available after consumption. where U is the utility of consumption. Multiple steady states result from a threshold externality. In contrast. Capra and T. δ = 1).80. at time t. The parameters ρ and δ represent the discount and depreciation rates. we used a random ending rule. we discuss the procedures and the results from the communication treatment. and a hybrid treatment in which subjects were allowed to communicate and vote. which is equivalent to having a value of the discount parameter. if K is lower than – – the threshold K . We implement communication by allowing subjects to send unrestricted messages to each other in real time.e.’s (2006). by creating a dynamic environment. Thus. respectively. the probability of continuation was 0. To implement infinite horizons A A in the laboratory. . In this chapter. K . before any extraction decisions are made. If the resource stock is higher – – than a threshold level. there is 100 percent depreciation (i. Experimental design Our design is based on Capra et al. voting. At any given period. then A is equal to – with A > – .2 Capra et al. our study attempts to answer the following question: what “kind” of messages can eliminate strategic uncertainty in a dynamic environment? The data from our experiments suggest that messages that contain positive feedback or verbal approval are important for successful coordination. Technology is given by AF(Kt) and Kt is capital stock (the amount of the resource). (2006) use three institutions in their paper: communication. we make the decision context much more realistic.

In our experiments. email is not as effective as a chat room because communication is not in real time. Communication In our experiments. and six) were conducted at Caltech.Threshold externalities 145 Table 7. these body and tone messages may be hard for the experimenter to interpret and even notice. and the resource stock. In a chat room environment. they can use body expressions and voice intonations to their conversation that add content to their messages. For example. In addition.77 if K ≥ 31 A ρ = 0. A lifetime consists of one or more periods.4 Table 7. On the other hand. there is some evidence that real time computer communication (chat room) is as effective as face-to-face communication.” A horizon represents a lifetime.1 Parameters of the model U(Ct ) = 400Ct – 2Ct2 F(Kt ) = Kt0. subjects were allowed to send unrestricted messages to each other in real time in a “chat room” environment before making trading and consumption decisions. CH = 70. and K H = 45 and CL* = 16 and * 3 KL = 9 when the threshold was not surpassed. Rocco (1996) showed that face-to-face communication and chats are equally effective at resolving social dilemmas. The sessions lasted no more than 3 hours and earnings ranged from $11.56 to $44. a chat room environment is not as realistic and “dynamic” as face-to-face communication. K*. – * * when the threshold.1 and under the assumption that the markets are competitive. Experimental procedures We organized a total of six sessions. the experimenter has a less ambiguous way to record and interpret messages. Because of the random ending rule. there is a lag between the time the message is composed and sent and the time it is received and read. There are advantages and disadvantages of using this form of communication vis-à-vis email or face-to-face. C*. Clearly.88 if K < 31 − = 16. each session consisted of several “horizons. five. K . In each session.5 Kt = 31 ) A − = 7.25 δ =1 For the parameter values described in Table 7.2 below shows the number of horizons and periods for each session.24. However. three sessions (sessions one. and three) were conducted at Emory University and three sessions (sessions four. . five subjects participated in the experiment. When people can talk and see each other. was surpassed. two. it is possible to derive the equilibrium levels of the resource consumption. We believe that the benefits from the added control in the computer setup outweigh the costs incurred from sacrificing reality.

H3(1). H3(5). Because they were not identical.1 shows the aggregate production function and the kink generated by the resource externality.e. but in all sessions. H2(4). H4(4) H1(6). H3(6).146 C. For each unit sold. H2(7) H1(8). The individual demand schedules can be found in the instructions (available upon request).M. The latter was fixed (perfectly inelastic) at a quantity equal to the aggregate amount of the resource. Notice that the production functions were lower when the threshold level of the resource was not attained. participants traded units of the resource in a call market. H4(2) At the beginning of each session. subjects sat in their computer terminals. they benefited from trading in the market. H2(7). the subject made an amount equal to the market price. plus a communication stage that preceded all decisions. H5(4) H1(3). subjects were asked to submit their demand schedules by specifying the quantity and price of each unit of the resource. subjects were heterogeneous and differed in their valuation of the resource and their individual production functions. subjects participated in a three-period practice exercise that did not count towards their earnings. . H4(3). In the communication stage. In each of our sessions. For each unit bought in the market. Their differing production functions and the aggregate production function are shown in Figure 7. each subject was endowed with five units of the resource and 10. an amount equal to the market price was subtracted from the initial endowment of tokens. The last panel of Figure 7. where participants could chat with each other in real time.2 Number of periods per horizon Session 1 2 3 4 5 6 Horizon (periods) H1(7). H2(1). Tanaka Table 7. H5(6) H1(11). H2(2). Thus. The aggregate demand was then derived by adding each subject’s individual demand. After the instructions were read aloud. H3(3). Capra and T. trading determined the amount of the resource available to each of them in the current period. before stage one of the experiment. H3(1). H4(2). In each horizon. In our experiments.1. When the call market opened. The market price of the resource was determined by the interaction between the aggregate demand and the aggregate supply. the initial conditions were such that the threshold level of the resource was not surpassed (i. H5(2) H1(3). In stage one.000 experimental tokens (called Yen). the computer displayed a chat room screen. Subjects were given 180 seconds to chat. H4(5). communication lasted less then 2 minutes. there was a total of 25 units of the resource at the beginning of each horizon). five subjects participated in multiple periods and in each period they faced a two-stage decision problem (stage one and stage two). H2(1).

1 Production functions. .60 50 40 Agent 1’s production function Above threshold Output 30 20 10 0 0 5 10 15 Units of input 40 35 30 25 Agent 2’s production function Above threshold 20 25 30 Below threshold Output Below threshold 20 15 10 5 0 0 5 10 15 Units of input 40 35 30 25 Agent 3’s production function Above threshold 20 25 30 Output 20 15 10 5 0 0 5 10 15 Below threshold 20 25 30 Units of input Figure 7.

45 40 35 30 Agent 4’s production function Above threshold Output 25 20 15 10 5 0 0 5 10 15 Units of input 40 35 30 25 Above threshold Agent 5’s production function 20 25 30 Below threshold Output 20 15 10 5 0 0 5 10 Below threshold 15 Units of input 20 25 30 220 200 180 160 140 120 100 80 60 40 20 0 0 25 Aggregate production function Output 50 75 Units of input 100 125 150 Figure 7.1 continued. .

Results In Lei and Noussair (2003) and Capra et al. coordination was partially successful because the threshold level of the resource was surpassed in 80 70 60 50 40 30 20 10 0 H1 H1 H2 Session 1 H2 Session 2 H3 H2 H4 H5 Session 4 Session 3 H2 H3 H1 H1 H4 H2 H3 H4 H5 H1 H3 Session 5 H5 H4 H1 H2 Session 6 H3 H4 Equil (H) Threshold Equil (L) Figure 7. Thus. consume more). the threshold was not surpassed. .5 Once the stage two decisions were made. In our experiment. there were seven periods. The resource left over then was mapped by the production function into units available for the next period.2 shows the average amount of the resource in each horizon for all six sessions. long run) that is often faced when making natural resource exploitation decisions. From Figure 7. if too many units were consumed in a given period. our decision task captured the tradeoff between current and future consumption (short run vs. (2006) an economy without institutions (e. However. This renders lower earnings for all participants in the economy. Weighting each period equally. In the first horizon (H1) of the first session.2 we observe that in sessions one and two. Subjects derived utility. the results are mixed. from the extraction or consumption of the resource. In contrast. the average resource amount remaining after consumption was 10. just above the low equilibrium level of nine units. and on how many units to save for future extraction. the resource could not grow at a pace that would make all better off in the future (i.2 Average resource amount (all sessions). each subject decided on how many of the resource (fish. and higher than the high equilibrium level of 45 units. the computer calculated the amount of the resource left over to determine whether the threshold level of the resource was surpassed. In this horizon. no communication) unambiguously converges to the “inferior” equilibrium. in the second horizon (H2).e. for example) to extract or consume. Figure 7. This amount is higher than the threshold (31 units).Threshold externalities 149 In stage two. This process was repeated in each period until the random end of the horizon.14 units.29 units.g. the average amount of the resource across periods left after consumption was 71. which later is translated into earnings. for example. and the threshold level of the resource was not surpassed. where subjects are allowed to communicate freely in a chat room environment.

. There were four horizons in this session. subjects face a tradeoff: consume today or save for tomorrow. In session three. Thus. The numbers above the marker represent the number of people whose messages contained language related to extraction. Most subjects in this session were either unaware of the coordination problem or unwilling to express awareness. respectively. it is important that subjects express through relevant messages the need to coordinate actions so as to restrain extraction and make sure that they pass the threshold. In contrast. he said: man . when chatting.150 C. got to save some. The left panel of Figure 7. There were only two instances where “one” subject expressed awareness of the problem and attempted to exchange information. Analysis of the chat transcripts The basic requirement for communication to work is that people exchange information relevant to the coordination problem or that they share knowledge. but not in all horizons. There are two reasons for why communication may “fail. each horizon is separated in the figure by a space. subjects do not know that all know that too much extraction is worse for all. Each marker represents the stock of the resource available in that period. t7. All except two sentences from the transcripts in session three excluded relevant messages. The next subsection reports the analysis of the transcripts. subjects successfully restrained consumption in all horizons. . (2) There is no common knowledge of awareness. In sessions three and four subjects were unable to coordinate their actions as to restrict extraction. but is clearly not a sufficient condition for successful coordination. no subject brings up this issue in their conversation. The horizontal line represents the threshold level of the resource. t3. This divergent pattern of the data tells us that communication does not resolve strategic uncertainty unambiguously. . In our setup. subject three said: K just keep getting smaller in the economy. in sessions five and six. there were two sessions where coordination failed: sessions three and four. we hope to determine whether these two conditions or additional ones are sufficient conditions for successful coordination. By looking at the chat transcripts. However. H4. In our experiments. . . Tanaka some. Capra and T. saving only makes sense when enough people save enough of the resource.e. that is.” (1) There is no awareness or expression of awareness for the need to coordinate actions. coordination or anything related to the coordination problem in that period (i. relevant messages). don’t consume all your K . in none of the horizons were they able to pass the threshold. This happened in periods seven and three of the second and last horizons.3 shows a graph of the resource stock in session three that was left over after extraction. consumption. This is what the “aware” subject said: H2. there was little evidence of awareness and of expression of awareness. but they failed for different reasons. Just the existence of an institution that allows subjects to communicate may be a necessary condition.M. that is. Averaging across periods.

Thus.” In this case. Indeed. there was an attempt to coordinate. The subjects were successful at passing the threshold. as expected.” “good. In the first period of this horizon. there was little relevant content in the messages sent by the subjects.3). For example. two players were responsible for 89 percent of the messages. Only subject four followed up. Only nine messages (11 percent) were sent by subjects one.4 shows the stock of the resource in sessions one and two. we can argue that coordination failed because of the lack of common knowledge of awareness. In contrast. are awareness and common knowledge of awareness sufficient conditions for eliminating strategic uncertainty in our environment? Figure 7. out of 84 messages sent by all subjects in seven periods. One of these subjects also took upon himself to convince aggressively others of the benefits from restraining extraction. successful coordination should require that most players express their agreement with the proposal to restrain extraction or keep the resource above the threshold level. subjects were able to restrain extraction in all . However. Thus. and five altogether. two other subjects expressed agreement to this proposal by sending relevant messages. awareness becomes common knowledge and strategic uncertainty should be reduced.3 Resource stock and number of subjects with relevant messages in each period and horizon. However. which resulted in more extraction. one subject expressed awareness and proposed lowering extraction.4). his messages had a weak echo. in the first horizon of this session. This happened despite the fact that there were two “aware” subjects who immediately expressed awareness. but their communication later collapsed. In the third horizon.” and “ok. in none of the periods of session four were subjects able to pass the threshold (see right panel of Figure 7. By communicating their agreement through relevant messages. three. Subject four was responsible for 23 (27 percent) of the messages. awareness and agreement were not enough to sustain coordination in session two of our experiment. Thus. awareness of the benefits from cooperation is not enough. In the first horizon of session two. As in session three. Clearly. The reasons for why communication failed in session four are quite different. the messages were not followed up with any relevant action or comment. 52 (62 percent) were sent by subject two. although communication in the first horizon of session one failed (see the right panel of Figure 7.Threshold externalities 151 100 80 60 40 20 0 0 0 Session three 100 80 Session four Threshold 0 0 0 0 0 000 0 00 00 0 0 01 1 60 40 20 0 Threshold 3 2 2 0 4 2 0 1 3 2 0 3 0 0 Figure 7. subjects must know that others know. Subject one said: “hihi. The horizon converged to the low equilibrium because communication failed.

In horizon one of session five. This conversation followed successful coordination in the first period of the second horizon (H2. Again. subjects were able to maintain the average resource above the threshold in all of the horizons. However. Capra and T. subject two: t2. positive feedback). awareness.5 Resource stock and number of subjects with relevant messages in each period and horizon. periods of the second horizon.M. seems to have been responsible for successful communication in the second horizon of session one. and acknowledgment of the benefits of coordination. subject three: SEE! good work. In the second horizon subjects not only expressed awareness and agreed on a proposed strategy.5 show the resource stocks in each period and horizon of sessions five and six. In sessions five and six. In these sessions the three ingredients.152 100 80 60 40 20 0 C. t1). subjects exchanged information 100 80 60 40 20 0 4 5 0 4 4 Session five 100 80 60 40 20 0 Session six 5 3 0 0 0 0 0 0 0 0 0 00 5 0 0 3 00 0 0 4 0 0 0 0 0 0 0 0 Threshold Threshold Figure 7. they also acknowledged that the strategy worked.4 Resource stock and number of subjects with relevant messages in each period and horizon. . The left and right panels of Figure 7. we should observe a similar pattern in the sessions where communication worked. We have included below an excerpt of the communication that shows the kinds of messages that imply acknowledgment (i. in addition to awareness and majority agreement. Tanaka Session two 100 80 60 40 20 0 Session one 5 2 3 0 1 0 0 3 4 2 1 0 0 1 3 0 3 1 2 1 0 0 00 1 2 0 0 0 3 2 0 Threshold Threshold Figure 7. wanna do it again? let’s let it keep growing. subject three: t2. majority agreement to coordinate.e. subject four: t2. were present. t2. respectively. if acknowledgment was important for successful coordination. the sequence of messages was the following: in the first period. the positive feedback acknowledging the benefits from coordinating actions. for example.

” and other subjects also expressed similar messages. subject three: t1. t2. wake me up when it starts. subject five: t2. right. they did not need to continue sending messages where consumption information was included. subject two: don’t consume the k. drill = consume 1? whee. t1. in the third period of the horizon. sure. subject three: so next round we try to keep? t2. but the resource did not pass the threshold. t2. t2. then in the second period. subject five said “awesome” and “I think we know what we should do by now. however. A similar pattern of messages was observed in session six. yep. in period one. subject five: yes. subject three: t1. this subject did not dissent when player three advised to keep the resource high and asked: “no dissent?” Finally. keeping lots is a good strategy. subject four: t1. subject five: nonsense.Threshold externalities 153 about the benefits of coordinating actions.” The rest of the period two conversation is shown below: t2. probabilistically not nearly as good as round 1. subject three: nowhere near.” subject one said “wow. This explains the many zeroes in the last horizons. for the first round of a set (we’ll probably do this again). An interesting issue here is that once subjects surpassed the threshold for a couple of periods. yes. After successful coordination in the first period of the first horizon. subject five: this is markov. t2. subject five: t1. should try to keep and not consume more than one or two. subject five: yes. lets see if we can trust another. t2. subject three: I think we know what we should do by now.” “awesome guys. t2. . subject four: barely. zzz. subject three: t1. t2.” This is acknowledgment of success or positive feedback. subject four: t1. its already round 2. Below is an excerpt of the process preceding the drop in relevant messages. players expressed satisfaction with the results. For example. only subject one did not express his agreement in period two. subject five: all keep. by 1. subject four: but yeah. subject four: well it’s round two. t2. subject four said: “umm looks like we didn’t get above 30 this time again. t2. t2. subject one: t1. subject four: def. Notice that most subjects also express agreement with the proposal not to consume too much. subject two: t1. subject three: did we at least increase total K? t2. subject four said “yay. right.

once subjects establish effective ways of coordination.12 0. and (5) the number of positive feedback messages in period t.08 0. Each session and each horizon is assumed to have a random effect.78 1.00 0.6 The third and fourth variables capture awareness and agreement. This implies that coordination efforts need to be repeated at least until successful coordination is consolidated. Kt . (3) the number of subjects with relevant messages in period t. However.02 0.14 0. coordination or anything related to the coordination problem in that period) has a positive effect on resource stock.57 0. (4) the number of subjects with relevant messages in the previous period t – 1.e.154 C. expressions of approval or positive feedback have positive effects on the resource stock. language that contains messages of extraction. for instance. they tend to stop sending relevant messages.53 0.3 Random effects GLS estimation of the effects of communication on the levels of resource stock Independent variables R2 = 0.31 2. The total number of messages had a negative impact on the level of resource stock.90 Std Error 0. the number of subjects who sent relevant messages in the previous period does not significantly affect the level of resource stock. Economists have recently realized that verbal reprimands or sanctions matter. It is easy to see why awareness and agreement (common knowledge of awareness) are important in reducing strategic uncertainty.99 –0.M. in period t.00 0. (2) the total number of messages in period t.90 Estimated coefficient Previous K Total messages Relevant messages Relevant messages in previous period Positive feedback Constant 0. however. Positive feedback in the form of acknowledging that they did well increases the benefits of successful coordination. consumption.83 0. we use a random effects regression to estimate the effects of kinds of communication on the level of the resource stock.04 0. The number of subjects who sent messages related to coordination (i. (2003). Finally. This happened because. why is acknowledgment that the strategy worked important to prevent coordination failure? We believe that subjects respond to both financial and non-financial incentives. Capra and T.3 shows the results of the regression. and thus is an important motivator for subjects.58 p > |t| 0. Table 7. show that subjects respond to nonmonetary punishment in the form of expressions of disapproval in voluntary Table 7.81 2. Our dependent variable is the total resource stock. Tanaka To test our hypothesis that positive feedback matters.00 0.95 . The independent variables are (1) the total resource in the previous period t – 1. Masclet et al. The regression pools all sessions and allows horizon-specific and session-specific disturbances.

send relevant messages). In the remainder of the sessions. a key lesson from our investigation. subjects failed to coordinate their actions to restrict extraction. First is the expression of awareness of the benefits from reducing extraction and surpassing the threshold level of the resource.e. Second is the common knowledge of awareness that arises when the majority of subjects express agreement to reduce extraction (i. However. which is consistent with the findings in static coordination games (see Chaudhuri et al. unlike the previous ones.Threshold externalities 155 contribution games. Indeed. 2001). It matters what “kind” of messages people send to each other.. we suggest that communication be structured so that the three ingredients be satisfied. and Lauren Munyan. choices converged to a low equilibrium (i.e. in which the amount of the resource in equilibrium depends on the ability of subjects to coordinate choices and reduce extraction. high resource extraction). If communication is considered a policy alternative to help people resolve coordination problems. Notes 1 The model was also studied by Lei and Noussair (2003) and Capra et al. it is likely that subjects would respond to verbal (non-monetary) reward or acknowledgment that they are doing the right thing in a coordination game. we present results from an experiment that captures important features of renewable resource extraction. we use a chat room environment. We study communication under this framework to determine under what conditions communication improves outcomes. We observe that in two of the six sessions we organized. Similarly. . We determine that there are three ingredients for successful coordination in our dynamic coordination environment. Brandts and Cooper (2006) find that communication is a more effective motivator than monetary incentives in weak-link games. In our experiment. Colin Camerer. is that the existence of communication alone cannot resolve coordination problems. The third ingredient is acknowledgment of successes. subjects were perfectly successful in avoiding the bad equilibrium. By contrast. the results were mixed. (2006) in different contexts. In these sessions. verbal rewards seem to matter in our dynamic environment. in two other sessions. These conflicting results motivated us to study the chat transcripts to understand better the kind of messages communication needs to be a necessary and sufficient condition for successful coordination. Conclusion In this chapter. is less obvious (at least to economists) at reducing strategic uncertainty in coordination games. Acknowledgments We thank Charles Noussair. Our framework differs from that of others in that we adopt a dynamic coordination game environment. This last ingredient. in which subjects are allowed to send unrestricted messages in real time before making decisions.

. Kay. Masclet. Equilibrium selection in an experimental macroeconomy. almost common and common knowledge of advice. Sefton. and M-C. DeJong.. Cardenas. 5 To facilitate decision making. Ahn. and T. C. 350–352. M. 2001. Cooperative efforts in electronic contexts: the relevance of prior face-toface interactions. 258–286. It is what you say not what you pay: an experimental study of manager–employee relationships in overcoming coordination failures. In: Steffen Huck (ed.. 2006. 549–570.. Communication in a commons: cooperation without external enforcement. Sovero. Lei. eds. K. Communication and cooperation in a common-pool resource dilemma: a field experiment. 2006. S. L. 3 For a more detailed derivation of the steady states. J. 366–380. C. Noussair. Communication in coordination games. Monetary and nonmonetary punishment in the voluntary contributions mechanism. and E.156 C. S. Emory University working paper. References Brandts. 585–608. MI: University of Michigan Press. C. C. Ann Arbor. Forthcoming in the Journal of the European Economic Association. we provided subjects with an option to simulate different consumption choices and see how these choices would affect their earnings. R. 6 The positive feedback messages expressed in period t referred to choices made in the previous period(s). 2003. T. 739–771. 2004. Walker. E. Experiments. Feiler. Isaac.. Atlanta. D. Rules. and C...M. Communication and free-riding behavior: the voluntary contribution mechanism. MI: University of Michigan Press. V. New York: Palgrave. 2002. Capra. Ostrom. and M.. Emory University working paper. 4 Subjects were recruited to participate in a 3 hour session. and D. 1992. Atlanta. Wang. Advances in Understanding Strategic Behaviour: Game Theory. and J. Ostrom. Ross. Chaudhuri. and J. 1991. please refer to Lei and Noussair (2003). Phoenix. A. 107 (2). Quarterly Journal of Economics. In: T. A. K. Economic Inquiry. and J. Palfrey (ed. New York. D. . and B. J. The impact of simple institutions in experimental economies with poverty traps. T. 93 (1). E. Cooper. V. Cooper. 287–322. 495–515. 26 (4). Walker. American Economic Review. and C. International Journal of Game Theory. Villeval. R. Gardner. If a horizon did not end by the end of the third hour. Lei. Ann Arbor. Schotter. American Economic Review. 2001. Subjects were also told that new horizons started only if at least 30 minutes were left before the end of the session. Clark. An experimental test of an optimal growth model. In no session was it necessary to continue a horizon on another date. Noussair. 2003. V. Proceedings of the 2nd Americas Information Systems Conference. Tanaka. R. 1988.). Tucker. 29 (4). L.. Arizona. Rocco. Georgia. R. Noussair. Noussair. and C. 1996. Games and Common Pool Resources. Camerer. Georgia. M.). Laboratory Research in Political Economy. 92 (3). Talking ourselves to efficiency: coordination in inter-generational minimum games with private.V. Tanaka 2 Lei and Noussair (2002) were the first to implement the RCK model in the laboratory. E. C. and Bounded Rationality: Essays in Honour of Werner Güth. When are Nash equilibria self-enforcing? An experimental analysis. Capra and T. Sopher. Walker. 1994. Ostrom. Starr Center for Applied Economics at New York University. M. Forsythe. subjects were told that they could continue the experiment on another day or be paid for the decisions made by other subjects who would finish their horizon. F.

. Given this situation. leadership may serve as a coordination device. It is not in the self-interest of each individual country to abate the Pareto efficient amount of emissions. climate protection policy generates not only environmental benefits but also economic profits.2 Therefore. The damage each country suffers depends on the aggregated emission of harmful material and not (only) on local emissions. . The pollution of the Baltic Sea or the acid rain problem between Finland and the former USSR are historic examples for the European region. The leading position of German climate protection policy is profitable. Obviously. environmental pressure groups often demand that local politicians take on a leading role and abate more emissions than is in the narrow self-interest of the country. Alternatively. this may cause emotional costs.1 It is not quite clear why countries should follow the good example of a leader even if it is not in their immediate self-interest. There are numerous examples for situations in which leadership of one country may have the potential to induce cooperation of other countries. because climate protection creates new jobs and opportunities to export . there is the hope that the leader himself can profit from the leading position. As an example we can quote the German Federal Minister for the Environment (BMU. they expect that “a good example” may encourage other countries to join the coalition of abating countries in order to overcome the social dilemma situation all countries are confronted with. . . 2002): We are the leader in international climate protection and we want to maintain our leading position . one may hope that the “good example” creates some kind of a social norm. The most prominent example is the climate change expected from global CO2 emissions. because parts of the total benefit generated by abatement cannot be internalized. If followers ignore this norm. However.8 Unilateral emissions abatement An experiment Bodo Sturm and Joachim Weimann Introduction Some of the most serious environmental problems can be characterized as international public good problems. one may speculate that cooperative solutions are only accessible if players (countries) manage to coordinate on the efficient outcome. Thus there are several . Furthermore.

Game theoretic models implicitly make very strong assumptions about what people know. However. Either subjects are confronted with the game as it is without any reference to environmental problems. This means that we gave subjects all the information the agents in the model are assumed to have. In van der Heijden and Moxnes. but to a greater total abatement. Once again. the presence of a leader in a repeated standard public good game did not enhance the overall level of cooperation relative to a situation without a leader. one has to decide how the experiment should be framed. In order to test the Hoel model experimentally. i. First. Gächter and Renner (2003) confirm that despite the significant correlation between the leader’s and followers’ contributions. the leader had a significant influence on the other subjects (followers).4 Thus it remains an open question whether “leadership matters”. not only the rules of the game but also the equilibrium solution and the consequences of all kinds of deviations from rational . However. leadership had a positive but only weakly significant effect on the aggregated contributions to the public good. Weimann ways in which leadership possibly is able to induce cooperative behavior of the followers. In this chapter. for the leader leadership does not pay. In a similar experiment by Güth et al.3 However. we decided to design the experiment analogously to the model.158 B. Hoel (1991) investigated the consequences of unilateral emissions abatement in a global public good game. In a seminal paper. two methodological questions have to be answered. as before. The issue of whether abatement on the part of the leader has an influence on the behavior of the followers is analyzed in an experiment by van der Heijden and Moxnes (2000). which has interior solutions and allows deviations from the equilibrium in two directions. for the leader leadership does not pay. there are some important differences between our design and the design chosen by van der Heijden and Moxnes. In their experiment. abatement of a single country (say country 1) above the equilibrium level. the assumption of selfish behavior of all countries except one rules out that “the good example” given by country 1 can infect other countries. In their experiment. It turns out that unilateral emissions abatement leads to a lower abatement of the other countries. However.e. we try to answer this question by testing the Hoel model experimentally. subjects played a standard public bad game where both the Pareto efficient outcome and the equilibrium solution for the simultaneous decision protocol and mixed sequential-simultaneous decision protocol had a boundary solution. or they are told the cover story that fits the story the Hoel model tells us. Total welfare increases if country 1 has the lowest marginal benefit from abatement. under the assumption that all countries decide simultaneously and that all other countries behave fully rationally and selfishly.5 Therefore. We decided on the latter because we wanted to test if unilateral abatement of harmful material induces further activities. Second. Normally. one has to decide how to inform subjects. the influence of the leader on the followers was not strong enough to make leadership profitable for the leader. A fair test of the model should take into account that it is designed to deal with environmental problems. (2007). they did not really test the Hoel model. Sturm and J.

chooses abatement Xi and possesses benefit and cost functions Bi (X ) and Ci (X i).e. with B i > 0.1) In the Pareto optimum (PO). respectively. . To test the model fairly. we had to ensure that these knowledge preconditions were fulfilled. Finally. But if this is the case. The sum of individual abatements is X = ∑i =1 X i .Unilateral emissions abatement 159 behavior are assumed to be common knowledge. the model is outlined and the parameters and functions used in the experiment are introduced. conclusions are offered. We would like to emphasize that this does not mean that the experiment is a “recommended play” game in the sense of Brandts and McLeod (1995). the Nash equilibrium is unique. The laboratory version of the Hoel model In the N-country version of the Hoel model. Every country i. and C i > 0.e.2) .6 The experimental design is then described. After this. Not informing subjects fully would mean not testing the model but testing the ability of students to find rational strategies in complicated games. i. agents in the Hoel model should be influenced in the same way because they possess the same information we gave our subjects. each country emits a global pollutant.N. N In the Nash equilibrium (NE) of the game with a simultaneous decision protocol. We therefore decided to teach the subjects before the experiment and to demonstrate the prediction of game theory as well as the Pareto solution. aggregated abatement is chosen such that social marginal benefit equals marginal costs. B ' (X ) = Ci' (X i ) with B ' (X ) = ∑i =1 Bi' N (8. country [IE10] abates to the amount at which private marginal benefit equals marginal costs. First. Because the game has complicated interior equilibrium solutions. In our experiment. Our aim was not to recommend one particular equilibrium among different existing equilibria but to make sure that the subjects perfectly understand that they are in a dilemma situation in which individual rationality leads to inefficient solutions. C i > 0. Bi' (X ) = Ci' (X i ) (8. . i. . the results are presented and discussed. we could not assume that subjects would be able to compute the prediction of game theory and the Pareto efficient solution in the laboratory during the experiment. . The chapter is organized as follows. i = 1. Recommendation of a particular strategy combination served in their experiment as an equilibrium selection device. B i > 0. One may argue that this can influence subjects.

N abate less than in the equilibrium given by (8. the other countries are informed of this decision.4) The standard case (equation 8. He also shows that under unilateral abatement B 1 > B j for all j = 2. and PO solution. Hoel shows that in the equilibrium described by (8. In the sequential treatments (seq). .7 Aggregated abatement in the SPE is lower than in the NE. In order to test the model experimentally. in the new NE for the simultaneous decision protocol it holds that B1' (X )+ h = C1' (X 1 ) and B 'j (X ) = C 'j (X j ) with j = 2. SPE. leadership generates a negative environmental effect. the cost and benefit functions have to be specified. and then decide simultaneously on their emission reduction. . . the shift from the simultaneous decision protocol to the sequential decision protocol alters the game theoretical prediction of individual behavior. the pure timing of action.1) is contained in this formulation with h = 0. In the subgame perfect equilibrium (SPE) of the sequential game. country 1 abates less and country j abates more than in the NE of the simultaneous game. matters.3).1 shows the specification of the model. In the simultaneous treatments (sim). country 1 decides first. the NE abatement is smaller than the PO abatement.. country 1 decides first. now has a strategic advantage because it may choose the point on the best response function of the followers that maximizes its own profit. all the countries simultaneously decide on their abatement. measured by a parameter h.. and in the appendix to this chapter we derive the appropriate NE. To test the Hoel model and the influence of leadership. the total benefit from abatement then is B1 (X ) + hX with h > 0 (8. Sturm and J. Although the cost and benefit functions remain unchanged. .e. We are seeking an answer to the question of whether “leadership”. the other countries are .160 B. . i. For country 1.e. . it is necessary for us to analyze the implications of a change in the timing of action from a simultaneous decision protocol to a mixed sequential-simultaneous (in the following “sequential”) decision protocol.. It may be the case that the inhabitants of this country have altruistic preferences or just love to see a “cleaner” world. . Hoel assumes that this country derives some kind of extra marginal benefit from total abatements. Therefore. In order to model unilateral abatement of country 1. i. we use a 2 × 2 factorial design varying the sequence of moves and parameter h.e. the leader..1) and that total abatements are greater in the case of unilateral abatement. Country 1. N (8. X PO > X NE. Table 8. N is a sufficient condition for an increase in total welfare.3) Given (8. . It is not important where this benefit comes from.4) country 1 abates more and countries 2. i.e. Weimann Because social marginal benefit is above private marginal benefit. i.

country 1 (j) can increase its profit by 24 percent (16). the profit for country 1 in the PO is 49 percent (44) above the profit in NE (SPE). Table 8. . Hypothesis 1 The parameter h has a significant influence on the abatement and profit of country i. country j can increase its profit in T III (T IV) by 16 percent (19) compared to the PO profit providing it chooses its best response. and PO. If all other countries choose their PO abatement.2 summarizes the treatments.Unilateral emissions abatement 161 Table 8.. seq) parameter h informed of this decision. The profit for country j in the PO is 57 percent (68) above the profit in NE (SPE). i = 1.. sim) treatment T II (h > 0. we had to specify the parameters of the functions in Table 8..5X 2) + hX Bj (X) = AX – 0. Table 8. Free rider incentives are also considerable. SPE. sim) seq treatment T III (h = 0..2 Experimental treatments Variable Sequence sim h=0 h>0 treatment T I (h = 0. . and then simultaneously decide on their emission reduction.1 in a way that ensures that the payoff functions are sufficiently steep. i = 1. the profit for country 1 in the PO is 73 percent (68) above the profit in NE (SPE).1 Cost and benefit functions Cost function of country i..8 The efficiency loss in the NE and the SPE is considerable in all treatments. we can now formulate the two central hypotheses that follow from standard game theory and that will be checked experimentally. N Ci (Xi) = cXi2 + T B1(X) = b(AX – 0. If all the other countries choose the PO abatement in T I. 5.5 X 2 Table 8.. For h = 0. .. both in the simultaneous and sequential treatments. country 1 (j) can increase its profit by 72 percent (16) compared to the PO profit providing it chooses its best response. numbers had to be rounded for the SPE and PO solutions. The parameters were chosen so that the NE has a solution in integers. If we define X iT as abatement of country i in treatment T ∈ {h = 0. For h > 0. seq) treatment T IV (h > 0. In T II and the same situation. . .3 summarizes the parameter values for T I to T IV and the abatements and payoffs subjects realize in the NE.9 Given this version of the Hoel model. N Benefit function of country 1 Benefit function of country j = 2. The profit for country j in the PO is 50 percent (62) above the profit in NE (SPE). h > 0} for a given . To implement the four treatments. The variation of h in the simultaneous case is a direct test of the model.

.80 Pareto optimum 2.30 22. .40 1 81.00 21.971 1 79.564 187. . payments include 12 EUR show-up fee. . .230 25.200 22. . .50 all 214.000 lab-dollars (LD) = 1 EUR.488 16.606 Nash equilibriumII T I and T III: b = 30/47. . 5 81. 5 all 1 Abatement 40 46 224 24.188 2. .59 Profits (LD) 24. h = 0. . . . h = 3000/47. . . . abatements.45 60.151 PaymentsI (EUR) 15. .619 21.000.45 17. . T= 40. .123 15. profits.54 37.70 T II and T IV: b = 30/47. N = 5 Country 1 2.611 166.16 286. 5 47. . N = 5 Country 1 2. T= 40.10 All 211. . III Mixed sequential-simultaneous decision protocol. 5 79. c = 3. .707 38.20 18.713 PaymentsI (EUR) 18. A = 500. . .312 Notes I 40.3 Summary of parameters.75 176.63 43.50 all 408.788 27.23 260.782 27.000. c = 3.20 all 397. . 5 48. .10 2. and payments Subgame perfect equilibriumIII 1 18.212 21.Table 8.44 12. II Simultaneous decision protocol. 5 all Abatement 30 47 218 Profits (LD) 11.974 40.63 60.07 156. A = 500.16 36.20 2.

The sessions were conducted between December 2003 and May 2005 at the Magdeburg Experimental Laboratory (MaXLab). π sim > π seq and j j ∑π i =1 5 sim i > ∑ π iseq . we can formulate hypothesis 2 for country 1 and j. After a preliminary analysis of the data.. i. we decided to conduct 12 additional independent observations in the sequential treatments. . i =1 5 If we assume the same notion for profit π iT. is in direct contrast to the above-mentioned idea that leadership may be a solution to the social dilemma situation countries are confronted with. we can formulate hypothesis 1 for country 1 and j. we have to distinguish between country 1 and j. seq} for a given choice of the parameter h. as follows: X 1sim > X 1seq . in treatment T ∈ {sim. Experimental design Each of the four treatments was originally played with six groups of five subjects.Unilateral emissions abatement 163 sequence of moves. j = 2.. X h =0 > X h >0 and j j ∑X i =1 5 h =0 i =1 < ∑ X ih=> 0 . All in all. . 240 subjects participated in the experiment. Each session lasted about 1 hour. . . i = 1. X sim < X seq and j j ∑X i =1 5 sim i > ∑ X iseq . . 1 i =1 5 10 If we assume the same notion for profit πiT we get: π ih >0 > π ih =0 . j = 2. 5. We get: π 1sim < π 1seq . Hypothesis 2 12 The variable sequence has an influence on the abatement both at the individual and the aggregate level. If we define X iT as abatement of country i. All subjects were undergraduate .5. that leadership results in less abatement for country 1 and a world with less aggregated abatement.. There were 16 sessions with three groups (15 subjects) playing the game in parallel.12 i =1 5 The most interesting point of hypothesis 2 is that the game theoretical prediction. .e. 5.. as follows: X 1h =0 < X 1h >0 . we have six independent observations for the simultaneous treatments T I and T II and 18 independent observations for the sequential treatments T III and T IV.. .

subjects were told that they should imagine that they were the head of a delegation from their country at an international conference on emissions abatement of a global pollutant. Each of the. Furthermore. we demonstrated the Nash equilibrium for the simultaneous game and the subgame perfect equilibrium for the sequential game graphically by means of best response functions. the aggregated abatement of all the other countries. the leader. subjects were informed about their individual and aggregated abatement. in total. The information to the subjects was organized as follows. At the beginning of the lessons. 5).e. they had to decide on the level of domestic abatement. each country could put in the expected abatement of the other countries. could put in its own abatement and then the profit maximizing response of the other countries was computed. the parameters. and then its profit maximizing response was computed. the individual profits of all countries. with the anonymity within the groups being guaranteed by the procedure. we decided to use the frame also employed in the Hoel model and to inform subjects comprehensively about the decision situation.16 Given all necessary information for their country (costs and benefits). The second element was the simulator. The subjects knew that in the test periods they were playing against four automated systems whose behavior would not change. and its application to finitely repeated games. the Nash equilibrium. Then the most important features of the decision situation were explained.164 B. First. Sturm and J. which subjects could use to evaluate the consequences of non-profit-maximizing actions. . the idea of the underlying social dilemma was illustrated by an example with three countries.15 The payoff simulator was identical for all subjects in a treatment and was visible both on the input and the output screen. country j used the same program as in the simultaneous treatment but country 1. Additionally. their role (country 1 or country 2. The experiment was fully computerized and anonymous. In the sequential treatments. Weimann economics students familiar with fundamental game theoretic concepts. . As already mentioned in the introduction. Before the experiment started. We showed that everybody was better off in the Pareto efficient solution but that there were strong incentives to deviate from the efficient solution.14 In the simultaneous treatments. and the aggregated profit for all expired periods. . Here the subjects could put in the expected abatement of the others and their own arbitrary abatement. . The subjects received written instructions about the rules of the game. i. The first was the profit maximizing function. 48 groups played the game ten times and subjects were informed about the number of repetitions. The abatement decisions in the equilibrium and the Pareto efficient solutions as well as the corresponding .13 The subjects were seated in soundproof booths and had no contact before. During the experiment. Second. questions were answered and all subjects played two test rounds against the computer. the expected profit and total abatement were computed for both cases. their computers were equipped with a payoff simulator that had two elements. and after the experiment. We conducted 15 lessons with 12 or 24 subjects. and the functional forms. For this purpose. we invited the subjects to attend a separate lesson held before the experiment on the same day or a day before the experiment was carried out. the idea of best response functions. during.

Unilateral emissions abatement 165 payoff implications were depicted. the output screen. subjects received a show-up fee of 12 EUR and were told that possible negative payoffs had to be settled using this fee. and total) as follows. Subjects were informed that in the experiment countries would have different roles (country 1 and the other countries). Total abatement is higher in the sequential treatments than in the simultaneous treatments but the difference is not significant. we must still reject the hypothesis that country 1 abates at the level of the SPE values in the sequential treatments. The lesson lasted about 1 hour. 5 percent level). However. Figure 8. At the end of the lecture. NE. and SPE values are marked.40 EUR to 27.30 EUR including the show-up fee. none of these differences between treatments is significa nt (exact two-sided M-W U-test. j. We summarize our findings with respect to the mean abatement (for country 1. At the beginning of the experiment. 5 percent level). The difference between the mean abatement and the SPE value in the sequential treatments is highly significant for country 1 (Wilcoxon signed-rank test. However. the difference between treatments is not significant. the theoretical prediction for country 1 is supported by our data in the simultaneous treatments. Results Abatement The payoffs per subject range from 14. the information on which role each subject would play and the timing of action was first given in the experiment. there is no significant difference for country j (5 percent level). Abatement is higher in the sequential than in the simultaneous treatments. the other countries j. 1 percent level). the input screen.1 displays the average abatement over all ten rounds for country 1. Due to the high variance of individual behavior. However. In other words. However.70 EUR. b c d The most striking observation is that country 1 does not reduce its abatement in the change from a simultaneous to a sequential decision protocol – as theory predicts – but raises its abatement. Observation 1: a The mean abatement of country 1 is ceteris paribus higher in the h > 0 treatments than in the h = 0 treatments. The PO. The average payoff of all subjects is 21. but this is not the case in . However. the difference between total abatement and the SPE value in the sequential treatments is significant (Wilcoxon signed-rank test. 5 percent level).17 There is no significant difference between the treatments for the mean abatement of country j (M-W U-test. and the payoff simulator were shown and explained using the above-mentioned example. and all countries.

4 T II (h > 0. Although we have only a relatively small number of independent observations per cell in Table 8. it is interesting to look at the correlation of the mean abatement of countries 1 and j. the sequential treatments.6 T III (h = 0.2 T IV (h > 0.sim) 51. Weimann Country 1 80 Abatement per period PO PO PO PO 60 40 NE 20 40.2.8 T IV (h > 0.1 T III (h = 0.0 T II (h > 0. Whereas the Spearman rank correlation yields .3 T I (h = 0.sim) 0 51.166 B.1 Abatement per period. the behavior of country j does not vary much between the treatments and the corresponding theoretical prediction is supported for all treatments. Sturm and J.seq) Figure 8.seq) 53.seq) Country j 80 Abatement per period PO PO PO 60 NE 40 NE SPE 20 51.sim) 0 NE SPE SPE 47. On the other hand.seq) PO SPE 52.7 T I (h = 0.sim) 45.

029) are obtained for the sequential treatments T III and IV respectively.5 T II (h > 0.1 continued. values of rS = 0.704) and 0.781 (p = 0.seq) T IV (h > 0.7 + 0.397) for the simultaneous treatments T I and II respectively.374 0 20 40 60 Abatement A1 80 0 80 Ind. i.687 Aj = 28.e. Abatement Aj 60 .seq) 262. there is a strong positive correlation of leader’s abatement with followers’ abatements in both sequential treatments.2 + 0.4 T IV (h > 0. T III (h > 0.429 (p = 0.2 T III (h = 0.Unilateral emissions abatement 167 Total 400 Abatement per period PO PO PO PO 300 NE 200 NE SPE SPE 100 247.2 Scatterplots for seq-treatments.sim) 251. rS = 0.575 A1 Adj R-squared = 0.1 T I (h = 0.446 A1 Adj R-squared = 0.000) and 0. SPE PO Fitted values 0 Figure 8. seq) 80 Abatement Aj 40 20 0 0 20 40 60 Abatement A1 Figure 8.sim) 251. obs. seq) 80 60 40 20 Aj = 25.200 (p = 0.546 (p = 0.

2 make this conclusion more clearly visible. The hypothesis cannot be rejected for any round of the simultaneous treatments (Wilcoxon signed-rank test.168 0.575 0.7 28. country j’s abatement in the last round of both sequential treatments is significantly below the SPE value (Wilcoxon signed-rank test. Remarkably.928 0.147 0.2 24.000 0.374 Note aj (a1) mean abatement of country j (1) over rounds. obs. R2 0.446 0.446 p(b1) 0. The SPE and the PO abatements are indicated to provide a better orientation and the line of best fit is obtained from the regression analysis in Table 8.4. N = 6/18 ind.552 0.4 Regression analysis Model aj = β0 + β iα1 b0 T I (h = 0. Nine of these cases appear in the first four rounds.004 adj. 5 percent level).9 25. The hypothesis that abatements for country j are equal to the PO level must be rejected in all 40 rounds of the treatments. seq) T IV (h > 0.002 0. The round-by-round analysis of the abatements confirms the findings of observation 1. The scatterplots for the sequential treatments in Figure 8. Observation 2: a b There is a significant and positive correlation between the abatement of the leader and the abatement of the followers. The same hypothesis must be rejected for country 1 in 17 of the 20 rounds of the simultaneous treatments b c .4 supports this result and yields a significantly positive coefficient for the sequential treatments. Figure 8. An increase in the leader’s abatement by one unit increases the total abatement of countries j by about two units.010 0. sim) T III (h = 0. The simple linear regression in Table 8.687 0. This observation is therefore an indication that (at least some) followers follow the leader’s example and that abatements of followers and leaders are positively correlated. The same hypothesis must be rejected in only 12 of the 40 rounds of all treatments for country j.002 b1 0. Observation 3: a The hypothesis that the abatement of country 1 is equal to the NE/SPE values must be rejected in each of the 20 rounds of the sequential treatments.3 shows the abatements in the four treatments round by round. Sturm and J. seq) 51. 5 percent level). sim) T II (h > 0.168 B.001 0.2 p(b0) 0.000 0. Weimann Table 8.

In order to gain more insight into the way subjects adjust their behavior. 5 percent level).3 is that subjects show a clear tendency towards the equilibrium strategy during the course of the game and that this tendency is more pronounced in the simultaneous treatments than in the sequential treatments. There are no other significant differences between the subsamples of abatement and the NE or SPE values.i − X tEq . 5 percent level).Unilateral emissions abatement 169 and in 17 of the 20 rounds of the sequential treatments (Wilcoxon signedrank test.i is i’s indiPO vidual equilibrium (the NE or SPE) abatement in period t and X t. This difference is significant except for country 1 in the simultaneous treatments (Wilcoxon matched-pairs signed-rank test.i . Therefore we split the mean abatement of both countries in two subsamples. Observation 4 summarizes our findings regarding both subsamples: a The average abatement for the last five periods is for both countries in all treatments lower than the average abatement for the first five periods. and even increasing. a subsample which contains the first five periods and a subsample with the last five periods.3 suggests that subjects presumably learn the equilibrium strategy – despite the fact that they know the equilibrium solution before the experiment starts. we use the coefficient 1 N ∑ X t . In the sequential treatments. Convergence and individual behavior The overall impression from Figure 8. 1 percent level).i N i =1 αt = X tPO − X tEq . Coefficient αt measures the mean absolute value of . The downward pattern of abatement for both countries in Figure 8. especially for country 1. b Based on this observation we may conclude that there is some learning of equilibrium behavior. abatement. The average abatement for the last five periods of country 1 is significantly above the SPE abatement in the sequential treatments (Wilcoxon signed-rank test. Nevertheless the surprising contradiction between the theoretical prediction regarding the abatement of country 1 and the corresponding behavior of this country as a leader preserves over the course of the experiment.i is the individual abatement of country i in period t. Xt. the downward trend to the SPE is interrupted by phases of constant.i is i’s individual PO abatement in period t.i as a simple measure of the deviation of individual behavior from individual Eq rationality. X t.

we observe that the αt values are higher for the sequential treatments than for the simultaneous treatments for both countries over almost all .4 decrease. the αt values in Figure 8. i. Sturm and J. At first glance. although not monotonically. αt summarizes the information about individual abatement behavior in Figure 8. the deviation of individual abatement from equilibrium abatement as a fraction of the difference between the equilibrium and PO abatement.e. We observe αt = 0 (1) for country 1(j) if all countries of this type play their equilibrium (PO) abatement.3 Abatement over periods. As expected.3.170 B. Weimann T I (h = 0. sim) 80 PO 60 Country j Abatement NE j Country 1 NE 1 20 40 0 1 2 3 4 5 6 Period 7 8 9 10 T II (h > 0. sim) 80 PO 60 Country j Abatement NE j NE 1 Country 1 40 20 0 1 2 3 4 5 6 Period 7 8 9 10 Figure 8.

and for the sequential treatments.Unilateral emissions abatement 171 T III (h = 0. on the other hand.18 The similarity between the αt values for the simultaneous treatments. seq) 80 PO 60 Country j Abatement NE j 40 Country 1 20 NE 1 0 1 2 3 4 5 6 Period T IV (h > 0. best response or NE/SPE . rounds. on the one hand. In order to analyze the structure of the deviations from the NE behavior and the SPE behavior. Because the NE and SPE are not at a boundary. the decrease in the αt values seems to be sharper for country 1 than for country j. Furthermore. we classify the individual decisions of countries 1 and j into three groups: cooperative behavior. is striking. αt measures departures from equilibrium behavior in both directions. seq) 80 PO 7 8 9 10 60 Country 1 Abatement 40 NE j Country j 20 NE 1 0 1 2 3 4 5 6 Period 7 8 9 10 Figure 8.3 continued. Abatements below the NE level or SPE level may therefore serve as a means of punishing subjects who behave selfishly.

Abatement below min(bR.6 0. seq) T IV (h > 0. Abatement between min(bR. sim) TIII (h = 0.2 Country j 0. Fraction of behaviour (%) 80 60 40 20 0 1 2 3 4 5 6 Period 7 8 9 10 Fraction of behaviour (%) 100 Country 1: sim treatments 100 80 60 40 20 0 1 2 Country 1: seq treatments 3 4 5 6 Period 7 8 9 10 Fraction of behaviour (%) 80 60 40 20 0 1 2 3 4 5 6 Period 7 8 9 10 Fraction of behaviour (%) 100 Country j: sim treatments 100 80 60 40 20 0 1 2 Country j: seq treatments 3 4 5 6 Period 7 8 9 10 Figure 8.2 0 1 2 3 4 5 6 Period 7 8 9 10 0 1 2 3 4 5 6 Period 7 8 9 10 T I (h = 0. Eq) and PO. Eq) with “bR” as individual best response to the aggregated abatement of the others in the current period and “Eq” as individual abatement in NE or SPE (each with +/– 20 percent). sim) T II (h > 0. Eq) and max(bR. Eq).6 0.4 0. Notes Co-operative behavior. i.5 Individual behavior of country 1 and j. .4 Alpha. seq) Figure 8.4 0.8 Alpha 0.1 0. abatement between max(bR.8 Alpha Country 1 1 0.e.

several behavioral patterns are striking. in the last rounds.e.e. Individual rational behavior prevails and becomes more frequent in the course of the experiment. On the other hand. This may be a reaction of some subjects to the frustrating experience made in the early rounds that the high abatement of country 1 does not motivate the other countries. Although this classification serves only an illustrative purpose. 5 percent level). a significant fraction of abatement decisions of country j is below the best response level. The hypothesis that the profit of country 1 is equal to the NE value or SPE value can only be rejected for treatment T III where the profit is below the SPE prediction (Wilcoxon signed-rank test. 1 percent level). leadership matters from the viewpoint of country 1. . However.6 shows the average profits earned by the subjects.19 Regarding country 1. i. i. However. on average. and abatements that are below the best response or NE/SPE. Profits and efficiency Figure 8.Unilateral emissions abatement 173 behavior. 5 percent level). On average. We summarize our findings concerning the profits below. one-third of the decisions can be classified as cooperative behavior. Observation 5: a Country 1 earns a significantly higher profit in the h > 0 treatments than in the h = 0 treatments (M-W U-test. not very successful. abatement above the individual rational level becomes less frequent even in the sequential treatments. These profits are not identical to the welfare measure used in the Hoel model because there the “extra profit” country 1 derives from abatements above the Nash level (measured by h) is not part of the welfare of country 1. The profit in T III is significantly below the profit in T I but there is no significant difference between the profits in T IV and T II (M-W U-test. to follow suit. Figure 8. This indicates that the efforts to induce cooperation through leadership are. The corresponding intervals are described at the bottom of Figure 8. there is virtually no abatement below the individual rational level for country 1.5 confirms the observation that cooperative behavior is more frequent and rather stable in the case of leadership. The positive aggregated welfare effect of “over-abating” of country 1 results because the welfare loss suffered by this country is overcompensated for by the welfare gains of the other countries. A reason for this behavior may come from the existence of some punishment behavior or negative reciprocity shown by the subjects in the role of country j. which shows the fractions of the three behavioral patterns in the simultaneous and sequential treatments. country j. Although the behavior of country 1 differs considerably between the simultaneous and sequential treatments. The proportion of this kind of deviation from individual rational behavior seems to be higher in the sequential treatments than in the simultaneous case. this kind of behavior clearly becomes less important over the periods. Remarkably.5. the behavior of country j seems to be rather stable for both kinds of sequence.

Sturm and J. seq) T I (h = 0. sim) 27.174 b B.433 T IV (h > 0. sim) T III (h = 0.557 44. seq) PO SPE 44.359 T III (h = 0.000 LD per period 50 40 30 NE 20 10 0 PO NE 13.000 LD per period 50 40 30 20 10 T II (h > 0. seq) SPE PO PO Country j 60 Profit in 1. The difference is significant in T I and in both sequential treatments (Wilcoxon signed-rank test.20 c The picture we get based on this observation is quite clear. 5 percent level). seq) 26.610 NE NE SPE PO PO PO Figure 8.116 T I (h = 0. 5 percent level). The difference is significant in T I and T IV (Wilcoxon signed-rank test. sim) 0 43.336 41. Although the abateCountry 1 60 Profit in 1. sim) PO SPE 9. Weimann Country j is able to increase its profits above the NE and SPE values in all treatments. .833 T IV (h > 0. Total profit is slightly above the NE and SPE values in all treatments.6 Profit per period.926 T II (h > 0.

We use the efficiency index Effi as an index to facilitate the comparability of results. seq) T II (h > 0. i.764 T III (h = 0.041 0. sim) 0.515 0.269 T IV (h > 0. sim) 0. Effi = π ireal − π iEq . seq) –0. if it is able to achieve the social optimal profit the efficiency is equal to one.311 0.223 0.000 LD per period 300 PO PO PO PO 200 NE NE SPE SPE 100 187. seq) 0. The index Effi measures the difference between the realized profit (πireal) and the equilibrium profit (πiEq) as a fraction of the difference between the maximal possible profit in the PO solution (πiPO) and the equilibrium profit for a country i.e.269 175.Unilateral emissions abatement 175 Total Profit in 1.182 T III (h = 0. sim) Figure 8. On the other hand.162 0.182 T IV (h > 0.187 0. the followers are able to increase their profit above the equilibrium prediction. sim) T I (h = 0. Therefore the efficiency index permits to compare the ability of both countries to realize Table 8.224 T II (h > 0.346 205.5 Efficiency index Treatment Country 1 Country j Total T I (h = 0. π iPO − π iEq If a country i realizes its equilibrium profit the efficiency is zero. Due to the asymmetry of profits in the equilibrium and the PO solution for both types of countries it is useful to normalize the realized profits of all treatments.6 continued.241 0.797 205. it turns out that country 1 is unable to increase its profits by using leadership.220 0. seq) 0 . ment behavior of country 1 with leadership differs considerably from that without leadership.

this could serve as a kind of focal point for the followers. 5 percent level). the rate of realized supra-equilibrium profits to maximal possible profits above the equilibrium are higher for the followers compared to the leader. Brosig et al. i. and then displays a finalround effect. neither the leader nor the follower. we cannot reject hypothesis 1. only the difference between the efficiency of T I and T III is significant (M-W U-test.176 B. This line of reasoning seems to be in line with our observations. Weimann profits above the equilibrium. Sturm and J. In terms of efficiency. The observations from the abatement section show that. Country 1 realizes in both sequential treatments a lower efficiency than in the simultaneous treatments. c d The interpretation of observation 6d is quite clear. There is no significant difference between the efficiency values of the simultaneous and sequential treatments for country j.5 depicts the results for country 1. the leaders take a chance going ahead with a “good example”.e. However. (2003) have shown that. decays during the course of the game.21 Discussion Our data support the Hoel model for the treatments with a simultaneous decision protocol at least for the second half of the ten rounds. However. None of the countries. this is not the case and cooperation breaks down after a few periods. Country 1’s . subjects try to coordinate their behavior in order to realize the efficient outcome but that this coordination is only successful if all subjects stick to their promise to cooperate. our findings for the simultaneous treatments are in line with the stylized facts of many public good experiments: abatement starts between the NE level and the PO level. i. in standard public good games. In the simultaneous treatments both countries do not differ in their ability to realize profits above the equilibrium. it seems fair to state that the Hoel model describes actual behavior surprisingly well. j and all countries. Observation 6: a b The overall efficiency is quite low and cannot be significantly enhanced in the game with leadership (M-W-U-test. On the other hand. countries j achieve efficiency values that are significantly higher than those of country 1 in the sequential treatments (Wilcoxon signed-rank test. 5 percent level). 1 percent level). We summarize our findings concerning the efficiency index below. Therefore. Table 8. in an environment where subjects act simultaneously. The most important question we sought to answer with this experiment was whether or not “leadership matters”. Normally. country 1 performs worse relative to country j when it becomes a leader. If the leader starts each round with the PO abatement. Having a leader may open a way to solving the coordination problem just mentioned. on average.e. are able to increase the efficiency in a game with leadership.

One of these groups even plays the PO abatement in all rounds. their reaction to the leader’s efforts is not sufficient to boost the profit of the leader above the equilibrium value. in these groups there are few followers who react cooperatively to the leader’s signal. on average.7) shows three interesting behavioral patterns in this context. In contradiction to the game theoretical prediction.7 Classification of 36 groups in the sequential treatments. leaders try to lead and set a “good example”. the identification of different types of behavior as an important result of experimental economics is of particular importance. i. average data may not be an appropriate means to analyze individual behavior. There are some groups (five of 33) that succeed in abating at the PO level in at least eight of the ten rounds.Unilateral emissions abatement 177 30 Number of groups 20 10 0 5 Successful leadership 28 Leadership with defection 3 No leadership Figure 8. it explains the fact that the mean profit of country j exceeds the SPE values in the sequential . Notes Successful leadership: Country 1 chooses the PO abatement (+/– 20 percent) and all countries follow with the PO abatement (+/– 20 percent) in at least eight of the ten periods.e.e. i. On average. in most cases (28 of 33) the leader fails to induce cooperation.22 A simple analysis of the group specific data (see Figure 8.24 However. Leadership with defection: Country 1 chooses the PO abatement (+/– 20 percent) in one or more periods and at least one country j does not follow with the PO abatement in each of this periods (+/– 20 percent). react only a little to these efforts by the leader.e.23 The reaction of the followers is quite mixed. In this context. including the last round. in 33 of 36 groups country 1’s abatement is near the PO abatement at least once over the periods. High efforts by the leader and the cooperative followers are exploited by the majority of defective followers. No leadership: Country 1 does not choose the PO abatement (+/– 20 percent) in any period. Moreover. abatement is significantly above the SPE abatement in the sequential treatments. This behavioral pattern is the explanation for the fact that the mean profit of country 1 does not exceed the SPE values in the sequential treatments although the mean abatement of country 1 is significantly higher than the SPE values. followers. i. However.

178 B. Hartmut Kliemt. Weimann treatments. Manfred Königstein. the subgame perfect equilibrium for the mixed sequential-simultaneous decision protocol. the probability that other followers free ride and cooperation breaks down very soon is high. the Hoel model describes the individual behavior surprisingly well in an environment with a simultaneous decision protocol and. Munich. Thomas Riechmann. Conclusion The primary objective of our experiment was to test the Hoel model and to analyze the influence of leadership. Martin Weber. The authors thank Jeannette Brosig. leadership by itself does not seem to be an appropriate tool to overcome social dilemma problems. Furthermore. these observations are in line with the related experimental research on leadership that was discussed in the introduction. . In particular. Based on these observations. Since the external validity of results gained in singular laboratory experiments is restricted to the specific laboratory environment. is gratefully acknowledged. second. the experiments show that countries that want to increase their own profit and the total profit of the group by showing leadership should not put too much hope in the effectiveness of their good example. even if leaders have a strong incentive to induce cooperative behavior. Appendix In this appendix we derive the solutions for the Nash equilibrium for the simultaneous decision protocol. Even if some follow this example. Sturm and J. The change of the efficiency values regarding country 1 and j from a simultaneous to a sequential game serves as another illustration of our main finding: by assuming leadership country 1 loses ground in relation to country j. first. All in all. Only the followers who free ride at the expense of the leader and the cooperative followers in cooperative groups can increase their profits. and the Pareto optimal solution based on the specification used in the experiment. we must reject hypothesis 2.25 However. and an anonymous referee for their helpful comments. but otherwise our data do not support the idea that leadership is an effective means to create stable cooperation. leadership matters a lot but is not able to increase the profit of the leader and to overcome the social dilemma situation all countries are confronted with. we have to admit that we are not able to make any recommendations for environmental policy purposes. Acknowledgments Support by the Deutsche Forschungsgemeinschaft (DFG) and the CESifo. we may come to the conclusion based on our results that.

^ k ≠ 1. maximizes the difference between benefit and costs of abatement given the abatement of all other countries. 2c + 1 Since country j and all other N − 2 countries k. N) are identical in their benefit and cost functions. maxπ j = A(X j + X − j )− 0.a.Unilateral emissions abatement 179 Nash equilibrium (NE) for the simultaneous decision protocol Country 1 maximizes the difference between benefit and costs of abatement given the abatement of all other countries.a. we can substitute abatement X−j with X−j = (N − 2)Xj + X1. The reaction function R1 (X −1 )= R1( X j) = 1 (bA + h − (N − 1)bX j) 2c + b (8. k ≠ j. we can substitute abatement X−1 with X−1 = (N − 1) Xj.a.5(X j + X − j ) − cX 2 + T j 2 Xj ( ) with X−j = ∑ N i≠ j Xi .a. 2cb + 2(N − 1)c + 4c 2 . The NE for the simultaneous decision protocol results as the intersection of the reaction functions (equation 8.2). . .5(X 1 + X −1 ) X1 [ 2 ] + h(X 1 + X −1 )− cX12 + T ( ) N with X–1 = j=2 Xj . 2c + b Since all other countries (J = 2. . X1.e.1) describes the best response of country 1 to the abatement chosen by country j. .e. . i. The reaction function of j is R j (X − j ) = 1 (A − X − j ) . i. .1) and (equation 8. describes the best response of country j to the abatement of country 1. j = 2. Country j.2). . The reaction function of j. and the interior solution for the NE is X 1NE = 2cbA + h(N − 1 + 2c ) . The reaction function of country 1 is 1 (bA + h − bX −1 ) . R j (X 1 ) = 1 (A − X 1 ) N − 1 + 2c (8. 2cb + 2(N − 1)c + 4c 2 X jNE = 2cA − h . maxπ 1 = b A(X 1 + X −1 )− 0.N. have identical benefit and cost functions. . Xj.

which maximizes its profit by backward induction. i.e. X SPE = .e. we have X1NE < XjNE. and ∂π NE / ∂h > 0. country 1 can use its first mover advantage by choosing a point on the follower’s best response function. For b = 1 and h = 0.5(X 1 + (N − 1)R j (X 1 )) X1 [ 2 ] + h X 1 + (N − 1)R j (X 1 ) − cX 12 + T [ ]( ) i. payoffs increase with a marginal increase in h. we can show that ∂π1NE / ∂h > 0. The solution for the subgame perfect equilibrium is A(N − 1 + 2c )− h 2cbA + h(N − 1 + 2c ) . we have the perfectly symmetric NE for the simultaneous decision protocol with NA . A and X NE = N + 2c N + 2c For b < 1 and a sufficiently small h. j 2 2 2cb + (N − 1 + 2c ) 2cb + (N − 1 + 2c ) 2cbA + A(N − 1)(N − 1 + 2c )+ 2ch . X 1NE = X jNE = Subgame perfect equilibrium (SPE) for the mixed sequential-simultaneous decision protocol Country 1 maximizes the difference between benefit and costs of abatement given the knowledge that the N − 1. Sturm and J.e. which is an element of the real payoff function of country 1 here. For profits. i.2) max π 1 = b A (X 1 + (N − 1)R j (X 1 )) − 0.180 B. This effect is independent of the relative size of the marginal benefit of country 1. ∂XjNE / ∂h < 0. ∂πjNE / ∂h > 0. We are interested in analyzing the effects of a marginal increase in h.a. and ∂X NE / ∂h > 0. It is easy to show that the results of the general model hold for our specification.e. ∂X1NE / ∂h > 0. Weimann and X NE = X 1NE + (N − 1)X jNE = (b + N − 1)A + h N − 1 + 2c + b . other countries will behave according to their best response function (equation 8. the country with the smaller marginal benefit from abatement (here country 1) abates less than all other countries in NE. 2 2cb + (N − 1 + 2c ) X 1SPE = and X SPE = X 1SPE + (N − 1)X SPE = j . i.

given by X iPO = A(b + N − 1)+ h A(b + N − 1)+ h . Thus coordination seems to be a necessary condition for cooperation. 56). a lower profit for country [IE51]. Unilateral reductions weaken the position of the leading country. the global planner has the following optimization problem: ⎛c ⎞ max π = π 1 + (N − 1) π j = (b + N − 1) AX − 0. Pareto optimum (PO) In the Pareto optimal allocation. X ⎝N ⎠ ( ) In the PO. For profits. Since all countries have the same marginal abatement costs. total profit from abatement is maximized. and XSPE < XNE. For abatement. and therefore ∂πPO / ∂h > 0. total emission reduction after international negotiations may be lower in the case of ex ante reductions of the leading country compared to the case without unilateral abatement. Regarding profits.Unilateral emissions abatement 181 We can show that X1SPE < X1NE. .e. in the PO all countries choose an equal abatement. i. Therefore. 5 See Andreoni (1995) for a public bad game. we can show that ∂π1PO / ∂h > 0. The influence of parameter h on abatements and profits is the same as in the simultaneous case. XjSPE > XjNE. 3 Hoel also shows that this positive result no longer holds if the unilateral abatement is followed by international negotiations on emission reductions. i. 2 Brosig et al. and (for the chosen parameters) a lower aggregated profit. independent of the protocol of play. Ci′ = 2cXi. and X PO = c c ⎛ ⎞ b + 2 + N −1 N ⎜ b + 2 + N − 1⎟ N N ⎝ ⎠ It is easy to show that XPO > XNE holds. the change from the simultaneous decision protocol to the mixed sequential-simultaneous decision protocol leads to a lower aggregated abatement.e.5 X 2 + hX − ⎜ X 2 + NT ⎟ . all countries adjust their marginal abatement costs to the marginal social benefit from abatement. 6 A more complete version of the model can be found in the appendix. it follows that ∂XiPO / ∂h > 0 and ∂XPO / ∂h > 0. (2003) was able to show in a standard public good experiment with communication that successful cooperation only occurs if subjects had the opportunity to coordinate their behavior by face-to-face communication. Notes 1 See Kaitala et al. the change from the simultaneous decision protocol to the mixed sequential-simultaneous decision protocol leads to a higher profit for country 1. (1992). 4 Hoel explicitly points out that “I do not take up the question of whether such action from one country might lead to similar behavior from other countries” (p. ∂πjPO / ∂h > 0. We do not deal with negotiations on emission reduction in this chapter but only look at decentralized decisions about abatement.

our frame provides a “worst case” scenario for game theory.182 B. 217–242. The difference in NE and SPE profits of country 1 between the simultaneous and sequential treatments is too small to get significant results (see Table 8. Barrett. 25 See Sturm and Weimann (2006) for a detailed methodological discussion. 24 Also see Figure 8. 4. (1995): Warm-glow versus Cold-Prickle: The Effects of Positive and Negative Framing on Cooperation in Experiments. 10 percent level). BMU (2002): Regierungserklärung von Bundesumweltminister J. 20 The difference is weakly significant for T III (Wilcoxon signed-rank test. (1994): Self-Enforcing International Environmental Agreements. Braudts. 1–21. Sturm and J.. the frame of the decision problem is irrelevant for individual behavior. 10 However. 18 However. A. Oxford Economic Papers. 15 Subjects who tried to maximize the collective profit given the expected abatement of the others could compute their “best response” with the help of the simulator.e. References Andreoni. S. 46. On the other hand. 13 We used Z-tree for programming. the hypothesis for country j is questionable due to the very small difference in the NE and SPE abatement of country j between treatments with h = 0 and h > 0 (see Table 8. Ockenfels. 22 See Weimann (1994) and Fischbacher et al. J.3). that they behave more cooperatively in a social dilemma with an environmental frame. 11(1). Quarterly Journal of Economics. German Economic Review. 17 If not otherwise stated all the following tests are exact and two-sided.B. 21 However. i. if we assume that people have “green preferences”.3). 36–63. 9 We assume that country 1 has no free rider option in the sequential treatments. we have to discriminate between the increase in profits in our experiment and the positive welfare effect that Hoel describes in his model for the simultaneous game. 12 Here the same problem as in Note 10 appears. 11 At this point. Weimann (2003): The Effect of Communication Media on Cooperation. the individual best response to the aggregated abatement of the others in the previous period.2 where these five groups can be easily identified. Trittin vor dem Deutschen Bundestag am 22 Marz 2002. i. We allowed for one decimal place in the sequential treatments. and W. The increase in profits is caused by a change in preferences due to parameter h. J. Games and Economic Behavior. The positive welfare effect assumes that country 1 voluntarily abates more than in the equilibrium and the resulting welfare loss is compensated for by the gain of all other countries. 110. Brosig. . 14 Instead of the term “best response” we use the term “profit maximizing response” because the latter is more neutral in our view. significant differences for the αt values can only be observed for few periods. 19 We get a similar picture if we generate this classification based on “myopic best response” behavior. and J. we have to admit that the treatment T III seems to be an outlier as country 1 is not able to realize even the SPE profit. 878–894. 16 From the viewpoint of game theory. 8 Subjects had to enter integers in the simultaneous treatments. Weimann 7 See the appendix. J. which is an element of the real payoff function of country 1.e. See Fischbacher (1999). Macleod (1995): Equilibrium Selection in Experimental Games with Recommended Play. (2001). 23 In three of 36 groups we observe behavior near the SPE for both the leader and the followers.

. 54. 91. M. (1994): Individual Behavior in a Free Riding Experiment. and E. 13/00. and J. 71. Renner (2003): Leading by Example in the Presence of Free-Rider Incentives. SNF Report No. Sutter. Gächter. Journal of Economic Surveys. (1999): z-Tree: A Toolbox for Readymade Economic Experiments. M. Working Paper. V. Weimann (2006): Experiments in Environmental Economics and some Close Relatives. Moxnes (2000): Leadership and Following in a Public Bad Experiment. Environmental and Resource Economics. (1991): Global Environmental Problems: The Effects of Unilateral Actions Taken by One Country. Gächter. Levati. University of St. U. Working Paper.. 1023–1042. S. 161–181. van der Heijden (2007): Leading by Example With and Without Exclusion Power in Voluntary Contribution Experiments. and E. and E. Gallen. Güth. W. Journal of Public Economics. S. 185–200.. Bergen. M. Sturm. 419–457. 20. 397–404.Unilateral emissions abatement 183 Fischbacher. Van der Heijden. Journal of Environmental Economics and Management. Weimann. Journal of Public Economics.. M. Kaitala. E. B. and E. Fischbacher. 2. U. 55–70. Economic Letters. J. University of Zurich. 21. Hoel. and O. Working Paper No. Pohjola. Fehr (2001): Are People Conditionally Cooperative? Evidence from a Public Goods Experiment. 20. Tahvonen (1992): Transboundary Air Pollution and Soil Acidification: A Dynamic Analysis of an Acid Rain Game between Finland and the USSR.

2001). the returns from the public good (Fisher et al. then a reduction in charity options (e. Cherry and David L. But the consensus has arisen almost entirely from experimental settings that present a single public good.) or quite similar (e. given that US charitable organizations have annual revenues of around $600 billion. anything that may increase . in the laboratory. and many others (see Ledyard. green energy program. Previous experimental work has consistently shown individuals deviate from the dominant strategy to make zero contributions in a linear public good game. Experimental research has highlighted many critical issues affecting the ability of a collective to provide public goods. But. punishment and rewards (Fehr and Gächter. Charitable organizations are but one example of public goods funded through voluntary contributions. neighborhood clean-up programs. The lack of attention directed to this issue is surprising considering life outside the laboratory generally has people contemplating a variety of public goods (Cornes and Itaya. merging charities) could increase total donations to charities. On the other hand.g. etc. 2005). 1995. In this chapter..g. member heterogeneity (Chan et al. 1995). 1999. An area that has received little attention is how individual contributions are affected by the presence of multiple. such as how contributions are affected by the size of the collective (Isaac and Walker. Consequently. if multiple public goods options is shown to reduce contributions. 2003).9 Voluntary contributions with multiple public goods Todd L. 2000.g. 1988). Dickinson. Cherry et al. Any evidence of reduced or destabilized contributions in the face of multiple options has significant implications. we extend the literature by examining whether the presence of more than one public good option affects total contribution levels.. or free riding. Evidence shows that people generally make positive though suboptimal contributions to the public good.. for a review). with the alternatives sometimes being diverse (e. three similar land conservation groups seeking your financial support). a larger number of charity choices may be advisable if a larger number of options is shown to increase overall giving. Dickinson Introduction The provision of public goods is a thoroughly examined issue in experimental and environmental economics. the question arises how the presence of multiple public goods will affect individual contributions. For example. competing public goods.

They include a treatment in which subjects may contribute to either a broadcast public good and/or an anonymous public good. This results from the fact that using the lottery for the less socially desirable public good appears to siphon contributions away from the more socially desirable public good. less socially desirable. and Moir (2005) all include multiple public goods environments in their experimental research. This treatment with two public goods options increased total contributions to public goods (i. Theoretical examinations of multiple public goods environments are less common (Kemp. none of the existing research has examined the pure effect of increasing the number of public goods options. Experimental economists have only recently become interested in examining the behavioral effects of having multiple public goods options. They speculate that a variety of factors likely contribute to their results.e. Blackwell and McKee (2003) examine a multiple public goods environment with a local versus global public good competing for subject contributions. where marginal incentives to contribute differ across public goods. That is. including altruism. A main conclusion of their research is that there may be complementarities in voluntary contributions across multiple public goods in environments such as theirs. learning. Cornes and Itaya. increased efficiency) relative to their most efficient single public good treatment. 1984. Existing research has quite thoroughly examined the voluntary contributions mechanism (VCM) for provision of public goods. he introduces a second public good along with rank-ordered social desirability of the two public goods. Moir finds that a second. 1996. 2003). though they each have a clearly distinct focus.Voluntary contributions 185 (or decrease) contribution levels by even a small amount would have a sizable monetary implication. Andreoni and Petrie (2004). Expanding on the experimental environment of Morgan and Sefton (2000). While lottery funding of public goods is shown to be efficient under risk neutrality. The marginal monetary incentives to contribute are identical for both public goods. 1986. We start with a traditional single public good VCM environment. Blackwell and McKee (2003). rising marginal benefits to global public goods seem to induce increased contributions. two) public goods into various experimental treatments. Bergstrom et al. Though researchers have now begun to incorporate multiple (i. which differ only in their information level.e. a result that does not occur when the public goods are both funded through a standard voluntary contributions mechanism. though most subjects chose to contribute to the non-anonymous broadcast public good. coordination.. This is the goal in the present chapter. and possibly reciprocity. Our Nash prediction in each stage of our five-round . Moir (2005) examines multiple public goods in the context of Morgan’s (2000) lottery provision mechanism for funding public goods. public good can decrease efficiency if funded through a lottery mechanism. Cornes and Schweinberger. but subjects still maintain contribution levels to their local group public good. Andreoni and Petrie (2004) study contribution levels when subjects are identifiable or not to other members of their group.

each individual i divides her endowment Ei into contributions to the jth public good account. In a second treatment. as defined by the marginal per capita return (MPCR) in Isaac et al.L. Thus. All environments share the traditional Nash free riding prediction (in the absence of any utility of group participation) and Pareto efficiency at full contributions. We also examine a third treatment.1 However. If individuals derive utility from coordinating contributions to the same public good. and our results indicate that coordination on the variable-MPCR public good occurs in a very deliberate and rational way. Theory and experimental design We examine the issue of multiple public goods in a linear public goods environment. and so total giving might decline. to this good is strictly less than the marginal incentive to contribute towards a second public good. In this framework. (1984). and so a comparison of contribution tendencies across these two treatments yields a measure of the pure effect of going from a single to a multiple public goods environment. indicating increasing marginal benefits of contributing. where we make the multiple public goods options heterogeneous by altering the marginal incentive to contribute across three public goods options – for one of the public goods options. j=1 m . Thus. the multiple heterogeneous treatment allows us to examine subject rationality and expectations. the marginal incentive to contribute is a monotonically increasing function of the number of total group contributions. On the other hand. in our third treatment one of the three public goods options is strictly dominated in the sense that the marginal incentive to contribute. then multiple options makes coordination more difficult in the absence of any communication. These three public goods are identical in every way. we have three public goods that are identical to the public good option in the single public good treatment – call this treatment multiple homogeneous.186 T. multiple heterogeneous. the effect of multiple and identical public goods options is ultimately an empirical question. Cherry and D. xij (0 ≤ ∑x j =1 m j i ≤ Ei ) and a private good. Ei − ∑ xij . reframing the VCM environment as one with multiple options may induce additional contributions if individuals consider it important to fund each public good at some level.L Dickinson baseline treatment is complete free riding – zero contributions to the public good. We nevertheless expect a positive level of voluntary contributions given the wealth of previous research documenting such over-contribution relative to the Nash outcome for this particular setup.

1.1 Experimental design Treatment Public Accounts MPCR Linear.Voluntary contributions 187 The n members of a group make their contributions decisions to the m public goods independently and simultaneously. 0. The sessions consisted of ten rounds with treatments varying within and across sessions. 0. Rounds 6–10 Multiple homogeneous Multiple heterogeneous Multiple heterogeneous Multiple homogeneous Number and MPCR of public accounts by treatment Baseline One Account (B) Multiple homogeneous Three Accounts (B) (C) (D) Multiple heterogeneous Three Accounts (B) (C) (D) Rounds 1–5 Treatment ordering and participation by group Groups 1 and 2 Baseline Groups 3 and 4 Multiple homogeneous Groups 5 and 6 Baseline Groups 7 and 8 Multiple heterogeneous Note a See Figure 9. 1984) from a contribution to the jth public good.1 provides an overview of the experimental design with treatment parameters and ordering. [0. and the monetary payoff π 0 for each i member i is π io = Ei − ∑ xij + ∑ α j X j .1) in which 0 < α j < 1 < nα j.04]a .6 Linear.. 0. Subjects were placed in groups of four. while the final five rounds presented parameters from a different treatment. The first five rounds of each session presented parameters from one of the three treatments.6 Linear. The constraint on α j ensures that the individually optimal contribution to the public good is zero. j =1 j =1 m m (9.45. Table 9.3 Nonlinear. 0. and each group participated in two of the three treatments.6 Linear. Thirty-two students from Appalachian State University participated in a public goods experiment consisting of three treatments and two sessions.6 Linear. 0. where α j is the marginal per capita return (Isaac et al.6 Linear. 0. and X j= ∑x k =1 n j k where X j is the total contribution to the jth public good. although the socially optimal outcome is achieved when all group members contribute their entire endowments to the public good.1. Table 9.

contributions were 79 percent of endowments in the first round and fell to 65 percent in the final round – significantly . We first review results from the single public good baseline treatment. 1. The second treatment introduced two additional public accounts to the allocation decision. accumulated individual payoffs were totaled and subjects were paid in private as they departed individually from the laboratory. C. but the payoff structure varied across the public accounts. but the three public accounts provided marginal per capita rates of return: MPCR = 0. Subjects were endowed with 15 tokens and randomly assigned to groups of four. The return from the private account remained unchanged from the previous treatments. The return from the private account remained unchanged from the baseline treatment.6 (account B). making the allocation decision.45. Group account payoff tables provided to subjects are available upon request.04] as represented in Figure 9.3 (account C). and payoffs were calculated and announced privately to each subject. Cherry and D. C. The three public accounts provided the same return with the marginal per capita return being identical to the baseline public account. the payoff structure did not change – only the number of public goods changed. but subjects did not know the identities of other group members.2 provides an overview of average group contributions by treatment. Individual contributions were recorded. After the final round. in this treatment. each with subjects receiving an initial endowment. and D). Treatment 2: multiple homogeneous. Therefore. and learning the subsequent outcomes and payoffs. Subjects decided how to allocate their endowment between one private account (A) and three public accounts (B. and MPCR [0. In the multiple homogeneous treatment. As in previous experimental research. Results from the multiple public good treatments indicate that the frame of additional public goods positively impacts total contribution levels. and D). we observe initial contribution levels significantly above the theoretically predicted zero and decline with repetition. or a public account B that generated a 0. Treatment 3: multiple heterogeneous. Subjects in this treatment therefore faced the decision of how to allocate their 15 token endowment between four accounts: one private (A) and three public (B. Subsequent rounds followed. The final treatment introduced variable payoff structures to the multiple public goods treatment.188 T.6 cent return to each group member for every token the group placed in this account.1 (account D). Subjects contributed 63 percent of their endowments in the first round and 54 percent in the final round. Results Figure 9. The baseline treatment follows the well-established literature in linear public good games. Subjects simultaneously and privately decided how to allocate their endowment between two accounts – a private account A that generated a 1 cent return to the individual for every token she placed in this account. MPCR = 0. Group affiliation remained unchanged across the session.L Dickinson Treatment 1: baseline.L.

2 1. Results reveal total contributions were significantly greater in both the multiple public good treatments than in the single public good baseline.8 more tokens to public goods in the multiple homogeneous treatment than in the baseline treatment.96. Estimates associated with the multiple homogeneous treatment. greater than the baseline (t = 3.6 0.1 MPCR by group account for multiple heterogeneous treatment.4 0.8 MPCR 0.6 0. Table 9.2 0 1 2 3 Round 4 5 Baseline Multihomo Multihetero Figure 9. and 3. We confirm these unconditional results by estimating the treatment effects on individual contributions using a two-way fixed effects panel model that controls for subject and round specific effects.41. p < 0.2 0 20 40 60 80 100 Group contributions as percentage of total tokens Account B Account D Figure 9. p < 0.2 Total contributions to group accounts by treatment. .0 0.58 more tokens in the multiple heterogeneous treatment. Note Contributions measured as a percentage of total group endowment. contributions were 75 percent of endowments in the first round and fell to 71 percent in the final round – significantly greater than the baseline (t = 4.001). which 1 0. Estimated coefficients indicate that subjects contributed 2.Voluntary contributions 189 1.001).2 presents the results for the total contributions model (column 1).4 Account C 0.8 Contribution 0. In the multiple heterogeneous treatment.

and D were 31.000 0. and 41 percent in the first round and 32.56 320 presents three identical public goods. 28.and round-specific effects.004 320 9. . and public account D provides a higher MPCR if contributions exceed 25 percent of total endowments.58 2.6 0. Figure 9.38 – 2.2 Individual contributions: two-way fixed effect estimatesa Coefficient Constant Baseline Multiple homogeneous Multiple heterogeneous F N Note a Individual. p-value 0.L Dickinson Table 9.190 T.3 Group contributions across homogeneous competing group accounts. suggest a framing effect that significantly increases total contributions. 17.80 3. the homogeneity of marginal per capita rates of return fails to provide any direction to subjects.4.8 Contribution 0. As one may expect.4 0. The allocation of contributions in the multiple heterogeneous treatment is illustrated by Figure 9.001 0. C. Recall the MPCR for public account B is strictly greater than that of public account C. We therefore should expect subjects to strictly prefer public account B to C. The proportion of total (public account) contributions going to public good B. and 51 in the final round.3 presents the allocation of contributions in the multiple homogeneous treatment and reveals no significant coordination by subjects. Cherry and D.L. Note Contributions measured as a percentage of total group endowment. and prefer D if they expect total contributions to be at least 25 percent of the endowments.000 – 0. 1 0. We now examine the allocation of contributions across the three competing public good accounts in the multiple account treatments.2 0 1 2 3 Round 4 5 Account B Account C Account D Account B + C + D Figure 9.

Conclusions The goal of this chapter is to fill a void in the literature on public goods experiments and contributions in multiple public goods environments. Subjects therefore appear quite able to understand and respond to the relative payoffs in this multiple public goods setting. An additional treatment that we examine also highlights the role that rationality and expectations play in the subjects’ giving behavior.9 percent in the first round. In the final round. Note Contributions measured as a percentage of total group endowment.2 Subjects also appear to learn across rounds. with accounts B and C receiving 26. As Figure 9.Voluntary contributions 191 1 0. it is important to understand the pure effect on cooperation of moving from a single to a multiple public goods environment. and they also seem to expect correctly others to identify and contribute to account D. Individuals are often in the position of choosing how to distribute their wealth among private versus public goods options.4 shows.7 percent. Our results indicate that more options for otherwise similar public goods will increase total dollar contributions towards . 100 percent of contributions went to public account D. Observed behavior reveals subjects understood the game and rationally responded to the heterogeneous incentives of the multiple public accounts. at least conditional upon the decision to contribute.2 0 1 2 3 Round 4 5 Account B Account C Account D Accounts B + C + D Figure 9. as is evidenced by convergence towards a single public good account when it is more efficient to do so.4 0. As experimental economists have started to become interested in these decision environments.8 Contribution 0. the proportion of contributions going to the public good D was 63.6 0.4 and 9. The implications of this research extend beyond the laboratory.4 Group contribution across heterogeneous competing group accounts. We uncover no evidence of confusion among subjects. Our results show clear evidence that total contributions and efficiency are increased when the number of public goods options are increased from one to three in an otherwise standard VCM environment.

Cherry. 2004. there is a private marginal incentive to contribute. we have separated and identified an important effect that has been previously hidden in the literature. 1995. So. Stephan Kroll. Shogren. References Andreoni. and R. “Public goods experiments without confidentiality: a glimpse into fund-raising. Notes 1 Once contributions reach 93 percent of the total group endowment.” Journal of Economic Behavior and Organization.. Schweinberger.” Experimental Economics.. Kenneth S. the variable MPCR account in multiple heterogeneous reaches MPCR = 1. “The impact of endowment heterogeneity and origin on public good contributions: evidence from the lab. Richard and A.” Journal of Economic Behavior and Organization. “Heterogeneity and the voluntary provision of public goods. At least in the context of voluntary giving. 2 Our design is not capable of determining whether the initial decision to “cooperate” is due to confusion or subject preference. and Jason F. 52(1): 115–131. James. 1986. disaster relief agencies may have private objectives that do not fully align with the public goal of providing the public good. Andreoni (1995) finds some evidence of both kindness and confusion. 2003. Blackwell. and so the real world incentive environment is undoubtedly more complex than what we study in this chapter. Stuart Mestelman. Cherry and D.192 T. if expectations are high that group members will contribute nearly all of their tokens to this one particular group account.” Journal of Public Economics. 1999.” Journal of Public Economics. 1996. and H. at the extreme. Of course. “Only for my own neighborhood? Preferences and voluntary provision of local and global public goods. James and Ragan Petrie. “Free riding and the inefficiency of the . Nevertheless. 2005.L Dickinson the larger cause. 57(3): 357–365. Stephan Kroll. For example. although our evidence on rational choice among public goods options in our heterogenous accounts treatment leads us to minimize the potential effect of confusion in our data generation. Acknowledgments We would like to thank the participants of the 2005 Experimental Economics and Public Policy Workshop at Appalachian State University. Andrew Muller.C. “On the private provision of public goods. Calvin and Michael McKee. Blume. Mike McKee. Chan. this effect of multiple options seems to indicate that more is better. T. Varian. Andreoni. multiple options for providing relief following an environmental disaster is predicted to increase the total amount of voluntary giving among private citizens. Timothy Perri. 29(1): 25–49.G. L..L. 2(1): 5–30. Cornes. “Cooperation in public-goods experiments: kindness or confusion?” American Economic Review. Bergstrom. 85(4): 891–904. Robert Moir. 88(7–8): 1605–1623. Todd L. and two anonymous referees provided helpful comments.

J. 1984. Department of Social Sciences (Economics). Dickinson.M.” Quarterly Journal of Economics. and S. Morgan. J. Handbook of Experimental Economics. M. 2000.” in John H. 1984. and J. “Financing public goods by means of lotteries. Robert. .” American Economic Review.. Moir.” Experimental Economics. 1988. Morgan. Fehr. “Cooperation and punishment in public goods experiments. 2005. M.” Review of Economics Studies. 43(1): 113–149. David L. Richard and Jun-ichi Itaya. 2000. Fisher. Walker. “Heterogeneous demand for public goods: behavior in the voluntary contributions mechanism. Kagel and Alvin E. 85(3–4): 249–266. Cornes. “Public goods: a survey of experimental research. 2003. Isaac. Schatzberg. Walker.C. the stick in work team motivation. J. “The carrot vs.” Public Choice. 111–194. John and Martin Sefton.” Working Paper.” Review of Economic Studies. 29(1): 70–91.).” Canadian Journal of Economics. 2001.” University of Nottingham Discussion Paper 03/21. 1995. Isaac. 2000. “Funding public goods with lotteries: experiment evidence. 90(4): 980–994. pp. Thomas. 1995. “A note on the theory of international transfers. John. Ledyard. “Group size effects in public goods provision: the voluntary contributions mechanism. 67(234): 785–810. “Divergent evidence on free riding: an experimental examination of possible explanations. Princeton: Princeton University Press.Voluntary contributions 193 private production of pure public goods. and S. Roth (eds. R. J.” Public Choice. Kemp.O. Walker. Gächter. Isaac. R. 67(234): 761–784.” Economics Letters. E.. University of New Brunswick. Mark and James M. “Models with two or more public goods. 103(1): 179–199. 4(1): 107–124. “Multiple public goods and lottery fundraising. 14(2–3): 259–262.

Participants are given an endowment of “tokens” to be divided between a private account and a public account. Understanding behavior in experimental implementations of the VCM game is critical for the work of economists with institutional and policy-oriented interests. is simply money. To explore how individuals behave in various public goods decision settings and to gain insights into how institutions might be better designed to encourage the provision of public goods. subjects contribute at levels far above this: on average. In repeated-round VCM experiments.e. In single-round VCM experiments where a public good contribution rate of zero is the unique Nash equilibrium. which is non-rival and non-excludable in consumption. Contributions to the private account are converted to cash and given to the individual. Vossler Introduction Public policy is frequently used to induce individuals to contribute to public goods when it may be in their private interests to free-ride off the contributions of others. Ferraro. . the individual free rides) while the social optimum is realized when everyone contributes their entire endowment to the public account. the individually rational contribution is zero (i. economists employ laboratory experiments. 40–60 percent of endowments. The cornerstone of experimental investigations on the private provision of public goods is the voluntary contributions mechanism (VCM). Paul J.10 Can public goods experiments inform policy? Interpreting results in the presence of confused subjects Stephen J. contributions start in the range of 40–60 percent but then decay towards zero (ending around 10 percent of endowments on average). but the sum of the marginal returns to the group is greater than the value of a token kept. If the marginal return from contributing a token to the public account is less than the value of a token kept in the private account. there seem to be motives for contributing that outweigh the incentive to free ride. Cotten. Contributions to the public account yield a cash return to all group members. including the contributor. Thus. The standard linear VCM experiment places individuals in a context-free setting where the public good. and Christian A.

but find significant conditional cooperation. 2001). Fischbacher et al. Marwell and Ames. is “confusion. (2002) find the opposite. Nowell and Tinkler (1994) find females are more cooperative. Whereas many studies find that economics students or economists are less likely to contribute to public goods in experiments (e. There are several possible reasons for these puzzling results. Cadsby and Maynes (1998) find no significant differences between men and women.g. 2005). Frey and Meier. 1998). Laury and Taylor (forthcoming) use behavior in a one-shot VCM experiment to predict behavior in a situation in which individuals can contribute to an urban tree planting program. Palfrey and Prisbrey (1997) find statistical evidence of warm-glow but no evidence of pure altruism. to estimate a pure altruism parameter for each subject. 2004). 1988. (2001) and Fischbacher and Gächter (2004) find no evidence of pure altruism or warm-glow. A particularly well-studied issue is whether contributions behavior differs between men and women (Eckel and Grossman. the authors find little or no relationship between subjects’ altruism parameters and subjects’ contributions to urban tree planting. and participants used in these experiments. A fourth motive. This chapter presents results from one new experiment and two previous experiments that use the .. Using the empirical approach of Goeree et al. Particularly troubling is the apparent lack of correspondence between contributions behavior in experimental and naturally occurring settings. implementation. 1996. which describes a situation in which an individual gains utility from the simple act of contributing to a publicly spirited cause.” We define confusion as behavior that stems from the failure of an individual to identify the dominant strategy of zero contributions. 1981. which describes a situation in which an individual’s utility function is a function of his own payoff and the payoffs of her group members. For example. Brown-Kruse and Hummels (1993) find that men contribute more than women..Public goods experiments and policy 195 Possible motives underlying contributions include: (1) “pure altruism” (sometimes called “inter-dependent utility”). including differences in the design. and (3) “conditional cooperation” (Andreoni. a public good. Investigations into the identification and relative importance of various motives for contributions in the VCM game and closely related games have led to conflicting conclusions. 1999. We focus on an alternative explanation: confusion confounds the interpretation of behavior in public goods experiments. 1990). Cadsby and Maynes. which the VCM literature often ignores but we are particularly interested in. which is a predisposition to contribute in social dilemmas but punish by revoking contributions when significant free riding behavior is observed. Fischbacher et al. attempts at externally validating this claim yield contradictory results (Yezer et al. Efforts to compare public goods contributions across different subpopulations have likewise led to mixed results. Laband and Beil. Andreoni. Goeree et al. More broadly. (2) “warm-glow” (often called “impure altruism”. and individuals do not understand how to utility-maximize in the context of the game. confusion behavior results from a failure for individuals to discern the nature of the game.

Houser and Kurzban find that confusion accounts for 54 percent of all public good contributions in the standard VCM game.” a novel methodology for detecting confusion through a split-sample design where some participants play with non-human players (automata) that undertake predetermined strategies or choices. perhaps through poorly prepared instructions or inadequate monetary rewards. and in attempts to use experimental results to improve policy design.J. Andreoni uses behavior from the ranking games to infer that both other-regarding behavior and confusion are “equally important” motives in the VCM. Thus contributions to the “public good” in this game do not increase aggregate benefits.196 S. but only confusion is present in the computer condition. Each experiment involves a slightly different public goods game and a different subject pool (with presumably differing abilities). Cotten et al. . Andreoni (p. in research that assesses the external validity of experiments. In each round the aggregate computer contribution to the public good is three-quarters of the average aggregate contribution observed for that round in the human condition. By making the reasonable assumption that other-regarding preferences and confusion are present in the human condition. The level of confusion in all experiments is both substantial and troubling.” which is similar to a standard VCM game except that each group consists of one human player and three nonhuman computer players (which we refer to as “virtual players”) and the human players are aware they are playing with computer players. To test his confusion hypothesis. or simply that many subjects are incapable of deducing the dominant strategy through the course of the experiment. “virtual-player method. Andreoni developed a VCM-like game that fixes the pool of payoffs and pays subjects according to their contributions to the public good. Prior evidence of confusion in public goods experiments Andreoni (1995) was the first to identify and test the hypothesis that confusion plays an important role in the contributions decisions of participants in public goods games. 893) hypothesizes that the experimenters may have failed to convey adequately the incentives to the subjects. and present results from a pilot study that uses “context-enhanced” instructions. Houser and Kurzban (2002) continued Andreoni’s (1995) work with a clever experimental design that includes: (1) a “human condition. The person who contributes the least is paid the most from the fixed pool. but merely cost the contributor and benefit the other group members. in studies that compare behavior across subpopulations. We conclude by proposing ways to mitigate confusion in standard public goods experiments. Specifically. and (2) a “computer condition. These experiments provide evidence that confusion is a confounding factor in investigations that discriminate among motives for public goods contributions.” which is the standard VCM game.

3). The authors estimate a logit choice model of noisy decision making with data from a series of one-shot VCM games (no feedback) in which the internal and external returns are varied. Similar to Palfrey and Prisbrey. Goeree et al.” While no point estimate was given of the proportion of contributions stemming from confusion in their experiment. no estimate of the fraction of contributions due to confusion is given. they present evidence showing that this confusion does not decline with experience. They find that approximately 54 percent of contributions are due to confusion.1 Palfrey and Prisbrey (1997) developed an experimental design that. and applied it to a single-round VCM game. on the other hand. (2002) use a VCM design in which group size is either two or four and the “internal” return of a subject’s contribution to the public good to the subject may differ from the “external” return of the same contribution to the other group members. when combined with a few behavioral assumptions. how much they would contribute to the public good. allows the authors to separate the effects of pure altruism. (2003) independently developed a similar design with virtual players.” they ask subjects to specify. “at most 17. for each average contribution level of the other group members. In response to the fifth study. Ferraro and Vossler point out that Fischbacher and . They find no evidence of pure altruism or warm-glow (no subjects stated they would contribute if other group members contributed zero). where they find that 52 percent of contributions across rounds stem from confusion. Overall. However. Fischbacher and Gächter (2004) design an experiment to test specifically for the presence of conditional cooperation. (1984) are attributable to error. Their design changes the standard VCM game by randomly assigning different rates of return from private consumption each round.” which is a standard VCM game with four-person groups. and confusion. The authors conclude that (p.5 percent” (p. In Fischbacher and Gächter’s “P-experiment. This difference stems from the atypical behavior that Hauser and Kurzban’s all-human condition exhibits (little decline in contribution) and two other aspects of their design that make it difficult to directly compare the human and computer conditions.Public goods experiments and policy 197 Ferraro et al. which enables the measurement of individual contribution rates as a function of that player’s investment costs. warm-glow. the authors use their model results to predict that “well over half” of contributions in the seminal VCM experiments by Isaac et al. This conclusion is alarming. In contrast to previous work. they claim confusion accounts for a smaller fraction of observed contributions to the public good. warm-glow effects and random error played both important and significant roles. They find that coefficients corresponding to pure altruism and decision error are both positive and significant. four of the five studies above that assess magnitude find that about half of all contributions stem from confusion. Fischbacher and Gächter argue that most contributions come from conditional cooperators. 842) “altruism played little or no role at all in the individual’s decision and. Ferraro and Vossler (2005) extend this design to the multi-round VCM. By comparing the responses in this experiment with those in their “C-experiment. unlike Hauser and Kurzban.

The virtual-player method The virtual-player method discriminates between confusion and other-regarding behavior in single-round public goods experiments. Despite this dispute. as the contribution profile for a virtual player suffices to ensure comparability. However. automata) that are preprogrammed to exercise decisions made by human players in an otherwise comparable treatment. Gächter’s characterization of conditional cooperators also describes the behavior of confused “herders” who simply use the contributions of others as a signal of the private payoff-maximizing strategy. in the typical multiple-round public goods game where group contributions levels are announced after each period. The virtual-player method makes each human subject aware that he or she is grouped with virtual players that do not receive payoffs and that make decisions that are exogenous to those of the human. Indeed. virtual players (i. The random assignment of participants to an all-human group or a virtualplayer group allows the researcher to net out confusion contributions by subtracting contributions from (human) participants in the virtual-player treatment from contributions in the all-human treatment. and (3) a procedure that ensures that human participants understand how the non-human. The method relies on three important features: (1) the introduction of nonhuman. Thus. (2003) for other applications). In single-round experiments where the decisions of other players are not known ex ante. (2) a split-sample design where each participant is randomly assigned to play with humans (the “all-human treatment”) or with virtual players (the “virtual-player treatment”). virtual players behave. the research on confusion in public goods experiments can be succinctly summarized: every study that looks for confusion finds that it plays a significant role in observed contributions. Ferraro and Vossler find that confused individuals use past contributions of virtual players as signals of how much to contribute. Thus the method neutralizes the otherregarding components of the human participant’s utility function and the motives for strategic play.198 S. it is important to exercise additional control as the history of play may affect contributions in the virtual-player treatment. any contributions made by humans in virtual-player groups can be attributed to confusion in the linear VCM game. the contributions from virtual players should have no effect on human contributions nor should they confound any comparison between all-human and virtual-player treatments. as long as participants understand their decision environment. Cotten et al. The additional control comes by establishing that each human in the all-human treatment . the proportion of confusion contributions (17. As such.2 Thus.5 percent) found by Fischbacher and Gächter may be best characterized as a lower bound estimate. one can argue that randomly selecting the profile of any previous human participant. with replacement.e.J. and discriminates between confusion and other-regarding behavior or self-interested strategic play in multiple-round experiments (see Ferraro et al.

confounds comparisons. and ranged from 2–12 cents. one playing exactly like human subject H1 and the other playing exactly like H3. The participants are told that inside the envelope are the choices for each round from the virtual players in their groups. For each decision task. To ensure that participants in the virtual-player treatment believe the virtualplayer contributions are truly preprogrammed and exogenous. Application of the virtual-player method to the Goeree. Group size was either two or four players. Note that having an imbalance between all-human and virtual-player treatments.Public goods experiments and policy 199 has a human “twin” in the virtual-player treatment: each twin sees exactly the same contributions by the other members of her group in each round. and H3. Holt. and Laury experiment The experiment of Goeree. and was either 2 or 4 cents. they can open the envelope and verify that the history of virtual group member contributions that they observed during the experiment is indeed the same as in the envelope. The subjects are informed that the reason we provide this envelope is to prove to them that there is no deception: the virtual players behave exactly as the moderator explained they do. The external rate of return refers to the marginal return to other players from one’s contribution to the public account. vary across tasks. Likewise. each subject has a sealed envelope in front of her. player V2 plays with two virtual players. which would occur if some participants do not have a “twin” or if a player in one treatment has multiple twins in other. and group size (n). not humans. and Laury (hereafter referred to as “GHL”) is a variant of the static linear VCM game that endeavors to test the significance and magnitudes of contributions stemming from pure altruism and warm-glow. and the other that makes the same choices H3 made. Subject V1 in the virtual-player treatment plays with two virtual players: one that makes the same choices H2 made in the all-human treatment. Each participant decides how to allocate 25 tokens between a private and a public account in each of ten “one-shot” decision tasks (referred to as “choices” in instructions). Post-experiment questionnaires are useful at assessing whether participants fully understand the nature of virtual players. the profit function of the individual i (in cents) for a particular decision task is given by π i = 5(25 − xi ) + mI xi + me ∑ x j i≠ j n . H2. where the internal (mI) and external rates (me) of return. without feedback. Participants in a group in the all-human treatment are labeled as H1. the only difference between the two treatments is that the player in the virtual-player treatment knows she is playing with preprogrammed virtual players. To illustrate. Formally. Thus. The internal rate of return refers to the marginal return to oneself from a token contributed to the public account. Holt. And so on. a token kept in the private account yielded 5 cents. consider a game that involves repeated interactions with groups consisting of three players. At the end of the experiment.

i.200 S. A total of 53 participants were recruited from a pool of undergraduate student volunteers at the University of Tennessee in the Spring of 2005. we pay participants based on three randomly chosen decisions instead of one. the conjecture is that contributions are largely attributable to warm-glow. which serves as a replication of the GHL design. The rates are varied in the GHL design because participants exhibiting pure altruism should increase their contributions when the external return or the group size increases.and four-member groups by selecting marked ping-pong balls after all decisions are made. This change increases the saliency of each decision. whereas 30 students participated in the virtual-player treatment.3 Experiment sessions consisted of groups ranging from four to 12 people. all players receive the same return from the public good. where xi ∈ [0. the external and internal rates of return are equal (mI = me). GHL randomly choose only one of the ten decisions to be binding using the roll of a ten-sided die and use a second. Since the internal rate of return in GHL is always lower than the value of a token kept. participants make decisions via paper and pencil. Such systematic correlations should be identifiable by observing patterns in individual contributions across the various decision settings. whose contributions are predetermined. The pre-assignment into groups shortens the length of both all-human and virtual-player treatments. Cotten et al. The instructions for all-human and virtual-player treatments are available from the authors on request. GHL assign subjects to two. We replicate the GHL experiment using the virtual-player method to explore whether conclusions drawn from the original study are robust after quantifying and netting out confusion contributions. Decision sheets are identical to GHL.e. Rather than engage our participants in a second experiment. The virtual-player instructions are similar with the exception of emphasizing that participants are matched with virtual players. a post-experiment questionnaire is given to collect basic demographic information as well as to assess understanding of the experimental design and decision tasks. In the typical one-shot VCM. with minor revisions. it is still the individual’s dominant strategy to contribute nothing. unrelated experiment to supplement earnings. so that full endowment contribution maximizes group earnings. Matching was anonymous. 25] denotes public account contributions from player i. Of these.J. We made two small changes in the way subjects were grouped and paid. subjects were not aware of the identity of the other members of . The all-human instructions are from GHL. The sum of the external and internal rates of return is always greater than 5 cents. Following the decisions. 23 students participated in the all-human treatment. Experiment instructions are presented both orally and in writing. but they show little correlation with external return and group size. As in GHL. If considerable contributions are observed. and participants were visually isolated through the use of dividers. We pre-assign participants to two. This is important for virtual-player sessions as it allows us to give each participant an envelope with the aggregate contributions of other players as well as earnings from virtual-player contributions for each possible decision selected.and four-member groups based on their subject ID number.

While µ measures dispersion and does not indicate the magnitude of confusion contributions. All sessions took place in a designated experimental economics laboratory. Results Goeree. For the sake of parsimony. This suggests that pure altruism is an important motive. the logit equilibrium model results for the all-human treatment suggest that pure altruism is an important motive. and we find this parameter to be statistically different from zero using a 5 percent significance level. The parameter µ is an error parameter.1 10 3 4 4 6 10. we refer the interested reader to the GHL study for details.8 11 4 4 2 2 5. Estimated logit equilibrium models are presented in Table 10. The pattern of contributions in relation to design factors is quite similar between this study and the GHL study.5 5 8 2 4 2 5. and the “combined” model considers both motives.7 6 10 2 4 12 12. the parameter α is a measure of pure altruism. which serves as a replication of the GHL study.9 9 7 2 2 6 6.2 4 5 2 4 6 9. In particular. Earnings ranged from $8 to $15 and the experiment lasted no more than 1 hour.3 12 . with contributions generally increasing with respect to external return and group size.2 5 2 2 4 4 10.1. all-human treatment results Decision task 1 Group size Internal return External return Mean Median 4 4 2 9.1 GHL application. statistical significance of this parameter does indicate decision error is Table 10. The parameter g measures warm-glow.Public goods experiments and policy 201 their group(s). We estimate logit equilibrium models with our data and concentrate on interpretations of estimated parameters and comparisons of parameters across treatments. Our estimates suggest that a participant is willing to give up between 5 cents (“altruism” model) and 15 cents (“combined” model) in order to increase another person’s earnings by $1. The “altruism” model considers the altruism motive but not warm-glow. Consistent with the contributions pattern observed in Table 10. Holt. and Laury application Table 10. the “warm-glow” model considers warm-glow but not altruism. which we find to be insignificant.1 presents mean and median contributions from the all-human treatment.2 3 9 4 2 6 8.3 for all-human and virtual-player treatments. To quantify formally the magnitude of altruism and warm-glow. GHL consider different theoretical specifications for individual utility and estimate utility function parameters using a logit equilibrium model.7 9 6 4 4 4 9.

For both specifications we fail to reject the hypothesis of equal Table 10. we find that estimated pure altruism parameters are statistically different from zero.1.5 6 4 4 4 7.2 GHL application.1 3.5 9 4 2 6 7. We now discuss the outcome from the virtual-player treatment and present two main results about the role of confusion. are generally smaller than in the all-human treatment but not strikingly so.5 . presented as Table 10.1. Result 1: Positive contributions stem largely from confusion and subjects use experimental parameters as cues to guide payoff-maximizing contributions. In particular.7 7.J.4 2 8 2 4 2 4. virtual-player contributions are approximately 75 percent of all-human contributions.7 0 5 2 4 6 7. Cotten et al.). Using the estimated “altruism” and “combined” models from the two treatments we test for equality of altruism parameters between the two (leaving other parameters unconstrained) using Wald Tests.202 S. From Figure 10. leading to behavior that mimics behavior motivated by pure altruism. Overall. one observes that subjects in the virtual treatment alter their contributions based on the same stimuli as subjects in the allhuman treatment. but that only confusion exists in the virtual-player treatment.1 7 4 4 2 2 2. the main conclusions drawn from GHL carry over in our all-human treatment model: pure altruism and confusion are important motives behind contributions whereas warm-glow is not. Specifically.8 10. as illustrated in Figure 10. mean contributions across all decision tasks are 6.8 tokens or 35 percent in the all-human treatment. Assuming that otherregarding preferences and confusion are present in the all-human treatment.7 tokens or 27 percent of endowment in the virtual-player treatment as compared to 8. Turning to the logit equilibrium models estimated from virtual-player treatment data.9 6 3 4 4 6 9. present (Goeree et al. Contributions in the virtual-player treatment.7 5. Put another way. this suggests that an alarming 75 percent of all-human treatment contributions stem from confusion.1 5 2 2 4 4 6. the two response patterns are parallel such that the difference between sets of contributions across decision tasks are approximately equal. virtual-player treatment results Decision task 1 Group size Internal return External return Mean Median 4 4 2 6.2. we find that a participant is willing to give up between 4 cents (“altruism” model) and 16 cents (“combined” model) in order to increase a virtual player’s earnings by $1.2 5 10 2 4 12 10.5 7 2 2 6 4. Perhaps more startling is the observed correspondence between all-human and virtual-player treatment contributions across decision tasks. Estimates of µ are indeed statistically different from zero at the 5 percent level for each specification.

470 (0.310* 32.718 230 Virtual-player treatment Altruism 0. by design.510 –823.054* (0.583 (1. Why would virtual-player participants respond to the same stimuli as all-human treatment participants? Confused subjects are using the changes in the parameters across decision tasks as a cue of how to behave.071 230 230 0.421. participants in the virtualplayer treatment are not exhibiting pure altruism.855). Of course.308 300 α g µ Log-L N –0.054) –668.050) –2.014) – Warm-glow – Combined 0. “combined” model: χ2 = 0.150) –824.311) –813.382* (3. unless one believes pure altruism includes preferences over the utility of fictional automata.148* (0.460) (7.497 –673. estimated logit equilibrium models All-human treatment Altruism Warm-glow – Combined 0.447) (11.1 GHL application.132* (5.383* (0.628) –671.269* (9.987) 21. p = 0.064) –1.033.021) – –1. The altruism parameter is picking up confusion about the role of the external return in the subject’s private payoff . * indicates parameter is statistically different from zero at the 5 percent level.034* (0. p = 0.647. altruism parameters (“altruism” model: χ2 = 0. 14 12 All-human treatment Virtual-player treatment Difference Contributions (tokens) 10 8 6 4 2 0 1 2 3 4 5 6 7 8 9 10 Decision task Figure 10.231 (0.3 GHL application.914* 24.148 300 300 Notes Standard errors in parentheses.Public goods experiments and policy 203 Table 10.163* (0.796) 11.769) 19.059) 28. comparison of all-human and virtual-player contributions.801* (1.

which has been attributed to altruism and “minimum profitable coalitions” (Cox and Sadiraj.4 to 0.7.204 S. Thus. what is the correct answer?).0 tokens.7 to 7. how many tokens should you have invested in each decision? (you may not have cared about making as much money as possible for yourself. but if you did. In a confusing situation. function. the MPCR doubles from 0. which is a nearly identical change of 5. Thus our results are consistent with the “MPCR effect” being related to confusion. subjects have to make ten contributions decisions for which the internal and external rates of return. A “stylized fact” from the experimental public goods literature is that an increase in the MPCR increases contributions. This behavior is similar in spirit to the “herding” behavior found in the Ferraro and Vossler dynamic VCM experiment. In the GHL experiment. mean contributions are 5.2 in Decision Task 4 and 9.J. Cotten et al. we also used a post-experiment questionnaire. In particular. 2005). it is quite likely confusion confounds their comparison. Such pervasive evidence of confusion may cause readers to doubt the validity of the virtual-player method. Subjects with positive altruism parameters are found to be less likely to contribute to the naturally occurring public good.8 from Decision Task 4 to Decision Task 6. and group size. Based on our logit equilibrium model results and observed correspondence between contributions both virtualplayer and all-human treatment contributions to changes in the marginal per capita return (MPCR). In this experiment. are all changing. even after controlling for experimental earnings and subject demographics and attitudes. most people look for cues to direct them towards the optimal behavior.9 in Decision Task 6 – an increase of 4. Recall that Laury and Taylor (forthcoming) run subjects through a GHL experiment and then ask participants to contribute to an urban tree-planting program. the lone design difference is analogous to a change in the MPCR in standard VCM experiments. .4 In the all-human treatment. The internal return is equal to the external return. Decision Task 4 and Decision Task 6 involve a group size of four. but these returns increase from two to four across the two tasks. Result 2: The common observation in public goods experiments that contributions increase with increases in the marginal per capita return likely results from subject confusion rather than altruism or expectations about the minimum profitable coalition.7 tokens – which is consistent with the results on MPCR changes in the literature. It should not be surprising that a confused subject will infer meaning from the changes in these parameters and decide that her behavior ought to change in response to them. We asked all subjects to answer the following question: If all you cared about was making as much money as possible for yourself. In the virtual-player treatment. In addition to the emphases placed in the instructions and the use of the sealed envelope. subjects use past group member contributions (human or nonhuman) as a cue of how to choose their own optimal responses. contributions go from 2.

etc. individual endowment is $12. For those in the all-human treatment we asked respondents to state the contributions level that would have maximized group earnings. it appears that an important issue with the public goods game is that some self-interested individuals are simply not able to deduce the dominant strategy. all-human treatment contributions are $2. Based on the all-human treatment results. We use raw data from this experiment to analyze further behavior according to gender (not reported in the original article). Table 10. virtual-player treatment contributions are also larger for females by $1.” Results from this experiment are presented in Table 10. and Poe We draw from previous experiments to strengthen our arguments about the confusion problem in public goods experiments. this confusion necessarily leads to what looks like other-regarding behavior. it appears as though confusion contributions are quite substantial. contributions from females are $0. such that we estimate confusion accounts for 54 percent of contributions.14 and virtual-player treatment contributions are $1.16.4. most of the purported difference between genders . 103).07. “our subject pool can be considered an ‘extreme’ environment in which to search for altruistic preferences: subjects were ‘economists in training. Rondeau. Using the entire sample.4 presents mean contributions by gender and treatment. All participants correctly stated 25 tokens or full endowment.50 for a correct answer. Thus. Total sample size is 85. public goods experiments are often used to make inferences about the behaviors of subgroups in the population (by gender.’ operating in an environment in which self-interest was being reinforced. Group size is 21. The first experiment is from Ferraro. A total of 13 out of 53 answered this question incorrectly. MPCR is $0. Thus.92 higher than males and this difference is statistically significant using a Mann-Whitney Test (p = 0. and there is a cap on returns from the public good of $7 each. other subjects who erroneously believe they are playing an assurance game will often answer “zero” to this question). However.” Cornell University undergraduates from an introductory economics class. Since decision errors can only be made in one direction (contributions are non-negative).76 each) – rather than full endowment – the dominant strategy is still for the individual to contribute nothing.Public goods experiments and policy 205 Subjects were aware that they would be paid $1. race. Rondeau.07). while the social optimum is for the group to contribute $100 (divided equally this is $4. (p. As discussed in the Introduction. suggesting that 25 percent of respondents were unable to discern the dominant strategy of zero contributions after participating in the experiment (note that this is a lower bound given that some subjects may only realize the correct answer after being asked the question and. Thus.24 (p < 0. culture. This study uses “Ivy League. who use the virtual-player method to study behavior in a single-round VCM-like game. whom all have prior experience in experiments. Thus. even with some of the world’s brightest young individuals as subjects.01). Ferraro.). As stated by Ferraro et al. and Poe (2003). as noted in Ferraro and Vossler.

93 for males versus $0.4 Ferraro et al. mean contributions All All-human treatment Virtual-player treatment Difference % confusion contributions 2.84 0.08 0.98 54% Males only 1. Cotten et al. What appears like a gender-effect is likely a gender-based difference in confusion for this specific sample. The sample consists of 160 subjects: 80 in an all-human treatment and 80 in a virtual-player treatment.5 Figure 10. Table 10. . who apply the virtual-player method to the dynamic VCM game. Subjects are undergraduate students from Georgia State University.2 presents mean contributions (measured as a percentage of endowment) by round for the all-human and virtual-player treatments. mean contributions. With 60 50 Percentage of endowment 40 30 20 10 0 1 3 5 7 9 11 13 Round 15 17 19 21 23 25 All-human treatment Virtual-player treatment Figure 10. Ferraro and Vossler The other prior experiment we draw upon is from Ferraro and Vossler (2005). (2003) VCM experiment. The first observation is that confusion contributions are considerable. They use an archetype multiple-round VCM game with group size of four.J.16 0. and feedback on group contributions after each round.61 for females).2 Ferraro and Vossler (2005) experiment.5.69 2.14 1. an MPCR of 0.206 S.77 0.93 47% Females only 2.61 78% disappears when confusion contributions are removed ($0.

the standard decay in VCM experiments over rounds is not due to learning the dominant strategy or a reduction in confusion. distort inferences about the role of other-regarding preferences. Clark (2002) found that having subjects play the VCM game with their own money had no discernible effect on their behavior. Further. this suggests 52 percent of total contributions in the standard VCM game stem from confusion. these results call into question the internal and external validity of this line of experimentation. Ferraro and Vossler carefully analyze the data using a dynamic pooled timeseries model and find that the reduction in contributions in the virtual-player treatment is largely driven by the decline in observed contributions from virtual players in previous rounds. and confound comparisons between subpopulations. given that decision errors are likely rather commonplace in many economics experiments. This finding in itself may not seem alarming. these confusion contributions do not simply amount to harmless statistical noise. dominant strategy is. Experiments discussed here involve payoffs that are on average much higher than student wages for this time commitment. Second. (2) poorly prepared instructions. Furthermore. confused participants in the virtual-player treatment simply use any available cue to help determine contributions. Overall. confusion just does not simply go away over the course of many repeated rounds. and additional evidence from the post-experiment questionnaire and focus groups suggests that this proportion is a lower bound. The results of Clark are . Do our results suggest we should just stop drawing inferences from public goods experiments? Certainly not. and (3) the inability of participants to decipher the dominant strategy. Instead. Thus.5 percent and 16. They find that 30 percent of respondents are unable to determine the dominant strategy of zero contributions. similar to the correlation between MPCR and confusion contributions in our GHL application. Ferraro and Vossler report responses to a question similar to the one in our GHL experiment concerning what the purely self-interested. but they do suggest we need to rethink how these experiments are implemented. average contributions still amount to 10 percent of endowment in round 25. Unfortunately. in an investigation of “house money” effects. we find that at least half of contributions in public goods games stem from confusion. Are inadequate monetary rewards a problem? We think not. we have shown that confusion contributions are sensitive to changes in design parameters. respectively. As additional validation of this result. recall that Andreoni (1995) cites three potential causes of confusion contributions: (1) inadequate monetary rewards. In particular.8 percent of endowment in the all-human and virtual-player treatments. As a starting point for discussion. while the virtual-player treatment contributions do decrease over rounds. Discussion Through the course of three different applications of the virtual-player method.Public goods experiments and policy 207 participants contributing 32.

1999. p. As a pilot study.J. Our results thus call into question the standard. one of the authors used such context-enhanced instructions in a standard. we could explain to participants that we are asking them for voluntary contributions for a public good and that the public good is simply an amount of money that gets distributed throughout the group. For instance. Thus. Subjects can be informed that it is perfectly reasonable to give nothing. consistent with the presence of a substantial number of individuals who are not clear about the appropriate strategy conditional on their preferences. Responses from post-experiment questionnaires we used. ten-round VCM experiment run in two sections of an undergraduate environmental economics course at the University of Tennessee in September 2005.” Our experimental evidence suggests that a bit of context could go a long way. that when faced with a naturally occurring contributions decision. as well as behavior. while our instructions – and instructions for public goods experiments in general – are not necessarily poorly prepared.6 We share the sentiment of Loewenstein (Loewenstein. however. effort is made to avoid context. Indeed. The experiment was being used to illustrate the free riding phenomenon (before the concept was formally introduced). since many subjects cannot figure out the dominant strategy (but all our GHL experiment participants figured out the social optimum) perhaps we can clue them in without altering their preferences for the public good. Cotten et al. but . even if they function quite adequately in familiar settings. the vast majority is quite capable at performing the necessary payoff calculations. Ferraro and Vossler report in a postexperiment focus group that just one-quarter of participants were able to figure out the dominant strategy by reading the instructions. F30). After students read the instructions.208 S. suggest that at least 30 percent of respondents simply are not able to figure out the dominant strategy of zero contributions. From our experience. The decision settings are presented using neutral language. who suggests “Subjects may seem like zero intelligence agents when they are placed in the unfamiliar and abstract context of an experiment. These instructions are available from the authors on request. the instructions are standard in experimental economics. the inability of individuals to decipher the dominant strategy does suggest the need for modifying how the game is explained. Standard instructions for this type of experiment use neutral language and do not reveal that the experiment is about public goods or that participants are being asked to make a contributions-like decision. “context-free” instructions used in public goods games. the focus group of Ferraro and Vossler reveals that many participants thought they were playing some sort of assurance game. and subjects go through simple exercises to assess their understanding of payoff computations. In particular. This has important consequences for the external validity of the experiment unless one can show confusion has similar effects and magnitudes in “real world” contributions. people recognize the tension between privately beneficial free riding and socially beneficial contributions. Are instructions “poorly prepared”? For the experiments discussed that use the virtual-player method. We believe.

and about the effects of alternative institutional arrangements that induce private contributions to the public goods. as opposed to after they make their decision. they cannot achieve their full potential as long as they are implemented in a way that leaves many subjects oblivious to the social dilemma that experimentalists are trying to induce. herders then further reduce. Consistent with our conjecture.Public goods experiments and policy 209 before contribution decisions were made. This rate of decay is quite low for a VCM experiment. but results are consistent with expectations based on our virtual-player treatment results. and (2) the automata contribute the average of what human condition members contributed. we believe that public good experiments will continue to play an important role in testing economic theory and designing public policies. and so on. Note that the standard VCM instructions provide information only on the payoffs associated with each level of group contributions. However. the findings of Oxoby and Spraggon (this volume) suggests that confusion also may be reduced by providing a payoff table showing the subjects’ payoffs given their decisions and the decisions of others. In conclusion. namely that it could systematically alter participant preferences for the welfare of others.4 tokens). and if participants behave differently when the contributions of other players are known ex ante. While there are likely tradeoffs associated with adding even generic context. context-free experiments: 30 percent from the post-experiment questionnaire in our GHL experiment. will continue to be impaired. the students were asked to write down the dominant strategy. Notes 1 The two potential design flaws are: (1) human subjects in the computer condition observe their group members aggregate contribution before they make their decision in a round. If the history of contributions affects both confused and other-regarding subjects. herding individuals are going to follow the group trend and so any reduction in contributions in early rounds really causes a downward spiral: conditional cooperators get an exacerbated signal of free riding and revoke contributions. through survey questions with monetary rewards for correct answers. it appears that investigation into instruction based modifications is warranted. Confused. The value of instruction enhancements could be tested using the virtual-player method. This figure is considerably below those from comparable. and the estimate from Vossler and Ferraro that three-quarters could not deduce the dominant strategy prior to the experiment. Without innovation in the design of these experiments. through debriefing sessions. Without the herders. our ability to draw inferences about behavior in collective action situations. the decay in average contributions over time should be relatively less steep. and through external validity tests. then such changes in design affects the comparability of the two . The pattern of contributions is quite similar in both class sections: contributions start at about 50 percent and fall to 40 percent by round 10. as in the human condition. Only three of 25 students (12 percent) failed to identify the dominant strategy of zero contributions (mean response was 0.

34 (4). Experimental Economics. An Assurance Game (also known as the Stag Hunt) is a game in which there are two pure strategy equilibria and both players prefer one equilibrium (payoff dominant) to the other. The less desirable equilibrium. Economic Journal. D.. 2005. Impure Altruism and Donations to Public Goods: A Theory of WarmGlow Giving. 1993. Due to the nature of the game.210 S. has a lower payoff variance over the other player’s strategies and thus is less risky (it is risk dominant). Cox. Indeed. Davis and Holt (1993. 1993. Paper presented at the Conference on Public Experimental Economics. 2005. Smith. 1995. This individual was subsequently dropped from the data set. In Handbook of Experimental Results. Brown-Kruse. NJ: Princeton University Press. P. Gender and Free Riding in a Threshold Public Goods Game: Experimental Evidence. 1998. 1990. Social Preferences and Voluntary Contributions to Public Goods.D. Plott and V. Cooperation in Public-goods Experiments: Kindness or Confusion? American Economic Review. some subjects erroneously view contributing their entire endowment as the most desirable strategy when everyone else in the group contributes their endowments too.” We only report their “VI” and “HI” treatments. J. however. 891–904.. Davis. 1988. and V. 100 (401). 291–304. 223–237. Journal of Public Economics. J. Vossler. 2 3 4 5 6 conditions. Experimental Economics. References Andreoni.J. and P. Hummels. and D. 464–477. C. Andreoni. this inclusion of this person should have no impact on the contribution level of the undergraduate participants.A. Journal of Economic Behavior and Organization. A similar use of virtual players was employed by Johnson et al. Andreoni. 332) define a “minimal profitable coalition as “the smallest collection of participants for whom the return from contributions to the [public account] exceed the return from investing in the private [account]. Volume 1. 2002. p. Subjects described this decision as “risky” because it leads to low payoffs if other players do not contribute their endowments.A. Grossman. 37 (3).. The Dynamics of Other-regarding Behavior and Confusion: What’s Really Going on in Voluntary Contributions Mechanism . and C. a graduate student was asked to participate as a last-minute measure to make the total number of participants divisible by four. Ferraro. Contributing zero was viewed as a payoff inferior choice but “less risky. Clark. Georgia State University. 85 (4). edited by C. Sadiraj. (2002) in a sequential bargaining game.. Cotten et al. Eckel. New York: Elsevier. Differences in the Economic Decisions of Men and Women: Experimental Evidence. Cadsby.” These subjects were unable to infer the dominant strategy in the VCM game. C. J. J. and C. Princeton. Holt. Maynes. 5 (3). 603–620.J. J. the intent of our virtual-player method is simply to have participants play with virtual players and not change any other aspect of the game. and E.B. 255–268. J. 2005. Gender Effects in Laboratory Public Goods Contributions: Do Individuals Put their Money Where their Mouth Is? Journal of Economic Behavior and Organization. In one session. 22 (3). Why Free Ride? Strategies and Learning in Public Goods Experiments. House Money Effects in Public Good Experiments. In the case of the VCM game.

Rondeau. Journal of Economic Behavior and Organization. Journal of Public Economics. Ferraro. Spraggon.. Kroll and J. 2002. E. Are People Conditionally Cooperative? Evidence from a Public Goods Experiment. 1984. U.. Tinkler. D. 1996. Does Anyone Else? Journal of Public Economics.. C.J. Economic Journal.P. and L. Frey. edited by T. Experimental Economics from the Vantage Point of Behavioural Economics. Ames. Goldfarb. 1997. Palfrey. 2001. Sen. and E. Camerer. 177–186. and R. 94 (5). 1717–1722. 83 (2). Marwell. P. 1999.O. Forthcoming. Oxoby. 1981. S. Journal of Economic Theory.L.L. Johnson. Social Comparisons and Pro-Social Behavior: Testing “Conditional Cooperation” in a Field Experiment. 92 (4). and J. The Influence of Gender on the Provision of a Public Good. Shogren (in this volume).N. 295–310. University of Zurich. and R. In Experimental Methods in Environmental Economics. 2007. S. Public Choice. 16–47. Poe.Public goods experiments and policy 211 Experiments? Experimental Laboratory Working Paper Series #2005–001. 109 (453). Houser. D. J. R. Laband.J. American Economic Review. and S.J... 113–149. Economic Letters. C. Rymon. Detecting Other-regarding Behavior with Virtual Players.. Cherry. and P.M. Revisiting Kindness and Confusion in Public Goods Experiments. 2004. Taylor. 397–404. S. Beil. and R. Detecting Failures of Backward Induction: Monitoring Information Search in Sequential Bargaining. 71 (3). C.S. 104 (1). Loewenstein. G. T. D. 2003. Does Studying Economics Discourage Cooperation? Watch What We Do. R. Are Economists More Selfish than Other “Social” Scientists? Public Choice. 15 (3). Department of Economics. S.F. Bruno and Stephan Meier. 1999. The Effects of Recommended Play on Compliance with Ambient Pollution Instruments. Goeree. Fehr. Private Costs and Public Benefits: Unraveling the Effects of Altruism and Noisy Behavior. . F25–F34. Gächter. Altruism Spillovers: Does Laboratory Behavior Predict Altruism in the Field? Journal of Economic Behavior and Organization. Economists Free Ride. Gächter. Andrew Young School of Policy Studies. 25–36. Fischbacher. 2004. U. Laury. Prisbrey. 1062–1069. Laury.. and J. Institute for Empirical Research in Economics.E. Not What We Say or How We Play. Nowell. 43 (2). and S. 99–109. 100 (1–2). and S. 829–846. Divergent Evidence on Free Riding: An Experimental Examination of Possible Explanations. J. A. Heterogeneous Motivations and the Dynamics of Free Riding in Public Goods. 87 (5). Anomalous Behavior in Public Goods Experiments: How Much and Why? American Economic Review. Holt. R.E. Kurzban. 1994. and T. 25 (1). Working Paper. 85–101. and G. 2002. G. Thomas. Yezer.J. Journal of Economic Perspectives. Isaac. 2002. Walker.M. Journal of Economic Behavior and Organization. and S. 51. Georgia State University. American Economic Review. 10 (1). 255–276. Fischbacher. Poppen.

and preferences for fairness (Fehr and Schmidt. Lueck. Fuelwood lots and grazing lands suffer from severe depletion around the world. groups. 1992a). the specific effects of costly monitoring of individuals and the role of endogenous sanctions are examined. at various times. some. Ostrom. 1992). though not all. To understand how these issues became a problem. Hardin. 1987. Sethi and Somanathan. 1994). Seabright. To understand that these issues are a problem. 1993). CPRs are successfully managed through a system of communal ownership and management. 1998: B1). Common-pool resources (CPRs).. however.1 This begs the question. cultural norms (Platteau. 1999. Abreu. Dutta and Sundaram. 1994a&b. In this chapter. Huberman and Glance. joint supply dependency (Hechter. like those above have a rich theoretical background (Gordon. 1993). theories of repeated games (Kreps et al. Institutional economics posits that rules and institutions can align individual self-interest with a more collective group goal. Fehr and Gächter. 1994. Each resource is characterized by costly exclusion and subtractable resource use (Ostrom. Ostrom.2 These sanctions are . 1998). Gardner. the interaction between individuals. and Walker (1994 – hereafter OGW) provide data from an economic experiment showing that the ability to impose welfare-reducing sanctions upon others reduces over-appropriation from a CPR. 1993. Excessive numbers of whale watchers have led to a reduction in the number of whales returning to – and consequently declining biodiversity of – the Bay of Fundy (Hawkins. Caputo and Lueck. 1968. 2000a).11 Spies and swords Behavior in environments with costly monitoring and sanctioning Rob Moir Introduction Depletion of resources is a serious global concern. Despite Hardin’s dire “Tragedy of the Commons” prediction of complete resource depletion. why do only some communally managed CPRs generate the cooperation necessary to succeed? Economic cooperation in games of collective action has been explained as an outcome of evolution (Bergstrom. 1996. 1988. one need only look at the socio-economic impacts upon the relevant population of users. 2003. and the resource must be examined. Bromley. 1954. 1982. Groundwater in the American Midwest is used faster than it is replenished. been closed and thousands of people put out of work. Fisheries on both Canadian coasts have.

and opt to do so. Following this.. Blaikie et al. Treatment variables include varying sanction costs. 1992. however. The results are then presented.Spies and swords 213 welfare reducing as they are costly to both the individual imposing the sanction and the individual sanctioned. With repetition. Both sanctions and communication significantly reduce CPR appropriation. and resource use is almost optimal. because . The remainder of this chapter is structured in the following manner. the results of relevant economic experiments are summarized. 1992). and allowing subjects to vote on the adoption of a sanctioning mechanism. varying sanction sizes. and India (Wade. cheating is minimal.4 In the seminal CPR experiment. contrary to the game theoretic prediction that neither treatment should cause a change. First. 1992b).. Sessions consist of a baseline treatment followed by a sanctioning treatment permitting a within-subject analysis. 1992). varying the right to communicate. While the choice to sanction is endogenous. sanction levels are fixed for a session. Third. subjects typically appropriate more than the Nash equilibrium prediction (Walker et al. 2000b).9 Second. 1992). OGW did not specifically include monitoring in their design. the experiment design. with the discussion and conclusions forming the final part of the chapter. is a key design variable in organizing a communally managed irrigation system (Ostrom. 7 OGW (1994) systematically address institutional issues of self-governance of a CPR in an experiment. and the monies are removed from the economy. 1992). Next.5 Monitoring has led to difficulties in formulating fishing policies in Chile (Peña-Torres.. Brazil (Cordell and McKean.8 The OGW design raises several important methodological issues. Monitoring. 1997). In these cases. appropriation tends towards the Nash equilibrium but falls far short of socially optimal appropriation.6. treatment conditions. and subjects are permitted to sanction only one individual per period. and is crucial in determining the degree of successful management of CPRs in Maine (Acheson. Previous experiments In baseline CPR experiments.. and how do such rules affect CPR appropriation? I present the results from an experiment designed specifically to examine the separate and joint roles of costly monitoring of individuals and costly sanctioning in a CPR environment. the size of the sanction and the number of sanctions to impose is exogenous. that is experiments with no external rules or structures. and predictions are outlined. Fehr and Gächter. few sanctions are imposed.g. Niger (Thomson et al. sanctions are not explicitly “rules based” in that subjects could sanction each other even if no “rule” is broken. I focus quite narrowly upon the institutional topic of selfgovernance of CPRs – can rules guide a group of individuals to control their joint use of a CPR. Particular attention is paid to related CPR experiments by OGW and Casari and Plott (2003) and to sanctioning experiments from the public goods literature (e. 1990). Japan (McKean. The effect of sanctioning is largest when groups are permitted to choose whether or not to adopt a sanction rule. no matter the number of offenders. 1988).3 In this chapter.

A number of public good experiments focus solely on monitoring and typically find that monitoring improves voluntary contributions and efficiency (Palfrey and Rosenthal. subjects receive information on aggregate CPR appropriation. there are important design differences (see Casari and Plott: Table 1. monitoring and fining are modeled as a single activity. while the other two remain isolated from the group and each other. strong fines are more effective. fines are fixed to be proportional to the deviation from an exogenously determined appropriation level. Uncertainty in monitoring reduces the significant efficiency-enhancing effects of face-to-face communication exhibited in OGW. as described in Acheson (1988). At the end of each period. While similar to both OGW and the design used in the present experiment. Schmitt et al. 1994. Most importantly. p. Casari and Plott (2003) implement an OGW design to study the ‘Carte di Regola’ mechanism used to regulate CPR appropriation in the Italian Alps from the thirteenth to the nineteenth century. (2000) examine uncertain monitoring in a CPR environment. subjects are aware of group appropriation before choosing to monitor. and the proceeds from the fines are returned to the individual doing the monitoring. 2004). This effect may be due to social sanctioning – the effect of having individual contributions and identity revealed to the group (Gächter and Fehr. monitoring is costly either in terms of effort – a person has to spend time watching the activities of others – or an independent monitor must be hired and paid for. it is impossible to separate social sanctioning from the direct effects of pecuniary sanctions. Six of them engage in face-to-face communication before making contribution decisions. Likewise. While both rewards and sanctions increase contributions. 1999). Holcomb and Nelson (1991 and 1997) examine the role of perfect versus imperfect monitoring in a duopoly environment and show that only after significant evidence of overproduction in the imperfect monitoring case do cartel arrangements break down.214 R. In all of these experiments monitoring remains costless. However. Andreoni and Petrie. 1995). Fourth. Finally. sanctions . 220 for a summary). monitoring is costless in public good experiments with rewards and sanctions. the role that monitoring plays in the OGW experiment is not explored. Moir communication takes place face-to-face. and implementing a costly monitoring/fining rule. It is difficult for the six communicating members of a group to determine if departures from any agreements made in the communication phase are due to cheating from within the communicating subgroup or because of outsiders’ actions. as subjects freely receive information on individual investments in CPR appropriation at the end of each period but before sanctioning takes place. 1999. While both versions of fines increase efficiency. subjects in the OGW experiment are presented with “rough” payoff tables that are known to change behavior in public goods experiments (Saijo and Nakamura. Casari and Plott implement weak and strong versions of fines and compare these treatment results to a baseline condition. Eight people are formed into a group. Cason and Khan.

Spies and swords 215 are more effective at increasing and sustaining contributions to the public good. Subject i is constrained by his endowment.5si − si where G = ∑gi is the aggregate investment in CPR appropriation. Sefton et al. With homogeneous individuals. wi = xi + gi + mi .12 Face-to-face communication. Further define si as the sum of all sanctions levied against subject i in a period. A fixed number of subjects (n = 8) are endowed with a fixed number of tokens (w = 20) each period. Subjects simultaneously select how to invest their tokens in a private activity (xi . Ones and Putterman. 1986. both repeated and one-shot. the period payoff in lab-dollars to subject i is πi = 3. even if it is ambiguous (e. Unlike OGW. It cost one token to obtain perfect information about the Market 2 investment of another individual in the current period. Communication permits coordination (though not necessarily at the optimal investment level)..5S (11. and an opportunity for social sanctioning. 2004). a CPR appropriation activity (gi . in collective action experiments greatly enhances efficiency. referred to as Market 2).g.53giG − 0. the creation of “rules”. Name calling. 2005. 2000b.68X + 80G − 0. 2007). Walker and Halloran. the size of a particular sanction is endogenously chosen by the subject. referred to as Market 1). Let sij be subject i’s sanction of subject j and si be the sum of i’s sanctioning activities in a given period so that i’s sanctioning costs are equal to 0. ‘some greedy idiot is ruining it for us all’) can lead to changes in behavior. (1985) provide subjects with the individual action necessary to achieve a socially optimal outcome. as is the number of sanctions to levy in a particular period.13 .1) where capitalized terms are aggregates and the social budget constraint is W = X + G + M.68xi + 80gi − 0. and in monitoring other individuals (mi).10 Let mij represent subject i’s monitoring of subject j and mi be the sum of i’s monitoring activities in a period. Holcomb and Nelson (1991 and 1997) have subjects anonymously pass notes to their partner while Isaac et al. Grouping subjects according to behavioral type influences net efficiency when sanctions are available (Yamagishi. The experiment Design The current experiment is based upon the OGW (1994) framework.5si.53G2 − 1. the aggregate payoff function is Π = 3. With these terms in mind. but sanctioning costs reduce efficiencies to baseline levels (Fehr and Gächter.11 Sanctions are expressed in laboratory dollars and subjects are required to pay one-half the value of the sanction in order to levy it against another individual.

X. your payoff increases. M-X. In fact. Subjects know that for each individual they choose to monitor they have one less token to use in making their own investment decision. X. the aggregate investment in Market 2 and her period payoff. M-X. then your payoff will increase if you allocate more tokens to Market 2. a subject knows her own investment in Market 2. At the end of each period. M-X. X. A subject can levy any size sanction against any number of offenders as long as her sanctioning costs do not exceed her current period payoff. her period payoff. I adopt a variant of the Isaac et al. at the known cost of one-half the value of the sanction. M-X. M-X. X. I. X. X. At the end of each period. M-X. After period five the following passage is read: You may have noticed in part 1. X. X. X. and the Market 2 investment decisions of any individuals she chooses to monitor. Using OGW terminology. X. X. Similar to the OGW stage-game terminology. subjects are required to select how much to invest in Market 2 (CPR appropriation) and the remaining endowment is invested in Market 1. After the monitoring stage. Subjects are informed that part 1 of the session lasts five periods and at that point new instructions describing part 2 are to be read aloud. a subject knows her own investment in Market 2. if each member of the group restricts their allocation to Market 2 to 9 tokens. the subject is permitted to sanction. the aggregate investment in Market 2. In the sanctioning stage. M-X. and the investment decisions of any individuals she chooses to monitor. X. The entire game is expressed by. the aggregate investment in Market 2. M-X.216 R. if everyone else is allocating 9 tokens to Market 2. then the group payoff (yours plus everyone else’s) will be maximized. this information phase is denoted as “I” when describing the entire game. I call this game “X”. Treatment conditions In the baseline treatment. procedure. X. Each subject selects the subject number of the individuals they wish to monitor. her period payoff (before sanctions). M-X. Subjects cannot earn negative payoffs in a period unless they are . X. X. M-X. X. In the monitoring treatment. The sanctioning treatment follows the same procedure as the monitoring decision expressed above. X. Prior to the sanctioning phase (identified as “S”). X. Subjects are then informed that the remainder of the experiment will last ten periods. subjects complete a monitoring phase (“M”) before making their investment decisions. any individuals they monitor and catch investing more than nine tokens in Market 2. X. a subject knows her own investment in Market 2. However. subjects choose their investments in Markets 1 and 2. Moir In order to mitigate social sanctions while still providing a focal point for the sanctioning rule. The entire game is expressed by. that if the other members of the group restrict their allocation of tokens to Market 2. I. All sessions are divided into two parts. X.

02 lab-dollars. X. Thus the SSPE in each of the three treatments is for each subject to invest 16 tokens in CPR appropriation. X. define the change in CPR appropriation as ∆Gkvt = Gkvt − avg(Gkv(1−5)). When the resource stock is time independent and constant then the unique symmetric equilibrium in the one-shot game (X) is a symmetric subgame perfect equilibrium (SSPE) in the repeated game (Selten. a subject knows her own investment in Market 2. Sanctions become public knowledge at the end of each period. M-X-S. and t > 5 is the period number. M-X-S. Extending the game to include costly sanctioning does not alter the SSPE prediction and zero sanctions are predicted (OGW. her period payoff. 1996). The final term. M-X-S. Sethi and Somanathan. or S) depending on whether it is a baseline. involves each subject investing 16 tokens in Market 2 and earning 209. First. each subject invests nine tokens in Market 2 and earns a period payoff of 417. then monitoring will not take place. is the average aggregate investment in the five periods prior to the communication .15 If seven subjects invest nine tokens each in Market 2. monitoring. At the symmetric social optimum (obtained by maximizing equation 11. Three summary variables are calculated and form the basis of comparison. M-X-S. X.1). I. M-X-S. Subjects are aware of the numerical identity of all sanctioned individuals and the amount each is sanctioned. X. Predictions In this chapter. I focus upon aggregate results – analysis of individual data is left for another paper. the investment decisions of any individuals she chooses to monitor. M-X-S. the aggregate investment in Market 2. M-X-S. M. M-X-S. X. the remaining subject could maximize his period profit by investing 20 tokens for a payoff of 720.04 labdollars. M-X-S. It is a simple corollary that if monitoring is costly and necessary for sanctioning.2) where G is the aggregate investment in CPR appropriation. The entire game is expressed by. At the end of each sanctioning stage. thus making cooperation difficult to sustain. (11. any sanctions she imposes.28 lab-dollars per period.Spies and swords 217 sanctioned by others. Resource appropriation at the SSPE exceeds the socially optimal level. v is the session number. or sanctioning session. M-X-S. 1994.14 Model solutions The unique symmetric Nash equilibrium for the one-shot game X. avg(Gkv(1−5)). This result is invariant to the inclusion of the information phase that takes place in period five. 1973). These sanctioning rules are enforced by computer software. and the total sanctions levied against every member in the group. k = (B.

Moreover. This is the primary variable for analysis. monitoring becomes more valuable in the sanctioning sessions.68 lab-dollars per token invested). .17 However. These predictions are summarized in Table 11. Because the information phase occurred in all three treatments. It “corrects” for any inherent cooperation within a group and concentrates upon the change in CPR appropriation because of the change in treatment. The right to sanction has monetary value. Given the SSPE prediction expressed earlier. Simply knowing that others can watch you may not affect your behavior.68M] − Π*)/(ΠO − Π*).1. and the sanctions. columns 2 and 3 respectively. (11. the SSPE prediction argues that neither monitoring nor sanctioning will take place. GrsE = ([Π + (1. the within-treatment prediction is that the passage read before period six will not change CPR appropriation behavior as compared to the first five periods. Prediction 2: CPR appropriations are not affected by the addition of monitoring or monitoring in conjunction with sanctioning.5)S + 3. The alternative hypotheses tested in these cases propose that the treatment reduces CPR appropriation and increases cooperation.1.16 Gross efficiency adds back monitoring costs (3. knowing that you may be punished if you are caught over-appropriating might cause a behavior change. sanctioning costs. Because monitoring is required in order to sanction. (11. the addition of sanctioning to the monitoring game does not affect CPR appropriation. Prediction 1: The information phase introduced before period six does not significantly alter CPR appropriation from the average behavior in the first five periods. The net efficiency gain over Nash in each period is calculated as NetE = (Π − Π*)/(ΠO − Π*). However. it is necessary to compare across treatments to test for treatment effects.3) where Π is the aggregate profit at the end of a period and the superscripts * and O refer to the aggregate profit realized at the Nash equilibrium and socially optimal levels of appropriation respectively. Summary efficiency results are presented. This prediction is summarized in Table 11.2. Moir phase. If sanctions serve to reduce CPR appropriation then payoffs increase. The alternative hypothesis tested is that the treatment reduces CPR appropriation.4) ∆NetEkvt and ∆GrsEkvt are defined as in equation 11.218 R. column 1.

. In addition to a fully detailed payoff table that allowed subjects to determine the payoff associated with their own Market 2 investment and the combined investment decisions of all others. The alternative hypothesis tested is that level of monitoring increases in the sanctioning sessions.1 Group contribution predictions Within treatment H0: ∆GBvt = 0 HA: ∆GBvt < 0 H0: ∆GMvt = 0 HA: ∆GMvt < 0 H0: ∆GSvt = 0 HA: ∆GSvt < 0 Baseline Sanctioning vs. monitoring H0: ∆GMvt = ∆GBvt HA: ∆GMvt < ∆GBvt H0: ∆GSvt = ∆GBvt HA: ∆GSvt < ∆GBvt H0: ∆GSvt = ∆GMvt HA: ∆GSvt < ∆GMvt Prediction 3: Total levels of monitoring in the monitoring and sanctioning sessions are no different from total monitoring levels in the monitoring-only sessions. At the end of the experiment subjects were paid privately according to the pre-announced exchange rate of 157 lab-dollars to $1 Canadian.Spies and swords 219 Table 11.5 hours of work were $24. subjects would earn $40. Subjects were also informed that their earnings were cumulated across both parts of the session. Subjects could enter hypothetical values for both their own investment and the investments of others and the resulting payoff was displayed. whereas if the social optimum was reached each period. Instructions for part one of the experiment were read aloud and questions were answered.18 Subjects were informed that there were five periods in part one of the session after which a new set of instructions would be distributed. If the Nash equilibrium was played each period. subjects were read the part two instructions and informed that after ten more periods the experiment would end. Eight subjects at a time were seated at sheltered computer terminals to ensure privacy and were aware that all members in the room were identical in terms of payoffs. subjects were provided with a scenario calculator on their computer screen. as subjects were identified only by a number that was private knowledge. However.25 (approximately double the minimum wage at the time of the experiment). anonymity of all actions was ensured.19 After period five. Procedure In total 120 students from McMaster University participated in 15 separate sessions – five groups of eight in each of three treatments. Average subject earnings for 1. subjects would earn $20.

1 Aggregate CPR appropriations by treatment.2 and Figures 11. The sanctioning treatment always shows a higher gross efficiency gain when compared to the baseline after period six. even in the last period. Instead. The SSPE prediction is that neither monitoring nor sanctioning will occur. The large difference between the gross efficiency and net efficiency measures in the sanctioning sessions (see Table 11. The average level of monitoring declines more quickly in the monitoring-only treatment when compared to the sanctioning treatment.220 R.2 columns headed by M).3) indicates that sanctions were used extensively.1 through 11. Moir Results Summary results are presented in Table 11. once monitoring and sanctioning costs are accounted for. the baseline and monitoring sessions quickly return to the Nash equilibrium. the net efficiency gain over Nash in the sanctioning sessions are not too different from those in the baseline treatment. it is evident in both the monitoring and sanctioning sessions that monitoring took place (see Table 11. There is a significant reduction in the appropriation levels in the sanctioning sessions. the SSPE prediction Baseline 160 Aggregate appropriation 128 2 4 5 3 3 4 3 3 3 2 2 4 3 4 2 2 1 1 5 5 4 5 5 4 2 1 5 1 2 3 4 5 1 1 1 5 4 3 1 2 3 4 1 4 3 5 3 2 3 5 4 2 5 5 1 3 4 5 3 5 4 4 1 5 2 4 2 1 3 1 1 2 1 2 2 Monitoring 2 5 5 2 3 4 2 5 1 5 4 4 3 4 4 3 1 5 4 4 3 5 3 1 4 3 1 3 1 2 5 4 3 2 2 4 1 3 1 5 1 1 1 1 5 1 5 5 5 3 4 2 3 1 4 2 2 4 3 2 4 3 3 2 2 5 2 1 2 4 3 5 1 2 72 1 Monitoring and sanctioning 160 128 5 3 3 2 4 1 2 4 1 5 2 1 5 3 3 4 5 4 2 1 2 3 5 5 3 5 3 2 4 3 2 5 2 1 4 3 1 4 1 4 5 1 5 3 1 4 4 1 2 2 4 5 5 3 4 2 1 1 3 2 3 5 4 4 5 5 2 3 3 2 4 1 1 1 2 3 5 7 9 11 13 15 Period Group ID Median spline 72 1 3 5 7 9 11 13 15 Figure 11. However. In some cases. despite the fact that CPR appropriation levels were approximately Nash in the monitoring session. the monitoring treatment exhibits the lowest gross and net efficiency gains over Nash. . it is evident that CPR appropriation starts between the Nash equilibrium and social optimum and trends towards the Nash prediction over the first five periods. Thus.2 and 11.3. From period six onward.1. Despite an initial “information” effect in period six. In Table 11. This is a reflection of the decrease in CPR appropriation in the sanctioning sessions. more than ten tokens on average were devoted to monitoring in a single period.2 and Figures 11.2 and in Figure 11.

22 0.0 121.8 10. 5 Monitoring was not available (n/a) in the first five periods of the monitoring and the sanctioning sessions.13 0.4 4.0 n/a 117.8 106.2 125.04 0.35 GrsE 3.2 n/a 120.4 Notes 1 These values are the average values for the five sessions within each treatment.6 10.18 0.4 127.0 125.5 1 1 1 3 1 2 5 2 2 1 2 1 1 4 4 2 2 2 1 5 4 5 1 2 3 5 3 3 1 5 5 3 1 1 4 2 2 3 3 2 1 5 4 4 4 5 3 5 2 1 3 2 4 4 5 5 4 4 3 3 2 3 4 3 1 5 3 5 5 4 4 2 3 4 1 5 Monitoring 2 2 5 1 3 4 3 3 2 1 2 5 3 2 2 4 2 2 3 1 2 1 4 2 1 3 4 3 4 4 5 5 3 2 5 5 1 1 1 1 1 1 5 3 1 5 1 4 5 3 2 4 5 4 3 4 3 1 4 3 4 4 3 4 5 2 4 1 5 5 2 53 5 2 Monitoring and sanctioning 1 0.4 5.6 3.61 0.2 116.0 5.3 and 11.4) account for gains as a fraction of that possible over-and-above the Nash equilibrium.65 0.2 Summary data (averages across treatments)1 Period Baseline G2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 105.2 8.4 114.36 0.21 0.34 0.65 0.0 4.42 0.8 n/a 122.4 120.8 10.04 –0.24 105.60 0. 2 Recall that the Nash prediction involves G = 128 and the social optimum involves G = 72.25 0.6 119.04 –0.4 n/a 131.07 –0.01 0.43 0.60 0.4 G 0.10 0.65 0.24 0.23 0.13 0.04 0.12 0.8 111.25 0.4 M n/a n/a n/a n/a n/a 9.Spies and swords 221 Table 11.43 0.47 0.11 –0.12 0.0 103.4 5.0 108.17 –0. 3 The efficiency calculations (see equations 11.0 128.16 0.02 –0.01 0.2 0.04 0.2 11.27 0.13 0.6 119.11 0.27 0.06 Sanctioning G 116.06 GrsE 0.17 –0.27 GrsE 0.27 0.49 0.2 111.6 119.8 127.40 0.17 0.5 0 0.65 0.0 131.13 0.29 0.19 0.2 110.4 104.6 125.4 10. 4 NetE and GrsE are identical in the baseline treatment as monitoring and sanctioning are unavailable.39 0.06 –0.25 0.5 0 0.00 0.0 7.4 124.14 0.0 n/a 118. .6 7.62 0.0 120.6 122.33 0.8 118.8 116.2 Monitoring M5 NetE 0.02 0.6 1.8 104.22 0.26 0.2 122.33 0.5 1 1 3 1 5 1 2 4 4 2 5 3 1 2 3 4 3 5 1 2 2 4 4 3 5 1 4 5 3 3 1 1 2 3 5 4 5 2 1 5 4 3 1 5 2 2 4 3 2 2 3 1 4 1 4 4 1 1 1 2 1 2 3 3 2 5 5 2 5 3 5 4 5 3 5 4 4 3 1 3 5 7 Period 9 11 13 15 Group ID Median spline 5 7 9 11 13 15 Figure 11.8 126.4 115.2 NetE 0.4 104.57 0.4 5.36 0.2 126.2 112.10 0.26 0.10 –0.0 123.4 9. Baseline 1 Gross efficiency gain over Nash 0.21 0.27 0.2 132.2 Gross efficiency gain over Nash by treatment.2 3.4 124.2 122.40 0.21 0.02 –0.

. However.004).702 and 0. time dependence. In each of the three cases. CPR appropriation is not reduced in either the baseline or monitoring treatments. in the sanctioning treatment. Moreover. are considered (p-values of 0.742 respectively). it is reasonable to expect that the introduction of the treatment (information. either averaged over periods six to 15 or from period 15 only. information and monitoring.2. this result does not carry over when data. Simply informing the subjects of the optimal symmetric solution to the problem does not significantly affect their long-term behavior.5 1 4 1 5 1 1 3 1 2 5 2 2 1 1 4 4 2 2 1 1 5 5 5 1 3 4 1 2 2 2 2 3 4 2 3 5 4 5 4 3 2 3 3 1 4 3 5 1 4 3 5 1 5 5 5 2 4 2 1 3 4 3 4 3 3 5 5 4 4 2 3 Monitoring 2 5 1 3 4 2 3 3 2 1 2 2 5 3 2 1 2 4 2 3 4 2 1 3 4 3 4 4 5 4 1 1 1 1 5 3 1 5 3 2 4 5 4 5 1 3 1 3 4 4 5 4 1 5 5 2 53 5 2 2 1 4 5 3 2 1 1 5 1 2 5 5 4 3 3 3 4 4 2 1 Monitoring and sanctioning 1 0.222 R. and uncertain data distribution – the exact randomization test for difference in means (Moir. this seems to be the case only in the sanctioning treatment.3 Net efficiency gain over Nash by treatment. CPR appropriation is significantly reduced.5 0 0.3. Due to data limitations – small sample size. Similar results are evident in the calculation of changes in gross efficiency gain over Nash (see Table 11. However. 1998) is used as the primary tool of analysis as it imposes few distributional assumptions.5 1 1 3 1 5 1 4 2 4 1 5 3 2 2 4 3 3 5 1 4 1 2 4 3 2 4 3 5 4 5 4 3 3 1 2 3 2 5 4 5 1 1 5 2 3 1 5 2 2 2 3 4 4 1 1 4 3 5 5 1 5 2 4 1 5 3 3 2 1 2 2 3 5 5 1 3 4 4 3 5 7 Period 9 11 13 15 Group ID Median spline 5 7 9 11 13 15 Figure 11.20 Result 1: Other than a small initial effect in period six. The remaining results address the specific predictions from the previous section. information and monitoring and sanctioning) might actually reduce CPR appropriation and increase efficiency. Moir Baseline 1 Net efficiency gain over Nash 0. Support for this result is presented in Table 11. This result is based upon the change in appropriation as defined in equation 11. which is net of any cooperation exhibited by the group in the first five periods. While we can reject the null of no change in appropriation in favor of an alternative that predicts a decrease in appropriation in period six (p-value = 0. broad support for the success of sanctioning in reducing CPR appropriation is evident.5 0 0.3). Consider the baseline treatment. receives at best only moderate support.

000 0.460).004 0.Spies and swords 223 Table 11.742 Monitoring p-value 1.000) nor when data from the final period is used (p-value = 1. Again.004 0. The considerably higher p-value for change in net efficiency gain suggests that a large number of sanctions were levied. 1998). This result is extended to analysis concerning efficiency values.980 Sanctioning p-value 0.004 0.000 0. though surprisingly in the opposite direction to that hypothesized.3 indicate that in all cases the null hypothesis of no change in appropriation is rejected in favor of the alternative that appropriation decreases (highest p-value is 0. The p-values reported in Table 11.024 0.004 0. Monitoring alone increases appropriation from the CPR. even in the last period when no further interaction is to take place.583 0.024 0. similar results occur when using net or gross efficiency gain data.004 0.004 0. CPR appropriation does not significantly decline in period six (p-value = 0.000 0.381 0.460 0.024 ∆G = 0 ∆G < 0 ∆NetE = 0 ∆NetE > 0 ∆GrsE = 0 ∆GrsE > 0 Note 1 The p-values were calculated using a one-sided exact randomization test for differences in means (Moir. 6–15 6 15 0.103) and the p-value of the change in gross efficiency gain in the same period (p-value = 0. information on its own does not change CPR appropriation.103 0. This certainly rejects the SSPE prediction and suggests that sanctioning might contain elements of punishment for offenses against moral norms. Once monitoring is available. subject behavior is changed.460 1. however.3 Within treatment comparisons Null Alternative Period(s) avg. In each pairwise comparison. five data points from each treatment were used – one for each session within a treatment. These exceptionally high p-values indicate that the opposite alternative hypothesis – CPR appropriation would increase – is supportable.980 1.703 Baseline p-value1 0.702 0.000). Very low p-values are associated with rejection of the null in favor of the alternative.21 In summary. Sanctioning is effective in reducing CPR appropriation.8 tokens – over two tokens per subject.024 0. Information combined with costly monitoring increases appropriation. On average groups reduced their allocation to CPR appropriation by 16. Very high values suggest that an opposite alternative hypothesis is supportable.024).000 1. . 6–15 6 15 avg. 6–15 6 15 avg. The average appropriation between periods six and 15 does not decline when compared to the first five periods (p-value = 1. An interesting comparison can be made between the p-value of the change in net efficiency gain in period 15 (p-value = 0.024).

224 R. 6–15 6 15 avg.052 0. Very high values suggest that an opposite alternative hypothesis is supportable.595 0.929 0. with no formal method of sanctioning. Table 11. In each pairwise comparison five data points from each treatment were used – one for each session within a treatment. Very low p-values are associated with rejection of the null in favor of the alternative. 6–15 6 15 avg.22 Monitoring. It is also possible subjects attempted to police the group by over-appropriating (allocating more tokens to Market 2) as a method of punishing non-cooperators. Sanctioning.016 0. All p-values are in excess of 0. 6–15 6 15 p-value1 0. Instead.012 Sanctioning vs. on the other hand.012 0. baseline Null ∆GM = ∆GB Alternative ∆GM < ∆GB Period(s) avg.4 Across treatment comparisons Comparison Monitoring vs. does not significantly reduce CPR appropriation when compared to the baseline treatment. significantly reduces CPR appropriation when compared to either the baseline or monitoring sessions. cooperating and earning lower profits while everyone else gained. monitoring did not reduce appropriation beyond the reduction already evident in the baseline sessions.4.091 0. Moir Information combined with costly monitoring and costly sanctioning decreases appropriation from the CPR.004 0. 1998). exhibits a detrimental effect in this CPR environment. 6–15 6 15 avg. Maybe they did not want to be the only “loser”.4 (monitoring vs.008 0. Result 2: Monitoring. on its own. Subjects did not seem to be motivated by a moral aversion to being caught cheating. the across treatment results are not all that surprising. baseline). In other words.004 0. Perhaps they felt justified in increasing their appropriation knowing that they were not the only over-appropriators. monitoring MS = MM MS > MM Note 1 The p-values were calculated using a one-sided exact randomization test for differences in means (Moir. Support for this result is found in Table 11.65.837 0. monitoring ∆GS = ∆GM ∆GS < ∆GM Sanctioning vs. monitoring enabled them to see what others were doing and react by increasing their own appropriation rates. Consider section 1 of Table 11. Given these results. . and that this signal went unheeded or caused others to reply in kind.659 0. baseline ∆GS = ∆GB ∆GS < ∆GB Sanctioning vs.

subjects initially make approximately equal investment in monitoring. Except for period six (in which the p-value is 0. This result is evident to a lesser degree when net efficiency gain is analyzed and is clearly evident when gross efficiency gain is analyzed (see Table 11. there is a noticeable decline in monitoring in both treatments. monitoring) of Table 11. baseline) and 3 (sanctioning vs. Likewise. In ten cases subjects were willing to absorb single period sanction costs in excess of $1 (the largest being $2. but the decline is not as severe in the sanctioning sessions.4 (section 4) and Figure 11. Under both treatments. Result 3: Monitoring is used significantly more often in the sanctioning sessions than in the monitoring-only sessions. Punishments were not always trivial. all p-values are significant at the 10 percent level.4.23 In sections 2 (sanctioning vs.24 In one case a single subject received sanctions totaling over $9 in one period.79) to impose a sanction.2). 20 sanctions were levied that exceeded $1 (Canadian) in value. .012).4 Group monitoring levels by treatment. Of the 50 periods in which sanctioning was available. Monitoring 25 5 2 Monitoring and sanctioning 20 Total monitoring 5 5 2 5 4 2 4 4 2 2 4 3 1 1 3 3 1 3 1 2 5 3 4 1 2 4 3 2 3 3 1 45 4 1 2 1 5 4 3 4 5 3 2 4 3 4 3 2 1 3 4 1 5 4 4 2 1 5 2 5 1 3 4 4 2 3 5 3 4 1 5 1 2 1 2 15 10 5 5 1 3 5 5 2 5 1 5 3 4 2 5 3 1 2 4 3 1 2 5 1 0 6 7 9 11 13 15 6 Period 7 9 11 13 15 Group ID Median spline Figure 11.4. subjects seem to concentrate less on over-appropriation and more on pecuniary punishment as a form of control in this environment.Spies and swords 225 Sanctioning significantly reduces CPR appropriation when compared to either the baseline or monitoring treatments. Support for this result can be found in Table 11.595) we can reject the null of equal monitoring in favor of increased monitoring under the sanctioning treatment (p-value = 0. As compared to the monitoring treatments.

at least with this subject pool. as compared to the OGW endogenous sanctioning result. However. an enforceable sanctioning rule is an effective device in controlling CPR appropriation. sanctioning leads to significant decreases in CPR appropriation. . monitoring on its own is not enough.226 R. This means that the results of many of the collective action experiments in public goods and oligopolies that involve costless information about the decisions of others in the experiment cannot be generalized as we once thought. Monitoring may alter people’s behavior but the effect can lead to excessive competition if sanctions are not available. ii social sanctioning and reinforcement of moral norms. and iii formulation of group solidarity or group pride. this may manifest itself as the subjects devising plans to get the most money from the researcher. It is empirically false to model monitoring as costless. and sanctions were endogenously chosen within a framework of rules specified by the experimenter. in this experiment. Even in this harsh environment. to curtail over-appropriation from the CPR. In the baseline treatment. To this extent. CPR appropriation rates approximate the Nash equilibrium prediction.25 It seems that the three elements in conjunction provide the cooperative effects we associate with communication. the method of communication was altered. Despite theoretical claims that monitoring and sanctioning should not take place. both before and after the reading of the passage. Using the information passage is not equivalent to the face-to-face communication that took place in OGW. In this experiment. The third effect can be thought of in the following way: because the group has put effort into finding an appropriation plan each member may be more willing to cooperate with the plan. Face-to-face communication seems to have longerlasting effects in OGW. Villagers must give up time to patrol CPR use or must hire guards to do so. the reading of the passage was meant to mimic (i) while avoiding (ii). monitoring was costly. The design follows that of OGW (1994) with a few important exceptions.26 Finally. In an experiment. I believe that face-to-face communication has three basic effects: i coordination (focus on a specific appropriation level or plan). Specifically. subjects make extensive use of both devices. In fact. The large amount of sanctioning in this experiment. monitoring alone leads to increased CPR appropriation. Costly monitoring is an important element in modeling CPR appropriation behavior. Moir Discussion and conclusion This experiment was designed to examine the role of costly monitoring and costly sanctioning as self-governance devices to control excessive use of a CPR. in which cheating upon group cooperation pays almost $2 Canadian per period. may be attributed to a combination of costly monitoring and the lack of important elements of communication.

g. Monitoring. Future experiments using the OGW design should look at how changes in the costs of monitoring and sanctioning affect resource appropriation. During the treaty negotiations Canadian and American fishers were able to monitor each other’s fishing effort (if not the actual catch itself). The software used for this experiment allows for these costs and the sanctioning rules to be modified on-the-fly. That said. . but with no treaty in place. monitoring. Other subjects commented that they would have been more eager to cooperate had they had the chance to engage in face-to-face communication. and should be altered. can lead to CPR appropriation rates that exceed levels when only minimal monitoring is available (e. so subject-selected values can be implemented. The Pacific salmon fishery on the North American West Coast is a good example of just such a policy breakdown.Spies and swords 227 Nevertheless. This resulted in a race for capture in both countries and severe depletion of the salmon stock. without the ability to impose sanctions. the introduction of an enforced sanction rule did serve the purpose of decreasing CPR appropriation. Many subjects commented that they cooperated long enough to establish a reputation and then dumped all their tokens into CPR appropriation for one period and then returned to cooperative behavior. the net efficiency gain over the Nash prediction is not significantly improved in the sanctioning treatment when compared to the baseline. These subjects used reputation to lull others into cooperative strategies and then earned high profits by defecting. This will allow for subjects to select their governance devices using various constitutional orders and allow research to continue in the important field of institutional economics. when resource users can only observe aggregate extraction rates). An important policy implication can be drawn from these results. and sanctioning behavior can be performed using a game theoretic approach like that outlined in Weissing and Ostrom (1991 and 1993). There are further insights to be gathered both from this data and from future experiments. These future studies have the potential to enhance understanding of CPR governance and more generally the governance of individual behavior in collective action environments. social or pecuniary. Still. Actions mediated by the computer and the experimenter remove the personal interaction according to some subjects. Another line of research involves analyzing this data to see if the endogenously determined level of sanctions is increasing in the size of the offense. While the resource is better managed. overall economic welfare is not necessarily improved. there were no rules to be broken and no principle for imposing sanctions. With the current data a more detailed analysis of individual appropriation. In 1997 a dispute arose between Canada and the United States regarding conservation efforts and this brought the Pacific Salmon Treaty into question. When the environment is expanded to include costly monitoring or an enforced sanctioning rule the strategies available to subjects increase in an almost exponential fashion. the theoretical prediction that sanctioning is a useless governance device is inconsistent with experimental evidence and empirical observation.

Second. Here I concentrate upon collective action experiments in which sanctioning and/or monitoring play a significant role. It is possible to design Dr. the value of which is not returned to the community. to a certain extent. monitoring may be a difficult procedure if the damage from overuse is not immediately apparent. (1995) suggest that subjects may be interested in share maximization or difference maximization. (1992) provides a large number of case studies that attest to both the successes and failures of communally managed CPRs. There are three main reasons for this. A good example of this would be fishers destroying the nets and traps of uncooperative individuals or outsiders (Acheson.228 R. R. Strangelove doomsday rules that involve group-wide punishment for any infraction (e. in which fishermen self-police.g. 5 In Ostrom (1992b: 69–76). Notes 1 Bromley et al. Andrew Muller. distinct from sanctioning. Ostrom (1998).g. Berkes (1992) provides an interesting analysis of fisheries in Turkey.. David Feeny. It is interesting to note that in the sessions in which subjects chose to adopt a sanctioning rule and set its level. monitoring allows individual transgressors to be caught and possibly punished. over time. the agreement upon an appropriation level and the selection of a sanctioning scheme) also affects the outcome. 7 A broader range of CPR experiments are summarized in Moir (1996). 2004). 1988). monitoring costs in terms of time and resources. A different line of research examines the role of collective monitoring when individual actions are too costly to observe (e. This suggests that not only is the ability to sanction important. they observe each other’s fishing behavior and mete out punishment as required. this chapter focuses specifically upon individuals monitoring other individuals just as OGW permitted individuals to sanction each other. 4 As an extension of OGW. Acheson (1988) describes a lobster fishery in Maine. 2 Ostrom et al. and an earlier version of this chapter.g. 8 Subjects in this final treatment were drawn from a pool of subjects experienced with an imposed sanctioning rule. Ito et al. Nonetheless. Another possible explanation may be because the difference in payoffs at the Nash equilibrium and social optimum in public good experiments has been large whereas in CPR experiments it has typically been small. (1994) summarize and expand upon experiment results first reported in Ostrom et al. 6 This result is contrary to the general finding in voluntary contributions to public goods experiments. but such rules are not fair. All remaining errors are my own. This hypothesis has not been tested. the sanctions were rarely imposed. Taylor et al. 3 I adopt the use of the word “sanction” to be a punitive device. and may lead to highly inefficient outcomes if the punishment is ever used. and Tasuyoshi Saijo and four anonymous referees. Stuart Mestelman. This difference in results may be due to the differences in the externality – public good experiments have a positive externality while CPR experiments have a negative externality – or because of income effects (Ledyard. and is able to ascertain some of the variables that affect success. I will use “fine” when any money is returned to the economy either directly to the individual catching a transgressor or to the community at large. Moir Acknowledgments I appreciate the helpful comments of Bram Cadsby. monitoring is listed as a key design principle. all boats are destroyed if anyone overfishes). (1992). First. but how rules arise through group discussion (e. . where subjects seem to show cooperation in excess of the Nash equilibrium prediction but. collapse towards the Nash equilibrium. 1995). Third. In particular. cooperation remained high. Elinor Ostrom.

004. a person could devote up to seven tokens to monitoring. Although total endowment is 160 and would permit negative aggregate profits. or ‘blind’ revenge” (p. 12 Solving equation 11. 17 Seabright (1993: 121) argues that history-dependent threats may allow groups to achieve efficient outcomes. Devices that embody such threats (e. 16 The calculation of benchmark values AO and A* assumes that M = S = 0. here the symmetric optimal individual investment is an integer. if the alternative hypothesis was that appropriation rates should fall. 10 The cost of monitoring is the investment opportunity forgone. monitoring and sanctioning) may thus serve to enforce efficient behavior while remaining relatively unused. so we may never see any history-dependence in observed behavior”. a p-value of one indicates that all observations in the sample show an increase in appropriation rates.1 for A= 0 when M = S = 0 implies that G > 151. A lobster fisherman may be unaware of infractions upon another boat (monitoring information is private). “but if the outcomes are achieved. but cut trap lines and the severity of the damage to equipment will quickly become public knowledge. in a duopoly environment. the largest subset of individuals (seven) commands only 140 tokens and thus cannot cause aggregate profits to be negative on its own. 190). the disequilibrium effects are large and subjects become extremely competitive over CPR appropriation. When testing against zero.htm) using a Novell network and the Windows operating system. no-communication sessions.Spies and swords 229 9 OGW state “[t]here is a nontrivial amount of sanctioning that can be classified as error. Fehr and Gächter (2000a) suggest that positive and negative reciprocity can explain the use of sanctions while Fehr and Falk (2002) extend the model to include an agent’s desire to avoid social disapproval (or gain social approval).000 implies that all observations directly contradicted the alternative hypothesis.g. available at: //socserv. 13 Holcomb and Nelson.68 lab-dollars).ca/econ/mceel/home. Cooper and Stockman (2002) conduct a sequential contribution steplevel public good experiment. They suggest that the latter three behavioral models do not explain their data. Strictly speaking. because subjects were always told the aggregate investment in CPR appropriation (G) and knew their own appropriation (gi). all equal to zero. we fail to reject the null in these . the threats do not need to be exercised. the resource is depleted. which means that all observations “agreed with” the alternative hypothesis. 20 These tests were conducted using software that allows me to calculate p-values for one-tail hypothesis testing. was used. 19 This experiment was conducted at the McMaster Experimental Economic Laboratory (Online. a second sample of five observations. The lowest possible p-value is 0. 176) in their exogenously imposed sanctioning.mcmaster. For instance. whereas Issac et al. Similar instances occurred in communication sessions (pp. then with only six tokens devoted to monitoring. 11 With seven other individuals. find that cooperation is not noticeably enhanced. 14 This is a reasonable model of society. 188–189) and were concerns expressed by subjects (p. Both the Nash equilibrium and socially optimal solution to this problem are independent of endowment so the “cost” of monitoring is equal to the return from investment in the private market (3. they could deduce the appropriation of the remaining individual. lagged punishment. The highest possible p-value of 1. using the Delphi programming language. find that this note passing facilitates tacit collusion between subjects. aggregate profits are negative. 18 Instructions are available from the author upon request. in a public good environment.. Fehr and Schmidt (1999) suggest that a model of inequity aversion could explain why sanctions are used and why they improve efficiency. Thus when G > 151. 15 Unlike the OGW design. The software was designed by the author and Andrew Muller. OGW results suggest that when a subset of individuals can cause negative aggregate profits. However.

Gt = β0 + β1M + β2S + β3t + εt was conducted in which. Success and failure in marine costal fisheries of Turkey.25. McKean. Thomson (eds. However. directly opposing the one actually tested. C. 88 (7–8).2.. “I am glad we talked in the hall without the experimenter. To put this in perspective.) Making the Commons Work: Theory. A.536. International Game Theory Review. Berkes. Perhaps this result is further evidence of the limitation of simple game theory in ultimatum games (see Roth. while in this design potential earnings increase by almost $2.10 in a period. Oakerson. The monetary gain to cheating is greater in this experiment and yet sanctioning is still effective. Blaikie. 2003. Acheson. Harriss.) Making the Commons Work: . D. 1988. .62 per period.. The difference in the payoff function between this experiment and OGW leads to an important difference in the gains from cheating when all others are investing in CPR appropriation at the optimal level..” Communication clearly led to group solidarity in this instance. and t = {1. P. any sanction a subject levies will reduce his cumulative profit and cannot lead to future cooperation in this game as it has ended.70) and significant (p-value = 0. Runge. D. Subjects were aware that the experiment would finish at the end of period 15. 1988.053) while the sanctioning coefficient is negative (−18. The management and use of common-property resources in Tamil Nadu. In: D. Public goods experiments without confidentiality: a glimpse into fund-raising. F. Additional information can be found in Moir (1996). .. M. 161–182. Bromley. R. On the theory of infinitely repeated games with discounting. and J. Thomson (eds. Her comments amounted to (I paraphrase). Peters. Hanover. Econometrica.00 in a period. India. D. McKean.W. a parametric regression. M.. M = 1 for monitoring-only sessions (and 0 otherwise). J. References Abreu. or approximately $1. P.66) and significant (p-value = 0. Bromley. In addition to the exact randomization test. It became a challenge to see how much money we could take from the experimenter. 1992.. Oakerson. The algebra of assortative encounters and the evolution of cooperation. Peters.F. In OGW. 1995 for a survey). it suggests that an alternative alternative hypothesis. 1605–1623. a payoff-maximizing reaction leads to a potential increase in earnings of $1.. San Francisco.F.W. J. T. 1991 and 1997). S = 1 for sanctioning sessions (and 0 otherwise). This may have a detrimental effect in CPR or public good environments as it may increase free riding behavior but it may have a socially desirable effect in the case of oligopolies by increasing output and reducing prices (Holcomb and Nelson.230 R. Gilles. Feeny. It was further reinforced by the comments of a subject who was participating in an unrelated experiment at McMaster. J. 1992. R. A sanction of $1 reduces this average period payoff by almost 62 percent. J. 211–228. The monitoring coefficient is positive (3. Bergstrom. J. 5 (3). P. C. Practice and Policy. NH: University Press of New England. Andreoni.. The Lobster Gangs of Maine. and Pain. R. Runge. .000). In: D. This issue was pointed out to me by Elinor Ostrom at a meeting of the Economic Science Association. Journal of Public Economics. Moir 21 22 23 24 25 26 cases.M. 2004. CA: Institute for Contemporary Studies Press. Gilles. 383–396. and J. could be supported.. Feeny. 10} is the period number normalized at 1 for period six. recall that the average subject payoff using the results of all subjects in all sessions was $24. and Petrie. Given sanctions are costly. 80 (4). The adjusted R2 for this cross-sectional time series is 0.

.H. (eds.. (28 August). E. Brazil. P. Feeny. The economic theory of a common-property resource: the fishery. Journal of Economic Behavior and Organization.. J. Saint John.. P. N. and J.. D. 1999. P. M. E. San Francisco.. Casari. Mikhailov (eds. J. Quarterly Journal of Economics. Peters. K. Peters. 1992. R. Theory of fairness. Fairness and retaliation: the economics of reciprocity. J. Oakerson. 26 (1). J. NY: Springer Press. Diversity and collective action. and Glance. 2000b. and Schmidt. 162 (December). A laboratory study of voluntary public goods provision with imperfect monitoring and communication. 14 (3). Journal of Economic Perspectives. and McKean. Fehr. 1968. Journal of Economic Behavior and Organization. E. Practice Policy. 817–868. 58 (2). 1991. T.. 51 (2). and Plott.K. Making the Commons Work: Theory. B1. Fehr. 46 (4–5). P. Experimental Economics.. and Gächter. E. Gilles. San Francisco. Oakerson. M. 1999. CA: Institute for Contemporary Studies Press.K. 183–205. 8 (3). S. 341–369. Bromley. Journal of Development Economics. Thomson (eds. CA: University of California Press. Making the Commons Work: Theory. 79–94. D. p. 1343–1348. In: H.H. Dutta..S. M. Oakerson.W.W. San Francisco. Caputo.. The commons. Modeling common property ownership as a dynamic contract. . 2002. In: D. Gächter. Thomson (eds. Feeny. M.K. G. D. 1987. 413–426. 39 (4). 1993. American Economic Review.. Journal of Political Economy. Gilles. 1992..F. Collective action as a social exchange. and Nelson. Fehr. The tragedy of the commons.).. The tragedy of the commons? Economic Theory. 217–247. M. B. 124–142. 159–181.) Interdisciplinary Approaches to Nonlinear Systems. 1998. and Falk. M. J. 980–994. 2000a. P. Runge. and Gächter. Cooperation and punishment in public goods experiment. Holcomb. 2003. CA: Institute for Contemporary Studies Press. 90 (4). Decentralized management of common property resources: experiments with a centuries-old institution. 44–64. competition and cooperation.. Holcomb. Runge.W. Learning to punish: experimental evidence from a sequential step-level public goods game. D. Natural Resource Modeling. and Nelson. McKean. Hawkins. Gordon. CA: Institute for Contemporary Studies Press.. R.. C.S. Practice and Policy. D. and Stockman. Huberman. J. newspaper. and Kahn. 533–552.. Science. J. 1992. Cooper.A. S. F.W.). 3 (3). M. Times Globe.M. Feeny. McKean.. Cason. 1993. A. and Lueck. Principles of Group Solidarity. Fehr.. J. 114 (3). C. Bromley. P.. Berkeley.R. and Sundaram. and J. Bromley. C. 225–245. Practice and Policy.S. In: D. 20 (3). S. Psychological foundations of incentives. 3–15. 39–51.F. 62 (2)... 687–724.. New York. 1999. property. Cartel failure: a mistake or do they do it to each other on purpose? Journal of Socio-Economics. C. S. M. R. 235–249. NB. Cordell. Hechter. Too many chasers. Hardin.. Haken and A.. and common-property regimes.. Peters. San Francisco. D. CA: Institute for Contemporary Studies Press.). 247–264. 1994. Practice and Policy. and Thomson. Runge. The role of monitoring in duopoly market outcomes.F. R. 5 (1). Sea tenure in Bahia. C. McKean. E. European Economic Review. D. Gilles. 1954. Bromley. Making the Commons Work: Theory.Spies and swords 231 Theory. Journal of Socio-Economics. and Fehr. 1997. 2002.

R. Journal of Economic Behavior and Organization.F. 545–565. T. Palfrey. Saijo. Ito. Public goods provision in an experimental environment. Ostrom. McKean. Lueck... The ecology of collective action: a public goods and sanctions experiment with controlled group formation. Repeated play. J. 1994. 245–252.P. E. Feeny. Roth. Gardner.. 1992. 495–521. L. Marine Resource Economics. and J. 253–280. 1994. H. U.M. 293–318.F.) The Handbook of Experimental Economics. Rules. The “spite” dilemma in voluntary contribution mechanism experiments. J. K. 1994a.. H. American Political Science Review.R.. Bromley... and Walker. J. Behind the market stage where real societies exist – Part I: the role of public and private order institutions. 61 (3). NJ: Princeton University Press.. E. 1985.. 26 (1).. and Putterman. and Gardner. 404–417. 1982. 92 (1). San Francisco. D. P.W. Peters. 753–817. The analysis of cooperation in collective action games. C.. Milgrom. In: D. Roth (eds. Traditional commons land in Japan.. and Nakamura. 87–100. Oakerson.H. 1994. Walker. ON: McMaster University.. T. 1992. Peña-Torres. 311–335. R. and Plott..E. Behind the market stage where real societies exist – Part II: the role of moral norms. Runge. 63–98. 1992a. 1997. Peters. and J.. Journal of Economic Behavior and Organization... Experimental Economics. J. E. Feeny. Review of Economic Studies. Moir. E.. R.P..F. Common property as an egalitarian share contract. In: J. San Francisco. D. MI: University of Michigan Press. The rudiments of a theory of the origins. CA: Institute for Contemporary Studies Press. PhD Thesis. American Political Science Review. Bargaining experiments. Ostrom. Roth (eds. CA: Institute for Contemporary Studies Press. 39 (3).. D. Kreps. 1995. Practice and Policy. Journal of Economic Theory.. McKean. The tragedy of the commons revisited: identifying behavioral principles. Moir. and Common-pool Resources. Games. C. 62 (4).. San Francisco. Journal of Development Studies. R.. D. Bromley. Saijo. 1995. Ostrom. 1994b. C. Ann Arbor. Moir Isaac. M. McCue. M.H. Roberts. M.E.W. J. Ostrom. R. P. 1992b. In: J. A Monte-Carlo analysis of the Fisher randomization technique: reviving randomization for experimental economists.. J. Platteau. Practice and Policy..) Making the Commons Work: Theory.R. E. Journal of Development Studies. 2007.. J. 1 (1). R. 1998. J. Ledyard. Kagel and A.) Making the Commons Work: Theory. Platteau. NJ: Princeton University Press. 93–108. Runge.E.. 535–560. Public goods: a survey of experimental research. A behavioral approach to the rational choice theory of collective action. 28 (3).. P. Ones. 1996. Crafting Institutions for Self-Governing Irrigation Systems. CA: Institute for Contemporary Studies Press.. 533–577. and Wilson.. Princeton. R. 86 (2). survival and performance of common-property institutions. 30 (3). A. J. . R. 1–22. 1998.232 R. 1995. 27 (2).) The Handbook of Experimental Economics. Oakerson. Gilles. Rational cooperation in the finitely repeated prisoner’s dilemma. Thomson (eds. Gilles. Journal of Economic Behavior and Organization. Journal of Public Economics. Kagel and A. T. Hamiton.. 253–348. In: D. 25 (1). 1995. M. cooperation and coordination: an experimental study. Ostrom. McKean. M. Princeton. and Une. and Rosenthal. 111–194. Covenants with and without a sword: selfgovernance is possible. 12 (4). Thomson (eds. The political economy of fishing regulation: the case of Chile. 30 (3). Journal of Conflict Resolution. 51–74.

Selten (ed. Peters. McKean. Scharpf (ed. M.. Managing local commons: theoretical issues in incentive design.. Randall. 188–262. M. San Francisco. D. E. Weissing. 2004. Southern Economics Journal. and J. and Pushkarskaya.W.A. P. Sohngen. and Ostrom.. 1990.. J. Weissing.. In: D. E. 1973. 113–134. A.pdf. In: D. available at: http://web. R.. 86 (4). Online. Wade. CO: Westview Press. R.M. 203–211. 1993. J. J. 51 (1). The provision of a sanctioning system as a public good. 110–116.and farmer-managed systems. Yamagishi. 387–428. CA: Institute for Contemporary Studies Press. Thomson (eds. and Oakerson.F. J. 19 (3). 1991. Oakerson.edu/cob/econ/research/papers/ecwps054shupp. 207–228. 1996. Thomson (eds. 1992. 1992. Feeny. San Francisco. Resource management in South Indian villages. P.) Making the Commons Work: Theory. Runge. Bromley. R. R. In: R. P. Feeny. Rent dissipation in a limited-access common-pool resource: experimental evidence.. Selten. Irrigation institutions and the games irrigators play: rule enforcement on government.. A simple model of imperfect competition where 4 are few and 6 are many. M. Oakerson. and J. Sefton.) Games in Hierarchies and Networks: Analytical and Empirical Approaches to the Study of Governance Institutions. C. 1993. K. Feeny..) Making the Commons Work: Theory. Shupp. 2000. American Journal of Agricultural Economics. R.) Game Equilibrium Models II: Methods. Practice and Policy. . 141–201. Berlin: Springer-Verlag. Runge. Practice and Policy. 7 (3). CA: Institute for Contemporary Studies Press. The effect of rewards and sanctions in provision of public goods.. 235–247. American Economic Review. Seabright. 86 (5).bsu.. and Somanathan. Journal of Personality and Social Psychology. E. J. D.Spies and swords 233 Schmitt. T. The evolution of social norms in common property resource use. Thomson. 7 (4). and Walker. Bromley. Manuscript: Ball State University..F. H. Peters. 1986. Taylor. 829–854. Rewards and sanctions and the provision of public goods in one-shot settings.W. Journal of Economic Perspectives. C.. McKean. and Halloran.. 2 (3).. 2004. Boulder. M. Experimental Economics.A. F.. In: F. and Walker.. Walker. E.. R. Collective action with incomplete commitment: experimental evidence. 766–788. D. and Markets. J. 129–160.. M.W. Journal of Environmental Economics and Management. Irrigation institutions and the games irrigators play: rule enforcement without guards. Swope. International Journal of Game Theory. and Ostrom. Sethi. R. Institutional dynamics: the evolution and dissolution of common-property resource management.. J. Morals. Gilles. P.. 66 (4).. F.. B. Group contracts for voluntary nonpoint source pollution: evidence from experiment auctions. and Ostrom. Gardner. 2005. 1196–1202. Gilles. Walker.. R.

on ways in which experimental studies could be adapted to be more useful. what were the early contingent valuation studies (a method that was firmly established as an important tool by the mid 1970s) but experimental studies in the field? The WTP/WTA divergence was first observed in these studies (for example Brown and Hammack 1973). which may ultimately have direct policy relevance. further. In considering this question. others provide insight into response to incentives under specific institutions. Indeed. After all. How useful are the findings from these chapters to advancing the field of environmental economics? It is my task to take a stab at this question and to opine.” contains five chapters that employ experimental methods to study the effects of alternative incentives with respect to public goods and common pool resources on the behavior of subjects. They . Capra and Tanaka take us to the world of decision making in an explicitly dynamic setting in the purview of a classic common property problem. As an environmental economist with a strong interest in non-market valuation. I find a fun paradox to this question as it appears to me that some of the earliest work in experimental economics was done by environmental economists and that this early body of work identified some key issues that helped spawn the behavioral economics literature. some experimental findings may have very immediate and direct policy relevance. Kling Part II of this book entitled “Common property and public goods. Bishop and Heberlein were experimenting with real money payments and numerous environmental economists were studying interviewer effects. Perhaps it is appropriate that we have come full circle to ask what the current research in experimental economics is contributing to the field of environmental economics and how those contributions might be further strengthened.12 Discussion Common property and public goods Catherine L. more generally. sequencing. The chapters in this part provide a rich base under which to consider this continuum. alternative payment vehicles. but only in a highly controlled environment that will be hard to extrapolate beyond the experimental setting. Nonetheless such contributions help identify key drivers. it is surely important to remind ourselves that there are many ways in which we can learn from any one study and that any one study rarely provides all the answers to any question. and the effects of many other institutional features of markets and the environmental goods being valued. as early as 1979.

they conclude that three forms of communication were necessary to achieve the optimum: (1) expression of awareness of the threshold effect. in addition to the small size of the subject pool (it would be a rare commercial fishery with only five fishermen). Indeed. Citing a dispute between the US and Canada that resulted in non-enforcement of the relevant treaty. Rob Moir also studies the behavior of subjects deciding how much of a common property resource to exploit in a dynamic setting. Moir argues that these results are relevant for fisheries using the Pacific salmon fishery as an example. there is the difficulty of interpreting causation in their work. it is not clear whether this communication is the cause or result of this successful coordination. and (3) acknowledgment of success. The authors have indeed focused on an important question that resource economists understand to be important. What do we learn from this experimental setting for regulation in fisheries or other common property resources? The focus of this work on the form and content of the communication between the subjects mirrors the important literature in fisheries related to information sharing (or lack thereof!) and control. Above 31. He finds evidence that monitoring without sanctioning does not move the group towards social efficiency (in fact it moves further towards the Nash solution). the third identified category.1 Interestingly. he argued that anglers monitored one another. subjects cannot communicate. While this analogy is compelling. and the stock size and consumption potential declines in the next period. the stock size increases and consumption potential increases commensurately. In his experimental design however. one would not want to jump to the conclusion that a policy recommendation of enabling anglers to punish one another is the solution. the identification of the lack of consequences associated with monitoring as a key problem does help us identify the sources of market failure and the directions that effective policy .Discussion 235 motivate their experiment as being relevant to the case of a renewable fishery where the current period’s fish stock is a function of last period’s stock in a discontinuous manner: if the previous period’s stock is less than 31. but in one treatment they can expend resources to learn about the level of use of the resource by other individuals in the experiment. there is less than full replacement. acknowledgement of success. However. there is increased communication of the sort they identify. but that monitoring in conjunction with sanctioning does. can only occur after successful cooperation and seems quite likely to be directly caused by that success. and in another treatment subjects can monitor individuals and punish (sanction) those who they have monitored. but were not able to impose sanctions and that significant overfishing occurred as a result. (2) recognition of this by the majority of the subjects. By analyzing the script of a chat room that allowed communication between the five subjects in each session. However. they seek to assess the type of communication that is most effective in solving the common property problem. While they make a compelling case that when subjects solve the coordination problem.

don’t bother monitoring if you aren’t going to put some teeth behind it. etc. subjects can contribute to three different public goods (that differ by differing marginal benefits of contribution). recall that the “public good” studied in this context is the contribution of a token that returns 0. Having thus criticized the conclusions. First. Cotten et al. and in a third treatment.L. If there is one thing that we have learned in over 30 years of non-market valuation studies with respect to the value of public goods. What does such an experiment teach us about the design of efficient provision mechanism for public goods? The authors indicate that: Our results indicate that more options for otherwise similar public goods will increase total dollar contributions towards the larger cause. preservation of natural habitats. In one treatment. the extent. they are comparing one public good with three identical public goods: there is no way to know whether the effect they find will be of the same sign when moving from a baseline other than one. it is that context matters. support for endangered species. Indeed. diabetes research. this conclusion reaches too far for at least two reasons. for example. form. Second. but suboptimal public good provision come from a VCM. They find that offering multiple public goods increases the total contribution to that public good.6 of its value to each of the group’s members. Cherry and Dickinson study the consequences of offering subjects multiple public goods in a voluntary contribution mechanism with an eye toward how such an offering affects the total amount of public good provision.236 C. directly raise and assess a potential limitation concerning the relevance of experimental findings regarding voluntary contribution . The value of a change in the provision of a public good varies with its initial condition. For example. or any other baseline level of provision. the number and quality of substitutes available. in a second treatment they offer three identical public goods (the “multiple homogeneous” treatment). six to seven. In my view. and specific type of change proposed. I hasten to add that the results are likely to hold in some cases and for some/many goods. the time needed for the change to occur. homeless populations. multiple options for providing relief following an environmental disaster is predicted to increase the total amount of voluntary giving among private citizens. aid to Africa. they replicate the standard findings that positive. Diminishing total contributions might set in when moving from three to four. the payment mechanism used to fund the provision. To conclude that what is learned in the context of a single experiment for a good that is quite sterile in its context will apply uniformly to all situations of public good provision overreaches. Kling could take. and almost any charitable cause that comes to mind. they are quite consistent with the simple observation that most charitable causes do indeed have multiple options for provision – there are multiple charitable organizations that collect funds for cancer research.

In the latter setting. experiments such as this provide important insight that can eventually add provide clear policy advice. Sturm and Weimann construct an experiment to test whether laboratory subjects behave as predicted by Hoel’s (1991) model of unilateral emissions abatement.Discussion 237 mechanisms for environmental policy. the fact that subjects do not respond to the incentive in an apparently logical way could be interpreted as prima facie evidence that inadequate incentives exist for full engagement in the activity. How relevant are these findings for predicting governments’ actions in international pollution abatement? The authors argue that there is little basis for making a claim to relevance as “the external validity of results gained in singular laboratory experiments is restricted to the specific laboratory environment. randomly matching some human subjects with a preprogrammed computer who will not earn rewards for anyone. do I have any suggestion as to how these studies might be of even greater relevance . For example. While each contributes valuable information. Indeed. but it may also be that other explanations can account for it. Their findings suggest that leadership does indeed matter and in a socially optimal direction. we have to admit that we are not able to make any recommendations for environmental policy purposes. it may be that this is really a test of whether the incentives are large enough to motivate subjects to consider seriously the consequences of their actions. additional possible explanations may arise. Their inference that this unexplained behavior is due to “confusion” may be correct. Hoel predicts that abatement by a single country above the equilibrium level will result in greater total abatement. their goal is to determine whether leadership matters. the study raises considerable concern about the ability to apply directly the findings of public good experiments to environmental policy. provides striking findings that support the contention that other contributions to the public good in this context cannot be fully explained by “other-regarding” behavior. although lower abatement levels by the other countries.” While this modest assessment is refreshing. As other researchers consider these questions. Sturm and Weimann test this hypothesis as well as one based on a variation of the game that has one player moving first. Their more interesting findings come in their sequential treatments where they find that an appointed leader will usually (33 out of 36 times!) choose a higher abatement level rather than the lower abatement predicted by Nash equilibrium. Each of these studies has clearly contributed to our understanding of environmental economics. Their findings largely confirm Hoel’s predictions – unilateral emissions abatement does generate lower total emissions. nonetheless. They consider the disturbing question of whether subjects are truly responding to the incentives of the experimental setting or are merely confused. While the authors note that their payment levels are in line with typical wage rates for students. although they represent very different locations on the spectrum of immediate relevance. this alone does not prove that the rewards are high enough. Their experimental design. In a study using international emissions abatement as the motivating public good.

environmental group representatives. In my view. Todd L. Mónica and Tomomi Tanaka. Dickinson.). Note 1 While the authors. I readily admit that this view is hardly novel. Gardner and Judd Hammack. . the right amount of information to provide respondents in a stated preference study. and T. it can entail increasing the payoffs to subjects. R. Brown. and increase otherwise. simplify too little and one is left with an overly complex model that does not identify the key policy variables or response. many experiments could move along the continuum towards the more immediate policy relevance side by adding more relevant context to their studies. of course. and it can entail using subject pools that are directly relevant to the questions being studied (fishermen. it has the potential for large dividends in increasing the direct relevance of the findings of experiments to environmental policy. “Voluntary Contributions with Multiple Public Goods” in this volume. use of two different values for “A” in their production function makes the threshold between rising and declining stocks discontinuous and larger than it would be for a single value. stocks would decline in later periods when Kt < A2. Specifically. “Communication and the Extraction of Natural Renewable Resources with Threshold Externalities” in this volume. There are of course many other ways in which relevant context can be added. Cherry. Kling to environmental economics? The degree to which the results of an experiment are relevant to a particular question or problem depends on the degree to which it captures the important features of the problem. etc. Heberlein.” Review of Economics and Statistics 55: 73–82. and David L. whether that is the right amount of realistic detail and financial incentives in an experiment. While this may make the job of the experimentalist more difficult. This can entail describing (as some of the experiments in this group of five did) the good in non-generic terms. indeed I suspect that Erica Jong summarizes it very well when she noted that “advice is what we ask for when we already know the answer but wish we didn’t. (1979) “Measuring Values of Extra-Market Goods: Are Indirect Measures Biased?” American Journal of Agricultural Economics 61: 926–930. (1973) “Dynamic Economic Management of Migratory Waterfowl. the standard conundrum of any model: simplify (assume away) too much and one is left with a sterile model (experiment) that cannot usefully inform policy.” It’s hard work to add context and it requires understanding a great deal about the institutions and history of a particular environmental problem to get it right.238 C. C. the ticket is to get the “right” amount of context. This is. Thus. References Bishop. or the right number and definition of explanatory variables in an econometric analysis. Capra. a single value would still generate the key threshold effects for their functional form.L.

and Christian A.” Journal of Environmental Economics and Management 20: 55–70. Rob. “Can Public Goods Experiments Inform Policy? Interpreting Results in the Presence of Confused Subjects” in this volume. Michael. “Unilateral Emissions Abatement: An Experiment” in this volume. (1991) “Global Environmental Problems: The Effects of Unilateral Actions Taken by One Country. Vossler.. Paul J.Discussion 239 Cotten. Bodo and Joachim Weimann. Hoel. Sturm. Stephen J. Moir. . “Spies and Swords: Behavior in Environments with Costly Monitoring and Sanctioning” in this volume. Ferraro.

.

Part III Regulation and compliance .

.

the literature has primarily focused on pollution and/or reporting decisions at the firm level. and voluntary programs such as Energy Star (see US EPA. consumer demand response. have received limited attention in the literature (see Shavell. and shareholder reaction. Thus. Evans. In the case of mandatory information disclosure programs. an understanding of firm incentives under such programs is essential to evaluating their performance. and motivating the emergence of new programs. and history of detected violations for example. We argue that a firm’s internal organizational structure alters the incentives faced by decision makers and therefore has the potential to affect their compliance decisions. the firm’s decision of whether to comply with reporting requirements. Helland. we first . These factors may affect the quantity of firm emissions. and Christian A. 1998). Following that. 1996. improving their design. 1990. We adapt a model developed by Gilpatric (2005) to examine these incentives and test the resulting hypotheses using experimental data. compliance costs. Vossler Introduction Publicly reported information on the environmental behavior of firms can increase the efficacy of private markets as a mechanism to control environmental malfeasance through liability for harm. Harrington. A number of factors have the potential to alter the effectiveness of environmental information disclosure programs in encouraging firms to adopt desirable behaviors. To date. 2001). 1984.13 Managerial incentives for compliance with environmental information disclosure programs Mary F. Gilpatric. Scott M. and the accuracy of reported emissions. Michael McKee. We seek to build on this literature by examining the pollution/reporting decision of individuals within the firm. Within the realm of environmental policy. These factors include features related to program design such as the timing of information release. Beard. firm characteristics such as size. examples exist of both mandatory information disclosure programs such as the EPA’s Toxics Release Inventory (TRI). Some of the factors that may affect the firm’s pollution and/or reporting decision such as financial status. firms are required to report information that is potentially damaging to them. Larson. The next part of the chapter motivates our work with an overview of the information disclosure and tournament literatures. and the existence of complementary policies such as liability rules. 1988.

Using data from the TRI. we derive testable hypotheses of behavior that vary with the payoffs received by division managers (based on rank and whether a manager is found to have engaged in malfeasance). Konar and Cohen (1997) find that future emissions were lower among firms with the largest stock price decreases on the day of the information release. examine opportunities for malfeasance in the context of information disclosure programs. which they maintain are more likely to employ a dedicated environmental staff. Konar and Cohen. we turn to the firm’s organizational structure and present a model where incentives of lower-level or division managers.. They find that both the smallest hazardous waste generators in their sample. 1995. investors react to information concerning the firm’s financial health such that the value of the firm (share prices) tends to fall when adverse environmental information (such as TRI reports) is made publicly available.F. 2001)? Khanna et al.244 M. We test these predictions using laboratory experiments and report results.1 The second question relates to the effect of public information disclosure on subsequent firm environmental performance. what is the reaction of investors to the release of information regarding a firm’s environmental performance (Hamilton. For example. Evans et al. Regardless of the reason. From this model. 1991). are less likely to violate. Their analysis suggests that facility size may be an important factor in compliance. and the penalty imposed on a manager caught engaging in malfeasance. which they argue have lower compliance costs. and the firms with the largest sales volumes. perhaps because the larger company provides . an analysis of TRI compliance of facilities in Minnesota by Brehm and Hamilton (1996) suggests that ignorance of the requirements of the regulation may better explain violations (measured as failure to report) than evasion. Motivation Previous studies of information disclosure programs have focused primarily on investigating two empirical questions. Estimates from the Government Accountability Office (until 2004 known as the General Accounting Office. Empirical analyses of compliance with information disclosure programs are less common due to the lack of detailed compliance data and the difficulties associated with detecting non-compliance through misreporting rather than through failing to report at all. Brehm and Hamilton also find that subsidiaries of larger companies are less likely to violate. First. 1998. the probability that malfeasance is detected. investors may perceive poor environmental performance as an indication of inefficient input use. we offer some conclusions and discuss the next steps in this line of research. then. In addition. investors may expect firms with poor environmental performance to face increased future compliance costs and greater risk of liabilities. In the final part of the chapter. are determined by a rank-order tournament. Khanna et al. However. who report to an ownermanager. GAO) suggest that approximately a third of facilities subject to reporting under the TRI during its initial years failed to report (GAO. (1998) list several motivations for this line of research.

while imposing potentially large future liabilities on the firm. firms face a tradeoff as compensation schemes that encourage greater managerial effort also generate an incentive for non-compliance.2 In the context of compliance with information disclosure programs (and other regulatory mandates). However. 1983). little work has explored incentives in a tournament setting when workers choose not solely how much work effort to exert. compensation mechanisms based on tournament structures can induce managers to engage in such malfeasance. failing to maintain equipment adequately. the program may have sufficient sanctions for noncompliance such that compliance is optimal at firm level assuming the firm can costlessly monitor manager behavior. Consider that many internal reward structures (including promotion ladders) imply that division managers are playing a rankorder tournament game. In general. Managers may be able to influence the mean output through choices other than work effort. However. All such activities may increase the manager’s output as observed by his employer. A substantial literature has compared the efficiency of tournaments with alternative compensation schemes. The most significant line of research regarding malfeasance in tournament settings involves the exploration of “influence activities”: behavior that arises . They maintain that this finding supports the argument that firms with more information (less ignorance) are more likely to comply. this type of malfeasance may take the form of a manager increasing division profits by illegally dumping waste. with regard to such factors as the risk-aversion of workers and the flexibility of the incentive framework to environmental uncertainty (e. When monitoring is imperfect it is likely that manager incentives are not perfectly aligned with those of the firm because there are opportunities to increase the probability of winning the tournament by engaging in activities that do not serve the firm’s interest. but some other aspect of the work that is undertaken.Managerial incentives for compliance 245 environmental and legal staff to the subsidiaries. Lazear and Rosen (1981) show that such mechanisms can induce efficient behavior when all managerial actions directed toward winning the game are in the form of productive effort. However. it seems reasonable that access to parent company environmental and legal staff may also reduce compliance costs thus increasing the likelihood of compliance.3 To the extent that non-compliance may improve managers’ apparent output or productivity and such behavior is costly to observe. to our knowledge the literature has overlooked the possible role of a firm’s internal organizational structure in creating a divergence between manager incentives and the objectives of an information disclosure program.g. While a potentially important consideration. However. to achieve full compliance it may be very costly to monitor the behavior of division managers. if purported performance is improved via fraudulent reporting or other malfeasance such as cost savings through higher and unreported toxics releases. Nalebuff and Stiglitz. such as through choice of production process or regulatory compliance. such as piece rates. or manipulating accounts to show larger current revenues at the expense of future revenues.

However. Prendergast and Topel. Such behavior is costly to the firm because it dulls a worker’s incentives to exert productive effort to win the tournament. we define malfeasance as a behavior that is inconsistent with the firm’s objectives. or other rewards is perhaps the most ubiquitous incentive mechanism within firms. 2003). the fact that the individual making decisions about whom to promote or otherwise reward benefits from the behavior at the expense of the firm). We focus on the incentives generated by a rank-order tournament compensation scheme (such as promotion ladders) for two reasons: (1) competing for promotion. Kim et al. when workers can influence the choice of superiors regarding who is promoted or otherwise rewarded in an organization through actions that are nonproductive. Evans et al. Malfeasance in the form of non-compliance with regulatory mandates. The firm may wish to limit emissions to avoid associated penalties and potential liability. including failing to disclose information accurately. 1988. Firm organizational structure and information disclosure Non-compliance and malfeasance in the context of information disclosure While the compliance literature has relied primarily on a framework that focuses on firm-level decision making.e. 1996. bonuses.. in which managers’ payments depend on their output. Gilpatric (2005) constructs such a model of the firm to examine the general case of corporate governance. then any compensation that rewards managers for higher output will generate both the intended incentive for them to exert greater work effort. Here. If managers can increase their apparent output (such as the profits from their division) by increasing emissions or reducing care (and thus increasing the probability of accidental emissions) and if this behavior is sufficiently costly for the firm to monitor and prevent such that monitoring is imperfect. in an evenly matched tournament individuals can face a strong incentive to cheat even if doing so achieves only a small output gain if this is sufficient to increase significantly . Environmental malfeasance of course also entails important social costs that do not arise from influence activities within a firm and that are clearly of significant concern to regulators. the firm may be an inappropriate unit of account. but also an incentive to engage in malfeasance. may provide managers with incentives to engage in malfeasance. ranging from ingratiation to bribery and sabotage of competitors (Milgrom and Roberts. any internal organizational structure that includes incentivebased compensation. 2002. Therefore.246 M. imposes direct costs on the firm that may significantly exceed those resulting from dulled incentives. for example) but on the advantage cheating provides relative to competitors. and (2) tournaments have the characteristic that players’ incentives to “cheat” depend not on the absolute gain from doing so (as would be true for piece-rate compensation. and Chen. The malfeasance we discuss here differs from influence activities because it does not derive from an agency conflict (i.F.

We do not explicitly model this form of non-compliance here. division managers are cheating or engaging in malfeasance. non-compliance may result from behavior on the part of the ownermanager. Note that even in cases 1 and 4 where the firm is compliant with the reporting requirements. the level of emissions need not equal the socially optimal level.1 illustrates the possible cheating and noncompliance cases where z = ∑ zi ∧ ∧ i =1 N represents the level of reported emissions based on the division manager reports and x = ∑ xi ∧ ∧ i =1 N is actual emissions of the firm. we assume that the owner-manager reports z = ∑ zi ∧ ∧ i =1 N to the environmental authority. First. Managers are said to be engaging in malfeasance or cheating if they (1) emit more than optimal from the firm’s perspective. Second. Let z represent the level of emissions that is optimal (at the firm level) to report to the environmental authority with z ≤ x. In the first three cases.Managerial incentives for compliance 247 their probability of winning. The owner-manager reports firm-level emissions to the environmental authority as required by the information disclosure program.4 By considering the emissions and reporting decisions of lower-level managers. each of which has a designated manager with the responsibility of reporting emissions for his division to the owner-manger. Table 13. Let x represent the firm’s (owner-manager’s) optimal total emissions level. we introduce several opportunities for non-compliance. malfeasance on the part of division managers may prevent him from doing so. as shown above. and/or (2) fail to report their actual emissions. Let z i represent emisˆ sions reported by the ith division manager and xi represent the optimal level of ˆ emissions for division i from the perspective of the division manager. The opportunities for malfeasance that arise when we extend the model to include internal organization may result in higher levels of overall emissions and/or more frequent misreporting as divisions compete to reduce current production costs. . even if the owner-manager wishes to report the level of emissions truthfully. In cases 2 and 3 the firm is misreporting its emissions and therefore is non-compliant with the information disclosure program. Assume that the firm is composed of N divisions. In order to focus on the effect of division manager-level decision making.

g. being fined or fired). It is costly for the firm to audit the behavior of division ˆ managers and it does so with probability η . Division managers are directed to emit no more than xi . being promoted) and may also face additional sanction (e. Malfeasance with managerial compensation based on tournament payoffs Gilpatric (2005) develops a model of cheating in a tournament in which identical contestants first choose effort then. we derive hypotheses regarding the likelihood of cheating on the part of managers who are playing a rank-order tournament game in terms of their financial compensation. One expects that a greater likelihood of cheating being detected or stiffer penalties if caught will deter cheating to some degree. we focus exclusively on the second case in Table 13.1 above. but increasing our understanding of .1 Potential cheating and non-compliance cases Case Relationship between x and x ˆ x> x ˆ x> x ˆ x= x ˆ x= x ˆ Relationship between x and z ˆ ˆ z=x ˆ ˆ z<x ˆ ˆ z<x ˆ ˆ z=x ˆ ˆ Are managers cheating? Yes Yes Yes No Is firm compliant with reporting requirement? Yes No No Yes 1 2 3 4 By adapting the model of Gilpatric (2005). However it remains an important question whether these effects are observed empirically and how well the model captures behavior. probability of cheating being detected. If managers are found to have “cheated” by emitting more than xi or by misreporting they are disqualified from winning the tournament (e. In what follows. Cheating is modeled as simply increasing output by a constant. number of contestants. The model developed in that paper shows how the likelihood of cheating depends on the payoffs at stake in the tournament. the variance of output. Table 13. The probability of cheating decreases as the probability of detection grows. where xi represents the optimal level of emissions for division i from the perspective of the owner-manager.F. The direction of these effects is generally quite intuitive.248 M. They are able to improve their output by increasing emissions up to a level of xi . the gain from cheating decreases. In this setting malfeasance always consists of both emitting more than is optimal for the firm and failing to truthfully report emissions (case 2 above). and the penalty associated with being found to have cheated. We make the following assumptions. after observing their opponents’ effort. Because managers face the same penalty if found to have cheated regardless of the magnitude of cheating there is no marginal deterrent and the manager’s decision reduces to choosing xi as directed by ˆ the firm or cheating by choosing xi . leaving additional discussion and experimental testing of the remaining cases to future research. or the cost of being caught cheating increases. choose whether to cheat. Evans et al.g.

a “high” distribution and a “low” distribution. not caught cheating) receives payoff w1 .Managerial incentives for compliance 249 exactly how behavior responds to changes in the competitive framework is quite valuable for understanding how competitive incentive systems can elicit effort while minimizing malfeasance and monitoring costs. his expected payoffs are identical).e. and a player caught cheating receives w2 − r. w1 − w2. Let r represent the outside penalty imposed on a player caught cheating. Here we illustrate this application of the model and show how the predicted probability of cheating is derived conditional on the underlying parameters of the model. In general player i’s probability of having the highest draw when he receives a draw from density function f(y) and faces N − 1 opponents k of whom cheat and who each receive a draw from a distribution G(y) if they do not cheat or H(y) if they do cheat is Pi = ∫ f (y )(G (y )) N −1− k (H (y ))k dy . those who do not win but are not caught cheating receive w2 . This is found by deriving the audit probability such that.6 If a player is found to have cheated he faces two possible types of sanctions: (1) a cheating player is disqualified from winning the tournament and receives the payoff associated with finishing last. Outside penalties represent such factors as a negative reputation arising from being found to have cheated. if a contestant believes his opponents will not cheat then he is indifferent between cheating and not (i. We first develop some notation.e. NC)S + w2. Finding P(C. (2) the player may face additional “outside” penalty in excess of any compensation at stake in the tournament.e. not cheating is a dominant strategy. The model predicts that cheating will be fully deterred (i.. We can now set the expected payoff from cheating equal to that from not cheating to solve for η a . Cheating entails choosing the high distribution. NC) is rather more complicated. NC) = 1/N. P(NC. Let P(.) represent player i’s probability of finishing first (but not necessarily receiving w1 since this probability does not account for the possibility of disqualification if cheating). In this symmetric contest. An “audit” of contest behavior occurs with probability η .5 Players choose a distribution of output. among two distributions. Then player i’s expected payoff if he cheats when his opponents do not is (1 − η )P(C. The contestant with the highest output who is eligible to win (i. The first argument of P denotes the action of player i and the second argument gives the action of his opponents. Let s represent the payoff spread. Contestants play only the second stage of the game in which they choose whether or not to “cheat”. a . To solve for strategies as a function of the tournament parameters we first identify the minimum probability of audit that will fully deter cheating. denoted y.. In this chapter we consider a special case of the model in which three contestants compete in a rank-order tournament. and if an audit occurs all contestants who cheated are discovered to have done so. NC )S + w2 − η r whereas player i’s expected payoff if he does not cheat when his opponents do not is P(NC. If the audit probability is greater than this value. which we will denote η a . This parameter represents the intensity of monitoring activity undertaken by the tournament sponsor.

This yields the existence of a symmetric mixed strategy equilibrium toward which behavior should converge over time if the game is played repeatedly and players update their beliefs regarding the rate of cheating among other contestants. NC . P(. which we denote ηb . as a function of the tournament parameters. C )− P(NC .F. NC )+ b(2. Evans et al. For expositional ease and consistency with our experimental application. players are indifferent between cheating and not cheating for audit probabilities in this range. If i does not cheat given each of his opponents cheat with probability ρ.2) Note that ηb < ηa. C . However. NC )+ b( .250 M.2.2. ρ)P(NC . ρ)P(NC . ρ)P(NC . player’s dominant strategy is not to cheat) if the probability of detection is at least ηa = P(C .. When player i faces N − 1 opponents the probability that k of them cheat given that each opponent cheats with probability ρ is defined by the binomial function b(k. N − 1. C ) ⎭ .2.2. NC )+ b( .. ρ ) P(NC . they do better to cheat.. C )]+ (S + r )/ S .. ρ. Note that the more a contestant believes his opponents will cheat the lower the payoff he receives from cheating and the higher the payoff from not cheating. now the first argument represents player i’s strategy and the second and third arguments denote his opponents’ respective strategies. the probability of winning (the first bracketed term) is the sum of the probabilities of winning given each possible combination of b(0.1) We can employ similar calculations to find the audit probability below which cheating is a dominant strategy. NC ) − 1 / N . ρ) P(NC . ρ) + w2 In the absence of an audit. ρ). For audit probabilities between ηa and ηb there is a unique symmetric equilibrium where each player cheats with probability ρ such that players’ expected payoffs for cheating and non-cheating are equal. C )] [P(C .2. In this context. let N = 3. C)⎫ 1 ⎬ ⎩+ b(2. In other words.2. NC ) + r / s (13. then his expected payoff is (1− η)(S)⎧ ⎨ } b 1 + η (S ){ (0. P(C . C )− P(NC . and if they conclude that they are cheating less frequently. This is the value where player i is indifferent between cheating and not cheating if he believes all his opponents will cheat. Cheating will be a dominant strategy if the probability of audit is less than ηb = [P(C . If contestants conclude that their opponents are cheating more frequently than with probability ρ they will do better not to cheat. (13. NC . We now illustrate how we solve for the equilibrium probability of cheating. NC .) continues to represent the probability that player i wins the tournament.

1982. C )− P(NC . indicating the probability of winning if an audit occurs. NC )+ b( . perhaps. We establish parallelism through ensuring that the essential features of the field environment are captured in the laboratory. ρ)[P(C . NC . It is the absence of such monitoring. ρ) P(C . 1987. ρ) [P(C . C . NC )− P(NC . Economics experiments allow us to control the parameters of the competition and observe all behavior by contestants to learn whether they respond as theory predicts. 2001). C . ρ) P(NC . NC . C )] ⎭ ⎩ ⎧b(0. with each term weighted by the probability of that occurrence. NC )+ b(2. ρ )⎭ (13. Cummings et al. A key to the use of the results of laboratory experiments to inform the policy debate is the precept of parallelism (Smith. NC . The second term in brackets. ρ) P(C .. NC . of course. Experimental design Our laboratory experiments are designed to test the responsiveness of the frequency of cheating to changes in the probability of audit and the imposition of . C )⎫ ⎬ − η r + w2 . r. ⎩+ b(2. ρ) P(NC . is that it is impossible to know for certain how much cheating takes place in any context without perfect monitoring of behavior.2. ρ) P(C .3) Solving this equation for ρ provides an expression for the equilibrium probability of cheating given values for η. NC . Plott. is similar except that cheating opponents are disqualified so the P terms are quite different (and in the final case where both opponents cheat player i wins with probability one if there is an audit).2.2. In the next section. NC . We can similarly find that the expected payoff to player i if he cheats is in this context is (1 − η)(S )⎧ ⎨ 1 b(0.2.2. Laboratory experiments Testing the theoretical model with field data is clearly problematic for a variety of reasons. ρ)[P(C . we discuss the results of experiments designed to test hypotheses that stem from the theoretical model. NC ) ⎫ η (S )⎨ ⎬ + ηr 1 ⎩+ b( .2.2. C )− P(NC . Most important of these. C )] ⎪ = ⎬ ⎨ ⎪ ⎪+ b(2.2. The experiments designed for this line of research focus on the strategic elements of the theory: we test behavioral arguments. that describes the circumstances the model seeks to capture. C .2. C ) ⎭ Setting these two expressions equal to each other (as they must be in equilibrium) and rearranging we have ⎧b(0. NC )]⎫ (1 − η)(S )⎪+ b(1. NC . and s.Managerial incentives for compliance 251 cheating and non-cheating opponents.

52]. We investigate audit probabilities that fall inside and outside the ηa and ηb ranges above.e.150 .32 give rise to unique mixed strategy equilibria. using expression (13. NC ) = ∫ ⎜ ⎟⎜ ≈ 0. we include treatments corresponding to η = 0. y ~ U[22.114.2.5. For r = 5 we include treatments corresponding to η = 0. r. If an audit occurs. The eligible (i. where the session is not designed to elicit a dominant strategy). In particular.2 there is a unique mixed strategy equilibrium whereas for η = 0.C) in hand. 22 30 ⎝ ⎠⎝ 30 ⎠ 2 With P(C.1) and (13. When r = 5 we have that ηa ≈ 0.234 and ηb ≈ 0.e.155. The unique mixed strategy .407 and ηb ≈ 0.3. C ) = ∫ ⎜ ⎟⎜ ⎟ dy ≈ 0. in a mixed strategy equilibrium where one exists (i. for r = 0. In particular. ⎟ dy + 22 30 30 ⎠ 30 ⎝ ⎠⎝ 2 In the case where i does not cheat and his opponents do we have that f(y) = 1/30 and H(y) = (y − 22)/30 for 22 ≤ y ≤ 45 and f(y) = 0 for y ≥ 45. we can apply the formulas from the theory section to obtain values for win probabilities for player i. As the audit probability 0.3) above and the parameters of each experimental session we can calculate the predicted frequency of cheating.2 and 0. The choice of a draw from the high distribution corresponds with the decision to cheat. 0. cheaters face an outside penalty. the win probability for player i given his opponents do not cheat is 45 ⎛ 1 ⎞⎛ y − 15 ⎞ 7 P(C . With η = 0. Using the experiment parameters above.45]. For the case where r = 0 we have that ηa ≈ 0. ρ. the group faces a random audit with probability η. such that the win probability for player i is 45 ⎛ 1 ⎞⎛ y − 22 ⎞ P(NC .NC) and P(NC. the winner receives a payoff of 19 lab-dollars and other participants receive 7 lab-dollars less any penalty if they are disqualified. respectively. which is equal to zero or five. 0. Evans et al. As in the model described in the previous section. non-disqualified) participant with the highest output wins the tournament and receives the highest payoff. we can solve for ηa and ηb using formulas (13. Further. In the case where i cheats and his opponents do not we have that f(y) = 1/30 and G(y) = (y − 15)/30 for 22 ≤ y < 45 and G(y) = 1 for y ≥ 45. The decision faced by the participant is whether to receive an output draw from a “low” distribution.2). Audit probabilities of 0.2 and 0.5 is greater than ηa it follows that for these parameter values there is a dominant strategy to cheat and not cheat.F.1.562 .1 is less than ηb and 0.3 there is a dominant strategy not to cheat. Participants are randomly assigned to three-player groups and play the role of division managers in a rankorder tournament. y ~ U[15. Thus. an outside penalty for managers caught cheating.252 M. and 0. for example by emitting more than permitted in order to increase productivity but falsely reporting lower emissions. cheaters are caught with a probability of one and are disqualified from the tournament.32. or a “high” distribution. respectively. Similarly.

00 1 2 3 4 5 6 0 0 0 5 5 0 equilibria are solved for using formula (13. Participants were drawn from a large pool of volunteers and represent a wide range of academic majors. As we observe.7) Payoff spread (s) 12 12 12 12 12 12 Penalty (r) Predicted prob. the experiment lasted either 20 or 30 periods. Experiment instructions are presented both orally and in writing.7. using Wilcoxon tests where the unit of measurement is cheating frequency for the individual over all decision . Repetition appears to be important here given some equilibrium predictions are predicated on mixing strategies. The decision to receive a high draw is not framed as cheating or malfeasance so as not to engender uncontrolled payoffs associated with ethical costs associated with cheating.10 0. and (6) his payoff. Decisions are made via laptop computers using software programmed in z-Tree (Fischbacher. Similarly. The design parameters and predicted cheating probabilities for the six treatments are summarized in Table 13.2. subjects were not aware of the identity of the other members of their group. (5) how many opponents were disqualified.3). (2) his output rank.2) (19. and participants were visually isolated through the use of dividers. (3) whether there was an audit.7) (19. ineligible) (19. η 0. and subjects received average compensation of approximately $15.7. Experiment results The results are summarized in Table 13.20 0. After each period. instructions use neutral language.20 0.7. The experiment lasted 30 to 60 minutes.7. Matching was anonymous. To thwart motivations for strategic play and efforts at tacit coordination.30 0.7.Managerial incentives for compliance 253 Table 13.2) (19.7. The experiments were conducted in a designated experimental economics laboratory.2 Design parameters by treatment Treatment N per contest 3 3 3 3 3 3 Audit prob.76 0. in each period participants are randomly and anonymously reassigned to tournament groups. the participant receives feedback on: (1) his output. (4) whether he was disqualified. Note that while we make analogies here to managerial decisions on environmental compliance.50 Payoffs: (Win.7) (19. we allow for possible learning through repeated play over T identical decision periods. 1999).7) (19. not win. we characterize the audit simply as a computer “check” of which distribution was chosen.32 0. the subjects do not behave exactly as the theory predicts.27 0. Sessions consisted of nine to 15 people.00 0.29 0. In particular. Due to time considerations. Although our theoretical model describes a one-shot game. of cheating (ρ) 1.00 0. The total of 96 undergraduate student subjects at the University of Tennessee participated in experiments in the summer and fall of 2005.3.

46 1. In both cases. the model controls for factors related to history of play as well as policy parameters. with η = 0.2.55 2. Overall there is a tendency toward an indifference between cheating or not. of No.036). predicted (z-statistic) 0. These variables correspond to signals of how background win probabilities change conditional on the decisions of other players. the effect of the penalty is not statistically significant. The effect of an outside penalty appears to be less pronounced. As participant behavior may be influenced by experience in prior periods. Evans et al. we find that predicted and observed cheating is statistically different for all treatments at the 5 percent significance level. cheating actually increases by 0.29 0. Nevertheless.3.74 0. However.12. prob.00 –3. we estimate the parameter covariance matrix using White’s robust “sandwich” estimator adjusted for clustering at the individual level. and to allow for possible distribution mis-specification. of subjects periods 1 2 3 4 5 6 15 18 18 18 12 15 30 20 20 20 30 30 Observed prob.54 0.10. Note that observed cheating probabilities vary very little across rounds such that the results of statistical tests do not depend on which periods are considered.00 0. and in particular the feedback received after each period.F.53 0. we include an indicator for the presence of a penalty and a variable corresponding to the audit probability.07 3. We turn now to a more formal analysis of individual behavior and estimate a probit model of the decision to cheat.94 3. and the proportion of prior “wins by cheating” whereby the participant won as a result of cheating.16 2. For instance.41 periods. for η = 0. with the exception of treatment 2. = 0. with observed rates of cheating below the predicted level when the theory predicts cheating the majority of the time (treatments 1 and 2) and observed rates of cheating above the predicted level when the theory predicts cheating a minority of the time (treatments 3 to 6).76 0. To account for unobserved subject heterogeneity. the penalty decreases cheating by 0.26 –1. Feedback variables include the proportion of prior “wins by opponent disqualification” whereby the participant won only as the result of competitors with higher output being disqualified. In terms of policy variables. our results are generally supportive of the theory as it predicts responses to changes in the audit probability.3 Observed cheating in experiments Treatment No. For example.42 0. This difference is statistically significant using a two-sample Wilcoxon Test (z = 2.1.27 0. Predicted Prob.63 0.254 M. actual cheating drops from 63 percent (treatment 2) to 42 percent (treatment 3) when the audit probability increases from 20 percent to 32 percent. Table 13. Wilcoxon Test: of cheating of cheating observed vs. As the effects of the exogenous audit probability and the two subjective win probabilities may have more pronounced .00 0.

4 Probit model results Variable Penalty Audit probability Audit in previous period Proportion of wins by opponent Disqualification in prior periods Win by opponent disqualification in previous period Proportion of wins by cheating in prior periods Win by cheating in previous period Constant Wald χ2 (7 d. and whether the subject won by cheating in the previous period. the presence of which at least partially explains why observed cheating is lower than predicted for low audit probabilities and higher than predicted for high audit probabilities.118 2244 Marginal effect (robust standard error) –0.4 presents our estimated probit coefficients and corresponding marginal effects.5.036) Note An asterisk indicates the parameter is statistically different from zero at the 5 percent level. which is about 0. the model suggests an effect inconsistent with theory: he is more likely to cheat in the current period. consider the mean audit probability across all non-penalty treatments.092) 0.152) 2. .116) –1. However.039) –0.107) –1. As predicted.185* (0. according to the estimated marginal effect. whether the subject won by disqualification. the model suggests that observed cheating would only increase (decrease) by 0. In particular.160) 0.127) 0.612* (0. Table 13. we also include three indicator variables corresponding to whether there was an audit.131 (0.409) 0.993* (0.056) 0.162) 131. the marginal effect suggests that changes in the audit probability has an effect on cheating less pronounced than predicted by theory.758) –0.185.28 with an associated cheating probability of roughly 0. the presence (absence) of an audit in the previous period increases (decreases) the probability of cheating by 0.06 for such a change in the audit probability. Table 13. If an individual was audited in the previous period.785* (0.20* 0. the presence of the penalty has no statistically significant effect on the cheating probability.051 (0.046) –0.13. This is the oft observed “gambler’s fallacy” behavior.851* (0.485* (0. Consistent with our non-parametric test results. However.) Pseudo R2 Number of observations Coefficient (robust standard error) –0. This result is surprising.Managerial incentives for compliance 255 short-term effects.299) –0..115 (0.321) 0.5. participants respond to the higher audit probability by reducing their probability of cheating.488* (0.160* (0.239* (0. In particular. but has a parallel in the law and economics literature where some studies find that increased penalties for criminal offenses (such as the death penalty) have little or no deterrent effect on crime rates (Katz et al. the theory predicts a decrease (increase) in cheating by about 0.109 (0.045 (0. 2003). If the audit probability is increased (decreased) by about 0.f.239* (0.

that of financial disclosure. such as through independent auditing. namely features of their organizational structures that induce greater non-compliance with environmental regulations and information disclosure programs. As these ceteris paribus interpretations are possibly confounded by the presence of the lagged indicator variables associated with the two subjective probability variables. As Alm and McKee (1998) have argued we can . Additionally.1 increase in the proportion of wins by disqualification the model estimates that cheating decreases by 0. a 0. To a large extent financial disclosure requirements are implemented through the requirement that publicly traded firms are subject to an independent audit by an outside auditor. The similarities of the underlying decision structure between compliance with corporate tax regulations and environmental reporting regulations suggests that we can gain some insight on how to design appropriate audits from the tax compliance literature. Nevertheless.256 M. may be an effective means of increasing compliance with information disclosure and other regulatory requirements. There is nearly a one-to-one relationship. One might also draw an analogy with enforcement of another type of information disclosure mandate. In particular. Our results can inform the debate on the efficient design of auditing procedures for verification of the information reported by firms as required under a mandatory disclosure program. Clearly this system is imperfect. Similarly. it remains true that mandating credible monitoring of internal firm behavior. Implications and extensions This chapter embarks on preliminary steps towards improving the design and implementation of environmental information disclosure programs. By examining manager-level emissions and compliance decisions. and others make clear. Audits of behavior by the regulators (i. we obtain predictions of firm characteristics. we note that the changes in cheating become 0.1 increase in the proportion of wins by cheating corresponds with an increase in cheating by 0. WorldCom. One implication of our model is that the optimal intensity of regulatory enforcement efforts depends on the magnitude of monitoring and enforcement within firms.1 and 0. merit greater regulatory scrutiny than those that more intensively monitor internal behavior. which is quite rational. the proportion of wins by disqualification and proportion of wins by cheating statistically decreases and increases. the SEC) are very infrequent.F. between changes in these subjective win probabilities and changes in cheating probabilities.08. and which have little internal monitoring of managerial behavior. Finally. the probability of cheating. respectively. we note that a participant who wins by disqualification is even more likely not to cheat in the following period.09. Evans et al. for a 0. as the recent high-profile accounting scandals involving Enron. Firms with managerial compensation systems that generate strong competitive incentives for cheating despite firm-level compliance with disclosure requirements and other regulatory mandates.e.09 when the indicator variables are excluded from the model.

Specifically. 2004. Chen and Chu. 2007. one can improve the efficiency of the audit process through the use of systematic or endogenous audits (selecting firms based on observable characteristics). 1999) argue that compliance in an emissions trading environment is independent of firm characteristics. 1992. The structure of the internal organization. Stranlund and co-authors (Murphy and Stranlund. For these programs. If the results of this literature apply to the setting of compliance with information disclosure programs. In this way. First.g. For many publicly held firms the general form of the compensation structure will be public information as will be the divisional structure. the owner-manager could benefit from releases that lower cost of production if the releases are reported to the public with a sufficient lag. will affect the propensity to emit and to under report. Our discussion of the effects of rank-order tournament compensation schemes on managerial reporting incentives provides some simple insights for the design of an auditing program. However. our work suggests that the method of compensation of divisional leaders and the number and size of divisions will affect the firm’s overall level of compliance. Alm and McKee. the emissions trading environment differs from simple information disclosure environments. In the case of emissions trading. Of particular relevance to the information disclosure programs is the lag in the audit process. 2005) has shown that individuals and firms respond in predictable ways to the elements of audit regimes.. 1998. our work suggests that there are systematic links between the organizational structure of the firm and its overall environmental malfeasance and reporting behavior.g. in particular the managerial incentives. considerable research in tax compliance behavior (e. Alm et al. then we would expect increased enforcement effort (such as the use of penalties and random audits) to increase compliance. 2006. evasion costs and benefits do differ across firms at the margin and these differences may be reflected in observable firm characteristics as suggested by our theoretical development and experimental results. Stranlund and Dhanda. 2004. targeted enforcement does not enhance efficiency because the market for permits yields an equilibrium price such that there are no differential incentives to evade. . The reporting requirements under the various environmental regulations are applied at the firm level. 1993.Managerial incentives for compliance 257 learn a great deal about managerial decisions (especially regulatory compliance decisions) from the extensive research work on tax compliance. A sufficiently long lag may allow the ownermanager to realize his payoff from the assets owned and exit the firm prior to the release of the information and the subsequent negative effect on the firm’s value. However. Second. This suggests that the reporting period should be shortened and audit resources optimized through the use of staggered reporting dates. This finding suggests that systematic audit rules will not be productive. division) level. Our analytical framework examines behavior at the sub-firm (e. Even if compliance with the reporting requirement is perfect (the firm reports exactly what is released). No such market occurs in response to mandatory information disclosure programs. the information concerning emissions would be provided to the market in a timely fashion and the anticipated effects on share values realized quickly.

Our research is directed to improving the performance of information disclosure programs through both the identification of firm characteristics that are more likely to be correlated with environmental malfeasance and incomplete information disclosure. 2 Nalebuff and Stiglitz (1983) term this issue the influence of prize on choice of technique. The former investigations will suggest ways the audit regimes can be improved while the latter will suggest design elements of the information disclosure program. non-compliance with reporting requirements could result in damages that are not easily reversed. Information disclosure programs. In the case of noncompliance with environmental reporting requirements. 4 Internal environmental auditing procedures may increase the validity of this assumption to the extent that the existence of internal records of division managers’ reports discourages the owner-manager from choosing to report a level of emissions inconsistent with . ex post actions will not likely make the public whole. David Bruner programmed the experiments. Stiglitz and Weiss (1981) address a related problem of the influence of the interest rate in bank lending on the risk involved in projects undertaken by borrowers. Notes 1 Of course the market will anticipate positive emissions levels in many cases and the reaction of the market will depend on the extent to which reported emissions differ from expectations. as well as the identification of the properties of information disclosure programs that enhance compliance.258 M. In the case of income tax evasion. The extent to which this potential is realized depends on the extent to which the information is accurate and timely. the evader can make the government whole through the payment of back taxes and interest. However. Evans et al. such as the TRI. The government may also argue for the imposition of additional fines given the incomplete detection and punishment regimes (much like punitive damages in tort litigation).F. 3 Of course the cost of monitoring the regulatory compliance behavior of managers can be thought of as simply one aspect of the total cost of regulatory compliance for the firm. All firms will wish to report only information that casts them in a favorable light and must be “encouraged” to provide truthful and timely information. Our point here is to separate the costs of implementing full compliance within the managerial incentive system from the direct costs of regulatory compliance (such as costs associated with using “cleaner” production technology). in some cases. An important distinction between tax compliance and compliance with information reporting requirements is that. have the potential to achieve significant improvements in the environmental behavior of firms. the liability system may apply additional penalties on firms that have violated the regulatory standards and failed to comply with reporting requirements. but do not model the problem. Acknowledgments We thank participants at Appalachian State University’s Experimental Economics Workshop and the 2005 ESA meetings in Montreal for comments on earlier versions of this research.

Managerial incentives for compliance 259
these reports. Anton et al. (2004) suggest that elements of the internal organization of firms’ environmental programs are important in explaining TRI emissions. 5 Gilpatric (2005) finds that in equilibrium all players choose identical effort in the first stage of the tournament and therefore play a symmetric cheating game in the second stage. Our focus here is on testing predicted behavior in this symmetric cheating game. Clearly players may not choose identical effort levels, and other circumstances may well occur that render contestants unequal when choosing whether to cheat, but we leave the study of behavior arising in such a setting to future research. 6 Gilpatric (2005) discusses how behavior differs when audits are independent and shows that correlated audits as discussed here (in which all players are audited or none are) more effectively deter cheating than independent audits of equal probability.

References
Alm, J. and M. McKee, 1998. Extending the Lessons of Laboratory Experiments on Tax Compliance to Managerial and Decision Economics. Managerial and Decision Economics, 19 (4–5), 259–275. Alm, J. and M. McKee, 2004. Tax Compliance as a Coordination Game. Journal of Economic Behavior and Organization, 54 (3), 297–312. Alm, J., B.R. Jackson, and M. McKee, 1992. Institutional Uncertainty and Taxpayer Compliance. American Economic Review, 82 (4), 1018–1026. Alm, J., M.B. Cronshaw, and M. McKee, 1993. Tax Compliance with Endogenous Audit Selection Rules. Kyklos, 46 (1), 27–45. Alm, J., B.R. Jackson, and M. McKee, 2004. Audit Information Dissemination, Taxpayer Communication, and Tax Compliance: An Experimental Investigation of Indirect Audit Effects. Presented at the 2004 ESA meetings in Tucson, AZ. Anton, W.R.Q., G. Deltas, and M. Khanna, 2004. Incentives for Environmental SelfRegulation and Implications for Environmental Performance. Journal of Environmental Economics and Management, 48 (1), 632–654. Beard, T.R., 1990. Bankruptcy and Care Choice. RAND Journal of Economics, 21 (4), 626–634. Brehm, J. and J.T. Hamilton, 1996. Non-compliance in Environmental Reporting: Are Violators Ignorant, or Evasive, of the Law? American Journal of Political Science, 40 (2), 444–477. Chen, K., 2003. Sabotage in Promotion Tournaments. Journal of Law, Economics, and Organization, 19 (1), 119–140. Chen, K.-P. and C.Y.C. Chu, 2005. Internal Control versus External Manipulation: A Model of Corporate Tax Evasion. RAND Journal of Economics, 36 (1), 151–164. Cummings, R., M. McKee, and L. Taylor, 2001. To Whisper in the Ears of Princes: Laboratory Economic Experiments and Environmental Policy. In: H. Folmer, H.L. Gabel, S. Gerking, and A. Rose, eds., Frontiers of Environmental Economics. Northampton, MA: Edward Elgar, pp. 121–147. Fischbacher, U. 1999. Toolbox for Readymade Economic Experiments. Technical Report IEW Working Paper 21, University of Zurich. GAO (General Accounting Office), 1991. Toxic Chemicals: EPA’s Toxics Release Inventory is Useful but Can be Improved. Washington: General Accounting Office. Gilpatric, S., 2005. Malfeasance in Tournaments. Working Paper, University of Tennessee, Department of Economics.

260

M.F. Evans et al.

Hamilton, J.T., 1995. Pollution as News: Media and Stock Market Reactions to the Toxics Release Inventory Data. Journal of Environmental Economics and Management, 28 (1), 98–113. Harrington, W., 1988. Enforcement Leverage When Penalties Are Restricted. Journal of Public Economics, 37 (1), 29–53. Helland, E., 1998. The Enforcement of Pollution Control Laws: Inspections, Violations, and Self-Reporting. Review of Economics and Statistics, 80 (1), 141–153. Katz, L., S.D. Levitt, and E. Shustorovich, 2003. Prison Conditions, Capital Punishment, and Deterrence. American Law and Economics Review, 5 (2), 318–343. Khanna, M., W.R.H. Quimio, and D. Bojilova, 1998. Toxics Release Information: A Policy Tool for Environmental Protection. Journal of Environmental Economics and Management, 36 (3), 243–266. Kim, S., C. Qin, and Y. Yu, 2002. Bribery in Rank-Order Tournaments. Working Paper, University of California Santa Barbara, Department of Economics. Konar, S. and M.A. Cohen, 1997. Information as Regulation: The Effect of Community Right to Know Laws on Toxic Emissions. Journal of Environmental Economics and Management, 32 (1), 109–124. Konar, S. and M.A. Cohen, 2001. Does the Market Value Environmental Performance? Review of Economics and Statistics, 83 (2): 281–289. Larson, B., 1996. Environmental Policy Based on Strict Liability: Implications of Uncertainty and Bankruptcy. Land Economics, 72 (1), 33–42. Lazear, E.P. and S. Rosen, 1981. Rank-Order Tournaments as Optimum Labor Contracts. Journal of Political Economy, 89 (5): 841–164. Milgrom, P. and J. Roberts, 1988. An Economic Approach to Influence Activities in Organizations. American Journal of Sociology, 94, S154–S179. Murphy, J.J. and J.K. Stranlund, 2006. Direct and Market Effects of Enforcing Emissions Trading Programs: An Experimental Analysis. Journal of Economic Behavior and Organization, 61 (2), 217–233. Murphy, J.J. and J.K. Stranlund, 2007. A Laboratory Investigation of Compliance Behavior Under Tradable Emissions Rights: Implications for Targeted Enforcement. Journal of Environmental Economics and Management, 53 (2), 196–212. Nalebuff, B. and J. Stiglitz, 1983. Prices and Incentives: Towards a General Theory of Compensation and Competition. Bell Journal of Economics, 14 (1), 21–43. Plott, C.R., 1987. Dimensions of Parallelism: Some Policy Applications of Experimental Methods. In: A.E. Roth, ed., Laboratory Experimentation in Economics. Cambridge: Cambridge University Press, pp. 193–219. Prendergast, C. and R. Topel, 1996. Favoritism in Organizations. Journal of Political Economy, 104(5), 958–978. Shavell, S., 1984. A Model of the Optimal Use of Liability and Safety Regulation. RAND Journal of Economics, 15(summer), 271–280. Smith, V.L., 1982. Microeconomic Systems as an Experimental Science. American Economic Review, 72(5), 923–955. Stiglitz, J. and A. Weiss, 1981. Credit Rationing in Markets with Imperfect Information, Part I. American Economic Review, 71 (3), 393–410. Stranlund, J.K. and K.K. Dhanda, 1999. Endogenous Monitoring and Enforcement of a Transferable Emissions Permit System. Journal of Environmental Economics and Management, 38 (3), 267–282. US EPA, 2001. The United States Experience with Economic Incentives for Protecting the Environment, EPA-240-R-01–001. Washington, DC: US EPA.

14 An investigation of voluntary discovery and disclosure of environmental violations using laboratory experiments
James J. Murphy and John K. Stranlund

Introduction
State and federal self-discovery and disclosure rules seek to encourage greater compliance with environmental regulations by reducing penalties for violations that are voluntarily discovered and reported to authorities. For example, the EPA’s Audit Policy reduces penalties “for regulated entities that voluntarily discover, promptly disclose, and expeditiously correct noncompliance.”1 Concurrent with the implementation of rules for voluntary discovery and disclosure of environmental violations over the last decade or so, a significant body of literature emerged that examines the conceptual properties of these rules (e.g., Malik, 1993; Kaplow and Shavell, 1994; Innes, 1999, 2001a, 2001b; Pfaff and Sanchirico, 2000). Taken as a whole this literature is noncommittal on the question of whether voluntary disclosure policies are worthwhile complements to conventional enforcement strategies. In fact, provided that the predictions about the performance of voluntary disclosure policies hold up under empirical scrutiny, it is clear that whether these schemes are worthwhile will depend upon the specifics of particular regulatory settings. Unfortunately, empirical analyses of the performance of voluntary disclosure policies are limited to just a few examinations of the effects of existing state and federal discovery and disclosure rules. For example, Stafford (2005) finds evidence that the EPA’s Audit Policy and state audit policies have had a positive effect on compliance among hazardous waste facilities. Pfaff and Sanchirico (2004) examine the effects of the Audit Policy on the number and form of selfdisclosed violations and find that the policy has encouraged self-discovery and disclosure of violations, but these reported violations are minor in comparison with the violations uncovered by conventional EPA audits. While econometric studies with field data are critical for understanding the effectiveness of existing policies, data limitations and the inability to vary these policies in a controlled setting can preclude direct tests of theoretical predictions. Moreover, experiments provide direct control over the parameters of interest, which allows researchers to perform sensitivity analyses that may not be possible outside of the laboratory. Therefore, in this chapter we report the results

262

J.J. Murphy and J.K. Stranlund

of a series of experiments designed to test fundamental hypotheses about the performance of voluntary discovery and disclosure policies. In particular, we address the following questions: How well do these policies perform in terms of motivating firms to voluntarily investigate whether they are in violation of an environmental standard and to disclose any violations they discover? How do voluntary discovery and disclosure policies affect the care that firms exercise to prevent environmental violations? Relative to conventional enforcement strategies, what are the effects of these policies on enforcement effort and environmental quality? We designed and conducted a series of experiments with seven treatments. Using a within-subject design, each subject participated once in each of the seven treatments. All experiments began with a conventional enforcement model as the baseline treatment. In this treatment subjects were responsible for making a costly decision about the level of care taken to reduce the likelihood of a violation occurring. Subjects did not incur any costs if a violation occurred; however, they were audited with a known, exogenous probability, and they were penalized if a violation was discovered. The elements of the conventional enforcement treatment were contained in the other six treatments. Each of the other treatments gave subjects the opportunity to voluntarily disclose their violations under different conditions. These treatments varied according to the penalty for voluntarily disclosed violations and whether it was costly for the subjects to determine their compliance status. Subjects responded strongly to the disclosure incentive. In each of our treatments involving an opportunity for voluntary disclosure, a significant number of subjects chose to disclose. As expected, the number of disclosers tended to fall as the automatic penalty for reported violations was increased. The policy significance of inducing a significant number of voluntary disclosures is well known – relative to a conventional enforcement strategy, the government can reduce the effort it expends to detect violations because it can focus these efforts on the subset of firms that do not disclose a violation (Malik, 1993; Kaplow and Shavell, 1994). However, reducing the penalty for disclosed violations to motivate more selfreporting also reduced the care that the subjects took to avoid these violations. Thus, we find strong evidence of a tradeoff between increased violation disclosures and reduced environmental quality. This does not mean, however, that every disclosure policy results in lower environmental quality. In fact, under the condition that subjects did not have to pay to discover their compliance status, we find that it is possible to induce a significant number of violation disclosures without affecting the deterrence of a conventional enforcement strategy (Kaplow and Shavell, 1994). However, attempting to induce increasing numbers of voluntary disclosures will at some point result in less deterrence relative to a conventional enforcement strategy and, hence, will eventually lead to reduced environmental quality. We also find strong support for a hypothesis of Malik (1993) that, relative to conventional enforcement, disclosure polices will result in more violations being

Voluntary discovery and disclosure 263 sanctioned, but fewer of these sanctions are for violations that are uncovered by the government. Sanctioning violations is likely to be costly. If the costs of sanctioning voluntarily disclosed violations are roughly equal to the costs of sanctioning violations that the government uncovers, then voluntary disclosure policies will tend to increase sanctioning costs. However, because fewer sanctions are applied to violations that the government uncovers, Malik (1993) argues that a voluntary disclosure policy may decrease sanctioning costs, in spite of the increase in the number of sanctions applied, because punishing disclosed violations is probably less expensive than punishing violations that the government uncovers. A firm that voluntarily discloses a violation is essentially admitting liability for being noncompliant. This admission can reduce the burden on the government to produce sufficient evidence for a finding of liability. Moreover, a firm that voluntarily admits liability is less likely to engage in costly efforts to challenge or otherwise avoid the imposition of a penalty.2 Although our results are largely consistent with the qualitative predictions of the existing theory regarding the role of voluntary disclosure in regulatory enforcement, we do observe one unanticipated effect. For each disclosure policy we examined, subjects who chose not to disclose their violations when given the opportunity to do so tended to exercise more care to avoid violations than under a conventional enforcement strategy. This is unexpected because a subject who chooses not to disclose a violation opts to face the identical random monitoring and penalty as under conventional enforcement. That the addition of a voluntary disclosure policy to a conventional enforcement strategy tended to induce nonreporters to exercise more care likely suggests a framing effect associated with the opportunity to voluntarily disclose violations. It is important to ask whether this effect is simply an artifact of the laboratory setting, or if there is some reason to believe that regulated firms would likely behave in this way. Since we see no reason to expect that this framing effect would hold in non-laboratory regulatory settings, our view is that it is probably limited to the laboratory. Despite this framing effect, our work provides strong empirical evidence of the fundamental tradeoffs inherent in voluntary discovery and disclosure policies. Thus, the policy significance of our work is clear. Both the theoretical underpinnings of this work and our experimental tests make it clear that any conclusions about the relative benefits and costs of voluntary disclosure policies will require detailed knowledge of monitoring costs, sanctioning costs, the harm caused by environmental violations, and firms’ costs of internal audits to determine their compliance status. Therefore, it is likely that the question of whether disclosure policies are a worthwhile complement to regular environmental enforcement will have to be answered on a case-by-case basis.

Theory and hypotheses
The theoretical underpinnings of our study are drawn from a simple model of an industry composed of n identical risk neutral firms.3 Each firm chooses a level of care to reduce the probability, p, of a violation of an environmental standard.

264 J.J. Murphy and J.K. Stranlund Each has a profit function v(p), with v' (p) > 0, and v"(p) < 0. Under conventional enforcement of the standard, firms do not have an opportunity to disclose their violations. The government randomly audits a subset of firms so that the probability that any firm will be audited is π. Uncovered violations are punished with a monetary penalty φ. A risk neutral firm chooses the probability that a violation occurs to maximize its expected profit, V(p, πφ ) = v(p) – pπφ . The interior choice of the probability of a violation is p *(πφ ), which is the implicit solution to v'(p) – πφ = 0. The main hypotheses of our work are devoted to examining the behavior of firms when voluntary disclosure rules are added to an existing conventional enforcement strategy, and the policy implications that flow from these behavioral hypotheses.4 All of the behavioral hypotheses are focused on the reporting and care decisions of firms under disclosure rules that would leave risk neutral firms indifferent between voluntarily disclosing their violations and choosing instead to face the random monitoring and penalty of conventional enforcement. These disclosure rules provide useful benchmark policies from which we can derive the relative merits of voluntary disclosure policies. Moreover, we conducted additional experiments to provide sensitivity analyses around these benchmarks. An obvious starting point is to ask whether a disclosure policy can motivate noncompliant firms to report voluntarily their violations to the government, and whether this has any effect on deterrence. Suppose that firms are given the opportunity to disclose voluntarily their violations to the government, and those that do so are penalized φd < φ automatically (the subscript d indicates voluntary disclosure). Under the common theoretical assumption that firms always disclose their violations when they are indifferent about doing so, a risk neutral firm will disclose a violation if and only if the automatic penalty for a disclosed violation does not exceed the expected penalty it faces if it fails to report the violation; that is, disclosure occurs if and only if φd ≤ πφ. From a theoretical perspective, setting φd = πφ so that risk neutral firms are indifferent between disclosing their violations and not doing so implies that all violations will be disclosed. Of course, in a laboratory setting we do not expect that indifferent subjects will always choose to disclose, nor do we expect that all subjects are risk neutral. Nevertheless, we test the following hypothesis: Hypothesis 1: Any voluntary disclosure policy that leaves risk neutral firms indifferent between disclosing their violations and not disclosing them will motivate a significant number of voluntary disclosures. When risk neutral firms have costless and perfect information about their compliance status, setting φd = πφ yields the care choice p *(φd) = p *(πφ), which implies no effect on deterrence. Therefore, we have: Hypothesis 2: Suppose that firms know their compliance status without a costly self-audit. A voluntary disclosure policy that leaves risk neutral firms indifferent

Voluntary discovery and disclosure 265 between disclosing their violations and not disclosing them will not change the care that disclosers take to avoid violations. However, complex regulations and production technologies may make it difficult for firms, particularly large firms, to determine whether they are in compliance with environmental standards without undertaking a costly self-audit of their operations (Pfaff and Sanchirico, 2000). With costly discovery, Innes (2001b) has shown that inducing voluntary discovery and disclosure of violations requires that the certain penalty for disclosed violations must be reduced to compensate firms for their discovery costs. However, doing so weakens deterrence. To demonstrate this result, suppose that a firm incurs a cost c to discover whether it has violated the standard. If the firm does not invest in discovery, then its expected payoff is the same as under conventional enforcement, that is, V(πφ) = v(p *(πφ)) – p *(πφ)πφ. However, if the firm has invested in discovery and has discovered a violation, then the firm will report the violation if φd ≤ πφ. Assuming that this holds, a firm’s choice of violation probability is p *(φd) if it invests in self-discovery and its expected payoff is V(φd) – c, where V(φd) = v(p *(φd)) – p *(φd)φd . Clearly, the firm is indifferent to discovery and disclosure if V(φd) – c = V(πφ). Since this requires V(φd) > V(πφ), the penalty for disclosed violations must be strictly lower than the expected penalty under conventional enforcement, thereby weakening deterrence. Consequently, p *(φd) < p *(πφ), and we have: Hypothesis 3: If firms must incur a cost to discover their compliance status, then a voluntary disclosure policy that leaves risk neutral firms indifferent between voluntarily discovering and disclosing their violations and facing a conventional enforcement strategy will motivate disclosers to decrease the care they take to avoid violations. Our final behavioral hypothesis deals with firms that choose not to disclose their violations. Obviously, firms that choose not to disclose their violations are simply choosing to face the conventional enforcement strategy. Therefore, the addition of a voluntary disclosure rule to a conventional enforcement strategy should have no effect on the care that non-disclosers take to avoid violations. Of course, this is true of any voluntary disclosure rule, leading to: Hypothesis 4: Add any voluntary disclosure rule to a conventional enforcement strategy. Those firms that choose not to report their violations will not change the care they take to avoid violations. From hypotheses 1 through 4 follow several important policy implications that reveal the relative merits of voluntary disclosure policies. First, if Hypotheses 2 and 4 hold, Kaplow and Shavell (1994) have shown: Policy Implication 1: If firms know their compliance status without a costly selfaudit, then a voluntary disclosure policy that leaves risk neutral firms indifferent

266

J.J. Murphy and J.K. Stranlund

between disclosing their violations and not disclosing them will not change the expected number of violations. But if discovery is costly and a disclosure rule motivates a significant number of firms to report their violations (Hypothesis 1), and if Hypotheses 3 and 4 hold, then we have the following result due to Innes (2001b): Policy Implication 2: If firms must incur a cost to discover their compliance status, then a voluntary disclosure policy that leaves risk neutral firms indifferent to voluntary discovery and disclosure will increase the expected number of violations. While a disclosure rule may or may not reduce deterrence depending on whether firms must undertake a costly self-audit to determine their compliance status, these rules will always allow the government to reduce its monitoring effort if they motivate a significant number of voluntary disclosures (Malik, 1993; Kaplow and Shavell, 1994). Since the government does not need to audit those that disclose their violations, it can focus its monitoring effort on the subset of firms that do not report a violation. Clearly, maintaining the same level of deterrence for those who do not report a violation requires fewer audits. This result holds regardless of whether firms must conduct a costly self-audit. Thus, if a significant number of firms are motivated to disclose their violations (Hypothesis 1), then we have: Policy Implication 3: A voluntary disclosure policy that leaves risk neutral firms indifferent to disclosing their violations will reduce the number of audits that are required to maintain the same level of deterrence for those that do not disclose their violations. While we expect that adding a voluntary disclosure policy to a conventional enforcement strategy will allow the government to reduce its monitoring effort, we also expect that more violations will be sanctioned. If firms do not have to pay to determine their compliance status, then Policy Implication 1 asserts that the expected number of violations will be unchanged. However, the voluntarily disclosed violations are sanctioned with certainty, whereas without the disclosure opportunity these violations would only be sanctioned with the probability of an audit. Thus, if a significant number of violations are disclosed (Hypothesis 1) when firms know their compliance status without cost, then a voluntary disclosure policy that leaves risk neutral firms indifferent to disclosing their violations and not doing so will increase the total number of sanctioned violations. This move toward more sanctions is reinforced when firms must audit themselves to determine their compliance status, simply because deterrence is weaker and the expected number of violations increases (Policy Implication 2). Thus, we have: Policy Implication 4: Any voluntary disclosure policy that leaves risk neutral firms indifferent about disclosing their violations and not doing so will increase the expected number of total sanctions.

Voluntary discovery and disclosure 267 Since sanctioning noncompliant firms is likely to be costly, it is possible that a voluntary disclosure rule could increase enforcement costs if the additional costs of sanctioning a higher number of violations outweigh the reduction in monitoring costs (Kaplow and Shavell, 1994). However, the effect on sanctioning costs is complicated by the fact that voluntary disclosure policies induce a shift from penalizing violations that are uncovered by the government to penalizing voluntarily disclosed violations. Malik (1993) argues that punishing violations that are voluntarily disclosed is probably cheaper than punishing violations that the government uncovers, because punishments for disclosed violations require less evidence and are less likely to be challenged. Although more violations are punished, fewer sanctions are levied for violations that are uncovered by the government. This is a simple consequence of Hypotheses 1 and 4 – firms that do not disclose their violations choose the same level of care to prevent their violations as under conventional enforcement (Hypothesis 4), but there are fewer violations that the government uncovers because only a subset of firms choose not to disclose their violations (Hypothesis 1), hence: Policy Implication 5: A voluntary disclosure policy that leaves risk neutral firms indifferent to disclosing their violations will reduce the expected number of sanctions that are levied on undisclosed violations. Before we move on to our experiments it is worth saying a few words about risk preferences. The theory we present, the hypotheses we test, and the implications that follow from these hypotheses are all based on the assumption that agents are risk neutral. We refrain from developing and testing a theory with more flexible risk preferences because the main motivation of this work is to test hypotheses from the existing theory of self-reporting in law enforcement. To our knowledge there is no theoretical work in this area that allows for flexible risk preferences. Thus, developing new theory to account for non-neutral risk preferences is beyond the scope of this chapter. Despite its limitations, the existing theory of voluntary disclosure policies with risk neutral firms provides a useful benchmark from which to judge the qualitative effects of disclosure policies on the variables of real policy significance, that is, the effects of these policies on overall deterrence and government enforcement efforts. Nevertheless, we do believe that all experimental studies that examine compliance behavior in various settings could benefit from information about subjects’ risk preferences.5

Experimental design
Our experiments were designed to test the hypotheses and policy implications presented in the previous section, and were conducted in a computer laboratory using software specifically developed for this research. In all treatments, subjects were responsible for making a production decision that yielded earnings, v. When they produced, there was a probability, p, that a violation would occur.6 Subjects could reduce the likelihood of a violation, but this was costly in terms

268 J.J. Murphy and J.K. Stranlund
Table 14.1 Experimental design Conventional enforcement CE Voluntary disclosure only D-H ($2.35) D-I ($1.50) D-L ($0.97) Voluntary disclosure with costly discovery CD-H ($1.50) CD-I ($0.97) CD-L ($0.60)

Note The conventional enforcement penalty in all treatments is $2.50. The reduced penalty, d, for voluntary disclosure is shown in parentheses.

of foregone production earnings, in particular, v(p) = 3.60 – [0.55/(0.30 + p)]. The computer screen presented each subject with a table that displayed all the possible violation probability/production earning combinations in 0.05 increments between 0.05 and 0.95. Table 14.1 summarizes the experimental design. The conventional enforcement treatment (CE) formed the baseline; the remaining six treatments built upon CE such that all features of CE were common throughout the experiment. Under the CE treatment, each subject knew that they would be audited with probability π = 0.6. If a violation occurred and was uncovered by an audit, then the subject incurred a fine of φ = $2.50. Subjects in this treatment did not have an opportunity to disclose voluntarily their violations. The middle column of Table 14.1 contains our voluntary disclosure only treatments: D-H, D-I, and D-L. In these treatments, subjects knew automatically and without cost whether a violation occurred. These treatments were identical to the CE treatment, except that subjects had the option to disclose voluntarily whether a violation occurred. If a subject chose not to disclose a violation, then she faced the identical enforcement strategy as CE (0.6 audit probability, $2.50 fine). However, if she chose to voluntarily disclose a violation, then she automatically paid a reduced fine, φd. The level of this fine is the distinguishing factor among the D-H, D-I, and D-L treatments and is shown in parentheses next to the treatment labels in Table 14.1. In treatment D-I, the automatic penalty for a voluntarily disclosed violation (φd = $1.50) was set such that a risk neutral subject would be indifferent between disclosing a violation if one occurred and facing the uncertainty of the conventional enforcement strategy. Note that this penalty equals the expected penalty under conventional enforcement; that is, πφ =φd = $1.50. To examine the responsiveness of the subjects to the voluntary disclosure incentive, we chose a higher disclosure penalty of $2.35 for the D-H treatment and a lower disclosure penalty of $0.97 for the D-L treatment. The final column of Table 14.1 contains our voluntary disclosure with costly discovery treatments: CD-H, CD-I, and CD-L. These treatments were the same as the voluntary disclosure only treatments, except that subjects did not know whether a violation occurred unless they paid $0.20 to find out. Those who chose not to pay the cost of self-discovery could not voluntarily disclose a

Notice that this automatic penalty for a disclosed violation in the CD-I treatment is lower than the expected penalty under conventional enforcement (πφ = $1. Subjects were given a copy of the instructions that the experimenter then read aloud. and were then given an opportunity to earn additional money in the experiment. . Amherst. and each treatment precedes and follows every other treatment one time. As with the voluntary disclosure only treatments. For the six treatments that included the option to voluntarily disclose a violation. we used the strategy method to ensure that we had an observation for each subject’s disclosure decision regardless of whether a violation occurred. this disclosure penalty makes a risk neutral subject indifferent between discovering and disclosing a violation.2. The parameters for the practice and real rounds were the same. This is necessary to motivate subjects to invest in self-discovery.7 The experimenter used an overhead projector to demonstrate the software while the subjects performed the same tasks on their individual computers. Every subject participated in all seven treatments. 30 subjects participated in each of the six sequences. since the disclosure decision was not costly. a column in Table 14. Like the voluntary disclosure only treatments.2). Each experiment lasted about an hour and a half.97.55 and $18.e. The sequences of treatments are provided in Table 14. Within a sequence (i. The Latin Square was constructed such that each treatment appears once in each sequence.e.27. we chose a higher disclosure penalty for CD-H and a lower penalty for CD-L to examine the responsiveness of subjects to the disclosure incentive. each subjected generated seven observations. once in each stage (i. It took about 30 minutes to complete the instructions and answer any questions.Voluntary discovery and disclosure 269 violation and therefore faced the identical enforcement strategy as CE. Earnings were paid in cash at the end of each experiment. Subjects were paid $5 for agreeing to participate and showing up on time. Conceptually. the automatic penalty for a disclosed violation was φd = $0. a row in Table 14. the costly discovery treatments varied according to the penalty for disclosed violations. with a mean of $14..88 (σ = 1. one for each treatment.2) there were seven stages.. A total of 180 students were recruited from the student population at the University of Massachusetts.49). data from the practice rounds were discarded. followed by one “real money” round. or face the uncertainty of random audits and potential penalties under conventional enforcement. forcing subjects to commit to this decision at the outset should not affect their behavior.50). In treatment CD-I. starting with conventional enforcement as the baseline. These additional earnings ranged between $10. In theory. The same experimenter conducted all sessions. one for each treatment. subjects had to decide whether they would commit to disclosing voluntarily a violation if one occurred. Thus. or facing the conventional enforcement strategy. Before it was revealed whether a violation occurred. A stage consisted of three practice rounds. The remaining six treatments were presented in one of six sequences using a Latin Square design to control for possible order effects.

did not submit a violation report.6)(180 − pd nd). all who ¯ did not commit to disclosure were subject to an audit. Obviously. Using the mean violation probabilities and the numbers of individuals who did and did not commit to disclosing their violations. the expected number of audits required to maintain the π = 0. and the ¯ expected number violations for those who did not. Table 14.3.3. nd is the number of subjects who committed to disclosure and. Note that the expected number of violations under CE is simply the mean violation probability for this treatment times the number of subjects.6)(180) = 108. hence. note first that individuals who committed to disclosure might have been subject to an audit if they did not experience a violation and. The second column of this table contains the mean violation probability by disclosure decision for each treatment. and the third column contains the number of individuals who did and did not commit to voluntary disclosure. These values are reported in the fifth column of Table 14.6 audit probability for those who did not report a violation. To calculate these values. respectively.270 J. are conducted with the data in Table 14. pnd (180 − nd) These values ¯ are reported in the fourth column of Table 14. the expected number of individuals who committed to disclosure but were subject to an audit because a violation did not occur is (1 − pd )nd.K.2 Sequence of treatments using a Latin Square Sequence ID Stage 1 A B C D E F CE CE CE CE CE CE 2 D-H D-I D-L CD-H CD-I CD-L 3 D-I D-L CD-H CD-I CD-L D-H 4 CD-L D-H D-I D-L CD-H CD-I 5 D-L CD-H CD-I CD-L D-H D-I 6 CD-I CD-L D-H D-I D-L CD-H 7 CD-H CD-I CD-L D-H D-I D-L Results The tests of the behavioral hypotheses and their policy implications that we specified in the second part of the chapter.3.3 also includes the expected number of audits necessary to maintain the π = 0. Stranlund Table 14. define these variables in the following way: for a particular treatment.6 probability of a random ¯ audit for those who did not disclose a violation is π[(1 − pd )nd + (180 − nd)] = (0. pd and pnd are the mean ¯ ¯ violation probabilities for those who did and did not commit to disclosing their violations. as well as sensitivity analyses with respect to the disclosure incentive.J. pd nd . For a particular treatment. we calculated the expected number of violations for those who committed to disclosure. Murphy and J. ¯ The required number of audits under CE is simply πN = (0. . Therefore.3. 180 − nd is the number of subjects who chose not to commit to disclosure. given 180 subjects. To show how we calculated the remaining values in Table 14.

no risk neutral subject would choose to disclose a violation in the D-H and CD-H treatments. expected payoff-maximizing agents.5 18. that is.8 93.547 0. yet a non-trivial minority of subjects did so.50) Disclose Not disclose D-L ($0.0 55.434 0. Since fines are levied on all reported violations. (0.497 0. On the other hand. as expected.1 36. and expected numbers of enforcement actions Treatment (disclosure penalty) CE D-H ($2.3.0 96.4 96.97) Disclose Not disclose CD-H ($1.467 0.Voluntary discovery and disclosure 271 Table 14.525 0.0 70. the expected number of fines for disclosed violations in a particular treatment equals the expected number of violations of those who committed to disclosure. Thus. fines for undisclosed violations are levied with the ¯ probability of an audit.0 96. Similarly.9 61. every risk neutral subject would choose to disclose their violations in the D-L and CD-L treatments.3 we report the ¯ expected numbers of fines for disclosed and undisclosed violations for each treatment.0 93. it is worth noting that.8 19.5 42.448 0.3 Mean violation probabilities.3 109. but a non-trivial minority of subjects chose not to do so.611 0.35) Disclose Not disclose D-I ($1. Before discussing the results contained in Table 14.97) Disclose Not disclose CD-L ($0.9 23.3 73. In the final column of Table 14.656 0.9 37.1 107. .1 Expected number of audits 108.4 57.6 89.468 0.6) ( pnd )(180 − nd).8 33.502 N Expected number of violations 91.448 0.9 13.1 84.5 42. the observed outcomes do not match specific predictions about violation probabilities and reporting choices based on a model of risk neutral.3 51. expected numbers of violations.1 28.50) Disclose Not disclose CD-I ($0. Violation probabilities tended to be higher than what a risk neutral subject would be expected to choose.6 80.536 0.701 0.650 0.421 0.5 117.7 61.1 48. pd nd.7 82.6 76.60) Disclose Not disclose Mean violation probability 0.9 18. we calculated the expected numbers of fines levied on disclosed and undisclosed violations.9 180 180 38 142 180 79 101 180 117 63 180 66 114 180 94 86 180 134 46 Finally.0 57. The expected number of fines under CE is the expected number of violations for this treatment times the audit probability.497 0.534 0.8 76. the expected number of fines for undisclosed violations is the expected number of these violations times the audit probability.616 0.2 47.9 36.2 28.8 64. Furthermore.9 86.9 23.8 81. as well as their sums.509 0.6 Expected number of fines 55.9 38.

p = 0. Instead.J. as in D-I. in all six voluntary disclosure treatments.616) is significantly higher than the mean violation probability made by these same subjects under CE (0. As predicted.9 However. simply because choosing not to disclose their violations means they are choosing to face the unchanged conventional enforcement strategy. p = 0. there is a roughly even split between the number of people who committed and who did not commit to disclosing their violations (79:101 for D-I and 94:86 for CD-I). their mean violation probability (0. Since deterrence is weaker for those who choose to discover and disclose. Hypothesis 4 asserts that these individuals should not change their violation probabilities when a disclosure policy is added to a conventional enforcement strategy. In both the D-I and CD-I treatments. as in CD-I. Although those subjects who committed to disclosing their violations under the D-I and CD-I treatments behaved as theory predicts. we preserve the within-subject comparison by using the non-parametric Wilcoxon signed-rank test for matched pairs. we should observe higher violation probabilities for these individuals relative to their choices under conventional enforcement (Hypothesis 3).64. . those who did not commit to disclosing their violations behaved unexpectedly. Murphy and J.534) is not statistically different from their mean violation probability under CE (0.506.518. Stranlund All of our hypotheses and their policy implications entail pairwise comparisons of the treatment effects of each disclosure treatment relative to conventional enforcement.61).00.K.035 with p = 0. ∆D-I = −0.272 J. We observe: ∆D-H = −0. when it is costly for subjects to discover whether a violation occurred. the disclosure penalty must be reduced below the expected penalty under conventional enforcement in order to induce discovery. Because each subject participated first in the conventional enforcement treatment and then once in each of the six disclosure treatments.02.053 with p = 0. the mean violation probability for those subjects who committed to discovery and disclosure under CD-I (0. Now consider the impacts of voluntary disclosure on the care taken by individuals to prevent violations in the D-I and CD-I treatments relative to their choice under CE (conventional enforcement). When self-discovery is costless.8 Behavioral hypotheses Hypothesis 1 holds as expected. Our results are consistent with this hypothesis: for those who committed to disclosure in D-I. and let p be the result of a Wilcoxon signed-rank test of the hypothesis that there is no change.02 with p = 0. Let ∆ t be the mean change in violation probability from CE for those who did not commit to disclosure in voluntary disclosure treatment t. those subjects who did not commit to disclosure tended to choose lower violation probabilities than under CE. ∆D-L = −0.01). there should be no change in the choice of violation probabilities by those who committed to disclosing their violations (Hypothesis 2). This confirms our expectation that a significant number of subjects would choose to disclose their violations under disclosure policies that make risk neutral individuals indifferent between disclosure and non-disclosure.

89. Policy implications In the absence of any framing effects.6 under CE vs. these reductions suggest a framing effect that is due to the introduction of a voluntary disclosure option that. our results strongly support the hypothesis that inducing costly discovery and voluntary disclosure will lead to weaker deterrence for those who . Policy Implication 1 asserts that the expected number of violations under treatment D-I should not be different from that number under CE. and this difference is not statistically significant (p = 0. Second.5 under D-I). so we need to determine whether the policy implications hold despite this effect.01. but this difference is not statistically significant (p = 0. Thus. because the incentives for exercising care to prevent violations are unchanged by the introduction of a voluntary disclosure policy if one does not intend to disclose a violation. Despite our failure to support Policy Implication 2.11. Table 14. the expected number of violations should be higher than under CE (Policy Implication 2) because those that disclose their violations increase their violation probabilities (Hypothesis 3) while those that do not disclose should choose the same violation probabilities (Hypothesis 4). the mean violation probability under CD-I is a bit higher than under CE (+0. when subjects incur a cost to discover whether a violation occurred under CD-I. ∆CD-I = −0. This is precisely what we observe despite the framing effect for those who chose not to disclose.047 with p = 0. our view is that it is hard to justify a belief that the framing effect that led to our failure to support Policy Implication 2 would actually motivate firms in the field.027). it appears that it is possible to add a disclosure opportunity to a conventional enforcement strategy without affecting deterrence – at least as long as there are no discovery costs.Voluntary discovery and disclosure 273 ∆CD-H = −0. This implies that the expected numbers of violations in these treatments are not significantly different (91.00. motivated non-disclosers to choose lower violation probabilities. which is inconsistent with Hypothesis 4. First. Hence. for some reason. We have already seen that the disclosers in CD-I significantly increased their violation probabilities by 0. However. Thus our data do not support Policy Implication 2. However. all of the policy implications follow directly from the behavioral hypotheses. the framing effect we have identified led non-disclosers to reduce their violation probabilities by 0. we have good reasons to continue to expect that disclosure policies when firms must invest in self-discovery would lead to less overall deterrence in non-laboratory settings.3 shows that the mean violation probability (and therefore the expected number of violations) under CE (0. These reductions are surprising. However.497).509) is about the same as the mean violation probability for both disclosers and non-disclosers under D-I (0.065 on average. on average. from their violation probabilities under CE.24. and ∆CD-L = −0. the framing effect (Hypothesis 4) is a potentially complicating factor.93).065 with p = 0. Clearly we reject the hypothesis of equal violation probabilities in four of the six voluntary disclosure treatments (including D-I and CD-I). Overall.079 with p = 0.23).

Policy Implication 3 asserts that if a voluntary disclosure policy motivates a significant number of violation disclosures.K. assuming that the framing effect for nondisclosers is unlikely to hold outside the laboratory. Similarly. To examine the sensitivity of our results to the incentive for voluntary disclosure.3. This is precisely what we observe. Thus. Stranlund choose to discover and disclose. we believe that our results do justify a continued expectation of weaker deterrence when firms are given the opportunity to discover and disclose their violations voluntarily.274 J. their policy implications.J. In fact.5 expected fines under D-I. it is not surprising that the results in Table 14. the discussion thus far has focused on the two treatments that were parameterized such that risk neutral agents would be indifferent about committing to disclosing their violations (D-I and CD-I). Policy Implication 4 suggests that the expected number of fines. the expected number of audits required to maintain a 0. Although these results indicate that voluntary disclosure policies can lead to reduced monitoring effort. This result necessarily holds if some violations are voluntarily disclosed (Hypothesis 1) and therefore automatically sanctioned. as shown in Table 14. as shown in Table 14.0).3) is significantly lower than under CE (108).0) are significantly greater than the expected number of fines under CE (55. since a significant number of subjects voluntarily disclosed their violations under both D-I and CD-I (consistent with Hypothesis 1). and. Policy Implication 5 asserts that the number of the potentially more costly fines for undisclosed violations will be smaller. the remaining subset that are subject to random audits will necessarily be smaller than under CE. and if there is no change in violation probabilities of those who choose not to disclose (Hypothesis 4). While voluntary disclosure policies are likely to result in a greater number of sanctions. we varied the reduced penalty for voluntarily reported violations. and possibly sanctioning costs.0 expected fines under CE. hence. Since a significant number of subjects chose not to disclose their violations in the D-I and CD-I treatments and there is only a small reduction in their mean violation probabilities in these treatments.4 are for violations that are uncovered by a government audit. Murphy and J. .5) and under CD-I (81. the expected total number of fines under D-I (70. could increase. Sensitivity analysis Our behavioral hypotheses. Of the 70.3 support this policy implication.0 expected fines are for undisclosed violations. Despite our observation that individuals who did not commit to disclosure tended to reduce their violation probabilities relative to conventional enforcement. Both of these are significantly lower than the 55.6 audit probability for those who did not report a violation under D-I (82.3. only 23. only 28.1 of the 81. under CD-I. then fewer government audits are required to maintain the same level of deterrence for those that do not disclose.7) and CD-I (73. it is impossible for the expected number of audits to increase: as long as some firms disclose their violations.

Table 14. The results in Table 14. Moreover. these values increase from 18.11 A similar pattern occurs in the costly discovery treatments. For the disclosure only treatments. but these lower penalties also provide an incentive for firms to exercise less care in avoiding violations.7 and then to 61.525 for D-L).611 as the disclosure penalty is reduced. 0. A Mann-Whitney test making the pairwise comparison of treatments D-H and D-L indicates that this increasing trend is statistically significant (p = 0. and on to 76.497 and then to 0.497 to 0.Voluntary discovery and disclosure 275 Our results suggest that the commitment to disclose a violation is accompanied by a potentially substantial decrease in the care that individuals took to prevent violations.3.5 and then to 109.8 for D-L.3 shows that within each of the six treatments that allow voluntary violation disclosure.7 to 82.2 for D-I.13 Perhaps more revealing is the resulting increase in the expected number of violations from 80.9 for the disclosure only treatments and similarly for the costly discovery treatments. thereby conserving monitoring effort. in four of the other five cases. the mean violation probability (and corresponding expected number of violations) is higher for those who committed to disclosure than for those who did not (e.9 for D-H to 42. this difference is significant at the 1 percent level: only for D-H is this difference not significant (p = 0. These patterns hold for the costly discovery treatments as well.10 However.12 Since the number of subjects who committed to disclose their violations also increases as the disclosure penalty is reduced. The same pattern holds for the costly discovery treatments. From Table 14.00.534 and then to 0.656 vs.8 for the disclosure only treatments. this overall trend is highly significant: the p-value for a Mann-Whitney test comparing D-H to D-L is 0. this difference is not statistically significant for the D-I treatment (p = 0. the mean violation probability for disclosers and non-disclosers combined increases in the disclosure only treatments from 0.448 to 0.656 as the penalty for disclosed violations is reduced. As one would expect. Of course. the expected total number of violations also increases.g. Largely because the number of subjects who committed to disclosing their violations and their violation probabilities both increase as the disclosure penalty is reduced. the expected number of required audits falls – from 96. the expected number of violations of those who committed to disclosure also increases quickly. For the disclosure only treatments. whether the reduction in monitoring effort results in a reduction in total enforcement costs also depends upon the impact on sanctioning costs.14 using a Mann-Whitney test for unmatched pairs).3 show that reducing the penalty for disclosed violations .00). Our results clearly suggest an important tradeoff inherent in voluntary discovery and disclosure policies: reducing the penalty that firms automatically pay if they voluntarily disclose a violation effectively induces more of them to report their violations.14 While the expected number of violations increases as the disclosure penalty is reduced. 0. Again.26). observe that the mean violation probability for those who committed to disclosure increases from 0. the mean violation probabilities of those who committed to disclosing their violations increases rather rapidly as the penalty for disclosed violations is reduced.5 to 89.

4 and then to 19. Stranlund increases the total number of fines in the disclosure only treatments from 55. The same pattern holds for the costly discovery treatments. However.K. as well as firms’ costs of auditing themselves to determine their compliance status. Although our results in this case fail to provide unequivocal support for the hypothesis that disclosure policies will lead to more violations when firms’ selfaudits are costly. Clearly. we doubt that adding a voluntary disclosure policy to an existing conventional enforcement strategy will leave deterrence unaffected. Consequently. Conclusions A key conclusion of our study is that. fewer of which are for violations that the government uncovers. our work highlights some of the essential tradeoffs inherent in voluntary discovery and disclosure policies. Murphy and J. Clearly. voluntary discovery and disclosure policies. While we have examined many of the essential aspects of voluntary disclosure policies. fewer penalties were levied on undisclosed violations. although lower disclosure penalties led to more penalties being levied. decreasing government monitoring effort. Generally. Thus. when firms must undertake costly self-audits to determine their compliance status. with a similar pattern holding for the costly discovery treatments. the number of fines levied on undisclosed violations falls as the disclosure penalty is reduced: 37 to 28. the costs of monitoring firms and sanctioning violations.276 J.9 to 70. whether voluntary discovery and disclosure policies are an efficiency-enhancing complement to conventional environmental enforcement will have to be determined on a case-by-case basis.6. when firms know their compliance status without cost.8 in the disclosure only treatments. any conclusion about the benefits and costs of voluntary disclosure policies will require detailed knowledge of the harm caused by environmental violations. this failure is due solely to a framing effect that we doubt would persist in field settings of environmental enforcement. In this case. motivating an increasing number of violation disclosures is associated with increasing incidences of noncompliance. and more sanctions. However. While voluntary disclosure policies can reduce government efforts to monitor the compliance behavior of firms. there are others that we have not considered. it is possible to motivate a significant number of voluntary violation disclosures without adversely affecting environmental quality. Namely. but that can be . their impact on the costs of sanctioning noncompliant firms depends on the relative costs of sanctioning voluntarily disclosed violations and sanctioning violations that the government uncovers.J. or opposition to. whether voluntary disclosure policies are worthwhile depends solely on their impact on government enforcement costs.5 and then to 96. Moreover. how total enforcement costs change with a greater incentive for voluntary discovery and disclosure depends on the relative costs of monitoring and sanctioning disclosed and undisclosed violations. it appears that there is little theoretical or empirical justification to warrant general support for. worsening environmental quality.

In the related economics literature this is usually referred to as self-reporting. product safety. Acknowledgments Primary funding for this research was provided by the US EPA – Science to Achieve Results (STAR) Program grant #R829608. and by the Center for Public Policy and Administration. have focused on voluntary reporting. Jason Shogren. or when they are able to engage in costly efforts to avoid government detection and punishment of their violations. clean-up of spills). Notes 1 US EPA (2000). We are also grateful to Jay Shimshack. Additional support was provided by the Cooperative State Research Extension. 3 Innes (2001b) provides a comprehensive review of the theoretical literature on voluntary discovery and disclosure policies. 2000. our results apply more broadly. Innes. 2001b). our framework can easily be adapted to examine Innes’ (1999 and 2001a) claims that there are additional benefits to voluntary discovery and disclosure policies when firms are required to undertake costly remediation (e. and an anonymous referee for useful comments and suggestions. the tradeoffs that we highlight will also manifest themselves when reporting is mandatory.. Although our experiments. “Discovery” refers to costly efforts by regulated entities to discover whether they are in violation of an environmental regulation. We follow the terminology used by the EPA. The model presented here is essentially the same as the one he employs to motivate his review. Mishra et al. Education Service. .Voluntary discovery and disclosure 277 addressed with straightforward modifications of our experimental designs. MAS00871. Massachusetts Agricultural Experiment Station. Maria Alejandra Velez and Elizabeth Gonzalez provided outstanding research assistance. Some also call these actions self-audits. Likewise. and federal sentencing guidelines (Kaplow and Shavell. University of Massachusetts Amherst. and the Department of Resource Economics under Project No. the use of disclosure policies extends well beyond environmental policies to regulations concerning occupational health and safety. For example.g. In addition. US Department of Agriculture. Our results apply to these contexts as well. Finally. the claims of several authors that self-discovery and disclosure rules might not be as effective as hoped because firms fear that the information they discover might improve the government’s own monitoring efforts (Pfaff and Sanchirico. 1994. while our study was motivated by voluntary discovery and disclosure policies to support compliance with environmental regulations. 2 Innes (1999) has argued that the fact that voluntary disclosure policies will tend to lead to a greater number of sanctioned violations could also imply improved environmental quality if firms are required to correct the harm caused by their violations. 1997) can and should be examined within our framework. Many environmental regulations require that firms report their compliance status to regulators. and most of the literature on self-reporting in law enforcement. “Disclosure” means voluntary reporting of violations to the authorities.

D. S. Self Enforcement of Environmental Law. We believe that this is an important area for future research for those who investigate compliance behavior in experimental settings. Journal of Public Economics. CD-L (p = 0. S. The Law and Economics of the Environment. 379–393. 9 Note that in Table 14.M. Therefore it is unclear whether risk preferences elicited with the Holt/Laury mechanism would be robust in predicting behavior in other settings. Innes. Innes. Risk Aversion and Incentive Effects. References Holt.50). 1999. and Organization. 1994. CD-L (p = 0.00). A. Just Who Are You Calling Risk Averse? Journal of Risk and Uncertainty. CD-I (p = 0. 20 (2).. Optimal Law Enforcement with Self-Reporting of Behavior. Remediation and Self-Reporting in Optimal Law Enforcement. 2000.. In: A. 7 Instructions are available upon request from the author. L. Innes. 24 (3). 12 The results of Mann-Whitney tests for the violation probabilities of disclosers only are as follows: CD-H vs. and Shavell.K. Malik. but a comparison of these values under D-I and D-L is significant (p = 0. 583–606.12). The mean violation probability under CE for the 79 subjects who committed to disclosure under D-I is 0.02). we framed the experiments as a production decision in which subjects chose the probability of an unspecified accident. Moreover. but there is evidence that risk preferences may be domain specific and not stable across institutions (Isaac and James.00). even though the subjects’ actions were about preventing and possibly disclosing accidents. 5 Unfortunately there is no consensus about how to elicit these preferences.J. CD-I vs. 1993...3 the mean violation probability under CE for all 180 subjects is 0.A. 2000). 10 This follows simply by combining Hypotheses 2 and 3. 2001b. 177–187. 92 (5). R.. 11 A pairwise comparison of the violation probabilities under D-H and D-I for disclosers only is not significant (p = 0. Journal of Law. 1644–1655. CD-L (p = 0. 2001a. Self-Reporting and the Design of Policies for Regulating Stochastic Pollution. Murphy and J.02). 8 For each hypothesis. CD-H vs. 13 A pairwise comparison of the violation probabilities under D-H and D-I for all subjects is significant (p = 0. 72 (3).509. Isaac. Heyes. Journal of Environmental Economics and Management.. 2002. . 150–184. 239–256. 17 (1). fixed conventional enforcement strategy. Instruments such as that presented by Holt and Laury (2002) may be useful. pp. American Economic Review. and Laury. CD-L (p = 0.00). CD-I vs.00). CD-H vs. Violator Avoidance Activities and Self-Reporting in Optimal Law Enforcement.05). 14 The results of Mann-Whitney tests for the violation probabilities of all subjects are as follows: CD-H vs. a matched pair t-test yielded the same conclusions. choosing instead to focus on the qualitative effects of adding various disclosure rules to an existing. instead of the probability of a violation of an environmental standard. 6 To avoid the possibility of introducing unwanted biases. Economics. as is a comparison of these probabilities under D-I and D-L (p = 0. R. For the design of optimal discovery and disclosure rules see Innes (2001b).K. R. 241–257. 103 (3).278 J. conventional enforcement is held fixed throughout this chapter. R. Cheltenham: Edward Elgar.. we do not consider the optimal design of voluntary discovery and disclosure rules.518. and James. Journal of Political Economy. We will continue to speak of violations throughout the chapter. ed. CD-I (p = 0. Stranlund 4 Thus. Kaplow.00). C.

65 (70). A. 2005. C.H.org/articles/articles/Stafford11FIN. Federal Register. Online. and Stinson.K.. Environmental Self-Auditing: Setting the Proper Incentives for Discovery and Correction of Environmental Harm. Vermont Journal of Environmental Law.. Small Potatoes: An Empirical Assessment of EPA’s Self-Audit Policy. 16 (1). Economics and Organization. D. S. Journal of Accounting and Public Policy. 187–214. US Environmental Protection Agency.html (accessed 14 September 2007.epa. Journal of Policy Analysis and Management.L. 23 (3). Incentive for Self-Policing. B. Newman. Stafford.P. Pfaff. Environmental Regulations and Incentives for Compliance Audits.Voluntary discovery and disclosure 279 Mishra. and Sanchirico. . 2004. Journal of Law.. 2000. 2000. Online.W. Big Field.htm (accessed 14 September 2007).. 1997. available at: vjel. 415–432. Does Self-Policing Help the Environment? EPA’s Audit Policy and Hazardous Waste Compliance. 16 (2). Disclosure. and Sanchirico. 189–208...W. Discovery. A. C. 19618– 19627. Correction and Prevention of Violations. available at:.gov/compliance/incentives/auditing/auditpolicy. C. Pfaff. 6.

a toll) in this environment and find this effective at reducing congestion. In addition to the political pressure traffic puts on city planners and other elected officials. but slower. even when subjects make the same decision for as many as 60 rounds. travel on city streets is an attractive alternative. DC.” The humorous contradiction in this comment raises the issue of how individuals actually respond to congestion. When subjects must simultaneously choose between the safe and risky options. congestion imposes huge welfare costs on commuters. However. Instead of providing innovative solutions to this problem.e. Many commuters face the daily dilemma of taking a predictable.15 Congestion pricing and welfare An entry experiment Lisa R. the number of commuters appears to be growing at a faster rate. The solution is suggested by a famous animal foraging experiment. We test the effectiveness of a user fee (i. Subjects must choose between a “safe” route with a fixed payoff and a “risky” route for which the payoff depends on the number of other users. in which two people stood on . Highway congestion was once just a problem for large cities like Los Angeles and Washington. there is significant congestion. It’s too crowded. With billions of dollars spent every year to add capacity. which is costly from a welfare perspective. technological advances have spawned new areas where congestion must be managed. reported by Harper (1982). resulting in large welfare losses. route or risking hours of gridlock on a potentially faster freeway. the simultaneous nature of the decision still results in a high variance in the number of entrants. Holt and David Reiley Introduction One of the most persistent problems facing cities is freeway congestion. but it is increasingly affecting smaller metropolitan areas. We also test information provision as a policy option and find that it reduces the number of entrants and the variance. like cell phones and the internet. Charles A.1 We study the problem of congestion in the context of a binary choice game. Anderson. In some cases. Ducks and “magic” Yogi Berra once remarked: “Nobody goes there anymore.

This successful coordination has been explained in terms of adaptation and learning (Meyer et al. with some switching back and forth. Moreover. which was more severe with large numbers of potential entrants. some experiments have produced too much entry into the more congestion-prone activity. However. as measured in grams per minute.Congestion pricing and welfare 281 opposite banks of a duck pond in the Cambridge University Botanical Garden and began throwing out five-gram bread balls at fixed intervals. Entry and congestion problems arise often when choices are decentralized. (1995). e. which is less time than it would take for most ducks to obtain a single bread ball. the ducks were in constant motion. beginning with Kahneman (1988).. as in the decisions of individuals concerning whether to congregate in a potentially crowded bar. The payout rate was twice as high on one bank (every ten seconds instead of every 20 seconds). and on factors that may increase efficiency. Psychologists and economists have conducted a series of similar binary choice experiments with congestion effects. the general result is that the amounts of over entry and under entry are small. using the argument that exogenous random noise in behavior may tend to pull entry rates towards one half. as long as they are not extreme. Ochs (1990). The flock of ducks sorted themselves to equalize expected payoffs.g. it looks like magic. Camerer and Lovallo (1999) conclude that entry can be affected by overconfidence. bridge.. 1999). This systematic (“inverse S”) pattern of over and under entry is reported by Sundali et al. who observed the payoff equalization and remarked: “To a psychologist. Erev and Rapoport.” There have been a number of subsequent experiments in which observed behavior tends to equate payoffs. Goeree and Holt (2005) provide a unified treatment of some (but not all) of these disparate results. 1992. Fischbacher and Thöni (forthcoming) conducted an experiment in which each entrant essentially gets a single lottery ticket with an equal chance for winning a money prize. For example. even after equilibrium was reached. Commuting time . A stylized model of congestion Consider a group of N commuters who must choose between a slow reliable route and a faster. To summarize. and that theoretical predictions are fairly accurate. but not otherwise. There was excess entry. In other words. since they observe over entry when post entry payoffs depend on a skill based trivia competition. which would result in over entry when the theoretical prediction is less than half and under entry otherwise. but potentially congested freeway. This latter case is known in the literature as the “El Farol” dilemma. The experiment reported in this chapter will focus on the welfare consequences of congestion. This stochastic element could offer an evolutionary advantage if it hastens the adjustment to changes in payoff conditions. so the expected payoff for entry is a decreasing function of the number of entrants. As Paul Glimcher (2002) notes. named after a popular bar in Santa Fe (Morgan et al. a change in the interval times resulted in a new equilibrium within about 90 seconds. 1998). or tunnel. the focus is on how to improve the lives of the ducks.

since entrants do not consider the effects of their own entry decisions on the other entrants. so the marginal social benefit of entry (shown by the dashed line) is below the average payoff line.50 3.1 shows entry payoffs for the risky route with a total of 12 commuters.50 for a larger number of people.282 L.50. note that the total payoff with x entrants and N – x non-entrants is: x(A – Bx) + (N – x)C.00 1.00 is fixed on the safe route and is an increasing function of traffic on the risky route.50 14. A = 4.50 and the payoff for the risky route is $4. It follows from these calculations that the optimal rate of entry is half of the equilibrium entry rate in this linear model. which is maximized when the marginal social value equals the marginal social cost: A – 2Bx = C.50 12.50) when two-thirds of the commuters enter the risky route. the marginal private benefit from entry is just the average payoff.50 6. if x = (A – C)/B. This equal payoff equilibrium outcome is not socially optimal.50 13.1 shows the locations of the free entry equilibrium and socially optimal entry levels. But entry imposes costs on others.50. Anderson et al. The payoff from taking the reliable route is C. Consider an example in which the payoff for the safe route is $0.50 –18.1 Payoff for the risky route Number of 1 entrants Average 4. Notice that the payoff for the risky route is equal to the payoff for the safe route (at $0.00 9. which constitutes an equilibrium. Individual entrants do not consider the cost of entry on other users of the risky route.00 2.00 –10. This marginal cost is the payoff from taking the safe route.R. as shown by the horizontal dotted line. shown by the solid line in the figure. Suppose that the average payoff for an entrant is decreasing in the number of entrants: A – Bx.50 –4. Hence. The socially optimal number of entrants occurs where marginal social . the free entry equilibrium prediction is for eight of the 12 commuters to take the risky route.00 –1. or equivalently.50. as the number of entrants increases.00 13.00 payoff per entrant Total earnings for all 2 3 4 5 6 7 8 9 10 11 12 3. average payoffs are equalized if A − Bx = C. as can be verified by substituting the payoff parameters into the formula for equilibrium entry derived earlier. which represents the opportunity cost for an entrant. or when x = (A – C)/2B. and B = 1).00 0. The marginal social benefit line is steeper than the average payoff line because.00 1. Figure 15. With free entry.00 –0.50 minus the number of entrants (C = 0. The marginal (private or social) cost is $0.50 –1. Table 15.50 0.50 2.50 1. and it equals the individual payoff from entry (“average payoff”) when there are eight entrants. so to them. To see this.50 9.50 12. where x is the number of entrants and A and B are positive parameters. Table 15. each additional entrant causes the value of entry to fall by $0.

since a nonentrant who attempts to enter will drive the number of entrants up to eight and hence will only earn $0. each earns $0. since it follows from Table 15. so none can be better off by taking the safe route. so that the intersection with the marginal cost line occurs at the optimal entry level of four. another Nash equilibrium with seven entrants. not realizing that the act of entry will reduce the payoffs from entry. Since the marginal private benefit of $2. The equal average payoff equilibrium that results from free entry is closely related to the notion of a Nash equilibrium. which is approximately 0. at four entrants in this example. with eight entrants.Congestion pricing and welfare 283 $6 Average payoff Marginal cost Marginal social benefit $3 Equilibrium $0 Optimum $3 $6 0 2 4 6 8 10 12 Number of entrants Figure 15. the effect of a $2 fee would be to shift the average payoff line down by $2 in a parallel manner. This person’s expected entry payoff can be calculated as a function of p using the formula for the density of a binomial distribution with N = 11. an entry fee of $2 corrects the externality.1 that a ninth entrant would earn zero. then the expected entry payoff for the remaining person exactly equals the exit payoff of $0.1 Benefits and costs of the risky route.64. There is. note that there is an asymmetric Nash equilibrium in which exactly eight people enter.50 exceeds the marginal social benefit of $0. This probability must be set to ensure that if the other 11 people enter with probability p. benefit is equal to the marginal private cost. The difference between this number and the two-thirds entry rate that equalizes expected payoffs is due to the fact that the number of entrants is finite. however. To see this.50 when p is 7/11.50. The experiment to be discussed in the next part of the chapter will evaluate the effects of both exogenous and endogenously determined entry fees. consider a symmetric Nash equilibrium in mixed strategies. It is straightforward to show that the expected entry payoff is $0. (In the free entry “competitive” approach. To see the intuition. each earning $1. with the probability of entry denoted by p.50 payoff from staying out. this nonentrant would enter anyway.50 by $2 at this point. In the figure.50. which is less than the $0. Conversely. think of an individual player who is .) Alternatively.50.

This is because a given player faces an expected 2/3(11) = 7. A congestion experiment Subjects were recruited from undergraduate classes at the College of William and Mary and at the University of Virginia. virginia.50*x. considering entry or not. so for that player to enter would create total expected entry of 8. For large numbers of players.67. Anderson et al. it turns out that an entrant earns an expected payoff of $0.R. the predictions are quite close. If the other 11 players randomize with probability 2/3.284 L. and for the parameters used in the experiment.econ.) Sessions lasted about an hour and the average person earned about $0.edu/admin. The player would be indifferent if exactly eight entrants are expected (that person and seven others). the risky route) or not. The experiment was conducted using the Market Entry program on the Veconlab website (online. except as noted below. The treatment parameters for all sessions and the resulting entry rates are listed in Table 15. To summarize.33 other entrants.2 shows results from a session in which subjects made this entry decision for 60 rounds. (The only exception was in the first session where the payoffs were all doubled. Even with 60 rounds . we will use the free entry prediction of 2/3 as the prediction. available at: veconlab. with 11 other players each entering with probability 2/3. The data for all sessions are online. exactly the same amount as the exit payoff. which is more than the entry level that equates the payoffs from entry and exit.33 rather than the $0.50 payoff in each round they did not enter.a. the probability of entry must be 7/11. In each round. plus a $6 show-up payment. each player earns an expected payoff of $0.1 in the Appendix to this chapter.50. depending on the treatment. As described above. there is significant variation from round to round. and a “free-entry” equilibrium that equates expected payoffs involves an entry rate of 0.e.virginia. a Nash equilibrium in mixed strategies with risk-neutral players is for entry to occur with probability 0.50 − 0. In order to get seven entrants out of the 11 others with a binomial distribution. There were 12 sessions with 12 participants in each. While the average entry rate is close to the prediction of 2/3. subjects faced a binary choice to enter the market (i. subjects earned a sure $0.64. Figure 15. To be willing to randomize. where x denotes the total number of entrants.edu/~cah2k/data/. The payoff for entry in a given round was determined by the total number of entrants in that round according to the following formula: $4. Therefore. then it turns out that the remaining player would prefer not to enter. Another perspective is to think about why randomizing with probability 2/3 is not an equilibrium.50 per round over 30–60 rounds.33. available at: people. since the session involved only 20 rounds.htm). the person must have the same expected payoff from entry (with probability 1) and exit. Indeed.50 that could be earned from exit. In the symmetric mixed-strategy equilibrium with all 12 players choosing entry with probability 7/11. these two approaches would be equivalent.

With a linear average payoff line.64. For example.3 0. free entry yields inefficient outcomes since entrants do not take the social cost of over entry into account.7 0. in which each person enters with a probability of 7/11.2) are represented by the black bars in Figure 15.50)/2 = $5. .2 0. so total earnings are $6. As noted above.00)/2 = $1. If the number of entrants is seven in one round and nine in the next. even over long periods of play.5 0. the probabilities associated with each number of entrants can be calculated.1 0 0 5 10 15 20 25 30 Round 35 40 45 50 55 60 Figure 15.1. Now consider how social welfare changes with some variance in the entry rate. the earnings are $0.50 for each person. However the average total earnings for these two periods is ($9.54 and N = 12). as shown by the gray bars in Figure 15.50 + $1. which results in large efficiency losses. As mentioned in the previous section.00 − $4. average of the two earning amounts is ($6. the welfare loss grows at an increasing rate.3. Result 1: There is considerable variation in the entry level. or 0.4 0.50.2 An entry game session with 60 rounds (me070804). there is a symmetric Nash equilibrium for this game. This effect is illustrated in the bottom row of Table 15. Using a binomial distribution (p = 0.3. The frequencies of the actual numbers of entrants for the 60 round session (in Figure 15.9 0.6 Equilibrium Entry rate 0. which show a mode at eight entrants. which indicates that outcome variability is about what would be expected. Although this was the longest session we ran.8 0. and variance will reduce average payoffs. the average entry rate is consistent with the theoretical prediction of eight. the amount of variation shown here is typical of the sessions with shorter durations. of play.Congestion pricing and welfare 285 Rate of entry 0. As the variance grows. the noise does not subside. with entry at six in one period and ten in the next period. At the equilibrium level of eight entrants. whether or not they enter. the total payoff will be quadratic and concave.

In two of the sessions.9 0.8 0.00 0 1 2 3 4 5 6 7 8 Number of entrants 9 10 11 12 Figure 15.30 Frequency 0.2 0.5 0. the average entry rate was 34 percent with the fee.4 shows results from a session with the optimal user fee of $2 per entrant.50 at the socially optimal entry level of four.05 0.4 Entry game with $2 entry fee (me062904). Note from Table 15. Anderson et al.3 0.R. the optimal entry fee was imposed in parts of six sessions.1 0 0 5 10 15 20 Round 25 30 35 40 Figure 15. Entry rates quickly fell when the user fee was imposed. the revenue collected from the fee was split equally between the 12 No entry fee Entry fee Equilibrium 1 0.10 0. Nash Data 0.15 0.4 0.25 0. .20 0.286 L. Figure 15.35 0. Imposing this $2 cost on entry reduces the private benefit to $0.50.1 that the private benefit from entry is $2. thus moving the free entry equilibrium prediction to the optimal level.7 Entry rate 06 0. Overall. Congestion tolls A common policy approach to congestion is to tax freeway use via a toll.3 Predicted and observed distributions of entry outcomes for a session with 60 rounds (me070804).

the entry fee revenue was not rebated to the subjects. People were allowed to enter in any order. and are greater than 0. but there is still considerable noise in the data.2. and social welfare is not maximized in these sessions. Result 2: With an optimal user fee. we made information about prior entry available to subjects as they were making decisions. The only exception was in Session 9. There is no overlap in these entry rates by treatment. entry is reduced to the socially optimal level on average. Specifically. the average entry rates were 33 percent and 38 percent. 35 percent. Notice that the rebate may reduce the variance around the optimal rate. 66 percent. In the two sessions with the rebate. and seven sessions that began with ten or more rounds of a “view” treatment that posted the number of prior entrants on each screen update. the number of entrants at any given point in time was displayed on all of the computer screens.01 level using standard non-parametric tests. 65 percent. The success of the entry fee did not depend on whether or not it was rebated to subjects.5 shows results from a typical session with endogenous entry order and the provision of prior entry information. If entry decisions are not made simultaneously. This difference is significant at the 0. but there is still noise. the variance in entry still persists in these sessions. The overall effect of imposing an entry fee. then coordination may be facilitated by improved information about current conditions. but even in this case. 63 percent. the provision of prior entry information tended to reduce variance of entry rates. Much like rush hour traffic reports. 35 percent. There was considerable variance in all of the no-view treatments. and the pattern with the view treatment was basically a flat line with an occasional “blip” as seen in Figure 15. is substantial and clear. In most cases. and in the sessions without the rebate.01 for the view treatments. the average entry rates were 34 percent.5. In the other four sessions. over entry was the result of two players clicking the “enter” button at precisely the same moment. All of the variances are less than 0. with or without rebate. 68 percent. so the result would be highly significant on the basis of standard non-parametric tests. Information and coordination Despite the success of user fees at moving usage towards the socially optimal level. . Even without an entry fee.01 for the no-view treatments. The combination of the entry fee and the information in the last half of the session resulted in the socially optimal entry rate in 16 of 20 rounds of play. the overall variance of entry rates was relatively low. 63 percent. all of the entry rates listed above with the fee are below the rates for the sessions with no-fee treatments: 69 percent. Figure 15. The variances for the first ten rounds of each of the 12 sessions are shown in Table 15.Congestion pricing and welfare 287 subjects. There were five sessions that began with ten or more rounds of a “no-view” treatment. and 35 percent.67). and 62 percent. where only four of the first ten entry rates were at the same level (0.

004 0.” be denoted by S(x).017 0.9 0. the free entry equilibrium that equates average payoffs from entry and exit is determined by the equation: S(x)/x = C + F. by view treatment Session 1 2 3 4 5 6 7 8 9 10 11 12 Variance 0. One issue is whether the fee setter has an incentive to set an optimal fee.R.001 0. 10 No entry fee and information Entry fee = $2 and information Equilibrium 0. the average earnings per entrant are given by S(x)/x. that is paid by each entrant. (If the surplus is a quadratic concave function. F.7 Entry rate 0.022 0.5 0. with S”(x) < 0.) The total earnings of the group as a whole are represented by S(x) + (N − x)C. When combined with the optimal user fee.6 0. the “surplus.2 Variances of entry rates in the first ten rounds. i. which increases welfare.002 0.001 0.0 0 5 10 15 20 Round 25 30 35 40 Figure 15. Voting and endogenous entry fees Consider the effect of an entry fee.042 0.288 L. Let the total earnings of entrants.1 0. When there are x entrants.2 0.e.3 0.006 0. Anderson et al. which is maximized by equating marginal surplus to marginal cost: S’ (x) = C. Table 15. which is maximized when S’ (x) = C. social welfare is maximized in most rounds.8 0.011 0.031 0. In contrast.035 0.4 0. this yields the linear average payoff model considered previously in this chapter.5 Entry game with information about other entrants (me071504).046 Treatment No No No No No No View View View View View No Result 3: Information about prior entrants reduces noise. Thus the revenue-maximizing fee under free entry is the efficient fee that . when the marginal value of the surplus equals the marginal cost. Multiply both sides of this equation by x to obtain an expression for the total entry fee revenue: xF = S(x) − Cx.

You are free to discuss any aspect of the process. those who entered and those who did not. Let me stress two things: all fees collected get divided up equally among all participants. since the fee that maximizes total fee revenue will maximize the 1/N share of this revenue. whether or not they entered. Figure 15. To facilitate this discussion. we will use a random device (throw of a die) to choose a person to chair the meeting.6 shows results from a session in which subjects were allowed to vote on an entry fee. All fees collected will be totaled and divided equally among the 12 participants. .6 Entry game with voting on entry fee (me063004). the optimal fee for the parameters used in the experiment is $2. At the end of those ten rounds.1 0 0 5 10 15 20 Round 25 30 35 40 Figure 15. except that you cannot talk about who enters and who does not. how much the fee should be. with all fee revenue divided equally among participants. One way to provide subjects in the experiments with the incentive to adopt an optimal entry fee is to split the fee revenues equally. and if so. Once the fee is selected.6 0.5 0. one subject was randomly chosen to be the “chair” and following instructions were read aloud: Now everybody should come to the front of the room. This person will call on people to speak. and we will have a meeting to discuss whether or not to require people who enter the market to pay an entry fee. and then when someone makes a motion that is seconded. which internalizes the externality at the optimal level of entry.7 No entry fee Entry fee = $1 Entry fee = $2 Entry rate 0. 1 0.8 0. As noted previously.Congestion pricing and welfare 289 maximizes total earnings for this model. The chair may vote to break a tie. it will be entered into the computer and will be in effect for the next 10 rounds.3 0.4 0.9 0. Voting sessions started with ten rounds of decision making with no fee.2 0. after which we may meet again to decide on a fee for the 10 rounds that follow. regardless of whether or not they entered. the chair will count the votes.

This modification would tend to add outcome variability even in settings with many more potential entrants. with only three votes. the subjects started with a $1 fee and adjusted it to $2 during the second round of voting. Smith for research assistance. However. Subsequently fees of $3 and $1. an entry fee of $1 was proposed and passed with very little discussion beforehand. In a second meeting following round 20. for some purposes. someone proposed an entry fee of $2. a little too forgiving in the sense that small increases in traffic often have “snowball effects” that increase congestion dramatically. entry behavior is approximately at a level that equalizes expected payoffs. the information based and fee based allocation mechanisms implemented in this experiment would have an additional efficiency-enhancing role if individuals differed in their values for lowered congestion and faster commutes. By charging the optimal entry fee. Result 4: Subjects have some success at finding the optimal user fee with discussion and voting. Anderson et al. but these near-equilibrium entry rates are inefficiently high. In the last round of voting they increased the fee to $1.R. Anyone could propose a fee and call for a vote.75 in the third round of voting and to $1. We wish to thank Angela M. followed by another vote on an entry fee for the next ten rounds.80. The linear congestion function used in this experiment is. Finally. A proposal of $0. In another session. The combination of the optimal entry fee and information about current entrants moves behavior very close to the socially optimal outcome. the $2 was reproposed and passed with seven of 12 votes. A motion to keep the $1 fee passed with a majority of votes.290 L. Summary We present results from a binary choice experiment based on a stylized model of congestion. Majority rule determined whether the proposed fee would be enacted for ten rounds. Also. In this particular session. they lowered it to $1.50 also failed to pass. Acknowledgments We gratefully acknowledge financial support from the National Science Foundation (SBR-0094800) and the Bankard Fund at the University of Virginia. outcomes move closer to the socially optimal level but there is still some under and over entry. the variability of entry from round to round introduces another source of inefficiency. Moreover.60 in the fourth round of voting. . An interesting extension would be to use congestion functions with nonlinear and stochastic elements. On average. as shown by Plott (1983). The chair presided over group discussions of the fee. but it only received four votes.50 were proposed and rejected.

63 0.67 0.50 0.5.5.80 (1/12) 0. 0.5 N = 12 4.65 0.00 (1/12) no 0.Congestion pricing and welfare 291 Appendix Table 15.68 0.67 0.67 0.00 (0) 1.5. 1.35 0.5.5. 0. 0.50 0.35 0.5 N = 12 4.66 0.5 N = 12 4.5.0. and data averages A. B. 0.5 N = 12 4. 0. 0.00 (0) no no no no no no no no no no no no no Session 6 me070804 UVA Session 7 me071404 W&M Session 8 me071504 W&M Session 9 me072104* W&M Session 10 me072804 UVA Session 11 me111004 UVA Session 12 me111504 UVA 4.67 0.62 0. 0. 0.67 0.5 N = 12 4.34 0.00 (1/12) no 0.5.00 (1/12) 2.00 (0) no 2.67 0. 0.57 0.40 0.00 (1/12) view no 0.00 (0) 0.52 0.00 (0) 2.69 0.60 (1/12) 1. 0.5.0. 0.00 (1/12) 1.33 0. 0. C N Session 1 me062304 UVA Session 2 me062404 UVA Session 3 me062904 UVA Session 4 me063004 UVA Session 5 Me070104 UVA 9.00 (1/12) 1. 0. 0.38 0.00 (0) 0.57 0.5.00 (0) 0. treatments.5.5. 1.33 0. 0.00 (0) 0.63 0.5.35 0.00 (0) 0.66 0.34 0.1 Sessions. 0. 0. 0. not me072104. .5 N = 12 4.42 0.0 N = 12 4.38 0.67 0.a.5.5.5.00 (1/12) 0.5 N = 12 view no 0.5.37 0.75 (1/12) 1.33 0.5.00 (0) 2.67 0.00 (0) view no 2.5.33 Average entry rate 0.67 0.67 0.35 0.33 0.5. 0.5 N = 12 4.38 0.00 (0) view no view no view view no no no no Note * The database name for this session was a temporary name.00 (0) no 4.00 (0) 2.5 N = 12 Rounds Entry fee (share) 1–10 11–20 1–20 21–40 1–20 21–40 1–10 11–20 21–30 31–40 1–11 12–20 21–30 31–40 41–50 51–60 1–60 1–20 1–20 21–40 1–20 1–20 20–40 1–15 1–20 21–40 View Voting no no no no no no no vote vote vote no vote vote vote vote vote no Predicted entry rate 0.37 0.67 0.35 0. 0.5.33 0.67 0.5 N = 12 4.5.67 0.00 (0) 1.33 0.5.68 0.50 0. 0.67 0.67 0.5 N = 12 4.33 0.00 (1/12) 2. 0.

201–213.’ and Reinforcement Learning in a Market Entry Game.” in Proceedings of the ACM SIGCOMM Workshop on Practice and Theory of Incentives in Networked Systems. Friedman. Glimcher. Daniel and Bernardo Huberman (2004) “Internet Congestion: A Laboratory Experiment. and Amnon Rapoport (1998) “Coordination. Lovallo (1999) “Overconfidence and Excess Entry: An Experimental Approach. 11–18. Fischbacher. Goeree. Tietz. Paul W. 575–584. G. Sethares (1999) “An Experimental Study of the El Farol Problem. Kahneman. Bell. 23 (May). R.” Economic Theory. Van Huyck. W. Raymond C. 146–175. Saving (1992) “History’s Role in Coordinating Decentralized Allocation Decisions: Laboratory Evidence on Repeated Binary Allocation Games. 203–218. et al. ” Quarterly Journal of Economics. Holt (2005) “An Explanation of Anomalous Behavior in Models of Political Participation. Colin. C. Harper. 30 (2). 306–318. presented at the Summer ESA Meetings. Portland. Decisions. 105 (May). John B.” Neuron. 93 (369). 89 (March). 33 (1).” Animal Behavior. and Christian Thöni (forthcoming) “Inefficient Excess Entry in an Experimental Winner-Take-All Market. Erev. Plott. and Charles A. Dylan.. Yan.292 L.R. Daniel (1988) “Experimental Economics: A Psychological Perspective. Turocy (2007) “Congestion Allocation for Distributed Networks: An Experimental Study. 100 (April). References Camerer. James A.” Discussion Paper. Morgan. 99 (2). Note 1 Experiments motivated by internet congestion issues are reported in Chen. Battalio. Anderson et al. USA. Chen. Sundali. Seale (1995) “Coordination in Market Entry Games with Symmetric Players. and Thomas R.. 121–143.” American Economic Review.” American Political Science Review. (1983) “Externalities and Corrective Policies in Experimental Markets. Oregon. eds. 177–182. Decisions: Choosing a Biological Science of Choice. 106–127. 292–316. ‘Magic.” Games and Economic Behavior.. Jack (1990) “The Coordination Problem in Decentralized Markets: An Experiment. (1982) “Competitive Foraging in Mallards: ‘Ideal Free’ Ducks. and William A. Anne M.” Organizational Behavior and Human Decision Processes. 64 (2). Selten.” Journal of Political Economy. 36 (2). Charles R. 223–232. New York: Springer-Verlag. and Darryl A. . Urs. and R. Donald J. 545–559. Jacob K. Amnon Rapoport. Tucson. (2002) “Decisions. Ochs. D. and D.” in Bounded Rational Behavior in Experimental Games and Markets. Meyer. Albers. (2007) and Friedman and Huberman (2004).” Journal of Economic Behavior and Organization. Laura Razzolini and Theodore L.” Economic Journal. Ido.

versus having the government auction pollution permits. When access rights in fishing commons are violated. 1994). they have sometimes been met with negative reciprocal acts resulting in property damage and. we show that individuals hold both reciprocal and distributional preferences over alternative regulatory solutions to social dilemmas in a laboratory experiment. 2004). for example. even homicide. In this chapter. In other social dilemmas. social preferences may be based on unconditional . with serious distributional consequences (Schlager. For example. such as pollution emissions. In the latter case. 2004). Additionally. Laurie T. These controversies often revolve around distributional issues. appropriation rules stipulating equal usage are common. and lotteries have been used to assign rights at some of the most productive fishery spots. In irrigation and fishery commons. 2004. in some cases. In a number of experiments. Ellermann. the rents are transferred from permit holders to the government conducting the auction. The importance of distributional concerns can be seen in several solutions to distributional conflicts that communities devise. subjects have been found to be motivated not only by self-interest. Increased attention concerning distributional consequences may therefore be called for. There is a growing experimental literature regarding what is often referred to as “social preferences. using so-called “grandfathering” allocation schemes (Harrison. 1994). making it harder to reach voluntary equitable agreements. Elisabet Rutström Introduction Economic analyses of regulatory solutions to social dilemmas focus more often on the efficiency than on the distributional consequences.” inspired in part by such distributional conflicts. in irrigation systems. Gregory George. 2002). leading to intensive lobbying by interested parties (Tietenberg and Johnstone.16 Social preferences in the face of regulatory change J. but also by a concern for payoffs to others (Charness and Rabin. Johnson. and E. The allocation of pollution permits has generated controversy. At issue has been whether permit allocations should be made to incumbent firms according to some historic performance and needs. the externality has broader geographical reach and a larger number of involved parties. users are often given equal time slots to extract water (Tang.

Charness and Rabin. preferences over the distributional consequences may reflect not just self-interest or unconditional distributional preferences. This common history is one of a social dilemma. namely.”2 According to this proposal. 1999). but they differ in the distributional consequences. 1993. Policy proposals such as Sky Trust are based on the premise that preferences over the distribution of the scarcity rents of the licenses exist and that voters (and interest groups) care about the way in which the policy solution distributes these licenses. Incentive compatibility is ensured by making distribution choices costly to the individual. carbon emission permits would be sold to companies and the income distributed to US citizens in the form of equal dividends.294 J. The regulation choices offered all guarantee that the social optimum is achieved. JRG investigate social preferences in a laboratory experiment. Participants express some willingness to pay for redistribution independent of the performance of the group during stage one. Rabin. a negative externality game.. 1994. Nevertheless. The design builds on the design in Johnson. and George (2006) (henceforth. Distributional preferences are elicited in a non-interactive setting. . Subjects are offered two choices: the default choice (which is free of charge) is to maintain the income distribution that was generated during the first stage but at the socially optimal aggregate income level. George et al. The propensity to select the egalitarian option is increasing in the tendency for an individual to cooperate during the first stage of the game. but also reciprocity.1 This chapter contributes to this literature by testing how distributional choices depend on whether participants have a common interactive history in a social dilemma. 2002). which cannot be explained based on reciprocity alone. but they may be affected by past interactions. such as in the JRG experiments.G. distributional preferences (Bolton and Ockenfels. where the first stage generates a common interactive history and the second stage elicits distribution choices. Rutström. The inspiration behind the design of the experiment is the proposal for a CO2 tradable permit market that includes a mechanism for redistribution of the scarcity rents. known as “Sky Trust. or on reciprocal motivations (Hoffman et al. 2000. The US Congressional Budget Office (2000) evaluated such a redistribution mechanism as one of several ways in which the government can distribute revenues from permit sales. JRG). Fehr and Schmidt. JRG control for the influence from self-interest and focus on the latter two as motivations for redistribution choices. the observed choice pattern reveals that there may be additional reasons for choosing the egalitarian trust fund based on unconditional distributional preferences. JRG argue that this is consistent with some form of reciprocal preferences. When the agents affected by the regulatory change have a history together. They find that the choices are consistent with some mixture of distributional and reciprocal preferences. In this chapter we extend the JRG experimental design in a way that more clearly allows us to infer distributional and reciprocal preferences separately. The distribution choices are elicited during a regulatory change that introduces solutions to this social dilemma. They introduce a two stage design. and the other choice is a costly alternative implementing an egalitarian income distribution at the same (socially optimal) income level.

there is no need to play it out. which demonstrates the value of removing reciprocal motivations altogether.1 displays the activities and the accompanying costs and earnings. can be implemented at no cost. where permits are auctioned out and the proceeds are redistributed to everyone in such a way that the future earnings distribution is perfectly egalitarian. In addition. Subjects are offered the opportunity to exchange this default for an alternative solution.” and where the rents of the permits are distributed in proportion to past profits. thus allowing us to identify the presence of unconditional distributional preferences. In every period of the game each player chooses from an identical list of discrete activity levels that are increasing in private earnings and external costs. This default solution models a situation where pollution permits are allocated according to “demonstrated needs. Experimental design The experiment in JRG consists of two stages. Subjects are simply paid the earnings that are determined by the regulatory choice. intended to model a public trust fund. . Table 16. Matchings into groups are anonymous and subjects remain in the same group for the entire experiment. A default solution. This motivational pattern could not be inferred from the JRG data alone. where in stage one subjects interact in a negative externality game in groups of six over ten rounds. In particular. We conclude that both reciprocal and distributional preferences play a role. The second stage consists of the introduction of two regulatory solutions to the externality problem. The solutions differ only in the implied distribution of income. We then pool the data from these new experiments with that of JRG and estimate the additional motivation to redistribute that is introduced by reciprocal preferences based on the common history. Since the regulatory choice defines the activity choices of any subsequent interactive game. or public trust funds. but aggregate earnings become unequally distributed across both groups and individuals within a group due to variations in play during the ten rounds. but only at a cost.Social preferences and regulatory change 295 We introduce a new experimental design that rules out reciprocal motivations. reciprocal preferences are stronger in groups that were less successful at cooperating. both constraining activities so that the social optimum is imposed. This alternative implements an egalitarian earnings distribution. keeping other aspects of the experiment the same. The language in the instructions is context neutral with no reference to pollution. All initial positions in the game are symmetric. pollution permits. even though they have to pay to enable such redistributions. unconditional distributional preferences are strongest among low income group members. implying that empathy towards others who are similarly positioned may be an ingredient in social preferences. with future earnings distributed in proportion to each subject’s aggregate earnings in stage one. Lower earning individuals in particular are more likely to take from the rich and give to other poor when they have experienced a relatively uncooperative group.

The social optimum is activity choice “d. First.1 Per period activity table for stage one (payoffs in cents) Activity choice a b c d e f g h i j k l m n o p q r Private earnings 255 265 275 285 290 295 307 326 337 348 365 378 387 390 393 394 395 396 Social cost 10 10 10 10 12 13 18 25 27 29 38 49 57 59 61 62 64 66 Social welfare* 1170 1230 1290 1350 1308 1302 1194 1056 1050 1044 822 504 270 216 162 132 66 0 SW per person* 195 205 215 225 218 217 199 176 175 174 137 84 45 36 27 22 11 0 Notes Only the first three columns were shown to subjects in the experiment. George et al. An essential feature of the design is that an individual’s policy choice in stage two does not affect his or her own earnings in this stage.296 J. only the earnings of the other five group members. Subjects submit a bid that is compared to a randomly selected value from a commonly known uniform distribution. In this table SW is calculated based on every player making the same activity choice. Table 16. * Social Welfare (SW) is defined as private earnings minus social cost for the agent’s activity choice. This is similar to Engelmann and Strobel (2004) and the multi-person experiments in Charness and .3 A bingo cage that contains 100 balls numbered one to 100 is used to select the random value. Those who prefer the alternative are then asked to express their willingness to pay (WTP) in percentage terms of their stage one earnings. all subjects are asked to express whether they prefer the default or the alternative regulatory solution. BDM) lottery mechanism (Becker et al. with a lowest bid of 1 percent and a highest bid of 100 percent. removing any incentives to bid above or below his true value. paying an amount equal to the drawn value. Incentive compatibility is assured since the subject pays the randomly drawn value rather than the bid. If the subject’s bid is higher than (or equal to) this randomly drawn value.G. In the third step.. Bids are restricted to be in percentages of stage one earnings. To make the WTP elicitation incentive compatible the Becker–deGroot–Marschak (henceforth. minus the sum of the social costs of the other five group participants.” The public choice mechanism used in the second stage is based on a three step procedure. one randomly selected subject gets his stated preference implemented for his group if his bid exceeds the drawn value. the bid becomes binding. 1964) is used.

50 $22.74 $2. and that earnings for the random dictator will be the same regardless of which solution he chooses. Group 4 Session 9 1 21.50 $22.00 .50 $22.50 $22.50 $135.00 $22.76 $28. JRG elicit choices from all subjects and select one of the six group members at random at the end of the experiment as the one whose decision will be imposed on the group and the one who will pay for it.4 5 14.4 The first column shows the earnings position a subject attained in stage one.0 5 13. Both of the schemes have the same aggregate group income for all groups ($135).36 $135.3 6 13.50 $22.50 $22.50 $22.50 $22.76 $135. or “grandfathered” distributional choice is populated by randomly selected representative distributions Table 16.50 $22.89 $18.2 2 22.0 $31.3 2 21.6 Total 100. With this design.50 $135. therefore eliminating efficiency as a motivation for choice. Group 3 Session 3 1 23.52 $19.32 $29. and the second contains his share of total stage one group income. The “random dictator” receives a fixed payment equal to what each group member would get if the alternative solution were selected. The third column shows corresponding earnings under the default (grandfathering) scheme. This guarantees that the alternative solution creates an equal income distribution in stage two.Social preferences and regulatory change 297 Rabin (2002). Table 16.00 $22. but it has no first stage game. The alternative solution gives the same earnings to all group members.2a illustrates typical distribution choices with numbers taken from two of the experimental groups as an illustration.2 4 14.0 JRG.44 $19.00 $28. whereas the default gives incomes that are proportional to the incomes earned during stage one. Instead.2a Typical distribution options in stage two Income position of group participant Participant’s share of stage one group earnings (%) Default solution payments Alternative solution payments JRG. so that incentives in stage one and two are commensurate. from highest to lowest. the default.9 6 2.0 Total 100. nor own-income comparisons across the two solutions. revealed preferences reflect neither a subject’s concern over stage two group efficiency.2 3 15.5 3 21.6 Our new experiment is identical to the second stage of the JRG experiment.5 4 17.62 $20.50 $22. and the last column shows earnings under the alternative (trust) scheme.31 $18.5 This income is set equal to ten periods of optimal play in the stage one game.00 $22.29 $30.

as shown in Table 16.50 19.11 29.67 35. from the JRG experiment. all choices over the two alternatives should be based on distributional preferences and not on reciprocity. whereas the . available at: exlab.72 36.2b.30 20. GroupEff is calculated as the combined earnings by all six members of a group divided by the maximum possible $135.bus. Table 16.xls” in the digital archive at ExLab (online.41 26.22 33.82 24.15 37. After subjects are randomly assigned to groups and income positions. Results We combine the data from the new experiments reported here with the data from JRG.26 25.29 31.96 30.51 17.11 35.edu). George et al.10 New experiment applied = YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES Note A complete table can be found in the excel file “distribution tables.75 36.ucf.91 16.G.89 38.41 11.19 23.93 27.298 J.96 34. the second stage is implemented exactly as it was in JRG. The data from JRG is based on 28 groups of six individuals.38 29.42 32.38 25. Each group in the new experiment received an initial income distribution that had been generated by some group during stage one of JRG. since there are no previous actions that can be rewarded or punished.2b Selected distributions for one-stage experiment JRG experiment GroupEff 11.90 37.87 18. By eliminating the common history.22 23.

26 1.10 19. It is measured as the percentage of the maximum possible group earnings ($135) that was achieved in stage one.51 0. The first three rows display the distribution choices made in the one stage experiment.50 22.20 1. Rank varies from 1 percent to 24 percent in our sample with a mean of 17 percent. We use two variables to capture the extent to which groups are cooperative in the JRG data.60 Median Standard deviation 0.33 Two stage experiment Propensity %WTP* $WTP* Rank (percentage) GroupEff GroupVar Rank-GrEff Rank-GrVar 0.00 45.84 4.20 28.60 $WTP 2.40 0. stage 1 and stage 2 results Mean One stage experiment Propensity 0.50 29.00 Note * These statistics are calculated based only on those who actually selected the trust and therefore bid a non-zero amount.3a summarizes results from the two experiments.02 0. namely.10 1.20 38. measures how uncooperative the individual was in stage one. It is measured as the variance in earnings within the group. this variable simply captures the randomly allocated group income level.60 46. how well group members coordinated in their action choices.50 2.88 4.07 1.00 0.00 1. The data from JRG (the two stage experiment) displays a great deal of variation in the cooperativeness of subjects during the interactive game. of stage one earnings within a group (in dollars).50 34.80 82. or efficiency.00 100.49 17.20 8.70 Minimum treatment Maximum 0.05 0.40 7.00 100. for the JRG data.40 3.43 1.11 16.70 27.00 2.00 0.3a Characteristics of subject pool from each experiment: characterization (N = 168).22 1. so it is normalized by group income. Perfect equality would have resulted in uniform ranks of 17 percent.00 1.80 465.Social preferences and regulatory change 299 new data is based on observations of 14 groups also of six individuals.00 8.44 44. .50 4. Rank.00 5.80 4. showing.48 %WTP 44.40 2. GroupEff measures the extent to which the group was able to achieve the goal of cooperation.45 1. It is calculated as the percentage of the group income that the individual received. followed by a summary of the behavior in stage one and the choices in stage two in the two stage experiment. Thus our total sample size is 252.00 37. efficiency.30 0. For the data from the new experiment. The earnings rank of a person within a group.20 11.40 0. Table 16. GroupVar measures the dispersion. GroupEff and GroupVar. Groups where only a few individuals cooperate are not considered to be as cooperative as groups where most or all Table 16.70 24.00 1. or inequality.

33. the distributions of these variables are similar by design. Since the randomly allocated stage one earnings in the new experiments were selected to match those of the JRG experiment. Bachelor’s Degree GPA Age 42% 71% 15% 27 hours $15 61% 2. Rank-GrEff is the interaction between Rank and GroupEff.50 in the social optimum. The data from the new experiment show very similar aggregate preferences for redistribution. The lowest income was $0.G. or 44. or 44.3b Characteristics of subject pool from each experiment: demographic questionnaire responses Macon State College (N = 72) two-stage experiment Male White African American Hours worked per week Hourly pay Highest expected education.00. but also the extent to which the grandfathering allocation results in an unequal income distribution. with a mean of $2. George et al. Thus. Table 16.20 per participant. Coordination is therefore an important property of cooperation.7 GroupEff varies between 11 percent and 38 percent with a mean of 28 percent.50 and the highest one was $12. Controlling for GroupVar.0 19 UCF (N = 84) one stage experiment 55% 73% 7% 13. GroupVar varies between 30 cents and $19.300 J.11. however. Efficiency is therefore quite low.9 28 UCF (N = 96) two-stage experiment 53% 64% 9% 21 hours $9 75% 3. but also the extent to which preferences are based on the degree of inequality in the default choice. the fact that these aggregate numbers are so similar . with an average WTP of $2. This panel shows the bids of the subjects who expressed a preference for the alternative and its resulting redistribution (44 percent of the subjects). Their average WTP was $2. resulting in stage one earnings that average only $6. Figure 16. GroupVar characterizes not only the degree of coordination in stage one. The same is not true in reverse. The coefficient on GroupVar can reflect not only a reaction to group cooperation history. we take GroupEff as a good measure of cooperation history. since the unequal distribution of the default choice in stage two is identical to the final income distribution in stage one.1 shows the bid distributions for the alternative solution in the two experiments. The left panel shows data from JRG.80. A total of 48 percent of participants selected the alternative solution.3 21 members cooperate.5 hours $6 70% 2. even if they achieve the same efficiency level.6 percent of their randomly assigned allocation.00. and Rank-GrVar the interaction between Rank and GroupVar. Since the new experiment is designed to eliminate motivations for reciprocity. We also construct two interaction variables.4 percent of stage one earnings. which can be compared to the maximum possible individual earnings of $22.

seems to indicate that there are no reciprocal preferences present. on the same set of covariates using a tobit specification. The coefficient on the interaction variable is therefore to be interpreted as the difference in effect between the history and no history treatments.10 The coefficients reported in Table 16. By combining the data from JRG with the new data and performing regression analysis we are able to say more about behavioral differences between the two experiments.Social preferences and regulatory change 301 Two stage experiment 30 30 One stage experiment 20 Percent Percent 10 20 10 0 0 1 2 3 4 5 6 7 8 9 Willingness to pay (dollars) 0 0 1 2 3 4 5 6 7 8 9 Willingness to pay (dollars) Figure 16.12 We estimate support for two measures of demand for redistribution: first. where it is used to capture the idea that the factors that cause someone to seek medical care are distinct from the factors that cause the doctor and patient to decide how much to spend. Tables 16.3a and 16. the propensity to select the alternative distribution. We employ a hurdle model specification where the first step is to regress the propensity to bid on a set of covariates using a probit specification.9 This specification is common in health economics. and second. for example.4 are marginal effects. in the conditional regression analysis below we show that there are some significant and interesting differences in behavior across the two experiments. conditional on the choice of bidding. Note Data shown does not include $0. For both measures. and infer possible motivations behind the choice heterogeneity reported in JRG.1 Distributions of bids for the alternative solutions. and the second step is to regress the actual bid.4 reports our regression results.3b report summary statistics of our explanatory variables and Table 16.8 Nevertheless.11 We report coefficients on the interaction variables with the history treatment (the JRG data) as the marginal effect controlling for the same variable’s effect in the no history treatment (the new data). the % WTP expressed for such choices.00 bids. we find a significantly weaker .

** 5% level.302 J.459) –0. We conclude that distributional preferences are not independent of the person’s position in the earnings hierarchy.G. the significant effects on the propensity to demand redistribution from GroupEff is absent in the new data.122) –0.15 (0.65 (0. when the group earnings are generated interactively with the history treatment we find that the demand for redistribution declines in the earnings of the group.019) Notes p-values in parentheses.071 (0.89 (0.044** (0.048 (0.662 (0.13 This is consistent with motivations based .109 (0.82 (0. This implies that highly ranked people are more likely to let other highly ranked people keep their money than are people with low rank.74** (0.004* (0.016) –0.388) –19.019) 0.312) 0. as indicated by the negative significant coefficient on rank.504) 0.087) –5.025) –0.548) 2. Hausmann specification tests do not detect significant endogeneity due to stage one choice variables used as explanatory variables in the stage two regressions.591) 0.010) 0.23*** (0.237* (0.007** (0.390** (0.0002 (0. as seen by the lack of significance on the interaction variable rank*history treatment.010) –0.072 (0.411) –0.026) –2.292* (0.020 (0.784) 10.811) –7. as shown by the lack of significance on GroupEff.144) –4.051 (0. demand for redistribution as the earnings rank of the individual increases. the effect on the demand for redistribution is the same.076) –0. ***1% level.020) 0.495 (0.6*** (0. with no additional effect from the history treatment.090) 0. George et al.913) 0.002 (0.026) 0. Whether receiving a high earnings rank through chance or through one’s own actions during an interactive game. even when the redistribution does not affect his or her own earnings. In contrast.396) –30.018) –0.385) –0.198 (0.113** (0.79 (0.33 (0.021 (0.588) –0.0032*** (0.552) –0.57 (0.126** (0.007) –1. * 10% level. Table 16.661) –0.066 (0.453) 2 – % WTP –9.74** (0.01 (0.11 (0.096 (0.016 (0. This is consistent with motivations that do not reflect reciprocity.8** (0.63*** (0.049) 0.010 (0.110** (0. On the other hand.442) 0.525) –0.479) 20.825) 0.018 (0.100) –0.014) 0.451) –1.443) 20.06*** (0.4 Regression results Variable Rank GroupEff GroupVar Rank-GrEff Rank-GrVar Rank*history treatment GroupEff*history treatment GroupVar*history treatment Rank-GrEff*history treatment Rank-GrVar*history treatment Session2 Session3 Session4 Session5 Inst Age Male White African-American Hours work per week Hourly pay Highest education GPA 1 – Propensity –0.012) 0.000) –5.001) –4.086** (0.195* (0.356) 8.134) 0.004** (0.

We find that distributional preferences depend on the position of the individual in the income ranking. The willingness to reward and punish increases with the extent to which the group failed to cooperate. but in the absence of reciprocal motivations. Lower earning individuals are more strongly affected by the lack of cooperation in the group than are higher earning individuals. we similarly find no significant effects for any of these variables. The aggregate preferences expressed in the two experiments appear to be very similar. These reciprocity effects due to the cooperativeness of the individual do not carry over to the % WTP measure.86). demand for the trust increases as inequality increases (the coefficient on GroupVar*history treatment is +0. They are more likely to take from the rich and give to other poor when they have experienced a relatively uncooperative group. We also see from the positive coefficient on the interaction variable rank-GrEff*history treatment that this decrease in demand weakens as the earnings rank of the individual increases. Nevertheless. and this effect is strongest among cooperative group members. our regression analysis shows that these aggregate numbers are misleading. In regressions using the dollar WTP as the dependent variable. Conclusions We find evidence consistent with both distributional and reciprocal preferences. with almost the same average propensity to select the alternative solution and very similar average willingness to pay. From this it appears as if there are no reciprocal preferences. such as in the no history treatment. This observation is consistent with empathetic preferences in favor of similarly positioned group members. as measured by GroupVar. We also find that reciprocal preferences depend on the relationship between the cooperativeness of the individual and the cooperativeness of the group. The two experiments discussed here were designed to identify social . When motivated by reciprocity.126). we find no evidence that dispersion in group income per se is a motivation to redistribute. In alternative specifications we use the standard deviation in group income or the coefficient of variation and find the same result. has an effect on the propensity for participants to redistribute. Thus. such that lower ranked individuals are more inclined to demand redistribution than higher ranked individuals. All the significant effects on the percentage WTP can be explained by distributional preferences alone. The dispersion in the stage one earnings within a group.Social preferences and regulatory change 303 on reciprocal preferences. This behavior is consistent with a wish to reward individuals who acted cooperatively and to punish those that did not. as shown by the insignificant effects on all variables interacted with history treatment. demand for redistribution decreases with dispersion (the coefficient on GroupVar is −0. Reciprocal preferences are indeed present and they are correlated with the cooperativeness of both individuals and groups. but not on their WTP.

2 US Sky Trust Inc.00. they were shown a table that showed the subject id number of each group member. . and Mark Schneider for assistance in conducting the laboratory experiments. George et al. and that further experiments that test the external validity of our findings may be called for before policy implications are clear. The regression results are robust with respect to the use of these alternatives.2 when they made their distribution decision. preferences under conditions of changing regulations of social dilemmas. i.G. as well as the coefficient of variation. available at: usskytrust. Instead. bus.ucf. Of the 28 groups. the last two columns from Table 16. We caution that our results are based on a context-free experimental design. 3 JRG restrict bids to be in percentages of stage one earnings in order to avoid both house money effects and incentive problems at the boundaries of the BDM value distributions. 5 Observations in Engelmann and Strobel (2004) imply that preferences over efficiency can otherwise confound distributive preferences. 6 Column 3 is simply column 2 multiplied by $135. 4 Subjects did not see distribution tables that looked exactly like those in Table 16.edu.org/ (accessed 26 July 2007). their own payoffs would be $22.50 no matter what they chose. All supporting documentation can be found Online.304 J. Charness and Rabin (2002) also report many cases where subjects tradeoff efficiency and redistribution. our findings should increase the interest in undertaking additional studies of distributional preferences in relation to regulatory change. Peacebuilding. Acknowledgments Johnson thanks the University of Denver Faculty Research Fund and the Political Economy Research Institute at the University of Massachusetts. All subjects in the group saw the same distribution table. Nevertheless. Bids are restricted to come out of stage one earnings so that they do not change the distribution of future earnings. Ligia Pinto. Rutström thanks the US National Science Foundation for research support under grants NSF/IIS 9817518 and NSF/POWRE 9973669. Bob Potter. 26 have a GroupVar of less than seven. and Environment Program of the University of Massachusetts. Unconditional distributional preferences are therefore purely consequentialist. Online. Ryan Brossette. They could therefore infer the redistribution consequences within their groups for each of the other participants and relate this to the outcome from stage one. The remaining two have values of 14 and 19. We find evidence of preferences in favor of redistribution that do not reflect selfinterested earnings maximization. Notes 1 The term “unconditional” refers to the fact that the behavior is unaffected by either evaluations of past interactions or expectations on future ones.e. with 21 of these 26 being less than two. Because of the skewness in this variable we also run our regressions based on the standard deviation of the group income. and the dollar amounts of the two distribution choices. On the same screen we included text reminding them that if they were selected to make the choice. Development. We are grateful to Anabela Botelho.2. available at: exlab. the dollar amount earnings from stage one. 7 The distribution of GroupVar is quite skewed.

G. Evaluating the Tobacco Settlement: Are the Damage Awards Too Much or Not Enough? American Journal of Public Health. American Economic Review. Bolton. The qualitative findings are robust with respect to all of these variations in specification. M. Charness. 2004.M..W.. 45–70. and Marschak.. 857–869. Tietenberg and N. P-values for both tests are 0. JRG test for endogeneity but find none. 2002. Inequality Aversion. respectively.. M. Ellerman. 166–93.. rank. Understanding Social Preferences with Simple Tests. and also the functional forms of the regression equations. and Cooperation.. eds. Tietenberg and N. and Schmidt. In terms of the latter we included the use of censored normal regression for the dollar WTP instead of a tobit regression of the percentage WTP. Tradable Permits: Policy Evaluation. We caution that the marginal effect for the interacted variables Rank-GrEff and Rank-GrVar cannot be interpreted as interaction effects in the sense of cross-partial derivatives since the model is nonlinear. Paris: Organization for Economic Cooperation and Development.34. (2002) for an application. D. Johnstone. We verify separately that the marginal interaction effect is qualitatively the same. using chi-square or Fisher Exact tests. 1964. June. We are able to estimate the interaction effect in a more limited specification of the model using the inteff command in Stata. Quarterly Journal of Economics. D. Tradable Permits: Policy Evaluation.M. We separately estimated the interaction effect of Rank-GrEff and verify the qualitative interaction effects indicated by the coefficient estimate on the interacted variables. In T. 2000. Design and Reform.. A. Behavioral Science.... J. Harrison. Design and Reform. 77–106. M. 92 (6). In T. 226–232. ERC: A Theory of Equity. Fehr. Ex-Post Evaluation of the Reclaim Emissions Trading Programmes for the Los Angeles Air Basin.43. Johnstone. D. 114 (3). See Norton et al.59 and 0. and incorporating the square of the variables GroupEff. 117 (3). Coller. K... and Ockenfels. measuring the dispersion in group income. 13 We caution the reader not to interpret the coefficient on the interaction variables in the probit regression as marginal interaction effects since this is a nonlinear model (see Note 11). The US SO2 Cap-and-Trade Programme. . 9 (3). 90 (1). eds. M.38 and 0. G. as well as the interaction variable between them. 12 We test alternative specifications with respect to measuring an individual’s cooperativeness. 1999. 2004. and Rabin. Wilcoxon Mann-Whitney non-parametric tests of differences in either the dollar amount WTP or the percentage WTP similarly do not show a significant difference with p-values of 0. 2004. 817–869. respectively. however. and Strobel. see McDowell (2003). 94 (4). Quarterly Journal of Economics. Competition. and McInnes. M.H. A Theory of Fairness. Engelmann. 817–868. E.Social preferences and regulatory change 305 8 The propensity to select the alternative solution is not significantly different across the two experiments. Measuring Utility by a SingleResponse Sequential Method. 2002. DeGroot. American Economic Review. References Becker. Jr. 984–989. 9 For a discussion of hurdle models.A. Efficiency. and Maximin Preferences in Simple Distribution Experiments. 10 See Coller et al. (2004) for a discussion of this issue. G. Harrison. Reciprocity and Competition. 11 Since the main explanatory variables are choice variables in stage one. G. Paris: Organization for Economic Cooperation and Development.M. We also ran the main regression model without including the demographic controls as an additional test for endogeneity.

Tietenberg and N.T.. Ann Arbor.. H. Ex Post Evaluation of Tradable Permits: Methodological Issues and Literature Review. 1281–1302.G..G. Walker. In E. and J. K. From the Help Desk: Hurdle Models. S. available at: cbo. L. Gardner. Ai.pdf (accessed 18 September 2007).C... 3 (2). Stata Journal. eds. 181–198. Schlager.. 4 (2). Norton. 247–265. 2003.. .L. 1994. Who Gains and Who Pays Under CarbonAllowance Trading: The Distributional Effects of Alternative Policy Designs.gov/ftpdocs/21xx/doc2104/carbon. McDowell. American Economic Review.. Games. In T. MI: University of Michigan Press. Rabin. Property Rights. Stata Journal. Games and Economic Behavior.. Walker. 83 (5). 346–380. Ostrom. 2000. 2004.. George et al. 1994. Johnson. Tang. Johnstone.306 J. and Common-Pool Resources. US Congressional Budget Office.. Design and Reform. eds. and Johnstone. J. Gardner. C. Ann Arbor.. and Smith. N. 178–184.. eds. and Anonymity in Bargaining Games. and J. Journal of Economic Behavior and Organization.E. Hoffman. Rules. Fishers’ Institutional Responses to Common-Pool Resource Dilemmas. MI: University of Michigan Press. Paris: Organization for Economic Cooperation and Development. A. Incorporating Fairness into Game Theory and Economics. R. Income Distribution Preferences and Regulatory Change in Social Dilemmas. 7 (3). Rutström. Institutions and Performance in Irrigation Systems. E. Wang. K. E. Preferences. 2006.. Online. Tradable Permits: Policy Evaluation. E. 1993.Y.. Ostrom. T. Games. In E. Computing Interaction Effects and Standard Errors in Logit and Probit Models. 1994. Rules. 61 (2). 2004. 9–44. Unpublished Manuscript. and George. M. 225–245. R. and Common-Pool Resources. 103–166. V. E. Tietenberg. Shachat. McCabe.

we investigate making the decision environment and the corresponding incentives clearer to decision makers (i. Fehr and Gächter. 2002) casts doubt on the ability of these instruments to induce polluters to comply with a standard. among participants who understand the concept of Nash equilibrium. the choices of participants are not individually optimal. As a result. we hope participants will not only be able to identify their dominant strategies... Poe et al. recent evidence from the laboratory (e.17 The effects of recommended play on compliance with ambient pollution instruments Robert J. Dickinson. In explaining the environment. 2000. Alpízar et al.e. there is significant inefficiency and inequality under these instruments. Oxoby and John Spraggon Introduction Segerson (1988) has shown that ambient pollution instruments are theoretically able to induce individual non-point source polluters to reduce their emissions to the optimal level.. but also realize the reward and punishment properties of the instrument. recommended play): we explain the environment and the concept of “marginal decision making” carefully to participants. Andreoni et al. 2001) have shown that providing subjects with the ability to punish or reward others results in more efficient outcomes. 2003. For example. One potential explanation for the inability of these instruments to induce individuals to the socially optimal outcome is that subjects do not understand the decision making environment. To the extent that subjects realize this nature of the instrument we may observe more subjects choosing to punish the other members of their group. these instruments induce individually and socially optimal decision making. Cochard et al. 2005. In each of these studies. while the instruments are able to induce the group to the target outcome. 2004. Other studies (e. 2004b.g. The environment investigated in this chapter is of particular interest because it differs from the standard public good environments so often investigated .. However. Oxoby and Spraggon (2005) have shown that.g. Individuals who choose to reduce their decision numbers below the individually optimal level are rewarding the behavior of others in their group while those who choose numbers that exceed this level are punishing group members. This ability to punish or reward other members of the group is implicit in the ambient pollution instrument. Spraggon. This would lead to less efficient and more inequitable outcomes. 2004. With this in mind.

1 In a threshold public good environment. We proceed as follows.e. Indeed. 1995) to explain non-Nash decision making. ambient pollution instruments can be designed so that preferences incorporating altruism and reciprocity lead to Nash play. relative payoff maximization may lead subjects to make choices that differ from Nash.g.. the regulator could explain the form of the instrument: everyone is fined if the ambient level of pollution exceeds the target and therefore everyone should reduce their emissions to the point where their marginal benefit of emitting one more unit is equal to the tax rate. fairness.g. The purpose of this chapter is to determine if this type of recommendation increases compliance with ambient pollution instruments. preferences embodying payoff inequities. efficiency is not improved by much. With enhanced explanations. Experimental design The underlying structure of the experiment is based on the non-point source pollution problem (e. authors typically appeal to alternate preferences (e. We then present our results. our focus is somewhat different. However. making it prohibitively costly to determine how much an individual source is emitting. However. 2003). although there is more compliance. 1999) and decision error (Anderson et al. 2002. the results are mixed: enhanced instructions improve compliance under the tax instrument but not under the tax/subsidy instrument. Oxoby and J. It does not seem reasonable to expect that an environmental regulator would know the socially optimal level of emission for any given firm.J. say. recommended play) at the aggregate level. As a result. or reciprocity. the next part of the chapter presents the experimental environment and relates it to the non-point source pollution problem. and at the individual level. Given our interest in non-point source pollution. We delineate these results by analyzing the effects of enhanced instructions (i. the question posed here is “Does explaining the environment more carefully reduce errors in decision making and lead to more Nash play or do these other preference explanations dominate when subjects better understand the decision environment?” While others have investigated the effects of recommended play on behavior. In these standard public good environments. we assume that the regulator can measure the ambient level of pollution and knows the potential . the work of Croson and Marks (2001) is most germane to our interests. Moreover. 1998.308 R. The chapter concludes by discussing the policy implications of this result and suggests that these results may be due to seemingly minor weaknesses in the instruments from the point of view of standard theory. In this environment. preferences embodying. 1988). Ledyard. Fehr and Schmidt. Spraggon experimentally (see Zelmer. Segerson. by participant type. Croson and Marks show that recommended play increases Nash play when agents are heterogeneous. That is. see Charness and Rabin. Non-point sources emit pollution into the air or water in a diffuse manner. what does seem reasonable is that the regulator could explain the instrument in detail to representatives from each polluting source.

002( xn − xn ) 2 (17. Subjects’ private payoffs are presented in a table and the ambient pollution instrument is presented as a function. Under an ambient pollution instrument. The tax and subsidy rates are chosen so that individual polluters choose the optimal level of emission where the marginal benefit of one more unit of emission is equal to the marginal cost (the tax) they pay on that unit. Both instruments involve a tax if the aggregate decision number (analogous to the ambient level of pollution) exceeds the target. if X ≤ 150. Subjects also have access to a calculator allowing them to determine their payoff from any feasible combination of their decision number and the aggregate decision number of those in their group. subjects choose decision numbers that are analogous to emission levels. 1988).2 The higher a participant’s decision number the higher is her private payoff. There is also a group component to participants’ payoffs such that the higher the aggregate decision number (the sum of decision numbers within a group) the lower the group payoff. Thus. where X ≡ ∑ xn n =1 N .3( X − 150) if X > 150. The tax instrument is presented as ⎧0. The instrument that we refer to as the tax/subsidy instrument also involves a subsidy when the aggregate decision number is below the target. The private payoff function Bn is given by max Bn ( xn ) = 25 − 0.Recommended play on compliance 309 sources of the pollution. Notice that private payoff is maximized when xn = x max .3( X − 150) Tn ( X ) = ⎨ ⎩0. The n maximum decision number can be thought of as the unconstrained emission level in the non-point source pollution context.3( X − 150) Tn ( X ) = ⎨ ⎩0 if X > 150. The quadratic payoff function was chosen for consistency with Nalbantian and Schotter (1997) and for mathematical simplicity. We investigate two ambient pollution instruments. The tax/subsidy instrument is presented as ⎧0. polluters are fined if the observed ambient level of emission exceeds a target level and are (potentially) subsidized if the ambient level of emissions is below the target. if X ≤ 150. To investigate ambient pollution instruments in an experimental environment.1) Where xn is subject n’s decision number and x max is the subject n’s maximum n decision number. the ambient pollution instrument implements the socially optimal outcome as a dominant strategy Nash equilibrium (Segerson.

a subject’s best response function is the same as for the tax/subsidy instrument if the sum of everyone else’s decision numbers is greater than or equal to the target minus the subject’s optimal decision (x max − 75).. We refer to these different types as medium and large capacity subjects. Spraggon (2004a) suggests that both decision error and alternate preferences are important explanations for this non-Nash behavior. In the same environment.1). each with two subjects having a maximum decision number of 100 and two subjects with a maximum decision number of 125.002( xn − xn ) 2 − 0.00 − 0. Since all subjects face this same incentive.3 The tax/subsidy rate (0. under the tax instrument. Oxoby and Spraggon (2005) find much more Nash play among . n However. This choice. Thus. Oxoby and Spraggon.00 − 0. Thus. Oxoby and J. Spraggon is the aggregate decision number (referred to as the group total). for the tax/subsidy instrument. for the tax/subsidy instrument each individual’s best response for any given X is * max xn = xn − 75. n n We refer to the outcome where subjects reduce their decision number from the maximum by 75 as socially optimal as this is the solution to the social planner’s problem (the difference between the individual benefits from emission minus the cost to society of the emission): 4 ⎡ 4 ⎤ SP = max ⎢ ∑ Bn − 0. the Nash equilibrium (where everyone chooses x* = x max − 75) is unique. Note that there is n n also a group optimal outcome under this instrument where all subjects choose xn = 0 and the total payoff to everyone in the group is maximized. 2002.2) Previous experiments in this environment (Spraggon.. determines the exogenous target (150).310 R. 2004b.. respectively.3∑ xn ⎥ . 2005) suggest that subjects do not choose the dominant strategy Nash equilibrium. x4 ) n =1 ⎣ n =1 ⎦ (17. coupled with the number of subjects per group (N = 4) and the form of the private payoff function (equation 17. if the aggregate decision of the other group members is below this level. under the tax/subsidy instrument medium capacity subjects should choose x* while large capacity subjects should choose x* = 50. Therefore.J. ( x1 . We utilize four subject groups.4 For the tax instrument.002( xn − xn ) 2 − ⎨ ⎩0 if X > 150..3( X − 150) max π n = 25.3) was chosen as the emission damage rate for simplicity. if X ≤ 150. each individual maximizes max π n = 25. Under the tax/subsidy instrument. the subject should choose a decision number just large enough to insure that the aggregate decision is equal to the target.3( X − 150). each individual maximizes ⎧0.

presumably in an attempt to achieve the group optimal outcome. For example.2) as a percentage of the difference between the optimal and minimum possible value of the Social Planner’s problem.5 Primarily. When subjects are given instructions that include a description of the marginal analysis they are much more likely to choose the dominant strategy Nash decision. which was expanded to include the marginal benefit from increasing the subject’s decision number by one for each decision number and an explanation of “marginal decision making. suggesting that decision error may be more important in explaining non-Nash behavior than alternate preferences. Efficiency is measured as the difference between the optimal and actual value of the Social Planner’s problem (equation 17. This definition of efficiency accounts for not only differences between the group total and the target. and a treatment with enhanced instructions. Each experiment consisted of 25 periods. the differences between the instructions lie in the description of the payoff function.Recommended play on compliance 311 subjects who have had a course in game theory. Data and method of analysis The data was collected from eight sessions conducted at the University of Calgary in the winter and fall of 2003 and two sessions conducted in the winter of 2005. if two large capacity subjects each chose 75 and two .” The following is the relevant part of the enhanced instructions: The purpose of the Group Payoff is to insure that everyone chooses a certain Decision Number. but also for reductions in total payoff due to subjects reducing their decision numbers by more or less than is individually optimal. If this is the case then providing subjects with a better explanation of the environment should result in more Nash decision making. Notice that by increasing your Decision Number by one you increase your Private Payoff by the number given in the third column of Table 1. However. As a result you maximize your Total Payoff by increasing your decision number to the point where increasing your decision number by one more will increase your Private Payoff by less than 0. by increasing your Decision Number by one you reduce the Group Payoff by 0. Moreover. Participants were recruited from the general university population. Results The results from the experimental sessions are striking. Subjects were also provided with hypothetical numerical examples and a question to test their understanding. Sessions took approximately an hour and a half and average earnings varied between (Canadian) $10 and $25. they are also much more likely to choose numbers that are below their Nash decision.3.3. In our experiments we conduct two treatments: a standard instruction treatment.

1 shows the aggregate decision numbers and efficiency calculated at the means of session means.036) [5] 91.18 (10. we follow Anderson et al.017) [5] 87. Spraggon medium capacity subjects each chose zero.013) [5] Mean group efficiency Note Standard errors are provided in parenthesis and number of observations are provided in square brackets. We begin by discussing the results at the aggregate level. .45 Upper bound 127.09 (0.20 42. Note that the mean aggregate decision number is closer to the target with the enhanced instructions for the tax instrument but further for the tax/subsidy.3472 for the tax/subsidy and p = 0. (1998) in assuming that errors in decision making result in a distribution of decisions. Analysis at the aggregate level Table 17.19) [5] 205. these distributions are reasonably normal (subject to the constraints of the decision space).69) [5] 183.48 229.15 90.00 154. The group totals are not significantly different from each other using either analysis of variance (p = 0.1356 for the tax) or the Mann-Whitney U-test (p = 0.68 222.3205 for the tax/subsidy and p = 0. While this data is not independent. The efficiency results are consistent with the group total results (although the enhanced instructions improve efficiency for both treatments).312 R.6 Finally we look at the distributions of individual decisions.18 (0. Oxoby and J. These statistics are independent for the analysis at the aggregate level and we have five observations in each cell. As it turns out.30 (8.59 (0.J.1 Mean aggregate decision numbers by treatment Treatment Mean group total 96. the group total would be 150 but the efficiency of this outcome would be only 89 percent. Again the Table 17. Means are calculated for each group of four subjects and the mean of these means is calculated for each treatment cell.26 211. This data is also independent.07) [5] 133.026) [5] 89.94 (11. We also compare the data by participant type (medium and large capacity) across the different treatments.77 (0.59) [5] Confidence interval Lower bound Tax/subsidy enhanced instructions Tax/subsidy standard instructions Tax enhanced instructions Tax standard instructions 66.1172 for the tax).50 (32.89 181.

4577 for the tax/subsidy and p = 0. tax instrument.Recommended play on compliance 313 Tax/subsidy standard 600 500 400 300 200 100 0 0 5 10 Period 15 20 25 Tax/subsidy enhanced Figure 17.2 Mean group totals by treatment and period.1 and 17. tax/subsidy instrument.4700) or the Mann-Whitney U-test ( p = 0. Notice that the mean Tax standard 600 500 400 300 200 100 0 0 5 10 Period 15 20 25 Tax enhanced Figure 17. differences are not significant using standard parametric tests (p = 0.3472 for the tax/subsidy and p = 0.3 show the differences in the aggregate decision number and efficiency through time for the tax/subsidy instrument.3472). .1 Mean group totals by treatment and period. Figures 17.

85 0.95 0. tax/subsidy instrument. . tax instrument.4 Mean efficiency by treatment and period. Tax standard 1 Tax enhanced 0.9 0.95 0.3 Mean efficiency by treatment and period.Tax/subsidy standard 1 Tax/subsidy enhanced 0.9 0.8 0 5 10 Period 15 20 25 Figure 17.8 0 5 10 Period 15 20 25 Figure 17.85 0.

3).2 and 17.5 36. . notice that the median decisions in both the enhanced and standard instruction treatments are below this prediction. group total is typically closer to the target under the enhanced instructions with the standard instructions but efficiencies are higher during the early period with the enhanced instructions.10) [5] Confidence interval Lower bound Enhanced instructions.93 32. Median is the median of medians by group.99 Upper bound 42. Analysis by participant type Recall that x* = 50 and x* = 25 are dominant strategies for large and medium n n capacity subjects facing the tax/subsidy instrument.05 8. For large capacity subjects under the tax/subsidy instrument (Table 17.88) [5] 35. Further.88 (4.98 11. large capacity Standard instructions. Similar results are observed for medium capacity subjects although decisions are somewhat above x* under the standard instructions. medium capacity 15.59 (4.Recommended play on compliance 315 Table 17.79 51. the results are more consistent with the enhanced instructions.13 6. Figures 17.4 show that group total and efficiency under the tax instrument are better in the treatment with the enhanced instructions during the early periods but are very similar in the later periods.2) decisions in the enhanced instruction treatment are significantly below the Nash prediction.2 Mean aggregate decision numbers.2 shows that these predictions are not as consistent with the Nash prediction as the results regarding aggregate decision making might suggest. large capacity Enhanced instructions. These results suggest that.44 25.65) [5] 31. medium capacity Standard instructions.69) [5] 19.5 26 Median decision Note Standard errors are provided in parenthesis and number of observations are provided in square brackets. However. Table 17. providing more information to subjects does not significantly affect the efficiency of either instrument. n Under the tax instrument.5 20.03 (9. at the aggregate level.72 (7. decisions are much higher than the prediction for both the large and medium capacity subjects under the standard instructions and medium capacity subject under the enhanced instructions (Table 17.12 61. under the tax/subsidy by treatment Treatment Mean decision 28.

01.46) [5] 35. Figure 17. Under the tax instrument subjects seem to choose slightly larger decision numbers.7 present the distributions of individual decisions by subject type and instrument. < 0.50 (5. medium capacity Standard instructions. respectively).01 for the Kolmogorov-Smirnov test for large and medium capacity subjects.01 for the Mann-Whitney test. large capacity Enhanced instructions. < 0.68 57.24 52.12 (2.6 provides the distributions for the tax/subsidy instrument.55 Upper bound 60.6 and 17. Spraggon Table 17. Under the tax instrument the decisions of large and medium capacity subjects are different. explaining the marginal nature of the decision making environment results in participants recognizing the collusive outcome under the tax/subsidy and choosing lower decision numbers as a result. p < 0.33 29.7 shows that the story is different for the tax instrument. Oxoby and J. large capacity Standard instructions.0. although they are above the Nash predictions. In summary. Notice that subjects seem to focus on the group optimal decision (zero) in all cases except the large capacity subjects under the enhanced instructions.95 50.09 (1. Note that for the tax/subsidy under both the standard and enhanced instruction treatments decisions are very similar between the large and medium capacity types.53 (5.46) [5] 43. Analysis by participant Figures 17. The enhanced instructions focus both the large and medium capacity subjects on the optimal decision.93 65. but this effect is somewhat mitigated by the enhanced instructions.316 R.5 depicts the time series of average decision number by treatment and subject type for both the tax/subsidy (T/S) and tax instruments. the distributions are significantly .75) [5] 59. Again. under the tax by treatment Treatment Mean decision 56. Median is the median of medians by group.5 49. Figure 17.3 Mean aggregate decision numbers.04) [5] Confidence interval Lower bound Enhanced instructions.7 Figure 17.011 for the Median test and p < 0. medium capacity 51. < 0.J.5 Median decision Note Standard errors are provided in parenthesis and number of observations are provided in square brackets.01.51 50 60 27.39 20. These distributions are significantly different using standard non-parametric tests (p < 0.

T/S standard instructions Large capacity 150 125 100 75 50 25 0 0 5 10 Period 15 20 25 Medium capacity T/S enhanced instructions Large capacity 150 125 100 75 50 25 0 0 5 10 Period 15 20 25 Medium capacity Tax standard instructions Large capacity 150 125 100 75 50 25 0 0 5 10 Period 15 20 25 Medium capacity Figure 17. .5 Mean decision by subject type and period. tax/subsidy instrument.

tax/subsidy.1 0 100 110 120 130 10 20 30 40 50 60 70 80 90 0 Decision Figure 17.6 0.3 0. .4 0. large capacity 0.4 0.2 0.5 Fraction 0. large capacity 0. Standard instructions.5 Fraction 0.6 Distributions of individual decisions.2 0.6 0.1 0 100 110 120 130 140 140 150 150 10 20 30 40 50 60 70 80 90 0 Decision Enhanced instructions. by treatment.5 continued.Tax enhanced instructions Large capacity 150 125 100 75 50 25 0 0 5 10 Period 15 20 25 Medium capacity Figure 17.3 0.

3 0.1 0 100 110 10 20 30 40 50 60 70 80 90 0 Decision Figure 17.3 0.4 0.6 0.7 Distributions of individual decisions.1 0 100 110 120 120 130 130 10 20 30 40 50 60 70 80 90 0 Decision Figure 17.2 0.6 0.2 0.2 0.5 Fraction 0.4 0. by treatment.1 0 100 110 120 130 140 140 140 150 150 150 10 20 30 40 50 60 70 80 90 0 Decision Enhanced instructions. tax. medium capacity 0. Standard instructions.6 continued.5 Fraction 0.Standard instructions. . medium capacity 0. large capacity 0.3 0.5 Fraction 0.4 0.6 0.

1 0 100 110 120 120 130 130 10 20 30 40 50 60 70 80 90 0 Decision Enhanced instructions.6 0.5 Fraction 0.4 0.5 Fraction 0.1 0 100 110 10 20 30 40 50 60 70 80 90 0 Decision Figure 17. medium capacity 0.4 0.5 Fraction 0. .6 0.3 0.7 continued. medium capacity 0. large capacity 0.3 0.3 0.2 0.2 0.1 0 100 110 120 130 140 140 140 150 150 150 10 20 30 40 50 60 70 80 90 0 Decision Standard instructions.2 0.Enhanced instructions.4 0.6 0.

2005. < 0. Combined with our results on the efficiency and compliance gains of recommended play in conjunction with these instruments. one could consider firm-specific tax rates based on firms’ differing costs or damage functions. Poe et al.9 Second. Standard non-parametric tests for these periods are also consistent with this contention except for large capacity subjects under the tax instrument for which the Mann-Whitney and Median tests do not suggest a significant difference. p < 0. and participants from the annual meeting of the Economic Science Association (Tucson.01) for the Kolmogorov-Smirnov test for large and medium capacity subjects. the watershed). This may be appropriate for non-point source pollution problems as a firm’s damage (for a fixed level of emissions) depends on distance from the resource in question (e.. Acknowledgments We thank Jim Murphy. two issues regarding these instruments should be considered. 2004.01... the instruments’ information requirements (on the part of firms and policy makers) are extreme as optimal tax/subsidy rates depend on both firms’ costs and the environmental damages they impose.01 for the MannWhitney test. First. following Segerson (1988).01. 2004b) suggests that these instruments are remarkably efficient at reducing the aggregate emission level to the target. the experimental evidence (Cochard et al.01 for the Median test and p < 0.0909. Providing a better explanation of the instrument to the participants does not necessarily result in more compliance at the aggregate level.. if firms are competitors in product markets. These distributions are unchanged when only the data from periods ten to 20 is considered. 2004. < 0. 2004b) are likely due to a combination of decision errors and strategic play in which participants attempt to take advantage of the group nature of these instruments. John Stranlund. Spraggon. but also increases the fines paid by competitors without reducing a corresponding reduction in their abatement costs. 2005. < 0. In terms of implementation. respectively). 2003) and the Experimental . Spraggon. 2002).Recommended play on compliance 321 different using standard non-parametric tests (p = 0. an ambient pollution instrument may provide an additional tool for competition: increasing emissions reduces a firm’s abatement costs and increases its fine. Despite this.10 These reductions in efficiency are primarily due to differences from optimal behavior at the individual (rather than aggregate) level.g. This may be critically important for pollution problems where there are a large number of polluters and it is financially infeasible to monitor effectively.8 Conclusion The policy implications of these results are clear. 2002. However. This suggests that the reductions in efficiencies observed in previous empirical studies of these instruments (Cochard et al. Poe et al. this provides reason to be optimistic about the ability of these instruments to mitigate not only the non-point source pollution but also other group moral hazard problems (Spraggon.

Collective versus random fining: an experimental study on controlling ambient pollution. Quarterly Journal of Economics. 2 The instructions are presented in neutral language to abstract from the subjects’ feelings about pollution. This outcome was only observed among one group of inexperienced subjects with the standard instructions.004 on the dummy variable for the standard instructions in the regression with dummy variables for tax versus tax/subsidy. 36–63. Spraggon Economics and Public Policy Workshop (Appalachian State University. P. 2003. Weersink et al.0001.002. and the four interactions). We thank Kendra N. Rabin. 7 A Tobit regression also suggests that the enhanced instruction treatment matters (p = 0. 0. but we feel that the non-parametric tests are more appropriate due to the non-normality of the data. Andreoni. 1998. and W. Journal of Public Economics. and C. 2002.000 for the tax instrument. J. 5 All instruction sets are available upon request from the authors. 2005) for helpful comments. 93 (3).0022. 817–869. but not independent for different subject types (or different subjects) within the same treatment. and C. The carrot or the stick: rewards. 10 Poe et al. Equilibrium selection in experimental games with recommended play.138. S. . F. 0. 1995.000.056. 2004. Upjohn Institute. Median and Kolmogorov-Smirnov tests for large and medium capacity subjects respectively are 0. 231–251. American Economic Review. Brandts. Environmental and Resource Economics. Games and Economic Behavior. Vesterlund..000. J. MacLeod.. W.322 R. Holt. J. 1992. 0. 9 Indeed. B. 70 (2). A theoretic analysis of altruism and decision error in public goods games.4618. Goeree.000 for the tax/subsidy and 0.E.J. capacity. 117 (3). K. Financial support for this project was provided by the Social Science and Humanities Research Council of Canada and the W. American Economic Review. Understanding social preferences with simple tests..0084. 0. A. 0. and Oxoby and McLeish (2004). Charness. Holt. punishment and cooperation. 0. 29 (2). A. 0. (2005).2 percent) when a simple market is combined with the ambient pollution instrument. 8 p-values for the Mann-Whitney. G. and A. These experiments involved a simpler environment where subjects choose between zero and 20. 893–902. McLeish for valuable research assistance. and M. Requate. References Alpízar. Schram. 0. Notes 1 Other experiments using recommended play include Brandts and Holt (1992). Anderson. J. 0.000. and L. 297–323. Brandts and MacLeod (1995). 3 The instrument is presented to subjects in this way to be clear about the instruments’ dichotomous nature. 6 Means calculated for a subject type (or a subject) are independent from the means calculated for the same subject type (or different subjects) across treatments. 11 (1). Oxoby and J. (1998) argue (although not theoretically) that these instruments are only appropriate for homogeneous firms. (2004) finds a much lower level of efficiency (56. 4 This outcome was achieved in the experiments of Cochard et al. 82 (5). Harbaugh. instructions. T. 0. Brandts. An experimental test of equilibrium dominance in signaling games. 1350–1365.

Spraggon. 6 (3). Exploring the performance of ambient-based policy instruments when non-point source polluters can cooperate.. Livernois. 24 (3). Kagel and A. Dickinson. Schmidt. Suter. J.. K. 39 (2). E. F. A theory of fairness. Experimental Economics. 2001. Game theory for playing games: bounding rationality in a negative externality experiment. E. J. Poe. Oxoby. Spraggon. Public goods: a survey of experimental research. Cooperation and punishment in public goods experiments. Fehr. 90 (4). 238–249. J. O. F. 427–456. 48 (2). 87 (3). 2004b. 7 (3). 2003. H. American Economic Review. 2005. 107–124.. J. and K. Uncertainty and incentives for nonpoint pollution control. 2002. and A.). 84 (3). and J. Fehr. R. 817–868. J. McLeish. American Journal of Agricultural Economics. Handbook of Experimental Economics. D.. 1997.. Journal of Environmental Economics and Management. and cooperation. Individual decision making in a negative externality experiment. 2004. The carrot vs. W. and J. Roth (eds. A. Segerson. Zelmer. Experimental Economics. Journal of Environmental Economics and Management. Shortle. J. J. 2004a. 15 (1). Ledyard. Gächter. Schotter. N. J. 2000. 2001. E. 1988. Experimental Economics. and M. Journal of Public Economics. Economics Letters. 1999. and S. L. Marks.. Linear public goods experiments: a meta-analysis. Croson. Testing ambient pollution instruments with heterogeneous agents. competition. 837–856. . Spraggon. Manuscript. 249–269. K. 299–310. Nalbantian. Canadian Public Policy. H.. 84 (2).. 86 (15). Productivity under group incentives: an experimental study. 30 (14). Willinger. Economic instruments and environmental policy in agriculture. A. Lakehead University. New Jersey: Princeton University Press. and K. 2005.. In J. Exogenous targeting instruments as a solution to group moral hazard. Oxoby. 399–405. Quarterly Journal of Economics. The effect of recommended contributions in the voluntary provision of public goods. 87–98. and C. 980–994. 2004. J. 111–194. American Economic Review. G. Economic Inquiry. 4 (1). Spraggon. Weersink. 1998.. M. R. Segerson. R. L. Schulze.Recommended play on compliance 323 Cochard. 314–341. Shogren. Xepapadeas. 393–422. 1203–1210. 1995. 114 (3). Environmental and Resource Economics. and A. D. 309–327. Efficiency of nonpoint source pollution instruments: an experimental study. the stick in work team motivation. Vossler. J. Specific decision and strategy vector methods in ultimatum bargaining: evidence on the strength of other regarding behavior.

and experimenting with alternative policy approaches through real world implementation can be costly in terms of time. the chapters in this part of the book describe a series of laboratory experiments designed to enhance our understanding of behavior in general. The chapters on regulation and compliance in this volume provide additional evidence in support of this view. in order to have confidence that the real world response to a . for reasons described below. Laboratory experiments of the type described in these chapters provide a very useful middle ground for investigating the likely impacts of proposed policies at lower cost. For policy instruments that are innovative and have never been tried before. when the penalty was put in place.” as in the example above.e. policy interventions to control externalities are a fundamental part of this field. However. the outcome of the experiment was exactly the opposite of what theory predicted.e. the number of late pick-ups increased instead of decreasing. which parents were apparently much more willing to bear.18 Discussion Regulation and compliance Kathleen Segerson In a recently published popular book entitled Freakonomics. money. i. analysts have two tools for investigating the effects of alternative policy options: (1) economic theory. Levitt and Dubner cite two reasons to explain this unexpected outcome: (1) the magnitude of the penalty that was imposed was too low (only $3). Without laboratory experiments. However. and (2) empirical analysis of actual real world policies that have been implemented. authors Levitt and Dubner (2005) describe an experiment in several Israeli day care centers in which a fine was put in place to try to discourage parents from being late in picking up their children. and in some cases political capital. or the provision/disclosure of information). Since environmental economics is at its core the study of externalities. the associated guilt) with an economic penalty. and (2) the day care had replaced a moral incentive or penalty for being late (i. but more specifically behavioral responses to environmental policy interventions (such as taxes on ambient pollution or congestion. the predictions of theory can be “wrong. Collectively. Clearly. it highlights the important role that experiments can play in testing theory and the effectiveness of alternative policy interventions in a relatively low cost way. Although this was a field rather than a laboratory experiment. economic theory is the only analytical tool available.1 Economic theory predicted that a penalty for late pick-ups would reduce the number of late pick-ups.

behavior can also be motivated by moral or social incentives. Similarly. it seems imperative that the experiment be conducted (albeit on a small scale) in a field environment that mirrors the real world context in which the policy might be used.e. Evans et al. they can differ in the nature of the hypotheses that the experiments are designed to test. and in some cases (e. consider whether individuals hold “social preferences” and respond to motives other than self-interest when making decisions. is an example of the former. Even when experiments focus on evaluation of alternative policy approaches rather than underlying preferences. for example. a laboratory experiment of the penalty for late day care pickups. However. especially if conducted with university students (including economics majors) who have little or no experience with day care. can play a key role in generating this type of information. While the results of such experiments can have important implications for the design of environmental policy.g. most economic analyses assume neoclassical preferences under which agents respond primarily to economic incentives (i. Experiments such as those conducted by George et al. these experiments do not show how other motives are likely to influence choices in a particular context. appeal to the “gambler’s fallacy” as an explanation of observed results that are inconsistent with (neoclassical) theory. After all. while they may illustrate the existence of motives beyond self-interest and the potential for these motives to be important in some contexts. the day care example) these incentives can dominate. Information about preferences is fundamental to all policy analysis. Some experiments are simply designed to teach us more about people’s preferences and how they make decisions. Both illustrate why it is important to understand the extent to which preferences driven by non-economic considerations affect choice. As Levitt and Dubner (2005) note.Discussion 325 policy intervention will mirror the response observed in the laboratory. They show that individuals have both reciprocal and distributional preferences and are motivated by empathy. There are at least three interrelated issues that arise in assessing the usefulness of laboratory experiments in understanding environmental policies. the authors (appropriately) caution that their results are based on a context-free experimental design. while the other four chapters all fall into the latter category. However. The first is the purpose of the experiment. It seems likely that non-economic motives are context specific. standard rewards and punishments). For this.e. George et al. since. the chapter by George et al.2 Thus. the context-neutral setting in the laboratory must be replaced with a more realistic consideration of the specific attributes and types of decision makers that characterize the particular context of interest. the experiments themselves focus on the nature of preferences rather than the incentives created by alternative policies. These differences . In this part. while others seek insight into specific policy instruments or interventions. one might expect perceived moral obligations to hinge on context. might very well have shown a response to the imposition of the late fee consistent with the (incorrect) predictions of theory. i. what we hope to learn from it.

Spraggon. they find that while on average observed entry levels achieve the target. Anderson et al. In practice. Of course. environmental quality. They ask whether voluntary discovery and disclosure policies are efficiency enhancing and not whether they achieve a given target. Oxoby and Spraggon ask whether various ambient pollution instruments achieve a first best. increases. Since most theoretical models abstract from many real world complications and in practice the first best will be extremely difficult to know. in evaluating whether experimental results are consistent with theory. as predicted by theory. I think it is perhaps overly optimistic to conclude from this that it would likely lead to a first best outcome in practice as well. If it is shown to be effective in a laboratory experiment. such as an efficient outcome or a predicted equilibrium outcome. policy design.e. and the implications of this for enforcement costs. This result is consistent with previous experimental work on the efficiency of ambient pollution instruments (see.e. Evans et al. while others seek to test whether the mechanism achieves a given outcome predicted by theory. also compare observed outcomes (levels of highway entry) under a congestion tax to the efficient outcome to determine whether this policy instrument is able to achieve a first best. In contrast. For example. Likewise. it is easier to show that a given policy instrument moves the equilibrium in a desired direction (i. Interestingly. Clearly. the chapter by Murphy and Stranlund simply seeks to test how a specified audit policy with voluntary disclosure affects (i. the first best). compare cheating levels observed in their experiments to those predicted by theory and show that observed rates are sometimes above and sometimes below predicted levels. there is then the question of whether these weaker tests are nonetheless useful for environmental policy design. Segerson constitute a second issue of interest. in fact. In contrast. leading to inefficiency. This would imply that the incentives we seek to create through the policy are.326 K. I would want first and foremost to know if a given policy mechanism is at least effective. As an environmental economist interested in policy design. enhances efficiency) than to show that it reaches a specified target (e. it is perhaps both too much and unnecessary to ask of any given policy instrument that it moves the equilibrium to the first best. decreases. . there is considerable variability around the target. 2002). perhaps we should be reassured when a given instrument at least has the predicted effect rather than the opposite effect as in the day care example described above. or leaves unchanged) a firm’s decisions regarding disclosure and care under different assumptions about the cost of self-audits. then there is hope that it would also be effective if put into practice (especially if the experiments were context specific). “working” in the sense of inducing the type (if not the exact magnitude) of behavioral response that is desired. while it would be nice to know that in addition the mechanism induced an efficient outcome in the laboratory. for example. and. ultimately.g. Some experiments are designed to test simply whether a given policy is “effective” in the sense of moving the equilibrium in a particular direction.

This is particularly true if moral or social incentives play a role in one context but not another. this could simply mean laboratory experiments that (1) use subjects drawn from the population of decision makers who would actually be faced with the incentives that are being tested. The chapters by Evans et al. the experiments in Murphy and Stranlund are couched specifically in the contexts of enforcement and compliance.g. 2006). is that the resulting experiment might also abstract from context effects that could be very important in practice. Stripping the experiment of context has the advantage of testing more fundamental principles. Although the Evans et al. For example. real world implementation of policies that “worked” in the laboratory may turn out to be very costly indeed. experiments seek to test cheating incentives. it is not clear whether a group of neighboring farmers would respond to alternative ambient pollution policies the same way that unrelated undergraduates in a laboratory would. The provision of public goods such as environmental protection is a context in which these considerations might play an important role for some people. and (2) are couched in the context of the real world setting in which their behavioral responses would occur (e. Similarly. Oxoby and Spraggon also use neutral language in their experiments to “abstract from the subjects’ feeling about pollution. thereby allowing for more generalization.. claim that their experimental results could provide insights into financial disclosure and tax compliance. Previous work on instruments to control ambient pollution has also been context specific (e..” In contrast. . George et al. of course. Without this crucial step. in their experiments George et al. Short of field testing. it seems likely that at least part of the objection of some environmentalists to the use of emission taxes is the view that. Likewise. is the role of context. Many experimental researchers go to great lengths to be sure that their experiments are described in “neutral” terms and hence devoid of specific context that might influence the outcomes. despite the fact that their work is motivated by a proposal for the distribution of carbon emission permits..3 The disadvantage. there appears to be a fundamental tradeoff. by creating a “license to pollute. use of neutral language avoids framing effects that could be specific to the laboratory. For example. In addition. pollution control).g. Vossler et al. Given this. Evans et al. it seems that it would be risky to move directly from laboratory results using undergraduates placed in a context neutral setting to implementation of a given policy approach in practice. the authors specifically avoid framing decisions as cheating or malfeasance to avoid payoffs associated with the ethical costs of cheating. The step that seems to be essential but currently missing from most experimental work is the “testing” of results in a context that more closely mirrors the real world context in which the policy would be used. do not make any reference to pollution. alluded to above.Discussion 327 The final issue. When deciding how much context to provide. and Oxoby and Spraggon are examples.” these taxes remove any moral stigma that might otherwise be attached to releasing pollution into the environment.

emergency room care) but no moral obligation to provide equal access to other goods (e. John.g.” Journal of Public Economics 84(2): 427–456.328 K.” Economic Inquiry 44(4): 599–613. 2000. “A Fine is a Price. Segerson Notes 1 See Gneezy and Rustichini (2000) for the original study. “Exogenous Targeting Instruments as a Solution to Group Moral Hazards. Christian A. Levitt. Steven D. 2002. luxury cars). some individuals might feel that society has a moral obligation to provide equal access to some goods (e. and Aldo Rustichini. and Kathleen Segerson. Gregory L. Freakonomics: A Rogue Economist Explores the Hidden Side of Everything. Uri.g. 29(1) (January): 1–17. “Communication and Incentive Mechanisms Based on Group Performance: An Experimental Study of Nonpoint Pollution Control. William D. 2005.. Poe.” Journal of Legal Studies. References Gneezy. 3 Murphy and Stranlund attribute one of their results that is inconsistent with theory to a framing effect. 2 For example. New York: HarperCollins. Vossler.. which they claim is likely to be limited to the laboratory. and Stephen J. . Spraggon. 2006. Dubner. Schulze.

Part IV Valuation and preferences .

.

g. We are able to induce preference reversals through the use of anchoring cues. in which subjects favor one object in a set when preference is elicited via one method (e. time-differentiated payments or interpersonal income distributions may simply expose the tenuousness of . are well documented. though more evidence is emerging from studies with time differentiated payments or with payments involving issues of equity and fairness across several recipients (see Seidl (2002) for a thorough review of preference reversals). we look for experimental evidence of preference reversal behavior in a static choice setting. Perhaps of greater interest. we focus on one particularly troubling individual decision making anomaly: preference reversals. consistent and systematic choices toward certain and risky events. or act as if they make. Shogren (2005) puts the problem succinctly: “Relying on rational theory to guide environmental valuation and policy makes more sense if people make. J. Traditional environmental economic policy analyses have relied almost exclusively on the assumption that people are capable of and in fact do make rational decisions. In an unpublished draft manuscript.g. choice among alternatives) but favor another object when preference is elicited via another method (e. we identify an asymmetry in these static reversals that cannot be explained by the traditional reversal explanations posited in the risky choice literature. thereby demonstrating the rank instability of preferences in a static choice setting. In an attempt to broaden our understanding of preference stability. Preference reversals. Most evidence of reversals comes from studies of preferences over lotteries with similar expected payoffs. inferred from auction bids). The experiment is formulated in a simple labor market context where subjects are asked their preferences for performing menial tasks and then later asked to formulate bids representing the minimum willingness to accept to perform the tasks subject to an incentive compatible elicitation mechanism.19 Preference reversal asymmetries in a static choice setting Timothy Haab and Brian Roe The economic analysis of environmental policy requires a thorough understanding and appreciation of the ways in which people make decisions towards certain or risky events. Background Evidence from studies of choice over gambles.” Here.

SR for static reversals) utilize a within-subject design. while many studies performed in richer choice contexts (hereafter GR for gamble reversals) feature within-subject design. low stakes gamble is called the P-bet. but average bid prices are higher for the larger set when both sets are evaluated jointly. coherent preferences that allow for intransigent orderings over consumption bundles regardless of the procedure used to elicit them. Other evidence of reversals in static situations has been documented in the psychology literature (see Hsee et al. which leaves unchallenged the base assumption that preference orderings are stable though cardinal representations of these may not be unique. They find considerable arbitrariness in terms of cardinal preference structures but considerable robustness of preference orderings. A common explanation focuses on anchoring and adjustment. When faced with formulating dollar bids that may determine if the subject actually plays the gamble. The bids are elicited from sports card enthusiasts and dealers spending their own money in a real market and the goods in question did not involve any of the complications outlined above. A smaller though intriguing trove of evidence is emerging that documents similar reversals in simple. The typical setup involves allowing respondents to choose between two such lotteries where the low odds. Roe extending a static preference theory to richer choice environments rather than problems with the axioms girding static choice theory. though few of these studies feature market pressures and realities like the List or Ariely et al. Because of its market setting. Ariely et al.332 T. later in the experiment. joint comparison are much more likely to reverse themselves than those who prefer the $-bet when. Those who prefer the P-bet in a direct. List (2002) reports that average bid prices for a smaller. Haab and B. 1979). Reversals tend to occur for only one subset of respondents: those respondents who prefer gambles with high odds and low stakes to gambles of equivalent expected payouts but low odds and high stakes. subjects anchor on the amount of the potential payout because is it also demarcated in dollars. static settings that more directly contradicts the core assumptions of utility theory: individuals are endowed with stable. Because the $-bet involves a higher dollar figure than the . Hence the GR literature reports inconsistency within the same person while SR studies report reversals on the aggregate. conclude that preferences are arbitrary but coherent. Ariely et al.1 Extensive analysis of within-subject reversals from the GR literature uncovers an unusual pattern within the anomaly. studies. high stakes gamble is called the $-bet and the high odds. nested set of sports cards are higher than for the full set when bids are elicited for each set separately. For example. many of the critiques lodged by economists against preference reversal studies are clearly avoided (Grether and Plott. (2003) report several experiments in which they elicit subjects’ willingness to accept compensation to experience unpleasant stimuli (noises. (1999) and Seidl (2002) for summaries). the minimum bids for selling each lottery are elicited. Few studies performed on preference reversals in static situations (hereafter. disagreeable drinks and mild pain) using incentive compatible bidding mechanisms.

We first introduce the two experiments and discuss the results. We then compare the results to those of other SR and GR studies and discuss similarities and differences. The anchoring implicit in many of the GR studies reinforces the preference ordering of those who initially choose the $-bet and offsets the preference ordering of those who initially choose the P-bet and from this pattern emerges the asymmetry. which lends credence to anchoring and adjustment mechanisms as part of the explanation. The experiment is formulated in a simple labor market context where subjects are asked their preferences for performing several menial tasks (among them. 1990) show that this asymmetry in reversals is often reduced. but subjects who prefer adding numbers rarely reverse themselves. The remainder of the chapter is organized as follows. however. We finish with some hypotheses of what could be driving this static asymmetry result and a discussion of the implications of the results for microeconomic theory and behavioral economics. the subjects are likely to state a higher price for the $-bet than the P-bet. that is subjects anchor their responses to the winning dollar payout for each bet. In the process of reversing preferences. the plethora of other explanations forwarded for gambling reversals do not seem to explain the asymmetry in static reversals we identify. we induce within-subject preference reversals in a static choice setting by using anchoring cues prior to eliciting selling prices. Unlike the asymmetries in gambling reversals. . Previous experiments have found that value elicitation experiments are more effective if they elicit subject willingness to accept to give up goods rather than the willingness to pay for non-endowed goods (Kahnemann et al.. we develop an experiment that asks subjects to rank order two goods in the presence of external stimuli. though typically not eliminated. Subsequent experiments that reverse the direction of anchoring (Schkade and Johnson. we elicit values for a good with which subjects are preendowed: their time. Experiment one To test the rank stability of preferences.Preference reversal asymmetries 333 P-bet and payouts for losing the bet are similarly small. reversing the direction of the anchors does not reduce this asymmetry. typing and adding) and then later asked to formulate bids representing the minimum willingness to accept to perform the tasks subject to an incentive-compatible elicitation mechanism. In this chapter. This increases the likelihood that those who chose the P-bet in the choice setting will be reversed and reduces the likelihood that those choosing the $-bet in the choice setting will be reversed. 1989) or state winning outcomes in units other than dollars (Slovic et al. Furthermore. The experiment simulates a simple labor market in which we act as an employer and ask subjects to indicate their “wage demanded” for two simple but potentially time-consuming tasks: addition and typing.. we identify an asymmetry in these static reversals where subjects who prefer typing words to adding numbers are likely to reverse themselves during the price elicitation stage. To avoid potential endowment effects. 1991).

Haab and B. subjects rank the five tasks assuming the tasks paid the same. on a small scale. Please note that the price chosen by the computer is completely random and does not depend on your wage demanded for each . The introductory section gets subjects to think about and rank their preferences over five tasks: typing a list of words. adding pairs of numbers. with the order randomized. and subjects are given an estimate of the time it will take to complete the full adding or typing task. Subjects type a list of 30 words. The expected time commitment for the experiment was less than 1 hour. After the task is chosen. or adding together 175 pairs of two-digit numbers and typing the sum into the computer. The time to complete each of the demonstrations is recorded. the computer will randomly choose one of the two tasks. Roe Subjects were recruited to participate in the computer-based experiment via posted flyers and emails sent to undergraduate and graduate students of several departments on campus. each task is demonstrated. working in a fast-food kitchen and telemarketing. The exact tasks to be performed consist of typing a list of 500 words into the computer. The lowest ranked (most enjoyable) of the adding and typing tasks determines the pretreatment preference. The procedure for establishing the minimum payment demanded utilizes a Becker– DeGroot–Marshak (BDM) incentive compatible elicitation method described to subjects as follows: We are going to ask you some questions to determine the smallest amount of money we would have to pay you to perform the tasks. Following the pre-treatment preference elicitation. A total of 71 subjects participated in the first experiment. and add together ten pairs of numbers. To familiarize subjects with the tasks and to allow them to form realistic expectations of the full task. On separate screens. We then provide a sample demonstration of two of the five tasks: typing and adding. in random order. mopping floors in an office building. Subjects are told that later in the experiment they will have the opportunity to perform one of two tasks for pay. and once the subjects begin. Subjects can modify this predicted time according to their own expectations. Subjects were told they would have the opportunity to participate in a computer-based experiment that would guarantee a minimum cash payout of $5 and would offer the possibility of earning more money during the course of the experiment. subjects are asked to imagine performing each task for one hour and then rank order the five tasks from most to least enjoyable if they were paid the same amount to perform each task.” After we have established your wage demanded for each task.334 T. The experiment is fully automated. We will refer to this as your “wage demanded. and elicit a pre-treatment ranking of the two tasks. the computer will randomly select a price and announce it to you. they have no contact with anyone other than questions of clarification to the moderator (a very rare occurrence). a series of value elicitation questions establishes the payment demanded for each of the two tasks. Following the demonstrations. The resulting subjects consisted of a mix of undergraduate and graduate students.

Following the value elicitation. Table 19. you will be asked to either a) perform the task for the announced price and receive payment upon completion of the task.Preference reversal asymmetries 335 task. For the value elicitation questions.2 A series of follow-up screens repeats the procedure. The two anchored treatment groups receive an anchoring question prior to the open-ended elicitation for each task. Subjects are told that they have no incentive to reveal an amount other than their true lowest willingness to accept. the total payment due subjects is calculated and subjects are dismissed to the hallway for a simple game of chance to win an additional $20 (1/16 chance of success).1 summarizes the two anchor treatments: add low/type high and type low/add high. the anchor question is “Would you be willing to type 500 words for $T?” The anchor amounts ($A and $T) vary depending on the randomly assigned anchor treatment. Depending on the payment demanded for each task. the subject either performs the full task for the announced price or does not perform the task and receives no payment for the task. Depending on your wage demanded for the chosen task. The no anchor treatment group receives an open-ended value elicitation question for each task of the form: “We would like to know the lowest amount that we would have to pay you to add 175 pairs of number together. and a quiz ensures subjects understand the procedure. or b) not perform the task and receive no payment. and the announced price. the computer randomly determines the task to be offered (a 50/50 chance for each). For the adding task. subjects are randomly assigned to one of three anchor treatment groups: 25 percent receive a no anchor control treatment and 75 percent are randomly split between two anchor treatments. Table 19. The price is randomly chosen from a uniform distribution over the range $2 to $12. Following the successful completion of the task. and the price to be paid for the task. the anchor question is: “Would you be willing to add 175 pairs of numbers for $A?” For the typing task.1 Anchor treatments for experiment 1 Anchor $A Treatment Add low/type high Type low/add high $2 $12 $T $12 $2 . No specifics are given to the subject regarding the distribution from which the random prices are drawn as such information would dilute the effect of anchors we provide later in the bidding elicitation. What is the smallest amount we will have to pay you to add 175 pairs of numbers?” A similar question is asked for the 500 word typing task (and the order of the two elicitations is randomized).

a preference reversal will be observed if the subject prefers the adding task pre-treatment but reports a strictly lower willingness to accept payment for the typing task. If preferences are rank stable. but in the post-treatment. If anchoring effects do not exist. Haab and B. we expect the post-treatment preferences for the two tasks to be independent of the anchoring treatment across subjects.2 Post-treatment preferences by treatment Post-treatment preference Prefer typing Indifferent Prefer adding Anchoring treatment Low add ($2)/high type ($12) No anchor High add ($12)/low type ($2) 15/28 8/18 3/25 8/28 6/18 7/25 5/28 4/18 15/25 .2 presents the number of participants preferring each task post-treatment broken down by treatment. Given the random assignment of subjects to the three treatment groups.to post-treatment within subjects. the anchor reinforces the pre-treatment ranking. If the same subject receives the add low/type high anchor treatment then the anchors are counter to the pre-treatment ranking. Rows represent the three experimental treatments. Table 19. Prior to the anchoring treatment subjects rank the adding and typing tasks in relation to each other and three other tasks. If preferences are rank stable across subjects. If the subject has a pre-treatment preference for typing and receives the type low/add high anchoring treatment. and we expect that the anchoring effects will not induce reversals from pre. Experiment one results We first examine the potential for across-subject reversals. and the predetermined anchors for each group. then we label this a preference reversal. Table 19. If in the pre-treatment ranking. Similarly. Columns represent the post-treatment preferences as revealed through the relative payments demanded for each task. the anchoring treatments will have no effect on the subsequent relative ranking of the two tasks as implied by the value elicitation. the subject prefers the typing task. the subject indicates a willingness to accept less compensation to perform the adding task. the anchor treatments can be either reinforcing or counter-balancing to the pre-treatment rankings. we expect the proportion of participants preferring each task after the anchoring treatment to be independent of the anchoring treatment.3 Table entries are the number falling in each preference category over the number offered each treatment. Roe Discussion of experiment one Of relevance are two preference rankings: the pre-treatment ranking and the ranking implied through the value elicitation procedure.336 T.

20 0.60 0.10 0. Of those. the proportion of subjects preferring the adding task. Of potentially more significance would be a demonstration that participants could be induced to reverse rank orderings of two goods internally. but not the relative ranking to other goods (i. the adding anchor appears to have no statistically significant effect relative to the no anchor treatment. Similarly. we find that multiple anchors can on average affect the relative ranking of goods. post-treatment.. we look at the incidence of preference reversals within individuals.1 Between-subject comparison: experiment 1 post-treatment preferences for tasks (by treatment).50 Proportion 0.04 0. The proportion of subjects indicating a post-treatment preference for the typing task is significantly higher for subjects receiving the high anchor for typing than for those receiving the low anchor for typing (p = 0.00 Prefer typing after treatment (26 of 71) Low add ($2)/high type ($12) Indifferent after treatment (22 of 71) No anchor Prefer adding after treatment (24 of 71) High add ($12)/low type ($2) Figure 19. who find that on average a single anchor can affect the location of preferences for a single good (preferences are arbitrary).Preference reversal asymmetries 337 0. Of the 71 participants in the first experiment.to post-treatment.002). It is apparent that there are large differences in the relative proportion of those preferring each task by anchoring treatment. is significantly higher for those that received the high adding anchor than for those receiving the low adding anchor (p = 0. Anchoring opposite to the pre-treatment preferences results in a 23 percent (and statistically significant) increase in the probability of .70 0. Pair-wise tests of differences in proportions support the assertion that anchoring significantly influences the ranking of the adding and typing tasks on average.4 Reinforcing the findings of Ariely et al. 21 (30 percent) reversed their preferences for the two tasks from pre. even in the presence of multiple anchors).1 gives a graphical representation of these results. Figure 19. preferences are coherent.002). The proportion of indifferent subjects does not vary significantly by treatment. 19 received a (randomly assigned) anchoring treatment that provided anchors counter to their pre-treatment preferences. Interestingly.30 0. To investigate this.e.

Haab and B.85. and an indicator for a preference reinforcing anchoring treatment. Because the pretreatment ranking in experiment 1 requires the subject to rank completely five tasks.303 (0.52.020) –0. For those with a pre-treatment preference for the typing task.” It is possible that the responses to the value elicitation questions . –0. Independent variables are indicators for counter-balancing anchors by pre-treatment preference group. To provide further verification of the asymmetry. Further. Roe Table 19.861 (0.036) –0. while reinforcing anchors do not significantly affect the probability of reversal relative to the no anchor treatment. a leading explanation for asymmetry in GR.765 (0. is controlled for within the experimental design. 0 otherwise Coefficient (p-value) Constant Adding preferred pre-treatment/counterbalancing anchors Typing preferred pre-treatment/counterbalancing anchors Reinforcing anchors Note Log likelihood = –36. it appears possible that an asymmetry exists in the preference reversal over the typing and adding tasks. The effects of reinforcing anchors. those that prefer typing pre-treatment seem to have more malleable preferences and are significantly more likely to reverse preferences in the presence of counter-balancing anchors. experiment 1: dependent variable = 1 if preferences reverse. Those that prefer the adding task pre-treatment appear to have well-defined and invariant preferences and are therefore uninfluenced by anchors opposite their preferences.3 summarizes probit results on within-subject reversals. we conduct a second experiment with slight modifications from the first. On the other hand. observations = 71. Chi-squared = 12.306) a preference reversal relative to the no anchor treatment.652) 0. Similar to the preference over gambles literature. counterbalancing anchors induce a statistically significant increase (32 percent) in the probability of preference reversal relative to the no anchor treatment. or counter-balancing anchors for subjects with a pre-treatment preference for the adding task. A second experiment The asymmetry anomaly is surprising in a static setting where anchoring.338 T. produced no significant change in the probability of reversal relative to the no anchor treatment. Table 19.517 (0. it is possible that the ranking of the additional three tasks distracts the subject from ranking the two tasks of interest: typing and adding. a possible source of confusion in experiment 1 is the use of the wording “wage demanded.5 The omitted category is the no anchor treatment. The dependent variable is a binary indicator for within-subject preference reversals.3 Probit results on within-subject preference reversals.

6 To circumvent these potential problems in experiment 2. we would have to pay you more to perform a less preferred task.4 reports the results of a probit model from experiment 2 reversals (1 = reversal). 15 preference reversals (26 percent of subjects compared to 30 percent in experiment 1) were observed. That is. Of those. Independent variables include indicator variables for pre-treatment preference for adding. After the anchoring treatment and value elicitations. The insignificant anchoring effect indicates that those that prefer adding in the pre-treatment are not induced by the counter-balancing anchors to .Preference reversal asymmetries 339 were based on the expected hourly wage for performing the tasks rather than the piece-meal rate for the two tasks as we anticipated. The same cannot be said for those that preferred adding in the pretreatment. Following the pre-treatment ranking of the two tasks. i. The significant coefficient of the anchor treatment in the typing preferred group indicates that those who prefer typing in the pre-treatment can be induced to prefer adding in the value elicitation by anchoring opposite to their initial ordering. you would be willing to accept a lower payment for that task. and 40 indicated a pre-treatment preference for adding (26 assigned to the anchor group). those who prefer typing (PT) receive a $12 anchor for typing and a $2 anchor for adding while those who prefer adding (PA) receive a $12 anchor for adding and a $2 anchor for typing. 70 percent of the sample was assigned to one of two treatment groups while the remaining subjects received no anchoring treatment.” Because we are interested in testing the asymmetry of preference reversals. and three from the pretreatment adding group that was anchored. Subjects assigned to an anchor treatment receive anchors designed to induce preference reversals. which task would we have to pay you more to perform?”7 Finally we introduce reminders to subjects before the pre-treatment ranking. Table 19. one preferred typing and one preferred adding pre-treatment.” and we repeatedly remind subjects that we are looking for lumpsum payments and not hourly wages. we phrase all question in terms of the “payment demanded. 18 indicated a pretreatment preference for the adding task (15 were assigned to the anchor group).e. Experiment 2 results A total of 58 subjects participated in experiment 2. Further. the pre-treatment ranking from experiment 2 asks subjects to rank directly just the typing and adding tasks in terms of which task would demand a higher payment: “If we were to pay you to either add 175 pairs of numbers. Of the two unanchored reversals. pre-treatment preference for typing and the same two indicators crossed with anchoring treatment. or type 475 words. Thirteen reversals were in one of the two anchored groups: ten from the pre-treatment typing group that was anchored. and before the value elicitation questions that their payment demanded responses will indicate to us a preference for one task or the other: “We believe that if you prefer a task. we modify the anchoring treatments from experiment 1 to be conditional on the pre-treatment ranking.

Rather. experiment 2: dependent variable = 1 if preferences reverse. the result is asymmetric. Subsets of subjects immune to preference reversals may simply self-select into the $-bet preference in the GR literature or choose adding in our study.34. and perhaps more damning than. Several common explanations for such asymmetries in the GR literature exist (anchoring theory. the findings of preference reversals in preferences over uncertain gambles. .340 T.172 (0. No reinforced anchors are offered. we find the existence of rank reversals in preferences over simple tasks with certain outcomes. Experiment 2 reinforces the findings of experiment 1. Those with a pre-treatment preference for the adding task appear to have less manipulable preferences over the two tasks. which has been the focus of much of the preference reversal literature. prominence theory.004) –0. but may rely upon other attributes (e. observations = 58. regardless of the mode of preference elicitation.411 (0. Subjects facing preference elicitations demarcated in dollars rely upon other dollar demarcated information within the problem. a With no constant. compatibility theory). as one selfselected group – those who prefer adding to typing in the joint preference elicitation – are rarely manipulated to reverse their preference via an anchoring mechanism. default categories are pre-treatment preference for each task with no anchors. that counter-balancing anchors can induce rank reversals for the typing task. Chi-squared = 5.622) 1. Subjects that are anchored counter to their initial preference ordering during a subsequent bidding elicitation show a tendency to reverse those preferences in the direction of the anchors. indicate a preference for typing in the post-treatment rankings.037) Notes Log likelihood = –30.4 Probit results on within-subject preference reversals. Haab and B. that preference reversals may be a simple issue of sample selection that arises from heterogeneity across individuals. it is always selfselected.g.465 (0. probability of winning) when facing other preference elicitation modes (choice).431 (0. Even when studies feature extensive experimental control. We control for such possibilities in our experiments in a way that no asymmetry should arise. Roe Table 19.63. 0 otherwise Coefficient (p-value) Adding preferred pre-treatmenta Typing preferred pre-treatmenta Adding preferred pre-treatment and counter-balanced anchors Typing preferred pre-treatment and counter-balanced anchors –0. Discussion and conclusions Complementary to. but yet they do. such as the value of payouts in a lottery. We hypothesize that instead of the phenomenon arising from heterogeneity of the method in which preferences are elicited. no research design can assign a subject’s first rank ordering of preferences.565) –1. An asymmetry arises in the static preference reversals. However.

hence. How does the core of microeconomic theory operate if preferences are. Cox and Grether. 1996). after controlling for anchoring treatments. For example. Our ability to use simple anchoring mechanisms to induce within subject preference reversals in an economically meaningful. 1990. We leave this and other design augmentations to future work and would not be surprised if the asymmetry we uncover in static preference reversals follows the .Preference reversal asymmetries 341 Alternatively.09 for those preferring adding pre-treatment in experiment 2.g. minimally. coherent sets of preferences (Grether and Plott. Both differences are significant at the 90 percent level indicating mild support for the hypothesis that in the absence of anchoring. unlike the preference reversal over gambles literature (e. however. are less likely to reverse themselves. time discounting or issues of interpersonal utility. the average difference between the payment demanded for adding and the payment demanded for typing is $3. Seidl. We restate with amplification many of the concerns raised by others who question the dogma of stable. even in choice settings that are not encumbered with the intricacies of probabilistic outcomes. simple choice setting further escalates the need for economists to reconsider the wisdom of treating preferences as stable in either a cardinal or an ordinal sense. our experimental work has not introduced repetition of the elicitation mechanisms into the design to see if increased familiarity with the choice and bidding procedures limits either the extent or asymmetry of reversals. constructed on the fly as people are bombarded with key economic and institutional stimuli inherent in the decisions that interest economists? A partial list of disturbing questions includes: which set of preferences is correct – those elicited via direct choice among alternatives or those inferred from transacted prices? Is a market mechanism that expends $X to match a marginal buyer to a marginal seller more or less efficient than one that expends $Y < $X to shape the preferences of a proximate buyer and seller such that they are both at the margin? Which preference elicitation mechanism should be used when gathering information that will shape public policy? Do we adopt a policy if those who gain under that policy can use their windfall to shape the preferences of those who lose such that the policy passes the potential compensation principle? Further experimental work to investigate the robustness of our results is warranted. Among the non-anchored subjects. Our data shows some credence for this explanation. static.00 more than those who rank typing above adding.91 higher for those preferring adding pretreatment in experiment 1 and $4. 1979. Tversky and Thaler. maximally. those who rank adding higher than typing pre-treatment provide an average absolute difference in willingness to accept approximately $4. those who prefer the $-bet and the adding task may simply have more intense.g. highly malleable and. less malleable preferences for these particular goods and. heterogeneity with regards to intensity of preferences may lead to reversal asymmetries. 2002). those preferring the adding task have more intense preferences for their preferred task than those preferring the typing task. It is clear. that much work remains before we understand the causes and implications of rank instability of preferences. e. In both experiments.

Cox.. Bohm. Eliciting Reservation Prices: Becker– DeGroot–Marschak Mechanisms vs. D. Blount. (1997) show that failing to describe the distribution helps alleviate some of the critiques lodged against the BDM procedure and yields bids equivalent to those elicited during a second price auction mechanism. Loewenstein. Kahnemann. M. Grether. 1979. 69 (4). and Sonnegård. “Coherent Arbitrariness”: Stable Demand Curves without Stable Preferences. they are classified as preferring adding. Markets. Acknowledgments We want to thank Jason Eslick for extensive help in programming the experiments contained herein. Lindén. 623–38. The Preference Reversal Phenomenon: Response Mode.. Economic Journal.. G. G. Markets and Incentives. 381–405. Preference Reversals Between Joint and Separate Evaluations of Options: A Review and Theoretical Analysis. 1991.. 6 Although. Notes 1 The exception from the SR literature is Ariely et al. 5 We assume that the effect of reinforcing anchors on the probability of preference reversal is the same for those preferring typing and those preferring adding. and Prelec. References Ariely. 7 (3). 1079–89. P. 576–90. Knetsch. the reversal results are qualitatively identical. Journal of Economic Perspectives... We also thank John Kagel and Kerry Smith for helpful comments on earlier drafts. C.. and vice versa. Peter Bohm et al. 5 (1). 2003. 2 Also..342 T. Haab and B. and Status Quo Bias: Anomalies. Economic Theory. Economic Theory of Choice and the Preference Reversal Phenomenon. with enough market feedback and experience. and Plott. 1999. Quarterly Journal of Economics. J... Psychological Bulletin. who collect multiple observations per subject but only report averages across treatment groups. Roe course of the asymmetries uncovered in preference reversals in gambles: they remain robust in many settings but. it was not possible to identify separately the effects of reinforcing anchors by pre-treatment preference. D. J. 193–206. 73–105. J.. when responses to experiment 1 are converted to an hourly wage rather than a piece-meal wage. and Thaler. D. 118 (2). it is unlikely that such inconsistencies would arise. R. 1996. D. American Economic Review. S. 1997. Hsee. and Grether.. Loewenstein. Loss Aversion. J. 125 (5). they shrink to the boundary of significance. as this was not the primary focus of their research. 4 The results of all of the pair-wise tests are available on request from the authors. 107 (443). 7 We reduced the number of words to be typed from 500 in experiment 1 to 475 in experiment 2 based upon average times recorded in experiment 1 such that both tasks would take about 15 minutes to complete. If the two payments demanded are identical then they are classified as indifferent. The Endowment Effect. and Bazerman. Even if their data were analyzed for within-subject reversals. Because of the small number of subjects preferring adding pre-treatment. D. . 3 If the payment demanded for the adding task is strictly less than the payment demanded for the typing task. C.

Preference Reversal. 203–31. J. P. Schkade. A. 2002. 201–11. Anomalies: Preference Reversals. 92 (5). Cognitive Processes in Preference Reversals.. Spring 1990.. 2005.. 4 (2). 16 (5). Valuation in the lab.. Organization Behavior and Human Decision Processes. Slovic. 1989. Unpublished draft manuscript. Insights in Decision Making: Theory and Applications. Shogren. 5–27. Tversky. J. Compatibility Effects in Judgment and Choice. University of Wyoming Department of Economics. E. D. Robin M. R. Journal of Economic Perspectives. 44 (2). Preference Reversals of a Different Kind: The “More is Less” Phenomenon.Preference reversal asymmetries 343 List. A... D. pp. C.. Chicago: University of Chicago Press.. American Economic Review. Journal of Economic Surveys. 2002. and Johnson. and Thaler. . 1636–43.. 1990. 621–55. Seidl. 1990. In: Hogarth. Griffin. and Tversky. ed.

20 Measuring preferences for genetically modified food products Charles Noussair. and France are the countries in which the highest percentage of respondents rejects GM foods. as well as social dimensions. Stephane Robin. Survey responses indicate that aversion to GMOs is based on both private considerations. Ireland. (2001) report that 79 percent of French respondents either agreed or mostly agreed with the statement “GMOs should simply be banned”. Of these. 38 percent indicated that they were in support of agrobiotechnology and 46 percent were opposed.. such as potential health risk and a preference for natural foods. Noussair et al. and Bernard Ruffieux Introduction The introduction of genetically modified organisms (GMOs) into food products has been a major political issue for over a decade in many parts of the world. there is a consensus among scientists that biotechnology has the potential to create products that will enhance nutrition. Between 30 percent and 65 percent of the respondents in every EU country reject every reason for buying GM foods listed in the survey. For example. However. A poll of Americans conducted by ABC News in June 2001 found that 35 percent believed that GM foods were safe to eat. Moon and Balasubrimanian (2001) report the results of a survey conducted of 2. increase crop yields. 46 percent in food packaging. The unfavorable view has been exacerbated by the spread of . 2003) indicate that a majority of Europeans would not buy or eat GM foods. the FDA in the United States. Regulatory authorities such as the FSA in the United Kingdom. 89 percent in livestock feed. while 52 percent believed that they were not. A total of 89 percent were opposed to the presence of GMOs in food products. and reduce the use of toxic pesticides and herbicides. and 46 percent in fuels. Greece. In the UK. The results of the fifth Eurobarometer survey on biotechnology and the life sciences (Gaskell et al. such as environmental effects and ethical concerns. on the basis of recommendations from the scientific community.600 consumers in the UK. 86 percent in medicine. Moreover. and the DGAL in France. surveys show a similar pattern. have recognized that the GMO products currently available are safe for the consumer and the environment. polling of consumers consistently indicates a high degree of hostility to the presence of GMOs in the food supply.

while contextual cues or small changes in information provided to survey respondents may change results dramatically (Ajzen et al. measure different variables. into milk production in the United States in 1993. 1991. The second study surveyed here (Noussair et al.. and Nyborg (2000) argue that survey and hypothetical contingent valuation measurement techniques for public goods do not accurately reveal participants’ willingness to pay. 2001). it may not be reflected in his willingness to pay because of the free rider problem (Stevens et al. (1995). In addition. It is known that individuals’ decisions can differ drastically between when they are hypothetical. Krutilla. the extent that actual decisions to purchase food products are affected by the presence of GMOs. The focus of the first study described in this chapter (Noussair et al. Surveys place respondents in the role of citizens. surveys and purchase decisions. rather than consumers.Measuring preferences for GM food 345 the “mad cow” epidemic. despite the hostility toward GMOs that is ubiquitous in survey data. analysts predicted a 20 percent decline in total milk consumption. Surveys indicated that a majority of consumers had a negative opinion of the technique. 1987. such as the preservation of GMO-free crops.. primarily on ethical grounds.. Blamey et al. a bovine growth hormone. and the initial introduction of GMOs without the public’s knowledge. 1998. and when they involve a real commitment to purchase (see for example Neill et al. 1995. List and Shogren. even if provision or preservation of a public good is valuable to an individual. Sagoff (1988). that are explicitly guaranteed to be GMO free. 1967). using experimental methods. 1994. there was no decrease in actual milk consumption after the introduction of the technique (Aldrich and Blisard. Thus the two instruments. We also consider buyer behavior with respect to different thresholds of maximum GMO content. 2002) is motivated by the fact that. there is reason to question whether the anti-GMO sentiment expressed in surveys would be reflected in actual purchase behavior. 1996). Cummings et al. may be particularly suspect. who make actual purchase decisions. 2004b) is to consider. most surveys do not inquire about actual purchase decisions at specific prices. who make judgments from society’s point of view.. and that contain GMOs. However. A well documented example of a dichotomy between surveys and consumer behavior was observed during the introduction of recombinant bovine somatropin (rbST). since in a democratic system public opinion must be taken into account in addition to the scientific merits of the policy and the market pressures in the economy. or List and Gallet. the lack of benefit that the first generation of GMOs provides to the consumer. as in a contingent valuation study or other survey. We study purchasing behavior of consumers using a laboratory experiment designed to elicit and compare the willingness to pay for products that are traditional in content and labeling. However. Furthermore. Brookshire and Coursey. The dichotomy between scientific recommendations and public opinion has complicated the formulation of government policy with respect to GMOs. . 1998). On the basis of the survey data. Surveys about preferences over public goods.

. in addition to corn and soybeans. and that the artificial setting of the laboratory does not drastically alter consumer behavior. For the few GM products that are available. 2001. The current policy of most major European retailers not to carry GM foods. Moreover. In the UK. where GM content must be indicated on the product label. 2001. 1999) suggest that experiments provide a good alternative method to study product demand in general. The experimental approach is particularly appealing here because of the absence of field data. demand for GMOs cannot be inferred from market data since GM content is not indicated on the labeling. Lusk et al. in the US. In contrast. Any food product sold in the European Union for human consumption that contains an ingredient that consists of more than 0. three types of corn are authorized for cultivation. previous works (see for example Shogren et al.. We use an experiment to consider whether the absence of a reaction in demand to the current labeling of products is due to the fact that most customers do not notice the labeling. Huffman et al. We are unaware of any previous estimates of consumer demand for the GMO-free characteristic in food products other than those obtained from experimental studies (see Lusk et al. the policy adopted by most European governments has been to declare a moratorium on approval of new GM products for cultivation and sale. one type of GM tomato is authorized for importation and used in tomato puree. for those few GM products that have been put on the European market. Furthermore. 2005). One type of corn and one type of soybean are authorized for importation. means that it is very difficult to estimate product demand for foods containing GMOs using field data from European countries. as of early .9 percent GMOs must be labeled “contains GMOs”.. in the United States. In particular. experimental methodology provides an environment to measure individual preference by controlling for noise and other confounding factors. and to allow the market to determine how much of each type of product is sold.346 C. Noussair et al. there is reason to believe that consumers are unaware of the labeling of GM content.. and thus do not realize that the product they are purchasing contains GMOs. which is not possible in the field. Currently in France. There is no GM produce currently sold in Europe and the only GM products for sale appear as ingredients in processed foods. where the vast majority of GM food is sold. For the few products that have already been approved. Background Policy issues: segregation and thresholds In response to the tension between scientific and public opinion on the issue of GM foods. However. to require labeling of products containing GMOs. No GM crops are grown commercially in the UK. their policy has been to segregate GM and GMO-free products at all stages of production. researchers in the laboratory are able to control precisely the information communicated about product characteristics. sales do not decrease when the label reveals that the product contains GMOs. which has resulted from pressure of activists and the media.

2000) for a discussion of policy issues relating to the labeling of GM products. the gene producers. Of this total. vegetables.Measuring preferences for GM food 347 2002. 2000. (1999) find that in general. 59 percent went to US farmers. See Caswell (1998. To account for the incentive of producers to make the labels indicating their products’ positive characteristics as prominent and those revealing the unfavorable characteristics as discreet as possible. as long as it is easily legible. (2000) estimate that the welfare gains from the adoption of Bt corn in the US for the year 1996 equaled $240. Lin et al. also estimate that the $60 million in savings constituted 20 percent of the overall welfare gain to all parties (US farmers. or (b) to ban labeling. some economists might view it as an inefficient policy.”. producers and consumers in other . Although the current policy of segregation and mandatory labeling is freemarket oriented in that it offers consumers a choice. and grains were being cultivated. and $9. is also considered sufficient.9 billion per year for coarse grains. the studies that have estimated the gains from the adoption of biotechnology in farming in the United States have found them to be considerable.7 billion per year for cotton. and the seed companies). segregation costs have been estimated at 12 percent of the price of corn and 11 percent of the price of soybeans in the year 2000. A note at the end of the list of ingredients. specifying the genetically modified origin.2 billion per year for rice. and positioning of information on packaging. when a product is classified under current law as containing GMOs. savings to farmers from the adoption of herbicide tolerant soybeans have been estimated at $60 million annually (Lin et al. $6. Falck-Zepeda et al. The size of the letters must be at least as large as those in the list of ingredients. The gene developer received 21 percent. Both of these policies have potential downsides.. which are subject to the same regulations as other products. according to the US Department of Agriculture. In the United States. Segregating the entire process of production is costly to farmers and firms throughout the production chain. it must carry in its list of ingredients the statement “produced from genetically modified. . Buckwell et al. For soybeans grown in the American Midwest. and primary processors. . it can be argued that the expenditure represents a deadweight loss. Banning new GMOs may be inefficient if there are welfare gains from the adoption of biotechnology that are foregone. US consumers 9 percent. especially in the upstream part of the chain. which consists of the seed producers. identity preservation for specialty crops increases final costs by between 5 percent and 15 percent. The two main alternatives to this segregation of the market are (a) to ban GM varieties entirely. regulators have imposed strict conditions on the size. color. 2001). In the US.3 million. Indeed. there are no specific regulations for biotech products. (2000) estimate the gains from the introduction of biotechnology at $1. farmers. Anderson et al. rest of world farmers and consumers. Since there is no hard evidence that the GMOs that regulatory authorities have approved are harmful either to health or to the environment. effectively making GM and non-GM varieties indistinguishable from each other from the viewpoint of a consumer. about two dozen different GM fruits. In Europe.

Since GMOs have lower production costs. Furthermore. must be specified. if a considerable segment of the market refuses to purchase products containing GMOs at any price. 22 percent to the gene developer. it could potentially cause a market collapse for entire products. The surplus calculation hinges on whether the actual purchase behavior of consumers corresponds to the polling data. Because of the ease of contamination throughout the production chain. Under many of their parameterizations. Noussair et al. or would purchase products made with GMOs if they sold at lower prices. it is impossible to make intentionally any product. Under a policy of segregation. On the other hand. below which a product is to be considered as . but another large segment would purchase GM products if they were cheaper. If a firm cannot disclose that its product uses no ingredients that contain GMOs. If. producers have an incentive to insert them into the food supply. This technological constraint requires the specification of a threshold above zero. corn syrup. banning GMOs is probably the best option. On the other hand. Lence and Hayes (2001). if the production tracks are not segregated or labeling of GMO content is interdicted. such as soy lecithin. 1970). the appropriate policy depends in part on the relative sizes of consumer and producer surpluses and the costs of implementing different policies. above which a product is considered to be bioengineered. as the expense of creating two tracks of production would not be justified. (2000) find that the surplus from the use of Round up Ready soybeans in the US in 1997 was distributed in the following manner: 50 percent to US farmers. as it is in the United States. if the large majority of consumers behave as if they are indifferent to GMOs. The GMOs currently on the market were introduced for agronomic reasons and the foods containing them are indistinguishable from conventional foods to the consumer in the absence of labeling information. and the seed producer 5 percent. without any trace of GMOs. and 12 percent to foreign consumers and producers. reducing social welfare by eliminating potential gains from trade. This could eliminate the entire market for many products. a “lemons” scenario may result (Akerlof. separation of the production tracks and the enforcement of mandatory labeling of products containing GMOs would be worth the expense. However. 8 percent to US consumers. it might replace ingredients that consumers believe may contain GMOs with those that cannot contain GMOs. the threshold level of GMO content. they will be unwilling to pay more for an unlabelled product than an amount that reflects the presence of GMOs. a large majority of consumers is unwilling to purchase products containing GMOs. countries 6 percent. This would cause the market for non-GMO varieties to disappear. the production tracks could be safely integrated with little social cost. provide estimates of potential welfare gains and anticipated costs for the United States from the cultivation of GM crops.348 C. as suggested by the polls. If consumers value foods containing GMOs less than foods that do not contain GMOs. both overall consumer and producer welfare is greater after the introduction of GM technology. From an economist’s point of view. using simulation techniques. 9 percent to the seed companies. Traxler et al. and cornstarch. in whose manufacture GMOs are already authorized.

A demand revealing auction has the advantage over the study of purchase decisions with field data that it allows an individual’s limit price to be measured directly. regardless of the risk attitude of the bidder and the strategies other participants use.. the greater is the cost of production of GMO-free products. 1998.. we use the second-price sealed-bid auction. Any bidder who submits a bid greater than the sale price receives a unit of the good and pays an amount equal to the sale price. for an exposition of induced value theory) to create limit prices for fictitious goods. The agent who submits the highest bid wins the auction and receives the item.. and above which the product must be labeled as containing GMOs. Cox and Grether. 1997.. Some of this research has used the technique of induced values (see Smith.. The dominant strategy of truthful bidding and the commitment of real money create an incentive to reveal truthfully limit prices. 1961). food safety (Hayes et al. and lotteries (Grether and Plott. 1964). Experimental economists have employed demand revealing auctions to study limit prices for goods as varied as consumer products (see for example Hoffman et al. 1993. List and Shogren. A methodological issue: how to measure willingness to pay? Willingness to pay information is typically elicited with a demand revealing bidding mechanism. Accurate willingness to pay information is particularly useful for new products because other sources of demand estimates on which to base profit or costbenefit calculations are not readily available. each subject simultaneously submits an offer price to purchase a good. both described later in this chapter. In a second-price sealed-bid auction. Fox et al. and Lusk et al. Buzby et al. 1998. In this research. a sale price is randomly drawn from a distribution of prices with support on an interval from zero to a price greater than the anticipated maximum possible willingness to pay among bidders.. also called the Vickrey auction (Vickrey. The increase involves the cost of producing very pure seeds. and cleaning storage and transportation containers. 2001). The lower the threshold. Bohm et al. There is a substantial literature studying the behavior of the two mechanisms in the laboratory when university student subjects are bidding for goods. Observing only whether or not an individual purchases a product at the current market price in a store merely establishes whether or not his limit price exceeds the current market price. isolating parcels of land. 1996). such as the Vickrey auction or the Becker–DeGroot– Marschak mechanism. 1982. 1998. 1995.. The marginal cost of lowering the threshold may be justified if consumers have a strong preference for a low threshold. Huffman et al.Measuring preferences for GM food 349 GMO free. but pays an amount equal to the second highest bid among the bidders in the auction. each member of a group of potential buyers simultaneously submits a bid to purchase a good. 2001. and List and Lucking-Reiley.. and the Becker–DeGroot–Marschak (BDM) mechanism (Becker et al. In a BDM. The experimenter offers a guarantee that bidders can resell goods . 2000). 1979. Afterwards.

53 percent were female. when the goods considered have induced valuations. including Irwin et al. (1980). should they purchase the items in the auction. the evidence that bids tend to differ from valuations is indirect. more rapid? We pose these questions under specific conditions. Rutström (1998) finds that the two mechanisms generate different mean bids for the same objects. France area. We evaluate and compare the auctions according to the following criteria. all of the studies show that there is heterogeneity in bidding behavior that leads to a dispersion of bids relative to valuations. if it occurs. In the case of auctions for goods with home-grown (and therefore unobservable) valuations. including Coppinger et al. (1997) find that bids in the BDM are sensitive to the choice of end points of the distribution of possible transaction prices. which suggests a bias in bidding either in the early or in the late periods. for more detail). The socio-economic level of the sample was representative of the French urban population. Several authors. These studies reach a variety of conclusions about bids relative to valuations. Irwin et al. Bohm et al. At the time of . at prices that are specified in advance. The two experiments comprised 26 sessions. (1982). and averaged 33 years. List and Shogren (1999) find that bids in the Vickrey auction tend to increase as the auction is repeated. The research we have conducted on preferences for GM products allows us to compare these auctions within a similar environment. For example Kagel et al. indicating that at least one of the two must be biased. and 112 participated in experiment 2. and each session took approximately 2 hours. (1987) and Kagel and Levin (1993) find that most winning bids in the Vickrey auction are higher than valuations. Kagel et al. (1) Does either or both of the systems contain a bias toward under or over-revelation of true valuations? (2) Under which system do individuals on average bid closer to their true valuations? (3) Under which system is convergence by repetition toward demand revelation. and some suggest that average bids are biased away from valuations. (1987) and Kagel and Levin (1993). 2004c. have studied the behavior of the Vickrey auction. The ages of the subjects ranged between 18 and 75 years. and when specific training procedures are in effect that our experience and intuition suggest would enhance the demand revelation performance of the mechanisms (see Noussair et al. Noussair et al.350 C. Furthermore. (1993) have studied the BDM process using goods with induced values. (1998) find that the BDM process is more successful at eliciting true valuations for certain distributions of sale prices than others. Methodology The participants The participants in the experiments were a demographically representative sample of 209 consumers in the Grenoble. such as consumer products. A total of 97 subjects participated in experiment 1. (1998) and Keller et al.. Of these. and other authors. Cox et al. when the population considered is a diverse sample of the population.

Therefore in principle. The experimenter then randomly draws a sale price from a pre-specified interval. Each bidder has a dominant strategy to bid truthfully an amount equal to his willingness to pay (Vickrey. This training was similar for each mechanism and proceeded in the following manner. At the time of recruitment. As mentioned earlier. The ability to redeem an item from the experimenter induced a limit price in the auction. The rules of the BDM mechanism are simple. The others do not receive units and make no payment.Measuring preferences for GM food 351 recruitment. depending on which mechanism was in effect for the session. Before the auction took place. We recruited only individuals who were regular consumers of the products we used in the experiment. indicating a price at which he offers to purchase one unit of the good offered for sale. and pays an amount equal to the second highest bid among the bidders in the auction. in which they bid for fictitious items. subjects received no indication that the experiment concerned GMOs or potential risks to the food supply. each subject received a sheet of paper that indicated an amount of money. Each subject simultaneously submits a bid to the experimenter in a closed envelope. The other bidders do not receive items and pay zero. 1961). The . since a subject’s payoff if he won the auction equaled the induced value minus the price he paid. we used Vickrey auctions to elicit willingness to pay information. subjects were invited to come to the laboratory to sample food products for a government research project. The agent who submits the highest bid wins the auction. we used the BDM mechanism to elicit willingness to pay information. Only individuals who made the food purchasing decisions in the household were permitted to participate. regardless of his risk attitude. No communication between subjects is allowed during the bidding process. should he purchase it in the auction. each subject simultaneously submits a bid to purchase a good. The BDM mechanism and the Vickrey auction In experiment 1. In experiment 2. in the BDM there is an optimal strategy for a bidder to bid his valuation. Subjects then participated in several BDM or Vickrey auctions. In a Vickrey auction. the mechanism has the ability to reveal bidders’ valuations. The induced value differed from subject to subject and was private information. The fictitious items had induced values. for which he could redeem a unit of the fictitious item from the experimenter. Any subject who submits a bid greater than the sale price receives an item and pays an amount equal to the sale price. each subject received 100 francs (roughly US$14) in cash. The training phase Both experiments 1 and 2 began with a training phase to help subjects to learn to use the dominant strategy for the mechanism employed. At the beginning of a session. from zero to a price greater than the maximum possible willingness to pay among bidders.

and asked subjects if they could identify their own valuations and to predict which subjects would be receiving units of the good based on the valuations displayed. the label of which was visible. The valuations in each period were randomly drawn from a uniform distribution whose end points differed in each period. After bidding. who is required to pay immediately the price determined in the auction from his current cash total. After subjects submitted their bids (and the experimenter drew a selling price in the case of the BDM).352 C. who were free to engage in open discussion on the topics: (a) which subjects received units in the auction? (b) how much did the winners pay? (c) did anyone regret the bid he submitted? After the discussion. the experimenter wrote all of the valuations on the blackboard. The first reason is that it made subjects aware that others’ valuations for goods could differ from their own. and (c) to show subjects that the auction involved transactions where real money was at stake. but there was no discussion as in the earlier induced value auctions. They buy products with labels and packaging already removed. To render this transparent. the sale price was drawn. a bottle of wine. the rules of the auction. in full view of all participants. The dominant strategy of bidding one’s valuation in the auctions is not at first obvious to most subjects. they taste products without knowledge of the information displayed on the label. a bottle of wine is given to each winner. all of the bids were posted. and that they are not in a hypothetical simulation. each of the winners received. We chose not to inform directly the subjects of the dominant strategy. The second reason was to provide an easier transition into the next phase of the experiment. The auction for the actual consumer product also serves to illustrate to subjects that they are spending real money for real products that they can keep after the experiment. (b) to reduce the biases and noise that tend to arise in bidding behavior. . We ended the training phase of each session with an auction of an actual consumer product. an amount of money equal to his induced value minus the price he was required to pay. Then the experimenter recorded the submitted bids on the blackboard next to the corresponding valuations. we used a technique intended to encourage subjects to come to understand the strategies that constitute optimal behavior on their own. A series of identical auctions was conducted using the same procedure. and the transactions were implemented immediately. and they buy products without knowing the sale price beforehand. He posed the following questions to the group of subjects. Noussair et al. The cash was physically placed on the desk in front of the subject after the auction. the winners were announced. The auctions continued until at least 80 percent of the bids were within 5 percent of valuations. as in the induced value auctions. Instead. and verify their comprehension of. inclusion of the auctions with induced values had three objectives: (a) to teach the subjects. There are two reasons that we added this auction to the training phase. However. where subjects would be placed in a situation that is different in three ways from typical market purchases.

Before bidding in the first period. and no information was given to participants about other players’ bids at any time. which we referred to as S. At the beginning of this phase. the GMO phase of the session. Random draw of the auction that counts toward final allocations. Additional information: “S contains GMOs” and “N is GMO free”. C. Additional information: “no ingredient in L contains more than 1 percent GMOs”. and “No ingredient in N contains any detectable trace of GMOs”. L. The sale price was not drawn for any period until the end of period 5. “No ingredient in C contains more than 1/10 of 1 percent GMOs”. C. and we informed subjects of that fact before bidding began. was conducted. subjects received a sample of each of the four products to taste. Additional information: general information about GMOs. but were close substitutes.1 shows the information made available to subjects at the beginning of each period.. Auction. and N during the sessions. No information was given about products L and C in period 2. Each of the following periods consisted of the revelation of some information about some or all of the products. without its packaging or labeling. Additional information: the brand names of the four products. and N. as outlined in Table 20. Table 20. 2004a). At the beginning of period 2. the phase of primary interest. Auction. 1997. They then indicated how much they liked the product on a scale where “I like it very much” and “I don’t like it at all” were at the extremes of the rating scale (see Combris et al. we informed the subjects that product S contained GMOs and that product N was GMO free. L.1 Sequence of events in GMO phase of an experimental session. we simultaneously auctioned four products. All four products were biscuits that are typically available in grocery stores and supermarkets throughout France. and the designation “organically grown” for product N. one for each product. “One ingredient in S (soy) is derived from an authorized genetically modified product”. Then the auction for period 1 took place. Auction. or Noussair et al. Recording of hedonic rating of the four products. Implementation of transactions for the period that counts. Auction. The four products were auctioned simultaneously. At the beginning of Table 20.Measuring preferences for GM food 353 Experiment 1: GMO phase After the training phase of each session described above. experiment 1 Period Period 1 Period 2 Period 3 Events Information: blind tasting of the four products S. followed by four simultaneous auctions.. subjects were required to taste each product. The products were different from each other. In the GMO phase. The GMO phase of the experiment consisted of five periods. Period 4 Period 5 Transactions .1. Auction.

subjects each received a sample of each of the four products to taste. At the beginning of period 2. Subjects then had 3 minutes to study the products. soy. (b) the criteria for classifying a product as containing GMOs. We also indicated to subjects that no ingredient in N had any detectable trace of GM content. period 3. Translated from the original French. in which we revealed information about the products and then conducted an auction for the products. Subjects then bid in the final round of auctions. corn is replaced with “genetically modified corn”. The products are made by a world leader in the food industry and are widely available in grocery stores and supermarkets in Europe.354 C. The data in the figure are normalized by taking each individual’s actual bid in period 1 as the base equal to 100. At the beginning of period 1 of the GMO phase in experiment 2. tracking that individual’s bids over time relative to his bid in period 1. we magnified and projected the list of ingredients of each product. we informed the subjects that no ingredient in L contained more than 1 percent GMOs and that no ingredient in C contained more than 1/10 of one percent GMOs. A second auction was then conducted for each of the goods. Only the data from those who bid greater than zero for the product in period 1 are included in the . consisted of three periods. (d) the food products sold in France that contain GMOs. we revealed the brands of the four products and the label indicating that product N was organic. the phase of interest. Then a simultaneous Vickrey auction for each of the four goods took place. At the beginning of period 3. We took care to provide an unbiased characterization and provided only facts without comment. without its packaging or labeling. and that S did contain a GM ingredient. The information consisted of (a) the definition of a GMO. called S and U.1 graphs the evolution of the average normalized bid over all subjects over the five periods of the GMO phase for the four products. but with the list of ingredients visible). including two identical bars. Noussair et al. and averaging across all individuals in each period. (c) the list of GM plants authorized in France. which was authorized in France. the list of ingredients for product S includes “corn”. we distributed one unit of each of the products to each subject in its original packaging (with the price removed. Experiment 2: GMO phase In experiment 2. Results Experiment 1: the impact of GMO information Figure 20. Before the last period. the GMO phase. At the beginning of period 4. In the list of ingredients for product U. exactly as it appeared on the packaging. Four chocolate bars were auctioned each period. and invited subjects to read the list of ingredients. and (e) the current French law regarding GMOs. subjects received a four-page handout containing background information about GMOs.

Measuring preferences for GM food 355 120 114. A sign test (eliminating the ties in which bids were the same in both periods) rejects the hypothesis that a bidder is equally likely to lower as to raise his bid at the p < 0. We observe that consumers.001 level.2 114.01.2 63. blind 50 Period 4.1 80 70 61.9 90.1 percent GMOs as consistent with the typical GMO content of conventional prod- Period 5. figure (no subject who bid zero for a product in period 1 ever submitted a positive bid for that product in later periods).6 Contains GMOs (product S) Threshold of 1% (product L) Threshold of 0. Both a sign test and a pooled variance t-test reject the hypothesis of equality at the p < 0. Our subjects appear to view a guarantee that no ingredient contains more than 0. brands .9 104.1% (product C) GMO free (product N) 60 Period 2. consumers typically act as if there is a low probability that products contain GMOs. raised their bid in period 2. Of the 83 subjects. indicating that. and only seven lowered it. with GMOs or GMO free Period 3. we revealed that product N did not contain GMOs and product S did contain GMOs.3 110. experiment 1.7 percent. background information Figure 20.1 Average bids for the four biscuits in each period of GMO phase.9 100 90 89.7 99. on average. who bid more than zero for product N in period 1. value the absence of GMOs. 41.9 102 94. consumers increased their bids for product N in period 2. Only four participants increased their bid for S after learning that it contained GMOs while 64 lowered their bid. thresholds Period 1.1 60.9 100. The GMO-free guarantee raised the limit price for product N of the average consumer in our sample by 8 percent. The relatively small increase for the GMO-free product suggests that in the absence of any information about GM content. revealing that product S contains GMOs lowered its average limit price by 39 percent. In period 2. In contrast.6 110 100 107. A pooled variance t-test also rejects the hypothesis that the mean bid for product N is equal in periods 1 and 2 at p < 0.001 level. Period 3 is designed to measure the impact of GMO content thresholds. The average premium for the GMO-free product over the product containing GMOs was 46.5 66. on average.

01. the GMO-free product.05 for the sign test).1 percent GMOs. The increase was greatest for the GMOfree product N. The bidding behavior for product L reveals that a product meeting a 1 percent threshold is viewed very differently from a product labeled as containing GMOs. which contained GMOs. the more conservative sign test is not significant. were significantly different from each other at p < 0. the 1 percent guarantee is viewed differently from the label “contains GMOs” and the 0. The distribution of background information about biotechnology in period 4 led to a slight increase in average limit prices. and the decline was statistically significant (p < 0.05 for three of the four products.1 percent threshold was valued more or less highly than a conventional product. Thus over half of our participants considered a product satisfying the 1 percent threshold as no worse than the conventional product.9 percent reduced their bid: 4. 20 bidders increased their bid while nine lowered it.1 percent guarantee more highly than a 1 percent guarantee. we fail to reject the hypothesis that an equal number of bidders raised and lowered their bids in period 5 . for each of the products. In period 3.01 for L and S. There was no consensus among the participants about whether a product meeting the 0. The information did not bring the prices of L. Furthermore. as well as for products L and S. However. The 1 percent guarantee was viewed as different from the 0. The average bid for product C was significantly lower in period 5 than in period 4 at p < 0. which was significant at p < 0. for all four products. to their levels before any information was revealed. The effect was significant at p < 0. and 40. and the 1 percent threshold appears to be seen as a higher level of GMO content than that of a conventional product. They value a 0. with a 1 percent threshold.9 percent of subjects increased their bid when informed of the 1 percent threshold.01. The mean normalized bids in period 3 for products N and C. However.1 percent guarantee. and we can reject the hypothesis that an individual was equally likely to raise and to lower his bid at the p < 0.05 level.356 C. we revealed that no ingredient in product L contained more than 1 percent GMOs and no ingredient in product C contained more than 0. We observed no significant change in the median willingness to pay for product C between periods 2 and 3 (p = 0. For product N. but the average bid for product L declined by 10 percent. we cannot reject the analogous hypotheses for the other three products. while 27. For all four products at least 57 percent of the bids were unchanged between periods 3 and 4.5 percent left their bid unchanged.4 percent reduced their bid to zero. though the pooled variance t-test indicates that the information increased the average bid. Revealing the brand names of the products in period 5 raised the average prices for three of the four products. A total of 17.01. or S. ucts (the unlabeled product historically available).1 percent guarantee is interpreted differently from “GMO free”. and the majority of participants did not change their bids.1 percent of any ingredient. A total of 33 percent increased their bid (by an average of 28 percent) after learning the maximum possible GMO content was 0.38 for the sign test). A pooled variance t-test of the hypothesis that the mean normalized bids for products L and C are equal rejects the hypothesis at a significance level of p < 0. Thus. Noussair et al.

Of our consumers. subjects observe the products as they are seen in the supermarket. That means that over 95 percent of our participants were willing to accept a level of GMO content that typically results from inadvertent co-mingling if the product is sufficiently inexpensive.4 percent bid zero at 0.9 percent of our subjects. We call this group the reluctant consumers. Our consumers can be classified into four categories. 36. We classify them as indifferent consumers. Presumably.1 percent. Before bidding in period 2.071 for product S. demonstrating behavior consistent with having a preference for GM foods. Nevertheless.070 for product U. Specifying a threshold did result in a lower incidence of zero bidding than the announcement “contains GMOs”. which does not contain GMOs and t = 0. There was no increase in price for product N from revealing that it was organically produced. Unwilling consumers bid zero for product S after learning that it contained GMOs. .1 percent did not change their bid for product S upon finding out that it contained GMOs.Measuring preferences for GM food 357 relative to period 4. We thus obtain the result that the labeling of products as containing GMOs does not affect the willingness to pay of consumers. and only 4. A pooled variance t-test fails to reject the hypothesis that the normalized average bids are different between periods 1 and 2 (t = 0. This group places negative value on GMO content or a claim for an equitable share of the surplus created by GMOs adoption. Another 4. perhaps because revealing its label exerted an offsetting negative effect.3 percent. Results from the second experiment: do individuals read labels? Figure 20. we have created more favorable conditions for the subjects to read and study the labels than exist in the supermarket. but did not go so far as to bid zero. we observe that average bids do not change between periods 1 and 2. They are willing to tradeoff GMO content and the price they pay. A total of 10. 18. which does contain GM corn). Thus a full 23 percent of bidders were willing to accept GMOs in their food at the same price as the conventional product. Indeed. They will lower (raise) their bid prices when faced with products with higher (lower) GMO content.7 percent of the subjects bid zero for the product with a maximum of 1 percent GMO content in any ingredient.2 shows the normalized average bid over all subjects for each of the three periods of the GMO phase of experiment 2. We also cannot reject the hypothesis that the bids are different from each other (t = 1. there is still a large group of consumers willing to buy them at the same price as conventional products and to allow them to establish a foothold in the marketplace. Subjects are seated and have no alternative activities for 3 minutes other than to study the labels.9 percent of participants were favorable. The average percentage of the decrease was 28.1 percent of the reluctant consumers exhibited a willingness to pay that was monotonic in the strength of the guarantee of the maximum GMO content.2 percent lowered their bid for product S when they found out that it contained GMOs. Despite the current unpopularity of GMOs in food.53). They comprised 34. A total of 42.

We use two measures for our comparison. normalized . an identical product without any indication of GMO content (product S) experiences an insignificant average decrease of 3 percent from the previous period (t = 0.2 Average bids for the two identical chocolate bars in periods 1–3. GM free Product U. experiment 2 (GM free corn corresponds to Product S. In contrast. However.7 70 80 60 Period 2: observe package and labeling Period 3: ingredients displayed Period 1: taste product Figure 20. Upon learning that product U contains GM corn. and 60 percent lower their bid by at least 5 percent. Comparison of the BDM and Vickrey processes The data from the training phase of the two experiments provides an opportunity to compare the BDM and Vickrey processes with regard to their demand revelation properties. S and U. are significantly different from each other in period 3 (t = 10. Noussair et al.358 C.0 90 95. GM corn to product U).3 percent compared to the previous period. the data change radically in period 3. 22 percent of our subjects boycott the product entirely by bidding zero.40). The first measure is the overall average bias of the mechanisms in period t.0 Product S. containing GM corn 72. 100. induces a substantial decrease in willingness to pay that is specific to that product. Thus the labeling “contains GMOs”. in which subjects bid while able to view the list of ingredients on large overheads.0 100 97.271). The bids for the two products. when it is actually noticed.37). The average willingness to pay for the product labeled as “containing GMOs” decreases by 27. The decrease is statistically significant (a pooled variance t-test for a difference in sample means between periods 2 and 3 for product U yields t = 2.9 97.

89 (10. The bias for period t is calculated as Σj[bjt − vjt]/vjtnt . The table shows that both auctions are biased in period 1.13) 3. In the Vickrey auction.79 (24. The second measure is the average dispersion.27 (24. defined for period t as Σj|[bjt − vjt]|/vjtnt .57 (11. the dispersion is the absolute value of the bias.60) 13.06 (22.19 (23. with a standard deviation (of the percentage difference between bid and valuation) of 28.38) 16.95) 11.57 (30.33 (20. as well as the last period of a session (which never exceeded period 6).9 percent lower than valuations.86 (20. under both the Vickrey and the BDM processes. The percentage bidding an amount equal to their valuations is also small in both auctions. Table 20.53) 32.87 (28. and nt is the total number of bidders in period t.25 (11.2 Deviations of bids from valuation.65 (26. Overall.9 percent.16 (32. This bias is larger and the dispersion is greater under the BDM mechanism. though 17 percent bid within 2 percent of their valuations in the Vickrey auction. It indicates the extent to which average bids are higher or lower than valuations. under the BDM mechanism. Only 2.94) 6.59) Period 4 –8. vjt is her valuation in period t. On average.4 percent of participants bid more than their valuations under the BDM process and 6 percent did so under the Vickrey auction. where bjt denotes player j’s bid in period t.69) +1.96) Last period –6.2 illustrates the average value of each measure over the course of the first four periods.42) . between 6 and 7 percent of subjects under both systems. with bids tending to be below valuations.75 (18.94 (20.76) 16. The average dispersion is equal to the average absolute value of the difference between bids and valuations.89) 41.62) –11. The standard deviations are indicated in parentheses.06 (11.58) –5. The table reveals the following patterns.24) –30.43) –0.32) 28.33 (26.Measuring preferences for GM food 359 by the valuation. Period 1 was a practice period that did not count toward participants’ earnings.49) 9.50 (27.76 (23. normalized by the valuation. 90 percent of subjects bid less than their valuations and only a very small percentage bid more than their valuations. induced value phase of both experiments (%) Period 1 BDM Average bias BBDM = Σj[bjt – vjt]/vjtnt Average dispersion DBDM = Σj|[bjt – vjt]|/vjtnt Vickrey Average bias BV = Σj[bjt – vjt]/vjtnt Average dispersion DV = Σj|[bjt – vjt]|/vjtnt –39. the period 1 average bid is Table 20.10) Period 2 –28.59 (21. bids are 39.89) Period 3 –12. For an individual bid.

The percentage bidding equal to valuations increases to over 10 percent overall and is slightly higher in the Vickrey auction than in the BDM. Both the average bias and the average dispersion are significantly greater in the BDM than the Vickrey auction at the p < 0. the first auction that counted toward subjects’ earnings.5 percent versus 31. whereas in the Vickrey auction the decrease is 63. 30. exhibits lower dispersion. Thus.4 percent). The dispersion between bids and valuations decreases in each period of the BDM. reaching 41. The overall bias decreases in each subsequent period for both processes. our measure of dispersion. The decline is mainly due to a reduction in the amount that agents underbid. the overall data from periods 1 and 2 suggest that the Vickrey auction is less biased. as well as repetition.1 percent of valuation for the BDM and −11.5 percent. Nonetheless. The overall dispersion shrinks in both systems but the decrease is steeper in the Vickrey auction (51.360 C.5 percent for the BDM and 68.5 percent for the Vickrey auction. The introduction of monetary payments.4 percent for the Vickrey auction in the last period. The average absolute difference between bids and valuations. and improves its performance more quickly over time. 87. The average absolute . In each period.3 percent. However in all of the periods. the overall trend is clearly downward. and not to a decrease in the percentage of agents underbidding. The magnitude of the average underbid is less severe in the Vickrey auction. the bias is significant at the 5 percent level in the BDM.05 (according to a pooled variance t-test). exhibits less dispersion. Though the measure does increase between periods 3 and 4 in the Vickrey auction.8 percent of bids in the BDM and 76. appears to improve decisions.7 percent in the BDM compared to 32.6 percent. These trends continue in subsequent periods.2 percent less than the corresponding valuation with a standard deviation of 32. is 41.01 level for both mechanisms.6 percent of the valuation for the BDM and 36 percent for the Vickrey auction. Beginning in period 4. but less so than in period 1. Noussair et al. The decline in the bias between periods 1 and 2 is steeper in the Vickrey auction than in the BDM. Thus. The bias is −28.1 percent of those in the Vickrey auction are less than valuations. Pooled variance t-tests indicate that the bias is significantly different from zero at the p < 0. The proportion of participants bidding less than their values is greater in the BDM than in the Vickrey auction. In period 2. the Vickrey auction is less biased. induces a greater percentage to reveal their exact valuations. The bias in the BDM decreases by 29. In the Vickrey auction. The average underbid is 44. reaching zero in the Vickrey auction and 6 percent in the BDM mechanism in the final period. The percentage of agents bidding an amount equal to their valuations increases from period to period under both processes. the bias is no longer different from zero at conventional significance levels in the Vickrey auction. and has a greater percentage of agents bidding within 2 percent of values than the BDM process. 77 percent of bids are within 2 percent of valuations and 90 percent are within 10 percent of valuations in the last period. both auctions remain biased. the bias in the BDM is significantly greater in magnitude than in the Vickrey auction at p < 0.05 level.6 percent in the Vickrey auction. in the practice period.

Whereas 35 percent of our subjects absolutely refused to purchase a product containing GMOs. if it were sufficiently inexpensive. .1 percent and 1 percent. the 0. A total of 89 percent of our participants were willing to purchase a product satisfying the 1 percent threshold. The remainder is willing to purchase GMOs even when no threshold is specified. This indicates that market demand is decreasing in GMO content. The third is to create two production tracks and introduce a labeling system (which could be voluntary or mandatory) to allow the consumer to identify the two varieties. In our experiments. Nearly one-quarter of participants showed no decrease in their willingness to pay in response to learning that a product contained GMOs.8 percent in the BDM. as 96 percent of our participants were willing to purchase a product.1 percent would make another 7 percent of participants willing to purchase products satisfying the threshold. Discussion This chapter surveyed three studies. The policy options available to address the arrival of biotechnology in food production can be grouped into three types.1 percent threshold was not considered equivalent to GMO free. which would cause gains from trade to be foregone. over the entire time horizon. Furthermore. The data thus argue against the banning of GMOs.1 percent GMOs. and the third compares two different techniques to elicit willingness to pay. The second is to integrate conventional and biotech varieties into one production chain. the maximum content that the European Union exempts from labeling.9 percent in the last period of the Vickrey auction compared to 11. at least for the time being. and the 1 percent threshold generated higher bids than the classification “contains GMOs”. However. In the Vickrey auction. the remaining 65 percent of our subjects were willing to purchase a GM product if it was sufficiently inexpensive. The two different thresholds. and could receive a welfare gain if GMOs make products cheaper. 0.01 in all periods except for period 4. in which no ingredient contained more than 0.Measuring preferences for GM food 361 difference in the last period is 3. the Vickrey auction generated data much closer to truthful bidding than did the BDM. The first two consider empirical questions related to the willingness to pay for food products with genetically modified content. the dispersion is significantly less than in the BDM at p < 0. Based merely on polls. we observe a wide range of revealed preferences. Our results for the first experiment show a sharp contrast to the predominantly negative views of French survey respondents toward genetically modified organisms in food products. Lowering the threshold to 0. generated significantly different bids and were thus were clearly perceived as meaningfully different. our experimental results indicate that only slightly more than a third of the population would be unwilling to purchase GM foods at any price. we would have concluded that the only policy action that would be feasible in France given current public opinion would be the complete interdiction of GMOs in food. The first option is to ban the use of GMOs in food products. Therefore.

The separation and labeling policy gives the market the role of transmitting information about the safety of GM products.362 C. We observe that the BDM is subject to more severe bias. We would like to thank Isabelle Avelange. the Vickrey auction is preferable to the BDM mechanism as an instrument for the elicitation of the willingness to pay for private goods. the segment that refuses to purchase GM products at any price (35 percent of participants in our sample) would experience a decrease in their welfare. neither auction could be made into a perfect tool to reveal valuations with our subjects. Our experimental protocol was effective in de-biasing the Vickrey auction over several periods. Our research supports the proposition that the Vickrey auction can be an effective tool for demand revelation with non-student subject pools. but less effective on the BDM. at least not during the time horizons that were available to us. The consumers who are willing to purchase GMOs if they are sold at a discount might be made better off. it remains unknown whether unbiased bidding for goods with induced values carries over to subsequent bidding for goods with home-grown values. Pierre Combris. our results weigh in favor of segmenting the market between products containing GMOs and products that are GMO free. Of course. as well as indifferent and favorable. the unwilling consumers could be assured of GMO-free varieties. With our techniques. and slower convergence to truthful revelation than the Vickrey auction. the sizes of each of the markets appear to justify the establishment of two separate production tracks. As long as the segregation costs are not greater than the welfare gains from market segmentation. 2004b. The data also reveal potential welfare costs to consumers from integrating the two production streams. and Steve Tucker for valuable comments and assistance. by providing an opportunity and an incentive for consumers to sample the lower cost products made with GMOs voluntarily. Sylvie Issanchou. The program “Pertinence économique et faisabilité d’une filière sans utilisation d’OGM”. Therefore. and would have to switch to products with ingredients that have no GM varieties. Our data suggest that a large fraction of consumers would do so. However. However. Our comparison of the valuation elicitation systems indicates that. . in our opinion. but also cautions that sufficient practice and appropriate training in the rules of the auction is important. consumers could benefit from the cost reductions that the first generation of GMOs provides. In this way. the Vickrey auction performs better than the BDM by the three criteria we have set for it. Noussair et al. Egizio Valceschini. as well as the French National Institute of Agronomic Research (Program on Consumer Behaviour) provided research support for this project. 2002. while price-sensitive reluctant. Acknowledgments This chapter is a summary of three articles (Noussair et al. 2004c). given the training methods and the procedures we have used in our study. Yves Bertheau. greater dispersion of bids.

. Smith. 39 (3).. 107 (441). 488–500. 1996. Journal of Agricultural and Applied Economics. G. 1–6. R.. R. Common. 1997. Labeling Policy for GMOs: To Each His Own? AgBioForum. C.. A... P. J.. Quality of Information. 3 (1).. Vol. G.. G. Cummings. . L. M. J. L. H.. C. A. 1998.. Combris. J. D. M. J... J.. Roberson. R. and Crutchfield. N.. and Rutström. Respondents to Contingent Valuation Surveys: Consumers or Citizens? Australian Journal of Agricultural Economics.. 1995. 1995. 1998. Should Use Of Genetically Modified Organisms Be Labeled? AgBioForum. B. and Titus. Research in Experimental Economics. Economics of Identity Preservation for Genetically Modified Crops. Buckwell.Measuring preferences for GM food 363 References Ajzen. Aldrich. S. C. CT. R. and Sherman. and Marschak. 77 (4).. 85 (1). Estimating the Economic Effects of GMOs: The Importance of Policy Choices and Preferences. Markets. A. and Barfoot. The Market for “Lemons”: Quality Uncertainty and the Market Mechanism. 226–232. 1987. W. V. M. H. Bohm. J. 1999. P. and Coursey.. V. and Visser. and Motivational Orientation. Estimation of a Hedonic Price Equation for Bordeaux Wine: Does Quality Matter? Economic Journal. Economic Inquiry. M.. American Journal of Agricultural Economics. 360–369. and Grether. Information Bias in Contingent Valuation: Effects of Personal Relevance. Harrison. 1996. Brookes. 22–24. 43–57. Belgium. A. 1997. Incentives and Behavior in English. 1079–1089.. Linden. 30 (1). D. 1970. M. C.. B. 1 (1). Economic Theory.. Measuring Consumer Benefits of Food Safety Risk Reductions. and Blisard. A.. J. and Quiggin. Cox. December (74701). S. Eliciting Reservation Prices: Becker–DeGroot–Marschak Mechanism vs. L. P.. The Quarterly Journal of Economics. V. Blamey. R. and Nelson. C. 554–566. Measuring Utility by a SingleResponse Sequential Method. G. 1–43. 1–22. 2. 305–309.. Food Biotechnology Communication Initiative.. 7 (3). 2000. In V. E. Journal of Environmental Economics and Management. J.. J.. and Rosenthal. and Smith. 2000. Ready. A. S. Markets and Incentives. 30 (1). Coppinger.. C.. Greenwich. 381–405. 263–288. Adelaide University Policy Discussion Paper 0035. Buzby. The Preference Reversal Phenomenon: Response Mode. L. Akerlof.. Brown. Dutch and Sealed-Bid Auctions.. 1964. 69–82. 18 (1). Homegrown Values and Hypothetical Surveys: Is the Dichotomous Choice Approach Incentive-Compatible? American Economic Review.. 260–266.. G. Boussels. Lecocq. Nielsen. Traxler. 9 (3). Falck-Zepeda. JAI Press. 84 (3). 82 (2). Smith ed. Current Issues in Economics of Food Markets.. J. Brookshire.. Surplus Distribution from the Introduction of a Biotechnology Innovation.. Fox. Theory and Behavior of Single Object Auctions. Anderson. 1998. M. J. J. G. D. L. G. Behavioral Science.. R. Measuring the Value of a Public Good: An Empirical Comparison of Elicitation Procedures.. Cox. and Sonnegard. Becker. Caswell. Degroot.. 1982. L. M. 390–402. Consumer Acceptance of Biotechnology: Lessons from the rbST Experience. I. 2000. K. 107 (443). Economic Journal.. 1980.. E. T. American Economic Review. Caswell..

J. D. A.. (October). S. a report to the EC Directorate General for Research from the project “Life Science in European Society” QLG-7-CT-1999-00286. Using Laboratory Experimental Auctions in Marketing Research: A Case Study of New Packaging for Fresh Beef. V. and Harwood. W.. R. 55 (6).. Chambers. Lin. Response to an Asymmetric Demand for Attributes: An Application to the Market for Genetically Modified Crops. A. Estimating Farm Level Effects of Adopting Herbicide-Tolerant Soybeans. J. A. 36 (2). 77 (1). 868–879. Noussair et al. N. B. Shogren.. American Journal of Agricultural Economics. List. Demand Reduction in Multiunit Auctions: Evidence from a Sportscard Field Experiment. Menkhaus. and Tegene. 272–285. Information Impact and Allocation Rules in Auctions with Affiliated Private Values: A Laboratory Study. 623–638.. . 2001. D. Lence. and Fernandez-Cornejo. Iowa State University. 1998.. 1995. Economic Research Service USDA. Econometrica. McKee. J. Washington. G.. C. Biotechnology: US Grain Handlers Look Ahead. Hoffman. J. Theory and Decision. J. and Plott. Hayes. 90 (4). M. Brussels.. D. (April). Economic Journal. M. List. S. J. 25–34.. W..0. Iowa State University. H. Valuing Food Safety in Experimental Auction Markets. and Third-Price Auctions with Varying Numbers of Bidders. F. Rousu. J. Independent Private Value Auctions: Bidder Behavior in First-.. D.. 1993. and Levin. Price. H. 1967. Calibration of the Difference Between Actual and Hypothetical Valuations in a Field Experiment. S. Europeans and Biotechnology in 2002. American Journal of Agricultural Economics. Economic Inquiry. 37 (2).. 193–205. 318–338. Second-. M.. R. D. A.. and Lucking-Reiley.. Oil Crops Situation and Outlook.. Chakravarti. CVM-X: Calibrating Contingent Values with Experimental Auction Markets. W. 103 (419). 961–972. R. E. H.. and Kliebenstein. McClelland. 1998. G. and Gallet.. F. J.364 C. D.. The Becker–DeGroot–Marschak Mechanism and Generalized Utility Theories: Theoretical Predictions and Empirical Observations. J. 1998. and Norden. C. DC: US Department of Agriculture. Gaskell.. D. Shogren. 69 (4). Hayes.. mimeo 2021.. Huffman. Grether.. Shin. Irwin. A.. N. H. J. 2003. Eurobarometer 58.. Cognitive Transparency in Decision Making. J. Conservation Reconsidered. 1275–1304.. J. R. Lin. and Whipple. The Value to Consumers of GM Foods in a Market with Asymmetric Information: Evidence from Experimental Auctions. 1993. J. American Economic Review. European Commission. Department of Economics.. American Economic Review. Field. American Economic Review. D. R.. and Levin.. 777–786. Shogren.. M. 2000. T. 40–53. Journal of Economic Behavior and Organization. A. 2000. Payoff Dominance vs. 83–97. Krutilla. 1993.. Marketing Science. 80 (3). Agricultural Outlook. C. Kagel. U.. D. 57 (4). 455–465. and Stares. mimeo. and Kliebenstein. J. J. Economic Theory of Choice and the Preference Reversal Phenomenon. 2001. F. D.. 20 (3). J. What Experimental Protocol Influence Disparities Between Actual and Hypothetical Stated Values? Environmental and Resource Economics. 2001. G. Economic Research Service. List.. 34 (2). J. Allum.. Fox.. 29–34. 1979. Kagel.. and Hayes. Segal. A. Schulze. Harstad. W. and Wang.. 241–254. 1987. G. J. 2001. W. and Shogren. J.. J. 12 (3). R.. L.. Keller... E. D.

A Meta Analysis of Genetically Modified Food Valuation Studies. Journal of Economic Behavior & Organization. A Multi-Attribute Model of Public Acceptance of Genetically Modified Organisms.. Fox... Lusk. Noussair. and Ruffieux. P. A. 145–154. C. American Journal of Agricultural Economics. 67 (4). S. R.. T. 2001. Genetically Modified Organisms in the Food Supply: Public Opinion vs. T.. T. S.. 1999. M. V. Journal of Agricultural and Resource Economics. Krannert Graduate School of Management Working Paper 1139... Hayes. 81 (4). Noussair. J. E. D. and Ruffieux. Ganderton. K.. S. Auction and Competitive Sealed Tenders. J. B. J. 16 (1). L. C.. American Journal of Agricultural Economics. and Taulman.. Journal of Finance. 30 (1).. 427–441. Noussair. Agribusiness. 102–120. 1998.. Nyborg. G. 75 (1). Purdue. W.. S. 42 (3). Journal of Agricultural and Resource Economics. H. International Journal of Game Theory. Consumer Behavior. Food Quality and Preference. Counterspeculation.. R. Roucan.... L. 393–402. Journal of Economic Psychology.. 114 (492). B. J.. 942–949. Southern Illinois University. 1991. Stevens.. B.. 28–44. R. 47–53. T. L. Echeverria. 72 (5). A. Kurlander. 305–322. R. L... C.. Lusk. The Economy of the Earth. C. Hager. Cummings. Robin. 2002. B.. Robin. F. B. W. mimeo. 1994. Land Economics. J. and Ruffieux. Smith. 27 (3). .. Rent Creation and Distribution From Biotechnology Innovations: The Case of Bt Cotton and Herbicide-Tolerant Soybeans in 1997. Alternative Calibration and Auction Institutions for Predicting Consumer Willingness-to-Pay for Non Genetically Modified Corn Chips. 70 (2). and McGuckin. 1192–1199. and Shogren. 1999. J. Jamal. J. Harrison. Traxler. 40–57.. Robin. 1982.. Survey. 923–955. Daniel.. 15 (4). Do Consumers Really Refuse to Buy Genetically Modified Food? Economic Journal.. Neill. Observed Choices for Food Safety in Retail.. F. D. S. H. 390–400.Measuring preferences for GM food 365 List.. American Economic Review. Noussair. L. 2004b. Robin.. J. 2000.. C. Measuring the Existence Value of Wildlife: What Do CVM Estimates Really Show? Land Economics.. and Nelson. 2001.. 2000. L. 8–37.. 16 (1). 25 (6). M.. 2001. and Balasubrimanian. Sagoff. Robin. Shogren. Noussair. and Auction Markets.. Falck-Zepeda. and More. T. W. B. Price Information and Bidding Behavior in Repeated Second-Price Auctions. 725–741. G. and Roosen. Rutström. M.. M. Vickrey. 1961. A.. Microeconomic Systems as Experimental Science. C. S. J.. 81 (5). 2005.. Moon. R. J. 21–32. 26 (1). and Ruffieux. and Ruffieux.. G. Cambridge: Cambridge University Press. 1998.. A Comparison of Hedonic Rating and Demand-Revealing Auctions. Home-Grown Values and Incentive Compatible Auction Design. G. Homo Economicus and Homo Politicus: Interpretation and Aggregation of Environmental Values. 2004a. Mark.. Hypothetical Surveys and Real Economic Commitments. Do Consumers not Care about Biotech Foods or Do They Just not Read the Labels? Economics Letters. E. 2004c. Glass. S. Revealing Consumers’ Willingness-toPay: A Comparison of the BDM Mechanism and the Vickrey Auction. and Lusk.. J.

21 An experimental investigation of choice under “hard” uncertainty
Calvin Blackwell, Therese Grijalva, and Robert P. Berrens

Introduction
Economists and other researchers have examined decision making under uncertainty in thousands of papers. The vast majority of this work has focused on situations where uncertainty can be described with some known probability. There is no consensus normative or positive theory of behavior regarding decision making under “hard” uncertainty, where the probabilities of a set of events occurring are unknown. This chapter uses induced-value experiments to explore several theories of decision making under hard uncertainty. Improving our understanding of decision making under hard uncertainty may have important implications for complex environmental policy (e.g. protection of endangered species and biodiversity, and global climate change). One specific motivation for this research into situations of hard uncertainty is to increase our understanding of the Safe Minimum Standard (SMS) approach to protecting renewable resources (Ciriacy-Wantrup, 1952; Bishop, 1978; Randall and Farmer, 1995). The SMS approach is commonly defined as a collective choice process that prescribes protecting some minimum level (safe standard) of a renewable resource unless the social costs of doing so are somehow unacceptable or intolerable (see reviews in Farmer and Randall, 1998, and Berrens, 2001). With acknowledged difficulty in defining minimum safety, and the determination of intolerable social costs left to the political or administrative process in any particular case (Batie, 1989; Castle, 1996), the SMS approach is typically viewed as a fuzzy concept (e.g. see Hohl and Tisdell, 1993; Van Kooten and Bulte, 2002), and existing somewhat on the periphery of the field of environmental and resource economics (Vaughn, 1997). Nevertheless, a variety of authors (e.g. Ciriacy-Wantrup, 1952; Bishop, 1978; Randall, 1991; Castle and Berrens, 1993; Toman, 1994; Castle, 1996; Farmer and Randall, 1998; Woodward and Bishop, 1997; Bulte and Van Kooten, 2000) have proposed some variant of the SMS as a either a pragmatic or preferred decision-making rule for complex issues involving both hard uncertainty and irreversibility.1 Take, for example, the question of biodiversity and endangered species protection. A particular species may or may not be essential to the continued survival of its ecosystem. Should this species be saved from extinction? Allowing this species’

Choice under “hard” uncertainty 367 extinction has an unknown probability of destroying the underlying ecosystem. What costs should be borne by the current generation to try to protect the species (Bishop, 1980)? A traditional cost–benefit analysis is severely hampered because science is unable to offer policy makers a defined set of probabilities over the possible outcomes (WSTB, 2004, p. 227). Because it is impossible to value the tradeoff between the increased probability of ecosystem destruction and the costs of saving the affected species, and because such decisions may be irreversible, many authors have proposed the SMS as the preferred decisionmaking rule, especially in pluralistic collective choice processes (e.g. Randall and Farmer, 1995, and see Arrow et al., 2000). But aside from positing the SMS as prescriptive rule, how do SMS-type approaches hold up in describing actual behavior? Randall and Farmer (1995) argue that SMS-type approaches will often emerge in pluralistic social solutions to complex problems involving potentially irreversible losses and a high degree of uncertainty. For example, it is commonly argued that the Endangered Species Act (ESA) of 1973, as amended, broadly mimics an SMS-type approach (Bishop, 1980; Castle and Berrens, 1993; Randall, 1991; Berrens et al., 1998; Berrens, 2001; WSTB, 2004). However, outside of the political economy of any particular piece of legislation, the open empirical question is how prevalent are SMS approaches in actual individual decision making under hard uncertainty? For example, devoid of any endangered species context, would individuals commonly choose SMS-type approaches under hard uncertainty scenarios? To understand such individual decision making, several authors have modeled SMS-type decision processes as minimax decision rules (i.e. maximizing the minimum possible gains) in a game versus nature in the presence of hard uncertainty (Bishop, 1978; Tisdell, 1990). However, Ready and Bishop (1991) demonstrated that such a rule might yield inconsistent outcomes (e.g. for preservation or for economic development) depending upon the structure of the game against nature. In particular, a minimax rule appears to give results that are inconsistent with Milnor’s (1964) bonus invariance axiom for game-theoretic decision rules, and ignores the costs associated with wrong choices, resulting in choices that seem inconsistent with the philosophical approach of the SMS (Palmini, 1999; Milnor, 1964). Palmini (1999) has attempted to reconcile some of this inconsistency by showing that although a simple minimax decision rule can lead to inconsistency, a minimax regret rule, which is perhaps more in line with the basic philosophy of the SMS, can eliminate some of the inconsistencies. If we take Palmini’s argument that the minimax regret rule provides a general underpinning to support SMS-type approaches as a starting point (and see WSTB, 2004, p. 233), then this presents a competing hypothesis that can be empirically tested. In this chapter, we use induced-value experiments to examine three questions. First, do participants use minimax, minimax regret, or some other type of decision rule (e.g. simple maximum expected value) to choose under circumstances of “hard” uncertainty? Second, do participants consistently use the same rule, or do they deviate depending upon the relative payoffs of the choices

368 C. Blackwell et al. involved? Third, are there outside factors that predict participants’ choice of decision rule and if so, what are they?

Literature review
Normative theories Several ways of describing risk and uncertainty have been proposed. Knight (1921) drew a distinction between “risk” and “uncertainty.” His distinction has generally been interpreted in the following way: “risk” is defined as a situation with more than one potential outcome, and the probability of any of the possible outcomes is known in advance. “Uncertainty” is defined as a situation with more than one potential outcome and where the probability of any of the possible outcomes is unknown. Other attempts to differentiate between types of risk include “soft” vs. “hard” uncertainty (Vercelli, 1999), and use of the term “ambiguity” (Camerer, 1999). In a review of models of decision making under hard uncertainty, Camerer defined ambiguity as “known-to-be-missing information” (Camerer, 1999, p. 56). This definition precisely fits the information environment where the SMS approach is typically endorsed. What is important about all these definitions of uncertainty is that in all cases the probabilities of an event occurring are unknown. Such cases will be designated “hard” uncertainty. Research on optimal decision making under hard uncertainty has not converged to produce a single, optimal decision rule for the decision maker to follow. Milnor’s (1964) axiomatic approach typifies much of the research in this area. He examines various strategies that might be used in a game against nature when no information is known about the probability event i might occur. The player chooses a pure strategy, and then nature chooses a strategy by some unknown process, resulting in some known payoff. In particular, Milnor examined four decision criteria, which he named as follows: LaPlace, Wald, Hurwicz, and Savage. The LaPlace criterion essentially assumes that the probability of each of n events occurring is 1/n; the player then should choose the strategy that maximizes the “expected” payoff. The Wald criterion is a minimax criterion, where the player chooses the strategy that maximizes the minimum payoff. Under the Hurwicz criterion, the player chooses an “optimism” parameter α, 0 α 1, and then chooses the strategy such that αA + (1 α) a is maximized, where A is the best payoff and a the worst payoff possible for a given strategy. The Savage criterion is the minimax regret criterion, where the player chooses the strategy that minimizes the maximum regret possible. Regret is defined as the difference between the best possible outcome given nature’s strategy, and the payoff the player actually receives. Arrow and Hurwicz (1972) made a second important addition to the literature. In this paper the authors describe four axioms of rationality regarding choice under hard uncertainty: independence of irrelevant alternatives, relabeling (choices should not change because states of nature are renamed), dominance (higher payoffs are preferred to lower payoffs), and irrelevance of

Choice under “hard” uncertainty 369 repetitive states. Arrow and Hurwicz then show that only choice criteria that are functions exclusively of the maximum and minimum possible values are consistent with these four axioms; the implication here is that when probability distributions are unknown, a rational decision maker will focus on the possible end points. More recent research has refined the insights of Milnor and Arrow and Hurwicz. For example, Maskin (1979) shows how a different set of “reasonable” initial axioms can lead to the maximin criterion for making decisions under hard uncertainty, while Barbera and Jackson (1988) introduce refinements of the maximin criterion to deal with situations in which the maximin criterion cannot distinguish between choices. A second strand of research takes as its starting point not hard uncertainty, but soft, i.e. a situation in which the decision maker has some, limited information about the probabilities associated with various outcomes. These theories allow for non-additive probabilities. Examples include Schmeidler (1989) and Gilboa (1987). A Choquet integration is used to make an expected utility maximization using the non-additive probabilities. These theories share similarities with rank-dependent expected utility theory (Quiggin, 1982). In a recent article, Nehring (2000) proposes another choice criterion, “Simultaneous Expected Utility Maximization” (SIMEU). This choice criterion shares with minimax regret the feature that it depends on not just the minimum and maximum possible payouts, but on payouts in between. Positive theories Although there has been considerable research on choice behavior in environments containing both hard uncertainty and risk, few strong conclusions have been reached about uncertainty. Since Ellsberg’s paper (1961) on ambiguity, much research (e.g. Camerer and Weber, 1992) has shown that experimental participants have “ambiguity aversion,” that is, when given the choice between gambles with known probabilities and gambles with unknown probabilities, subjects will pay a premium to avoid ambiguity. Hogarth and Kunreuther (1995) show that experimental participants behave differently when facing decisions under risk as opposed to “ignorance” (lacking information on both probabilities and payoffs). Several researchers have reviewed the empirical literature on decision making under hard uncertainty, but are unable to recommend one model over another. In a 1992 review, Kelsey and Quiggin do not advocate a particular model. Camerer (1999) discusses numerous theories of decision making under hard uncertainty, but arrives at no conclusions regarding which are best. None of the models he discusses has been tested in an environment in which all the choices are characterized by hard uncertainty. Vercelli (1999) is also unable to recommend clearly one theory over any others, although he does emphasize the success of Choquet Expected Utility theory.2 As this brief review makes clear, there is still much we do not know about decision making under hard uncertainty. To wit, while there are a number of

370

C. Blackwell et al.

competing conjectures, there is no consensus position. Given this backdrop, we attempt to implement some exploratory induced-value experiments.

Experimental design
To ensure that our exploratory investigation is not tied to any particular description of uncertainty, we constructed five alternative scenarios that are posed to each participant. The basic experimental protocol was as follows. Subject pool. A total of 57 undergraduate students from Weber State University (Ogden, Utah) participated during the fall of 2003 and 24 undergraduates from the College of Charleston (Charleston, SC) participated during the summer of 2004. All subjects were recruited from the general campus population. Incentives. Participation took approximately 45 minutes. Participants received a $5 show-up fee, plus performance-based payoffs ranging from $10 to $30. We paid the participants one-fifth of their experimental earnings in cash. Setting. Participants received instructions and made decisions at a computer in a campus computer laboratory. In addition to the computer, participants were given a paper version of all materials presented on the computer (surveys and scenarios).3 Initially, participants were asked to fill out a short demographic survey. Then, participants were introduced to a sample scenario, included as Figure 21.1. Participants were asked to take a brief quiz to ensure comprehension of the decision task. The participants then received feedback on their quiz answers and were allowed to ask questions. Once all participants’ questions had been answered, the incentivized portion of the experiment began.

Decision maker Prospect 1 A1 $20 B1 $10 Prospect 2 A2 $25 B2 $8 Prospect 3 A3 $32 B3 $2

Figure 21.1 Basic decision tree.
Note During the experiment you may be asked to decide which of three prospects to take. All three prospects have two potential outcomes, A or B. The chance of either of these outcomes occurring is unknown. If you choose Prospect 1, then you will either earn outcomes A1 ($20) or B1 ($10). If you choose Prospect 2, then you will either earn outcomes A2 ($25) or B2 ($8). If you choose Prospect 3, then you will either earn outcomes A3 ($32) or B3 ($2). The chance that A1 occurs if you choose Prospect 1 is the same as the chance that A2 occurs if you choose Prospect 2 or A3 occurs if you choose Prospect 3. The figure below describes the potential outcomes of your choice.

Choice under “hard” uncertainty 371 As noted, the decision task consisted of five different scenarios.4 Although each of the scenarios is framed differently, they all have some basic commonalities. In all the scenarios each participant plays a sequential game against nature. The participant plays first, then nature. The participant must choose among two or three alternatives. Each prospect is risky, i.e. each prospect has two possible outcomes, one of which will result after nature has played. In Figure 21.1, nature chooses either event A or event B. As the probability of either event occurring is not given to the participant, from his/her point of view the situation is one of hard uncertainty. A great deal of care was taken to ensure that the probabilities of any particular event occurring were unknown to the subjects.5 The process by which nature selects one event or the other is as follows: each choice by nature is made by a draw from an opaque bag. Participants were not allowed to examine the bags in any way. Participants are told that the bag holds an unknown number of pink and white candies. The correspondence between the color of candy drawn and the actual outcome is determined by the participants via majority rule. For example, in the practice scenario participants would be asked to vote for one of two possibilities: (1) if a pink candy is drawn, then event A occurs, or (2) if a pink candy is drawn, then event B occurs. Allowing participants to select the association helped to ensure transparency from the participants’ point of view, i.e. that the actual probabilities were not skewed heavily to the “bad” outcomes. Each participant made choices for five different scenarios. All participants in a given treatment received the five scenarios in the same order. After making their selections for all five scenarios, each scenario’s risk was resolved, i.e. the correspondence between event A and the color of candy was resolved by vote, and a candy was drawn to determine the actual outcome for each scenario. Participants were then asked to fill out a follow-up survey while the experimenters calculated each participant’s earnings. After completing this final survey the participants were paid the earnings in cash and the experiment ended.

Theoretical predictions
The experiments were designed to distinguish between three choice criteria: minimax (MM), minimax regret (MMR) and “expected value” max (EV). To illustrate these three choice criteria, examine Figure 21.1. The MM criterion will choose Prospect 1, as it has the maximum minimum value (min[Prospect 1] = $10, min[Prospect 2] = $8, min[Prospect 3] = $2; max[min[Prospect 1], min[Prospect 2], min[Prospect 3]] = $10 for Prospect 1). The EV criterion assigns equal probability to each outcome occurring and then chooses the option with the highest “expected value.” For Scenario 1, this choice criterion selects Prospect 3 (EV(Prospect 1) = 1/2 ($20) + 1/2 ($10) = $15, EV(Prospect 2) = $16.50, EV(Prospect 3) = $17). An individual who uses the MMR strategy (minimizing the maximum possible regret) will choose Prospect 2. We define regret as the difference between the best possible outcome given nature’s resolution of the uncertainty and the outcome chosen by the individual. Table 21.1 presents

372

C. Blackwell et al.

Table 21.1 Regret analysis Option 1 1 2 2 3 3 Outcome A B A B A B Best $32 $10 $32 $10 $32 $10 Actual $20 $10 $25 $8 $32 $2 Regret $12 $0 $7 $2 $0 $8 Max regret $12 $7* $8

Note * Indicates the minimum of the maximum regrets.

the regret analysis, where it is shown that the prospect with the smallest maximum regret is Prospect 2. Using the analysis presented above, for each scenario, the first choice corresponds to the MM criterion, the second choice to the MMR criterion, and the third choice to the EV criterion. Treatments Tables 21.2 and 21.3 show the payoffs for each scenario under Treatments 1 and 2. These treatments differ only by the relative payoffs associated with the different prospects. Prospect 1’s payoffs remained unchanged from Treatment 1 to 2. Under Treatment 2, Prospect 2’s “good” payoff (the higher of the two payoffs) is 90 percent of Treatment 1, while the “bad” payoff is 93.75 percent of Treatment 1. For Prospect 3, Treatment 2’s payoffs were 88.75 percent of the “good” payoff, and 100 percent of the “bad” payoff. These changes to relative payoffs alter the predictions made by the EV criterion. Under Treatment 2, EV makes no distinction between any of the three prospects (i.e. the expected value of all three prospects is equal). The difference between Treatments 1 and 2 was made to induce participants to use more conservative choice criteria.

Table 21.2 Outcomes for treatment 1 Scenario 1 2 3 Framing Travel No context Fishing Train $20 $10 1 $10 $5 Lemon Bay $50 $25 Outcomes/choices Auto $25 $8 2 $12.50 $4 Mango Bay $62.50 $20 Plane $32 $2 3 $16 $1 Persimmon Bay $80 $5

Choice under “hard” uncertainty 373
Table 21.3 Outcomes for treatment 2 Scenario 1 2 3 Framing Travel No context Fishing Train $20 $10 1 $10 $5 Lemon Bay $50 $25 Outcomes/choices Auto $22.50 $7.50 2 $11.25 $3.75 Mango Bay $56.25 $18.75 Plane $28 $2 3 $14 $1 Persimmon Bay $70 $5

It seems reasonable to classify MM as the most conservative choice criterion, as it evaluates prospects purely on the basis of the worst that could happen (WSTB, 2004, p. 233). MMR is less conservative because the decision maker is concerned not merely with bad outcomes, but also with good ones. EV is the least conservative because it applies no factor to adjust for risk (e.g. if the real probability of the “bad” outcome is 90 percent, on average, EV will do much worse than the other two criteria). In Treatment 2, participants should be more likely to choose strategies consistent with MM and MMR than under Treatment 1. This behavior should occur because the risk–reward aspect has changed. Unlike in Treatment 1, there is no increase in the “expected value” in return for an increase in the variance of payoffs. This fact should make both MM and MMR more attractive, and allow a finer test of which of the two criteria is preferred.

Experimental results
Table 21.4 shows the overall results. As the most conservative strategy available, the minimax (MM) criterion was not heavily chosen. In only one out of six cases did more than 50 percent of respondents choose this criterion, and in all other cases it was below 25 percent (Table 21.4). The minimax regret (MMR) criterion was chosen by more than 50 percent of respondents in five out of six cases. Overall, the MMR was chosen slightly more than 50 percent of the time. Dropping the most anomalous case (Scenario 2, Treatment 2) raises the MMR selection to over 60 percent of the time. Treatment 1 Participants clearly did not choose randomly, and a χ2 test of proportions confirms this conclusion (H0: participants choose any strategy with probability 1/3; p-value less than 0.0001 for all three scenarios). Scenarios 1–3 were, in part, designed to test for consistency; note that the values of the outcomes in Scenario 2 are one-half of the values of the outcomes for Scenario 1 and that the

374

C. Blackwell et al.

Table 21.4 Participants’ selections for scenarios 1–3 Scenario Choice Consistent with . . . Treatment 1 percentage of participants selecting (%) 23.6 60.0 16.4 16.4 65.5 18.2 12.7 70.9 16.4 Treatment 2 percentage of participants selecting (%) 23.1 50.0 26.9 73.1 11.5 15.4 23.1 61.5 15.4

1 2 3

Train Auto Plane 1 2 3 Lemon Bay Mango Bay Persimmon Bay

Minimax Minimax regret EV max Minimax Minimax regret EV max Minimax Minimax regret EV max

Note For Treatment 2 any choice is consistent with the EV max criterion as all three prospects have the same “expected value” using the 1/n weight.

values of the outcomes in Scenario 3 are 2.5 times the values of the outcomes for Scenario 1. The proportion of participants making choices consistent with MM, MMR, and EV is not different across the three scenarios (a Friedman test of proportions of H0: proportion choosing MM in Scenario 1 = proportion choosing MM in Scenario 2 = proportion choosing MM in Scenario 3, yielded a p-value of 0.525). The aggregate data across Scenarios 1–3 seems to indicate stable criteria use by participants. To examine this stability further, we looked at individual consistency across the first three scenarios. To do this, we grouped participants by their “uncertainty tolerance,” ordering their choices from least uncertainty tolerant (choosing MM) to most uncertainty tolerant (choosing EV). The scale we created assigns a score of 0 for every choice consistent with MM, 1 for every choice consistent with MMR, and 2 for every choice consistent with EV, resulting in a scale varying from 0 to 6. For example, if a participant made choices consistent with MMR on all three scenarios, she would receive a score of 1 for each scenario, making her score on our uncertainty scale a 3; or if a participant who made two choices consistent with EV and one consistent with MMR would receive a score of 2(2) + 1 = 5. We removed participants whose choices did not seem consistent, which we defined as making choices that correspond to each of the three criteria. By removing these inconsistent data points, we removed the possibility of a single score referring to different sets of choices (for example, a participant who chose consistent with MM, MMR, and EV would receive a score of 0 + 1 + 2 = 3, similar to the first example above; thus, by removing these data points the redundancy in scores is removed as well). Figure 21.2 shows a histogram of the distribution of scores on this scale. Approximately 7.3 percent of participants fell into this “inconsistent” category. We conclude that most participants choose relatively consistently, i.e. their

Choice under “hard” uncertainty 375
40 35 30 Frequency (%) 25 20 15 10 5 0 0 1 2 3 4 5 6 Treatment 1 Treatment 2

Uncertainty scale score

Figure 21.2 Histogram of participant criteria selection for scenarios 1–3.

preferences regarding uncertainty are relatively stable. For example, approximately one-quarter of the sample chooses MMR in all three scenarios, and approximately one-half chooses MMR in at least two out of three scenarios. Further, most participants favor MMR (ranging from 60 percent to 71 percent) in each of the three scenarios.6 Treatment 2 The data from Treatment 2 is less clear than Treatment 1. We cannot reject the null hypothesis of random choice among all three prospects for Scenario 1 (for the χ2 test of proportions p-value = 0.191), although this hypothesis can be rejected for Scenarios 2 and 3 (both p-values 0.01 or less). The proportion of participants making choices consistent with MM, MMR, and EV is different across scenarios (a Friedman test of proportions of H0: proportion choosing MM in Scenario 1 = proportion choosing MM in Scenario 2 = proportion choosing MM in Scenario 3, yielded a p-value of 0.009). Under Treatment 2, behavior generally seems to be more random. Examining the “uncertainty tolerance” scale data for Treatment 2 (see Figure 21.2), behavior is more “conservative,” but also less smoothly distributed than under Treatment 1. Under Treatment 2, 12.5 percent of participants were classified as “inconsistent” (i.e. they made choices consistent will all three choice criteria). Comparison of Treatments 1 and 2 It appears there is some difference between behavior in Treatments 1 and 2, and that this behavior is changed in a predictable way, i.e. as Treatment 2 payoffs involve the same risk and less reward than Treatment 1, we see more conservative choices made. To draw statistical conclusions regarding the observed behavior, we

Two other results from Table 21.482) 5. Blackwell et al. which presents results from a simple model including only treatment parameters (Model 1).286 (4. = 1 if data taken from Scenario 2 Dummy variable.792 (2. inducing participants to choose more conservatively. = 1 if participant is a member of the Republican Party Participant’s declared hourly wage Dummy variable.376 C.333 (0.432) 0. perhaps because MM has a high guaranteed payout.416) 0.424) 23. = 1 if participant indicated s/he thought the actual probability of each outcome occurring was 50% 0 to 10 scale of risk aversion from Holt and Laury (2002). = 1 if data taken from Treatment 2 Dummy variable.325 (0.) 0.464) 0. although some non-intuitive results also appear.453) . = 1 if married Dummy variable.247 (0.7 Variable definitions and descriptive statistics are presented in Table 21.938) 0.486) 7. risk attitude is correlated with Table 21. Generally the regression analysis confirms the observations made earlier.6. = 1 if white Dummy variable. also towards EV. = 1 if male Dummy variable.234 (0.5.377 (0. but curiously. Dev. use a multinomial logit regression model.083) 0.747) 0.234 (0.515 (4.472) 0. First. Results of the regression analysis from both a trimmed and extended model (including demographic variables) are presented in Tables 21.714 (0. we see that participants move towards MM under Treatment 2.688 (0. respectively.6 and 21.333 (0.779 (0. = 1 if participant has previously participated in a psychology experiment Participant’s age Dummy variable. a higher score indicates more risk aversion Dummy variable.364 (0. = 1 if participant was recruited from Weber State University campus Mean (St.472) 0. Scenario 2 pushes choices towards MM.6 also fit well with intuition. Participants choosing among the prospects of Scenario 2 are also more likely to choose consistent with MM under Treatment 2.469) 0.5 Variable definitions and descriptive statistics Variable name 50–50 RiskAtt Scenario 2 Scenario 3 Treat 2 SocSci PsychExp Age Married Male Kids White Republican HrWage WSU Definition Dummy variable.801) 0. = 1 if participant majors in a social science Dummy variable.432) 0.247 (0.7. = 1 if data taken from Scenario 2 Dummy variable. In Table 21. = 1 if participant has one or more children Dummy variable.

when participants assessed a subjective probability of 50 percent of each outcome occurring.623 (0.4047) –0.4553) –0.3720) 231 –187.5514) –0.0448) 0.614 (0.093 (0.455 (0.361 (0.5042) 0.5638) –0.469 (0.6749) Note Numbers in parentheses are standard errors. i. * and **indicate significance at the 5 percent and 1 percent levels.0593) –0.210* (0.161 (0.4818) –0.231 (0.0893) 0.320** (0.4364) 0.4514) 1.8701) 0.267 (2.256 (0. they were more likely to choose EV.5828) 0.181 (1.431** (0.095) 0.6323) –0.111 (0.7186) 0.508 (0.4119) 0. respectively.559 (0.230 (0. Table 21.5058) 0.4488) –0.508** (0.Choice under “hard” uncertainty 377 Table 21.4149) –0.2822) 0.840* (0.585 (0.0852) –0. the choice of EV max corresponds to choosing airplane in Scenario 1.2524 Choice = EV max –3.583** (0.5215) –0.111 (0.3611) 0.4924) 1.5292) –0.143 (0.730* (0.6474) 1.879* (0.293* (0.325 (0. respectively.4866) –0.257 (0.6167) –0.450 (0.7 Multinomial logit model 2 Variable Constant 50–50 RiskAtt Scenario 2 Scenario 3 Treat 2 SocSci PsychExp Age Married Male Kids White Republican HrWage WSU No. * and **indicate significance at the 5 percent and 1 percent levels. behavior in a sensible way. Prospect 3 in Scenario 2 and Persimmon Bay in Scenario 3. under the same circumstances participants were also more likely to choose MM.6 Multinomial logit model 1 Variable Constant 50–50 RiskAtt Scenario 2 Scenario 3 Treat 2 No.6589) 0.5546) 1. In Treatment 2.135 (0. observations Log likelihood Restricted log likelihood Choice = minimax –2.196 (0.1337) 0.026 (0. Second.e.027* (0.076* (0.108 (0.4432) 1.4299) –0.191 (0.4639) 0.177 (0.1140) 0.2524 Choice = EV max 0. . Puzzlingly.594 (0.0785) –0. a lower level of risk aversion is correlated with increased choice of EV.125 (0.5769) 231 –172.5321** –221.881* (0.312* (0. observations Log likelihood Restricted log likelihood Choice = minimax –5.413 (0.7065) 0.5327) 0.4509) Notes Numbers in parentheses are standard errors.5187** –221.4531) 1.537** (0.

Scenario 1 involves a decision about travel with weather as the source of uncertainty. this axiom seems a likely starting point for more research. This result seems to imply that at least one of the axioms of rationality postulated by Arrow and Hurwicz (1972) is being violated by a fairly large number of participants. The results of a more extended model (Model 2). A fruitful analysis would seem to use the minimax regret criterion as a starting point. are presented in Table 21.7. but the criterion used was mixed within the pool and was affected by the risk–reward structure that varied by treatments. We suspect these participants have learned not to trust experimenters (many psychology experiments involve deception). but look for some other parameters to model to help improve predictive power. including demographic variables. Blackwell et al.378 C. Because the minimax regret violates the independence of irrelevant alternatives assumption. the more recent work of Nehring (2000) . Interestingly. the most conservative strategy. Here we see again that participants are more likely to choose MM under Treatment 2 than under Treatment 1. We also see that lower levels of risk aversion lead to a higher likelihood of choosing EV. there is the possibility that framing effects may be tainting our results. Discussion and conclusions Before discussing the conclusions and implications of this research. as with any experiment. Most of the socio-economic and demographic information we collected does not help predict behavior with one notable exception: previous participation in a psychology experiment. and generalizing from our results should be done with care. The scenarios are analytically similar. For example. we should mention some of its limitations. because the five scenarios are framed differently. while Scenario 3 asks the decision maker to make a different sort of travel decision. For example. Across this set of exploratory experiments. the decisions participants make are artificial. and the uncertainty comes from the decisions of other fishermen. In addition. Milnor’s Hurwicz criterion has the decision maker choose a parameter α to weight the minimum and maximum values in an expected value-type maximization. supporting the notion that MM and MMR are the more conservative decision criteria. In addition. Students who were previous participants are more likely to choose MM. these two framing effects may also have had an impact on participants’ decisions. participants in our experiments did not seem to use the minimax choice criterion particularly often. and so try to guarantee themselves the highest payoff possible assuming the uncertainty will always be resolved to the experimenters’ favor. a strong majority of participants (approximately 90 percent) showed some degree of consistency in their choice of decision strategy. but the tradeoff made is some ambiguity about interpretation of our results. and the decision to frame them differently was made in order to maintain participant interest. and the results appear more consistent with MMR than minimax. Second. we did not randomize the order of the scenarios or the physical position of the choices on the screen. First.

to simulate better the decision policy makers or legislators face (say. a decision maker might pick auto with 40 percent probability and airplane with 60 percent probability in Scenario 1) that is a function of all the possible payoffs in the decision-making process. For example. when choosing only for himself. For example. 1995. A particularly important set of future treatments would be to vary the impact of the decision from one individual to groups of individuals. as one of the original motivations for this research. that same individual may use minimax regret instead. 2004). But experimental methods are being applied in an increasingly complex set of tasks and contextrich settings (Harrison and List. as to future research into decision making under hard uncertainty. but when making a decision that will affect the group or a greater collective. The size of the benefits (and hence the regret) could also interact with the group versus individual treatment effect. Given our mixed results. a majority of decisions under hard uncertainty in this stylized experimental setting were consistent with the MMR criterion. This approach differs from Arrow and Hurwicz (1972) in emphasizing that the decision maker should weight all the payoffs. such as those suggested by Milnor. an individual may simply use the minimax strategy. 1996. the literature on SMS-type approaches emphasizes that they are collective choice process that are particularly appropriate in complex settings with pluralistic perspectives (Randall and Farmer. B. This provides at least some modicum of support for Palmini’s (1999) argument for a minimax regret justification for policy approaches such as the SMS. Future experiments should investigate the impact of partial information on decision making (for example. One is differentiation among other competing approaches. but it is unclear how to operationalize “conservative” – as minimax or minimax regret! It is quite possible. and pluralistic moral perspectives. More specifically. or in potential revisions). Berrens. 2001).Choice under “hard” uncertainty 379 presents the possibility that decision makers may use a sort of mixed strategy (for example. The impact of more than two events occurring (say A. 1986) have argued that individuals may distinguish between their consumer preferences versus their citizen preferences. under the Endangered Species Act. The impact of increasing the number of prospects beyond three should be investigated. More generally. and more complex criteria. and C) would allow for a much richer decision environment. Castle. it remains an open question whether participants are using some alternative choice criterion than the three we have presented. however. It is expected that this treatment would induce more “conservative” decisions.g. the subject is told events . that an individual may use a different strategy if their choice affects the entire group. not just the minima and maxima. Future experimental designs may need to seek ways to add or emphasize irreversibility. and that there may be a variety of criterion present in any given population. one treatment could ask one participant to choose a prospect for all the participants in the group. A variety of authors (e. Sagoff. like the Hurwicz criterion. While there is no predominant (clearly superior) decision strategy in our results. collective (group) choice. there are some obvious areas that merit investigation.

3 The experiment is online. 71 (5): 1083–1101.. Maskin. Rogers and Sinden. 1997). In closing. A.edu/tgrijalva/SMS/SMSpage. Random amounts of pink and white candies were placed in each bag.. and Hurwicz. Carter and J. We hope that this exploratory research stimulates additional research into decision making under hard uncertainty. 2006). and Tietenberg. Drucker. Levin. P.. leximin and the protective criterion: characterization and comparisons. American Journal of Agricultural Economics. One important example of the relevance of such understanding is in debates over the “rationality” of Safe Minimum Standard (SMS) approaches to the protection of endangered species and biodiversity (Farmer and Randall. 5 The probabilities were unknown to the experimenters as well. Starrett. Solomon et al. Farmer. Oxford: Basil Blackwell. S.F. In: C. 2003). 1401–1406.J. 6 We have also tested for differences between the responses of Weber State University and College of Charleston undergraduates and found little evidence of a difference. Uncertainty and Expectations in Economics: Essays in Honour of G. Sterner. 7 Four participants had missing values for either RiskAtt or 50–50. If there is a trend in the literature.J. and the probability of event A occurring is 10 percent. References Arrow. certainly.weber. These theories allow the sum of the decision maker’s subjective probabilities over a distribution of outcomes to be other than one. or C will occur. L.. Daily. and incrementally improving our understanding. K-G. Maler. experimental economics offers a low-cost tool for learning about behavior.htm (accessed November 2003). E. . But. 46 (1). such as expected impacts on the further future due to global climate change (Woodward and Bishop. 34–44. of which Choquet is an example. 1998 and 1999. S.S. or the gap between rapid introduction of nanotechnologies and the expected social and environmental impacts (Munysiwalla et al. B. and Jackson.L. 2005.. Arrow. D. Barbera. 1994. there is considerable room for improving our understanding about how people make choices under situations of hard uncertainty.. 34. S. 1972. See Woodward and Bishop (1997) for a discussion. Sustainable development: challenges to the profession of agricultural economics. 1–11. T. ed. Dasgupta.. 1998). available at: faculty. during or after the experiments. Journal of Economic Theory. Managing ecosystem services. M. Notes 1 Case study applications of the SMS remain limited (e. Blackwell et al. 1980. Global climate change is another problem that is characterized by uncertainty and irreversibility.. there are other important arenas.L. 2000. decreasing the data set by 12 observations (three data points for each participant). and no attempt was made to count the candies before. 1988. K. G.. Bishop. Shackle.. 4 Although participants made decisions for all five scenarios. 2001.g. T. Maximin. K. Environmental Science and Technology.... Batie. Ford. it is towards a greater acceptance of non-additive utility theories. 1989. but no information about B or C is given).. Given such dilemmas. An optimality criterion for decision-making under ignorance. for the purposes of this chapter.. Berrens et al.380 C. we will focus only on the first three scenarios. 2 See Schmeidler (1989) for a description of Choquet Expected Utility.

and Laury. M. ambiguity and the Savage axioms. Farmer. Camerer. 147–161. and Schmidt. . D. 75 (4). 5 (4)... 1998. Journal of Economic Surveys. M. Berkeley. and Tisdell. Pluralism and pragmatism in the pursuit of sustainable development. E. American Journal of Agricultural Economics. 92 (5). In: W. Holt. E. Bulte.. R. 30 (3). UK: CAB International. R. 104–116. 287–302. Hogarth. 1993. Getting the safe minimum standard to work in the real world: a case study in moral pragmatism.. 1992. 1961.. Risk. Environment and Development Economics. 2 (2).. Ellsberg. 108–130. 2004. Resource Conservation: Economics and Policy. Uncertainty and Profit. 1993. How useful are environmental safety standards in economics? The example of safe minimum standard for protection of species. 1952. Kelsey.Choice under “hard” uncertainty 381 Berrens.. 9. Recent developments in modeling preferences: uncertainty and ambiguity. Conservation Biology. 77–94. C. S...C.A. 1980. 1999. 325–370. and Kunreuther.. endangered species and the safe minimum standard. Field experiments. I. 113–120. R. 2000. 643–669. G. 65–88.. Drucker. 1009–1055.. Camerer.. 38 (2). The rationality of a safe minimum standard. 1992. Phillips. Berrens.. Economics and Environment.. 1978. C. American Economic Review. ed. Economic analysis. S. M. 60 (1). 1996. M. C. 133–153. 2001. 461–474. 45. Theories of choice under ignorance and uncertainty. 15–36. Wallingford. F. H.. Luini.. McKee. Biodiversity and Conservation. 11 (1).. Expected utility with purely subjective non-additive probabilities. 74 (3). P. Harrison. Boxall. Journal of Risk and Uncertainty. Journal of Mathematical Economics. and Weber. R. E. Risk aversion and incentive effects. Risk. 1998. and W. Ecological Economics. In: L.C. White eds.. McKee. J. Northwest Environmental Journal. Uncertain Decisions: Bridging Theory and Experiments. CA: University of California Press. Brookshire. 42 (4). Ciriacy-Wantrup. Berrens. Transactions of the North American Wildlife Conference. endangered species and biodiversity loss. models and applications.. 1644–1655. 28 (2). and List. 209–226. Ecological Economics.M. Ambiguity-aversion and non-additive probability: experimental evidence. M.. R. MA: Kluwer Academic Publishers. 208–218. A. Bishop. Journal of Economic Literature. D. 6 (2). An application of the use of safe minimum standards in the conservation of livestock biodiversity. Implementing the safe minimum standard approach: two case studies from the US endangered species act.A. 1995. 74 (2).K. R. Journal of Risk and Uncertainty. Castle. 1921. Boston.. Castle. Environmental Conservation. and Randall. A. Quarterly Journal of Economics.. 1987. Bishop. 53–79. Land Economics. M. G. A.C. 2002. Forestry. Endangered species and uncertainty: the economics of a safe minimum standard. Luckert. Knight.V. 2001. Decision-making under ignorance: arguing with yourself. 10 (1)... C. and Van Kooten. Endangered species: an economic perspective. 10–18. 168–181. M. Boston. and Farmer. C. D.. 2006. 1–9. R. and Quiggin. 1999.. Adamowicz. J. MA: Houghlin Mifflin. Land Economics. W. 14 (1)... Incorporating distributional considerations in the safe minimum standard approach: endangered species and local impacts. Hohl. Gilboa. Farmer. 16 (1). Economic science. The safe minimum standard of conservation and endangered species: a review. and Berrens.

.. costs and the safe minimum standard of conservation. K.. Land Economics. 571–587. Blackwell et al... The Handbook of Environmental Economics. J. Daar. New York: Wiley. Van Kooten. Luini. J. Quiggin. 1994. Safe minimum standards for environmental choices: old-growth forest in New South Wales. 70 (4). 64–68. Benefits. 11 (3). How to decide when experts disagree: uncertainty-based choice rules in environmental policy. A. 1999. 50 (1–2). 1995. In: M. Milnor. 14. and Bulte. The Florida Manatee and ecotourism: toward a safe minimum standard. National Research Council of the National Academies. Rogers. Ecological Economics. Econometrica. 101–115. A. Economics and the debate about preservation of species. Malden. Decision making under ignorance with implications for social choice. . 205–240. 3 (4). 12.. 2003. R. Sagoff. Theory and Decision. and Farmer. Bromley. 2 (1). Subjective probability and expected utility without additivity.. and the game theoretic foundations of the safe minimum standard: a reassessment. D. 1990. 77–90. Munysiwalla. Ambio. Solomon. Valuing Ecosystem Services: Towards Better Environmental Decision-Making. R. Mind the gap: science and ethics in nanotechnology. 1986. 1979. and Bishop. 1997.. 2002. 48 (3).. Cambridge. Toman. 1982. and Singer. Ecological Economics. 1999.. and Sinden.C. M. C. Tisdell. 1991. 323–343. MA: Blackwell. risk aversion. J. Ethics. In: D. P. 2000.. The value of biodiversity. A theory of rational choice under ignorance. B. Boston.T. Ready. E. Vaughn. Randall.. D. 309–312. Theory and Decision. The Economics of Nature: Managing Biological Assets.. R9–R13. G. crop variety and genetic diversity. 41 (2). M.... 1997. (Water Science and Technology Board) 2004. and Halvorsen. 399–413. Journal of Economic Behavior and Organization. Corey-Luse.C. Economics and sustainability: balancing trade-offs and imperatives. 96 (2). Randall. 73 (2). Vercelli. 57 (3). In: L. K.. C. Values and preferences.. MA: Blackwell Publishers. 20 (2). ed. 29 (3).. Journal of Environmental Management. The recent advances in decision theory under uncertainty. 237–260. R. A. 1989. 89–99. Seigfried Von Ciriacy-Wantrup and his safe minimum standard of conservation. Land Economics.C.382 C. A theory of anticipated utility. Nanotechnology. 73 (4). WSTB. 301–316.. 319–337. 1964. A. E. 1994. Shubik. American Journal of Agricultural Economics. and Bishop.C. Maskin.. Choices. Endangered species and the safe minimum standard. 492–507. Games against nature. R. A. Palmini. G. Uncertainty. Schmeidler. 26–44.. M. Ecological Economics. Washington DC: National Academies Press. Uncertain Decisions: Bridging Theory and Experiments. 1991. 2005. 463–472. ed. 20–31. Woodward..C. MA: Kluwer Academic Publishers. M. Game Theory and Related Approaches to Social Behavior. ed. Nehring. 30–33.

it is the experience in the market that makes consumers rational. and if rationality can spill over from one market to another. and Jason F. A rational consumer has experience in markets. If preference reversals are a sign of irrationality. such as an environmental good. Shogren Introduction Rationality on the part of economic agents is presumed in economic models. Heiner. 2006). Do these reversals lend credence to the theory consumers are not rational? Can we achieve rationality in this context or are preference reversals persistent? The observed differences between assumed optimizing behavior in economic theory and the actual behavior of people can be impacted through repeated market transactions with large enough sums of money at stake for the individual (see Smith. 1989.22 Rationality spillovers in Yellowstone Chad Settle. 1983). 1955. 1979). But the limitations reduced cognition puts on the types of rationality assumed in economics may or may not be a binding constraint. Shogren. Recent research has shown rationality can be increased in situations in which preference reversals are prevalent (Cherry et al. (2003) has shown rationality spillovers exist in laboratory experiments. 1990. if rationality can be learned through market experience. While the work of Cherry et al. has experience with the available bundles of goods.2 Constraints on individuals’ cognition and humankind’s physiological limits on cognition play a role in the ultimate ability to process information (see Simon. 1993. 2003). these laboratory experiments may need to be applied to specific problems. high severity event. one being the inconsistent choices consumers make when choosing preference between two bundles of goods and then being asked for their willingness to pay for the two bundles – preference reversals (Grether and Plott. Applying this method to a . The laboratory experiments have shown rational behavior can spill over from one market to another.. and has clearly defined preferences over those bundles. It is possible economic agents could learn from market experience and cognition isn’t the limiting factor. from a market context to a hypothetical context with a low probability. Todd L. Cherry. Market experience can then be used as a tool to help consumers make more rational decisions. but rather inexperience is. Smith and Walker.1 The phenomenon of preference reversals shows a potential failure in economic theory. This rationality has come into question in the literature for a variety of reasons.

took a break. Lake trout are an exotic species to Yellowstone Lake and are a predator of the native and popular species. river otter.4 Not only are cutthroat trout expected to decline in number. which can easily be changed in future experiments. A detailed description of the experiment as presented to each participant in the experiment is followed by the results of the experiment and the conclusion. cutthroat trout. We wish to determine preferences for and values of seeing species in and around Yellowstone Lake that might be affected by the introduction of an exotic species into Yellowstone Lake. issues dealing with control of subjects are important. as well as many other species in Yellowstone National Park. but are also an important food source for grizzly bears. specific problem requires that we take the theory from the laboratory to the field and target people who are interested in these non-marketed goods. Cutthroat trout are not only important to fishermen who come to Yellowstone Lake to fish for cutthroat.3 This research is an attempt to take the rationality spillover design from the laboratory into the field. given the current technology and given we are unable physically to be in the same room as the participants. Settle et al. whether it is done in a laboratory or on the internet. we cannot only elicit values. some are due to this experiment being conducted over the internet and beyond our control. In the internet experiment we did not control for: • • • • whether people started the experiment.384 C. We then give the reader the specific challenges of gathering data in this setting since this experiment was designed to gather information for a larger project. Specific issues listed below were due to the particular experimental design used. the number of people who could be looking at the screen at the same time helping the participant think about how to make his or her ch