Intergenerational Wellbeing and Public Policy: Girol Karacaoglu Jacek B. Krawczyk Anita King

Girol Karacaoglu · Jacek B.
Krawczyk ·
Anita King
Intergenerational
Wellbeing and
Public Policy
An Integrated Environmental, Social,
and Economic Framework
Intergenerational Wellbeing and Public Policy
Girol Karacaoglu Jacek B. Krawczyk
• •
Anita King
Intergenerational Wellbeing
and Public Policy
An Integrated Environmental, Social,
and Economic Framework
123
Girol Karacaoglu Jacek B. Krawczyk
School of Government Flinders University
Victoria University of Wellington Bedford Park, Adelaide, SA, Australia
Wellington, New Zealand
Anita King
The New Zealand Treasury
Wellington, New Zealand
ISBN 978-981-13-6103-6 ISBN 978-981-13-6104-3 (eBook)

https://doi.org/10.1007/978-981-13-6104-3
Library of Congress Control Number: 2018967443
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd., part of Springer Nature
2019
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained herein or
for any errors or omissions that may have been made. The publisher remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
The truth is the whole. The whole, however, is
merely the essential nature reaching its
completeness through the process of its own
development.
George Wilhelm Friedrich Hegel
(1770–1831). Preface, Phenomenology
of Mind, 81–82.
Girol Karacaoglu would like his contribution
to this book to be dedicated to Angela,
Melissa, Michael, and Natalia.
Jacek B. Krawczyk dedicates his contribution

to Diana, Aniela, Clara and Viktor.
Anita King dedicates her contribution to

Richard, Peter, and Christina.
Preface
This book is about the design, implementation, and evaluation of public policy
when the objective is to improve and sustain intergenerational wellbeing. Wellbeing
is defined in terms of the ability of individuals and communities to live the lives
they value. Wellbeing has several dimensions (collectively referred to as “com-
prehensive consumption”) and a number of sources (“comprehensive wealth”). It is
argued, and demonstrated, that unless an integrated environmental, social, and
economic policy framework is adopted, public policy cannot achieve its objective.
When we introduce radical uncertainty and complexity into the environment, the
policy framework changes significantly. Satisficing1 replaces optimising as the
modus operandi of policy, and building systemic resilience emerges as a primary
focus for delivering the policy objective. Then, a wellbeing-focused public policy is
no longer about creating and delivering predetermined outcomes. The substantive
involvements of communities in the definition of desired outcomes, as well as in the
design, implementation, and evaluation of public policies aimed at delivering those
outcomes, become critical.
Philosophical and Institutional Foundations
The ultimate purpose of public policy is to improve the lives of individuals and
communities, as they choose to live them, now and into the future.
We do not know how each and every individual wishes to live his or her life, nor
do we wish to pass judgement on how they should be living their lives—as long as,
that is, they do not prevent others from living the lives they value. There are an
infinite number of possible lives, shaped by personal circumstances, including
capabilities, opportunities, and preferences, as well as cultures, religions, political
arrangements, geographical surroundings, and so on.
1
This neologism was introduced by Simon (1955).
ix
x Preface
After quoting Thomas Jefferson (“The care of human life and happiness […] is
the only legitimate object of good government”), Layard and O’Donnell (2015) go
on to write “What should be the goal of public policy? We agree with Thomas
Jefferson. What matters is the quality of life, as people themselves experience it.
And the best judge of each person’s life is that same person. Is she happy with her
life; is she satisfied? In a democracy that should be the criterion for good policy”
(ibid., p. 77).
We adopt this value judgement and the associated ethical value system as a
foundation for good public policy. In doing so, we firmly ground the axiomatic and
philosophical underpinnings of this book in the works of Adam Smith (see Smith
(1790), Smith (1937)) and Amartya Sen (see Sen (2009), Sen (2017)), who jointly
stand “in the tradition of classical political economy, where a separation of ethics
and economics was unthinkable” Eiffe (2010) (p. 17).2 Sen (2017) is a must-read
for an understanding of the theoretical underpinnings of the argument that, in a
policy context, we cannot develop and provide economic advice in an ethical void.3
In this book, we associate an individual’s wellbeing with his or her ability to live
the kind of life he or she values, on a sustainable basis. Good public policy con-
tributes to individuals’ wellbeing by enhancing their capabilities and opportunities
(i.e., substantive freedoms) to do so.4 Whether actual public policy pursues this
purpose or not, and whose wellbeing is the focus of public policy, depends on the
institutions that exist in a society. These institutions are the outcomes of social
decisions, made by humans, chosen for their consequences; they reflect the evo-
lution of the distribution of power in society. The distribution of power in turn is a
reflection of the distribution of political and economic resources (see Acemoglu
et al. (2005) and Acemoglu and Robinson (2013)).
Clearly, providing the “enablers” (in the form of improved capabilities and
opportunities) to achieve a better life is simply a necessary condition; it does not
guarantee the achievement of better outcomes. Indeed, “the key idea behind the
capabilities approach is that individuals differ in their ability to transform resources
into wellbeing or ‘flourishing’” (Chakravarty and Lugo (2016), p. 246). If, to take
an example, despite best efforts in the formulation and implementation of public
policy, youth suicide keeps increasing in various communities around the country,
should this be a concern for public policy as well? Our answer is an emphatic,
“yes”. We pay particular attention to this issue, of adjustments to how policy should
2
See Markey-Towler (2016) for an attempt to demonstrate mathematically that economic policy
cannot be developed in a political and philosophical vacuum.
3
Needless to say, there are numerous other possible philosophical perspectives and ethical value
systems that can serve as a platform for public policy. See Hausman et al. (2016), Sen (2017), and
Sen (2009) for detailed discussions of some of these alternatives.
4
Needless to say, in a democracy, it is unlikely that the electorate will be agnostic about the lives
some people may choose to live, if sustaining those lifestyles requires considerable taxpayer
support. We wish to thank Bryce Wilkinson for raising this point. We address some of these
complexities in Chap. 5. In particular, we suggest there that giving substantive voice to local
preferences in the design and implementation of public policy, and providing appropriate funding
in support, may be one way of addressing this dilemma.
Preface xi
be designed and implemented to achieve desired outcomes (especially social and

environmental outcomes), in Chap. 5. This is precisely why we put so much
emphasis on both individual and community wellbeing in the design and imple-
mentation of public policy. When, despite best efforts, public policy does not
achieve the desired environmental, social, and economic outcomes, it is the
communities that bear the brunt of the associated costs. And, as we shall argue in
Chap. 5, in a world of complexity and radical uncertainty, communities also emerge
as the most effective and efficient instruments of public policy.
Analytical Framework
Suppose we do have the institutions in place to pursue good public policy. How
then do we resolve the dilemma presented by our desire to improve people’s lives,
without knowing, or wanting to pass judgement on, how people wish to live their
lives? In other words, how would we know whether we are succeeding in achieving
our objective?
To assist with answering this question, we start with a broadly defined “well-
being function” as the basis of our analysis, and as a guide to wellbeing-enhancing
public policy. In doing so, as a first step, we follow Arrow et al. (2012) and define
the object of interest for good public policy as intergenerational wellbeing.
Although we do not know, nor do we wish to pass judgement on, what lives
individuals and communities value, we have the benefit of numerous studies,
covering a large variety of countries and cultures across time, to help us identify the
broader influences on individual and community wellbeing. In other words, based
primarily on robust, survey-based, empirical evidence, we have a broad sense of the
common elements of what individuals and communities value (see Boarini et al.
(2014), Fleche et al. (2012), Smith (2015)). Arrow et al. (2012) refer to these
collectively as “comprehensive consumption”. They include not only standard
marketed consumption goods and leisure, but also others such as health status,
environmental quality, personal security, social connections, and so on. These
correspond to the components of the OECD’s Better Life Index (BLI) (see Durand
(2015)). They are the “functionings” (beings and doings) that people have reason to
value (see Sen (2009), Alkire (2016)).
The ultimate source of comprehensive consumption is “comprehensive wealth”
(see Arrow et al. (2012)). This refers to the stocks of capital assets, broadly defined,
that yield income and other sources of wellbeing now and into the future (ibid.,
p. 320). These assets include economic (or built) capital, human capital, natural
capital, and social and cultural capital.
It is by expanding, enhancing, and protecting, these capital assets, as well as
ensuring widespread access to them across society and across generations, that
good public policy can push out (i.e., expand) our collective “wellbeing frontier”, as
we will explain in Chap. 1. To repeat, whether good public policy will be pursued
xii Preface
or not will depend on the institutions that prevail. These institutions in turn reflect
the evolution of the distribution of power in society.
We can assess whether we are pursuing and achieving good policy objectives
through four (complementary) mechanisms. First, by establishing whether more
resources are being made available to support better lives—say, more funding for
housing. Second, by assessing whether these resources are actually being converted
into higher capabilities and opportunities for a better life (i.e., functionings)—say,
better (quantity and quality of) housing. Third, and as a complementary step to the
first two, by simply asking people, through various survey-based assessments of
subjective wellbeing, whether their lives are actually improving or not. As a subset
of, and complement to, the third mechanism, we can also specifically target the
most disadvantaged people in the population, and ask them whether their lives are
improving.5
Where intergenerational wellbeing is the objective of public policy, there needs
to be a fourth assessment specifically targetted at evaluating whether public policy
is creating the platform for sustainable intergenerational wellbeing. To anticipate
the discussions of Chaps. 1 and 3, in this latter context, good public policies are
those that expand the “wellbeing frontier”.
Complex notation we sometimes use is carefully explained when we introduce it.
Independently, the Glossary (from page 235) lists the majority of symbols used in
the book.
Extensions
The first part of the book, comprising the first four substantive chapters, is essen-
tially based on an expanded version of a standard intertemporal optimisation model.
It asks and answers the question, “how far can we stretch this standard model, to
serve as an analytical vehicle for a public policy that is focused on enhancing
intergenerational wellbeing”? The main “stretch” is in the construction of a stylised
model that integrates environmental, social, and economic influences (and, in
particular, externalities associated with these influences) on wellbeing.
5
One may legitimately ask, of course, why we would bother with all these three steps, instead of
simply going directly to the third one, and asking people how they are feeling about their lives.
One of the several answers to this question is adaptive preferences; by way of example, consider
an athlete who becomes resigned to the fact that she will never represent her country in the
Olympics, but is grateful for any improvement in her performance over time. The contributions in
Adler and Fleurbaey (2016), including Dolan and Fujiwara (2016), Clark (2016), and Alkire
(2016), provide very useful critical surveys of the literature on both the objective and subjective
ways of measuring progress in wellbeing. See Alkire (2016) in particular, which has numerous
references to Amartya Sen’s foundational work, about the necessity of using all three mechanisms,
or steps, listed above in assessing the effectiveness of a policy that has improving wellbeing as its
objective. In this broad context, Stiglitz et al. (2010) also provide an extremely useful reference.
Preface xiii
Then, in the second part of the book, in Chap. 5, we introduce two critical factors
that cannot be ignored: radical uncertainty (see King (2016), Chap. 4) and com-
plexity (see Arthur (2014)). Under radical uncertainty, with the set of possible states
of the world and/or their probabilities unknown, optimisation at an individual or
policy level becomes impossible. Under complexity, in a social context, actions and
outcomes cannot be separated—there is “reflexivity”: there are no external (ex-
ogenous) social or economic facts that are independent of our actions. Individuals’
actions and interactions with others influence the social and economic outcomes
that emerge, which in turn lead to reactions, and on it goes (see Soros (2003)).
Such extensions do not alter the objective or domains of public policy—but they
do have a major bearing on the primary focus and instruments of policy.6 The
primary focus of policy switches to building resilience to systemic risks. This serves
two purposes. First, it attempts to protect the environment, society, and economy
from catastrophes. Second, it aims to create a fertile environment that nourishes
creativity. There is also a parallel, and complementary, increase in the reliance on
local- or community-based solutions and implementation mechanisms in addressing
public policy challenges.
Readership
The primary audience for this book are policy analysts and advisers, and academics.
The book is also intended to serve as supplementary material, for undergraduate
and graduate courses on public policy, wellbeing economics, welfare economics, as
well as economic growth and development.
The first (Introduction and Overview) and last (Themes) chapters of the book are
written for general readership; they do not require any particular technical back-
ground. The remaining chapters of the book include some technical material. An
introductory part, for each section of each chapter, presents the basic material in
words and with the help of diagrams. This is followed by a mathematical exposi-
tion, which serves two purposes. First, it provides a tool for policy analysis and
numerical simulations. Second, it is intended to demonstrate that one does not need
to compromise rigour to develop rich, multi-dimensional, policy-informing models.
Readers can skip the technical parts of each chapter without losing the substance
of the argument. Readers who wish to read the technical parts of each chapter need
to be familiar with calculus, including partial differentiation, as well as intermediate
micro- and macroeconomics. Readers who would like to follow all our derivations
and apply the proposed mathematical formulae should be familiar with difference
and differential equations. For those who need it, we suggest supplementary
readings on the technical parts of each chapter.
6
See Schmidt (2015) for an extremely useful discussion, from an evolutionary perspective, on the
implications of complexity and fundamental uncertainty for the justification, design, and
implementation of public policy.
xiv Preface
Where appropriate and relevant, the substantive chapters include references to

the empirical literature in support of the main assumptions that underpin the various
analytical models that are used throughout the book. At the end of each substantive
chapter, we also offer some challenges for further consideration that build on the
material presented in the chapter. These come in two distinct parts. Part one is
non-technical, targeted at readers who do not wish to study the mathematical
sections of the chapter. Part two contains challenges of a more technical nature,
based on the mathematical contents of the chapter. These two types of challenges
should help student–readers to extend what they have learned in each chapter. To
help with that, we suggest some references for further reading.
Organisation of the Book
The book has seven chapters organised into two parts. A common thread that runs
through all the chapters is the specification of intergenerational wellbeing as the
primary objective of public policy. A second, and complementary, thread that runs
through the book is the absolute imperative of adopting an integrated approach to
environmental, social, and economic policies. These common threads are woven
together through numerical simulations, examples, and case studies, to demonstrate
the interaction between environmental, social, and economic variables towards
enhancing intergenerational wellbeing on a sustained basis. This is where the use of
formal or mathematical models throughout the book comes in handy. The main
purpose of introducing these models is to identify and formalise the key interde-
pendencies between environmental, social, and economic influences on wellbeing.
In the absence of such formal models, it is very difficult, if not impossible, to be
able to quantify and simulate the influences of alternative policy interventions on
wellbeing.
Part one of the book comprises the first four chapters.
Chapter 1, Introduction and Overview, is the introductory chapter and provides
an overview of the contents of the book.
Chapter 2, Stylised Model, introduces stylised mathematical models of several
environmental, social, and economic processes, crucial for coordinated policy
formulation. Section 2.2 is on Consumption, Wellbeing, and Wealth; Sect. 2.3
covers Production and Wellbeing; Sect. 2.4 is on Ideas and Technology; Sect. 2.5
on Finance - Money, Credit, and Banking; Sect. 2.6 on International Connections;
Sect. 2.7 on Pulling the Pieces Together - Market Clearing and Equilibrium; and
finally Sect. 2.8 on Environmental, Social, Economic Externalities.
Chapter 3, Wellbeing-Enhancing Public Policy—Stylised Model, attempts to
extract policy guidelines based on the insights gained from the stylised model of
Chap. 2. Here we focus more exclusively on the formulation of public policy. We
also provide an example of how the model works, through a numerical simulation
of the impact of a policy move that is intended to reduce the level of poverty. One
Preface xv
of the main motivations behind this particular example is to demonstrate the

interaction of environmental, social, and economic influences on wellbeing.
Chapter 4, Operational Policy-Simulation Model introduces, and demonstrates
the purpose and uses of, a policy-informing simulation model that can be used to
operationalise the public policy insights suggested in Chap. 3. The chapter also
discusses ways of interpreting and communicating the results of the simulations
performed by such a model. The model is based on the eleven critical influences on
wellbeing, which are the components of the OECD Better Life Index (BLI). In
order to simulate the outcomes of policy actions, the model has to assume particular
linkages among the eleven components of the BLI, as well as between them and the
capital assets that provide a critical source for these ingredients. The nature of these
linkages (e.g., between health, leisure, productivity, production, and pollution) is
informed by, and consistent with, the stylised model of Chap. 2. The primary
purpose of these simulations is to demonstrate how a set of policy interventions,
informed by the trade-offs and complementarities between environmental, social,
and economic influences on wellbeing, can be designed to achieve a sustainable
improvement in wellbeing.
Part two comprises the last three chapters of the book.
Chapter 5, Wellbeing-Enhancing Public Policy - Extensions, explores the
implications of complexity and radical uncertainty for the design and implemen-
tation of public policy. The focus of policy shifts from directly influencing well-
being by manipulating key influences such as health, environmental quality, and
housing, to enhancing resilience. Resilience has two key dimensions, the capacity
to absorb shocks, and the ability to adapt in the aftermath of shocks.
Chapter 6, Viability Theory for Policy Formulation, formalises the concepts
introduced in Chap. 5. It introduces and demonstrates the uses of viability theory as
an alternative, and promising, modelling device for numerical policy simulations
when we do have satisficing (rather than optimising) agents, and the focus of public
policy is primarily on investing in building resilience. If achieving resilience is
defined as a problem of keeping some key indices in a closed set, for example
pollution below a certain level, then knowledge of initial states from which the
closed set can be attained, given the system’s dynamics and the strength of
instruments, is a viability theory problem. We demonstrate how various policy
instruments can be used to ensure that the whole integrated system remains within
sustainable bounds. Allowing for constrained instrument adjustments is an impor-
tant feature of solutions based on viability theory. For example, in the context of
policy making, the speed with which taxation can change may be constrained by an
election promise. Varying this speed and/or the bounds for the key economic
indices will produce solutions of different levels of resilience. Focus on resilience,
as the primary domain of public policy, is the key link between Chaps. 5 and 6.
Chapter 7, Themes, summarises the main themes of the book.
Readers will note, as they go through the book, that there are quite a few
references to New Zealand in our policy discussions and examples. There are two
reasons for this. First, although it is a small country, certainly in terms of its
population size of less than five million people, New Zealand is one of the very few
xvi Preface
countries in the world that has chosen to adopt a wider wellbeing framework in its
design, implementation, and evaluation of public policy. The ongoing intense effort,
on the part of the New Zealand public sector, towards operationalising this aspi-
ration—partly in preparation for the Government’s 2019 “Wellbeing Budget”, may
be of interest to a wider international audience. Second, all three of us lived and
worked in New Zealand for a considerable period of our lives—in fact two of us
(Anita and Girol) still live and work in New Zealand.
Wellington, New Zealand Girol Karacaoglu

2018 Jacek B. Krawczyk
Anita King
Acknowledgements
This book uses some results from earlier work, as well as joint research carried out
with different co-authors.
In particular, parts of Chaps. 1–3 draw on the work that Girol Karacaoglu started
when he was the Chief Economist of the New Zealand Treasury (see Karacaoglu
(2015)).
Similarly, Chap. 4 draws on the work that Anita King is currently doing at the
New Zealand Treasury (see King (2018)).
Jacek Krawczyk thanks his research assistants Messrs. Wilbur Townsend and
Maximilien Driscoll-O’Keefe for their input into the creation of dynamic inequality
models and Professor Kunhong Kim from the Auckland University of Technology
(formerly from Hallym University, Korea) for his work on adapting a growth model
to a macroeconomic pollution control model. Jacek also wishes to thank Kunhong
and Professor Ken Judd from Stanford for multiple discussions, which have led to
improvements in Jacek’s understanding of the links between viability theory and
macroeconomics.
xvii
Contents
Part I
1 Introduction and Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 Justification for Public Policy . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Purpose of Good Public Policy . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4 Domain of Good Public Policy . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.5 Policy Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.6 Delivery Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.7 Time Consistency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.8 Measuring Progress on Wellbeing . . . . . . . . . . . . . . . . . . . . . . . 19
1.9 Measuring the Quality of Public Policy . . . . . . . . . . . . . . . . . . . 22
1.10 Empirical Evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.11 Methodology and Structure of Presentation . . . . . . . . . . . . . . . . 24
1.11.1 Why Mathematical Models . . . . . . . . . . . . . . . . . . . . . . 24
1.11.2 What Types of Models . . . . . . . . . . . . . . . . . . . . . . . . . 25
2 Stylised Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.2 Consumption, Wellbeing, and Wealth . . . . . . . . . . . . . . . . . . . . 30
2.2.1 Generic Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.2.2 Stylised Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.3 Production and Wellbeing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
2.3.1 General Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
2.3.2 Mathematical Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
2.4 Ideas and Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
xix
xx Contents
2.5 Finance - Money, Credit, and Banking . . . . . . . . . . . . . . . . . . . . 58

2.6 International Connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
2.6.3 Empirical Evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
2.7 Pulling the Pieces Together - Market Clearing . . . . . . . . . . . . . . 68
2.7.2 Mathematical Model - Market-Clearing Conditions . . . . 68
2.8 Environmental, Social, Economic Externalities . . . . . . . . . . . . . . 71
2.8.4 Motivation: Government . . . . . . . . . . . . . . . . . . . . . . . . 85
3 Wellbeing-Enhancing Public Policy . . . . . . . . . . . . . . . . . . . . . . . ... 89
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... 90
3.2 Three Steady State Mathematical Models . . . . . . . . . . . . . . . ... 91
3.2.1 Mathematical Model: Laissez Faire - A Benchmark
Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
3.2.2 Mathematical Model: Laissez Faire - Stylised Model . . . 93
3.2.3 Mathematical Model - Stylised . . . . . . . . . . . . . . . . . . . 95
3.3 Policy Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
3.3.1 Policy in the Benchmark Model . . . . . . . . . . . . . . . . . . 97
3.3.2 Policy in the Stylised Model . . . . . . . . . . . . . . . . . . . . . 97
3.4 Comparative Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
3.5 Policy Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
3.5.2 Example: Familias en Acción Urbano . . . . . . . . . . . . . . 104
3.5.3 Numerical Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . 105
4 Operational Policy-Simulation Model . . . . . . . . . . . . . . . . . . . . . . . . 111
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
4.2 Model Overview and Structure . . . . . . . . . . . . . . . . . . . . . . . . . 113
4.2.1 Links between Different Influences on Wellbeing . . . . . . 113
4.2.2 Modeling Strategy: Direct Influences . . . . . . . . . . . . . . . 114
4.2.3 Modeling Strategy: Production Sector . . . . . . . . . . . . . . 121
4.2.4 Modeling Strategy: Market Clearing . . . . . . . . . . . . . . . 125
4.2.5 Modeling Strategy: Rest of the World . . . . . . . . . . . . . . 125
4.2.6 Modelling Strategy: Government . . . . . . . . . . . . . . . . . . 125
4.2.7 Modelling Strategy: Calculation of Indexes . . . . . . . . . . 127
4.2.8 Modelling Strategy: Calculation of Wellbeing . . . . . . . . 128
Contents xxi
4.3 Policy Simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

4.3.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
4.3.2 Using the Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Part II
5 Wellbeing-Enhancing Public Policy - Extensions . . . . . . . . . . . . . . . . 139
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
5.2 A Radically Different Policy Framework . . . . . . . . . . . . . . . . . . 142
5.3 Key Policy Focus is on Building Resilience . . . . . . . . . . . . . . . . 144
5.4 Critical Role of Political Economy and Institutions . . . . . . . . . . . 146
5.4.1 Political Economy Considerations . . . . . . . . . . . . . . . . . 146
5.4.2 Institutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
5.5 Case for Localism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
5.6 Setting Policy Priorities in a Complex and Uncertain World . . . . 154
5.7 A Simple but Complex Example . . . . . . . . . . . . . . . . . . . . . . . . 155
6 Viability Theory for Policy Formulation . . . . . . . . . . . . . . . . . . . . . . 159
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
6.2 The Basics of Viability Theory . . . . . . . . . . . . . . . . . . . . . . . . . 163
6.2.1 An Introduction to Viability Theory . . . . . . . . . . . . . . . 163
6.2.2 The Viability Kernel – The Largest Viability
Domain in the Constraint Set . . . . . . . . . . . . . . . . . . . . 164
6.2.3 Viability Versus Optimality . . . . . . . . . . . . . . . . . . . . . 167
6.3 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
6.3.1 Sustainable Emissions . . . . . . . . . . . . . . . . . . . . . . . . . 174
6.3.2 Sustainable Inequality Policy . . . . . . . . . . . . . . . . . . . . 188
6.3.3 Viability with Inequality and Emissions Constraints . . . . 199
6.4 Resilience in a Viability Theory Model . . . . . . . . . . . . . . . . . . . 212
6.4.1 How to Accomplish Resilience? . . . . . . . . . . . . . . . . . . 212
6.4.2 Kernel Width . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
6.4.3 Faster Instrument Adjustments . . . . . . . . . . . . . . . . . . . 214
6.4.4 Risk Aversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
6.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Appendix 6.1: Empirical Evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
Appendix 6.2: A Method for Finding Viability Kernels . . . . . . . . . . . . 218
7 Themes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Part I
Part I of the book, which includes the first four chapters, is based on extensions of
the standard dynamic optimisation model.
We demonstrate that the standard model can accommodate a wider wellbeing
framework. The dimensions of wellbeing are represented by a broader set of con-
sumables, referred as “comprehensive consumption”, which are sourced from a broad
set of assets (“comprehensive wealth”).
The expanded formal models provide useful tools for identifying the justifications
for policy interventions. They also provide guidance on the types of interventions
that would be most effective and efficient in enhancing intergenerational wellbeing.
What is critical, especially for a wellbeing-focused public policy, is not only
the diversity of consumables and capital assets as sources of wellbeing, but also
the interdependencies between them. The design of public policy can exploit the
complementarities and substitutabilities between and among these goods and assets
towards enhancing intergenerational wellbeing.
We need quantitative models to be able to identify the full set of these interre-
lationships. We also need such models to quantify the signs and magnitudes of the
same. This is critical in designing policy interventions. It is also critical in prioritising
investments in the components of “comprehensive wealth” to increase the wellbeing-
impact of these interventions most efficiently.
These quantitative models are operationalised by first calibrating the key parame-
ters, and then performing numerical simulations of the effects of policy interventions.
Chapter 1
Introduction and Overview
Abstract This chapter provides an overall introduction to, and overview of, the
whole book. It starts by framing the purpose of good public policy as improving the
wellbeing of individuals and communities. Wellbeing is defined in terms of people’s
abilities to live the kinds of lives they have reason to value. Public policy can pursue its
purpose by focusing on improving the capabilities and opportunities (i.e., substantive
freedoms) of individuals and communities to pursue the lives they wish to live. This in
turn requires an increase in, accompanied by a broadening access to, “comprehensive
wealth”. We ask, and answer, the question of why we need public policy at all in
this context. Our answer is much wider than the one that is based on the standard
“market failure” argument. Following an explanation of the several justifications for
public policy, we specify the domains of public policy in terms of the dimensions of
a collective “wellbeing frontier”. This is then followed by a discussion of appropriate
policy instruments, including institutions, that can be used to extend this frontier. The
common thread that runs through the chapter, as well as the whole book, demonstrated
with the help of examples and numerical simulations, is the interaction between
environmental, social, and economic influences on wellbeing. The chapter ends with
an overview of the relevant empirical evidence.
Key Points of the Chapter
This chapter shows how to:
• Critically assess the deficiencies of income (or national income at the aggregate
level) as a measure of the human condition.
• Define wellbeing at both a conceptual and an operational level - and discuss the
sources of wellbeing and how they may be measured.
• Explain and discuss the linkages between “comprehensive consumption”, “com-
prehensive wealth”, and wellbeing.
• Evaluate the differences between a public policy that is focused on improving
national income versus national wellbeing on a sustained basis.
• Critically discuss the fundamental cases for public policy - why the government
should be involved in trying to improve people’s lives.
• Differentiate between the indicators and domains of public policy, and how these
relate to wellbeing.
© The Author(s), under exclusive license to Springer Nature 3

Singapore Pte Ltd., part of Springer Nature 2019
G. Karacaoglu et al., Intergenerational Wellbeing and Public Policy,
https://doi.org/10.1007/978-981-13-6104-3_1
4 1 Introduction and Overview
• Appreciate the nature, and significance, of empirical evidence in informing the

choices of wellbeing indicators.
• Critically assess the differences between a “welfare state” and a “wellbeing state”,
and the implications of these differences for the role, domains, and instruments of
public policy.
1.1 Introduction
Our broad objective is to make it possible for individuals to live the kinds of lives
they value, in the present and into the future - without compromising others’ rights
to do the same. This is what individual and community wellbeing is all about.
Working with this broad canvas as background and context, this book is organised
around, and tries to answer, the following questions:
• Why is there a place for public policy at all in improving individual and community
wellbeing?
• If there is a place for it, what is its purpose?
• What are its domains – i.e., what should its focus be?
• What are its instruments?
• How does it operate - what kind of delivery mechanisms are used?
• How do we ensure that policies are time consistent – they evolve with the public’s
preferences?
• Are the answers to any of the questions above affected by fundamental uncertainty
and complexity; if so how?
• How do we know if public policy is working - is it achieving its objective?
We explore, and provide answers to, these questions with the aid of a formal
analytical framework and a set of models. The formal models we introduce and
use are critical in identifying the interdependencies between various instruments of
public policy, in pursuing our main objective of improving wellbeing. We also need
these models to operationalise the theoretical framework, and assess the wellbeing-
implications of various policy interventions.
1.2 Justification for Public Policy
Why does the government have any role to play at all, towards helping enhance
wellbeing on a sustained basis?1 After all, individuals – partly through voluntary
cooperation and exchanges with others – can and do invest in their own economic
1 Note that, “the government” is the generic term we use to refer to any collective agency through
which we may wish to organise the delivery of certain products or services; it does not necessarily
refer to the central government.
1.2 Justification for Public Policy 5
and human capital, and make all sorts of arrangements to manage associated risks
(partly through the purchase of insurance contracts), to sustain their own and their
families’ wellbeing into the future. They also voluntarily form communities and
associations to create the collective goods and services they value.
The standard economics answer to this first question rests on “market failure”.
As (Gintis 2014, p. 1) points out, in traditional economic analysis based on “market
failure”, there are two distinct roles for economic policy: regulating market dynamics
(e.g., stabilizing business fluctuations and preventing catastrophic breakdowns) and
influencing the equilibrium allocation of economic resources (e.g., financing social
infrastructure, providing public goods, and regulating industrial effluents). In other
words, system-wide, resilience-enhancing roles; and marginal, static, and dynamic
allocative, roles. The “market failures” that underpin these roles fall under five cat-
egories: increasing returns-to-scale; public goods; externalities; ensuring product
quality (information asymmetries); merit goods (ibid., pp. 3–4).
Governments fail too. This happens for all kinds of reasons, which include the fact
that they are subject to political forces that lead them to favour special interests rather
than the general good. Governments do not have at their disposal all the information
required, concerning the people’s preferences and the operation of the economy, to
be able to design optimal solutions. In any case, “the market economy is a complex,
dynamic, and adaptive system more like a natural ecology than a man-made machine.
The complex economy cannot be controlled, as the planners would like, but it can be
influenced by very carefully formulated and judiciously applied ‘rules of the game’
that move market dynamics in preferred directions” (ibid., p. 5).
A very recent and useful discussion of the case for public policy in traditional
economic analysis is provided by Tirole (2017), especially in chapter six. After
reviewing the various categories of market failures, Tirole proceeds to discuss the
various types and sources of state (or government) failure as well. He concludes, on
the basis of his careful analysis, that “the market and the state are not alternatives
but, on the contrary, are mutually dependent. The proper functioning of the market
depends on the proper functioning of the state. Conversely, a defective state can
neither contribute to the market’s efficiency nor offer an alternative to it” (ibid., p.
162). He then proceeds most usefully to outline a process for reforming the state
so that it does perform its complementary function, to the most useful services of
markets, towards the common good (ibid., pp. 169–173).
The only economic rationale for public policy is not market failure; nor is eco-
nomic life the exclusive domain of public policy. Mazzucato (2016), for one, empha-
sises the wider role of public policy in the economic domain (going beyond simply
fixing market and/or system failures), by drawing on a vast literature on the subject.
She too, just like Tirole, but in a wider context, emphasises the complementary roles
of the state and markets. Mazzucato (2015) argues that when governed effectively
and efficiently, “the State’s visible hand is firm but not heavy, providing the vision
and the dynamic push (as well as some ‘nudges’) to make things happen that other-
wise would not have” (Mazzucato 2015, p. 2).2 Such actions are meant to increase
the courage of private business. This requires understanding the State as neither a
‘meddler’ nor a simple ‘facilitator’ of economic growth. It is a key partner of the
private sector – and often a more daring one, willing to take the risks that business
won’t” (ibid., p. 31).
A related and broader role for public policy in the economic domain is grounded
in the interface between complexity theory, evolutionary theory, and radical uncer-
tainty (see Antonelli 2016; Colander and Kupers 2014; Roos 2015; Schmidt 2015).
As Schmidt (2015) forcefully argues, referring to the “fallacy of failure thinking”,
the debate about “market failure” versus “government failure”, as it relates to the
common good, is a logically and empirically unresolvable ideological debate. From
this perspective, the search for optimal policies based on an equilibrium framework
is a meaningless and futile exercise. In a world of radical uncertainty and complexity,
there is no exogenously given “optimum” (equilibrium) benchmark against which
we can assess the performance of markets, the government, or any other set of insti-
tutions; so “failure” cannot be determined. It is a given that markets, governments,
and communities are all “imperfect institutions”, whatever that may mean. The key is
to acknowledge this,3 as well as the complementary strengths of all these institutions
towards improving overall wellbeing, when they do function reasonably well. We
should strive to design public policy in a way that enhances the governance of all
these institutions towards improving their collective efficiency and effectiveness in
improving individual and community wellbeing.
Indeed, when improving wellbeing, rather than simply economic outcomes, is
specified as the objective, the justification for public policy becomes much richer.
People’s wellbeing is influenced by environmental, social, and economic factors and
their interactions. These interactions are the source of a whole range of externalities
that provide additional justifications for active public policy.
1.3 Purpose of Good Public Policy
We start with three observations as the foundation of our analysis for the purpose
of good public policy, when the broader objective of policy is to enhance individual
and community wellbeing on a shared and sustained basis.
First, individuals constantly strive to either preserve or improve their lives, subject
to their capabilities and opportunities (i.e., substantive freedoms), and conditioned
2 “Most of the radical, revolutionary innovations that have fuelled the dynamics of capitalism –
from railroads to the internet, to modern-day nanotechnology and pharmaceuticals – trace the most
courageous, early and capital-intensive ‘entrepreneurial’ investments back to the State”.
3 In an approach to policy design based on viability theory, explained in Chap. 6, the evolution of
economies is influenced by viable policies.

1.3 Purpose of Good Public Policy 7
by the incentives that surround them.4 In doing so, they are substantially influenced
by prevailing institutions.
Second, both the choice of lives people value, and how they pursue their valued
lives, are conditioned by the social context in which they live. “Rationality” is thus
a fundamentally social concept, reflecting the interdependence of individual and
community wellbeing. Individuals are rational in an adaptive sense, operating in a
social context, and bounded by decision-making costs, and various psychological,
computational, and other constraints. As conditions change, if superior alternatives
become available, most individuals will adapt their behaviour to exploit these superior
alternatives to improve their lives (see Savioli and Patuelli 2016); others will be
content to continue living the way they are.
Rationality is not restricted to self-interest and consistent choice. Both prudent
and ethical behaviour are key ingredients of rationality, motivated by a human instinct
for self-preservation, as well as a human drive to improve own lives. Self- (as distinct
from purely selfish) interest is clearly a motivating force, based however on empathy
as a foundation for both moral judgement and an instinct for survival. Thus, being
rational has to do with reflecting about one’s options and the consequences of one’s
actions, and having good reasons for one’s choices (see Eiffe 2010, p. 8 and Sen
2004). As Costanza et al. (2017) express it, “most of the assigned values to goods,
services, people or cultural achievements have to do with the survival of the humans
as individuals, many with the wellbeing of individuals and social groups, and some
with the ethical considerations humans make about other people’s and other species’
rights to live” (p. 8).
Third, although we do not know, nor do we wish to pass judgement on, how
people want to live their lives, we have the benefit of numerous studies, covering
a large variety of countries and cultures across time, to help us identify the main
influences on individual wellbeing. In other words, based primarily on robust, survey-
based, empirical evidence, we have a broad sense of the common elements of what
individuals value (see Boarini et al. 2014; Fleche et al. 2012; Smith 2015).
Based on previous works summarised in Atkinson (2015), Sen (2009), Stiglitz
et al. (2010), O’Donnell et al. (2014), Braunerhjelm and Henrekson (2016), Feldman
et al. (2016), Gough (2015), McCloskey (2014), we specify the purpose of good pub-
lic policy as enhancing (not necessarily maximising or optimising) the capabilities
and opportunities of individuals to pursue the lives they have reason to value (i.e., to
increase their wellbeing),5 helping them remove the obstacles they face in this pursuit
and, in doing so, making sure that we do not blunt the incentives of individuals to do
the best they can for themselves (see Gleisner et al. 2012).6
4 This is not to deny the existence of many individuals who, for various reasons, are simply incapable
of making such deliberate decisions. Good public policy, through various means, such as institutions
for the mentally disabled, can help them.
5 Or, to use the language of Phelps (2013), to provide opportunities and capabilities for individuals to
flourish and prosper, towards creating an inclusive society that will in turn underpin mass flourishing.
6 Needless to say, limits will have to be imposed on those who want to live their lives in total disregard
for the wellbeing of current and/or future generations. Such limits must be ethically based, and able
to withstand intense and rigorous public scrutiny. We thank Jonathan Boston for raising this point.
We start with a broadly defined “wellbeing function” as the basis of our analysis,
and as a platform for exploring wellbeing-enhancing public policy. In doing so, as
a first step, we follow Arrow et al. (2012) and define the object of interest for good
public policy as intergenerational wellbeing. “When scholars adopt intergenerational
wellbeing as the object of interest, their presumption is that at any given date social
wellbeing is not only the wellbeing of the current generation, but also the poten-
tial welfare of the generations that are to follow. The point is to ask whether the
society under study is functioning sufficiently well to ensure that some measure of
intergenerational wellbeing does not decline” (ibid., p. 318).7
Arrow et al. (2012) formulate wellbeing as a function of “comprehensive con-
sumption”. The components of comprehensive consumption include not only stan-
dard marketed consumption goods and leisure, but also others such as health status,
environmental quality, personal security, social connections, altruism, and so on.
They broadly correspond to the components of the OECD’s Better Life Index (BLI)
(see Durand 2015). They are the “functionings” (beings and doings) that people have
reason to value (see Sen 2009; Alkire 2016).
The OECD’s Better Life Index (BLI) (see Durand 2015; Van Zanden et al. 2014)
focuses on the main influences on, or spheres of,8 individual wellbeing presented in
Fig. 1.1.
These spheres are classified under quality of life (health status, work-life bal-
ance, education and skills, social connections, civic engagement and governance,
environmental quality, personal security, subjective wellbeing) and material condi-
tions (income and wealth, jobs and earnings, housing). The identification of these
spheres is based on extensive consultations with the wider public (see Watson et al.
2016) and/or informed by empirical work (see Di Tella and MacCulloch 2008; Au
et al. 2015; Benjamin et al. 2014; Ward 2015). It is also very important to note
the broad correspondence between the main influences on wellbeing highlighted
here and the “universal characteristics of needs satisfiers” (p. 1196), or the 11 cat-
egories of “intermediate needs” (p. 1202), that Gough refers to (see Gough 2015).
Indeed, as Gough emphasises in referring to the work of Alkire (2005), there are
significant overlaps with Alkire’s 39 lists of dimensions of human development as
well. Thus, although we recognise the significant philosophical differences between
“needs based”, “preference based”, “subjective wellbeing based”, and “capabilities
based” approaches highlighted by Gough (2015), as well the various contributions to
Adler and Fleurbaey (2016), the policy package that is suggested by our integrated
7 This is very much an extension and adaptation of the standard dynamic optimisation model to
the problem at hand. In the second part of this book, especially in Chaps. 5 and 6, we explicitly
and deliberately introduce radical uncertainty and complexity into our analysis, and explore their
implications for the design and implementation of public policy.
8 We make reference to “spheres of wellbeing” as a tribute to the great work by Walzer (1983).
Walzer emphasises that some of the spheres of justice are incommensurable. We suspect what he
really means is that they are not comparable or rankable, irrespective of how much reason we bring
to bear on relevant comparative valuations (see Sen 2017, p. 456). Alternatively, we can argue that
various spheres of wellbeing are complementary for the purposes of overall wellbeing. We thank
Geoff Bertram for bringing Walzer’s work to our attention.
1.3 Purpose of Good Public Policy 9
Fig. 1.1 OECD BLI
approach and associated model would be perfectly consistent with all four (see
Chap. 3).
If people (or the government as the representative of the people) are able to assign
some desirability limits to each sphere of wellbeing, then keeping “comprehensive
consumption” inside these limits can be technically interpreted as a viability theory
problem, as we explain in Chap. 6.
The ultimate source of comprehensive consumption is “comprehensive wealth”
(Arrow et al. 2012). This refers to the stocks of capital assets, broadly defined, that
yield income and other sources of wellbeing now and into the future (ibid., p. 320),
to be conceptualised as the discounted present value of our overall future stream of
wellbeing. These assets include economic (or built) capital, human capital, natural
capital, and social and cultural capital. “[...] Because the determinants of intergener-
ational wellbeing are the multitude of capital assets the economy has inherited from
the past, the criterion function for sustainable development reduces to a weighted
sum of the stocks of those assets – the weights being the marginal contributions
of the stocks to intergenerational wellbeing. The weights are therefore the assets’
shadow prices, and the weighted sum is the economy’s wealth” (ibid., p. 318). This
assumes weak sustainability,9 and an increase in comprehensive wealth is referred
9 Refer to footnote 15 for a discussion on when and why we may want to differentiate between weak
and strong sustainability.

to as “genuine saving” (or “comprehensive investment”) (see Hamilton and Hepburn

2014; Ferreira et al. 2008).10, 11
In our model, these capital assets provide individuals’ capabilities and oppor-
tunities (i.e., substantive freedoms) to achieve their valued functionings (see Alkire
2016). They cover both “internal capabilities” (such as the skills embodied in individ-
uals) and “external capabilities” (i.e., the social and political institutions that inhibit
or promote the expressions of these skills) (see Heckman and Corbin 2016). That is
precisely why the distribution of “comprehensive wealth” across society and across
generations is such a critical determinant of the distribution of effective power, that
ultimately defines these capabilities and opportunities.
It is by expanding, enhancing, and protecting, as well as distributing across society
and generations, these capital assets, that good public policy can push out (i.e.,
expand) our collective “wellbeing frontier”, as we will explain shortly in the next
section.12
1.4 Domain of Good Public Policy
What then is the domain of a wellbeing-focused public policy (i.e., what areas should
it focus on)?
This is a good point to distinguish between a “welfare state” and a “wellbeing
state”.
A “welfare state” focuses on delivering welfare itself through redistribution of
income and other direct sources of wellbeing, funded by taxation and borrowing
(which eventually needs to be repaid through future taxation). A “wellbeing state”,
on the other hand, is focused on enabling individuals and communities, through
appropriate investments in the components of comprehensive wealth (the capital
assets), to look after themselves, again funded by taxation and borrowing. The role
of a “welfare state” is to deliver welfare directly to those in need – agency is typically
assumed to reside with the government. A “wellbeing state”, on the other hand, aims
to expand the opportunities and capabilities (i.e., substantive freedoms) of individuals
to enhance their own wellbeing, underpinned by appropriate incentive structures.
Agency for improving wellbeing rests primarily with the individuals. Citizens have
10 For an earlier specification of the same type of model of an economy which produces (possibly)
multiple consumption goods representable by an index, using multiple types of capital, over an
infinite time horizon, see Weitzman (1976), Asheim and Weitzman (2001), and Weitzman (2009).
11 To avoid “circular definitions”, so that to compute the capital shadow prices we have to first define
what wellbeing is, but then we define wellbeing as a weighted sum of capitals, we need to be able to
define and measure wellbeing independently. Adler and Fleurbaey (2016) provides a rich resource
for various ways of measuring (subjective) wellbeing (see, in particular, Clark 2016). In addition,
we have various objective wellbeing indices based on the critical influences on wellbeing similar
to those used to construct the OECD’s Better Life Index.
12 In New Zealand, where all three authors of this book live (or used to live), we refer to this as “the
investment approach to public policy”.

1.4 Domain of Good Public Policy 11
a responsibility in being actively engaged in pursuing what they want; unless their
circumstances make it inevitable, they cannot be passive recipients of what they value
(see Dalziel and Saunders 2014; Sen 2009).13
A wellbeing state is the steward of “comprehensive wealth” (i.e., all forms of cap-
ital assets) on behalf of current and future generations. While individuals do manage
their own assets (including human capital), it is only a representative government
that can take a systems-view, and deal with systemic risks and opportunities.
Let us start with the key influences on wellbeing identified in Sect. 1.3 above
and note that the broader role of a representative government is embedded in the
multi-dimensional intergenerational wellbeing function that we referred to there. A
fundamental assumption underpinning this wider concept of wellbeing from a public
policy perspective is that, in addition to their individual incomes and consumption of
private goods (including leisure and good health), individuals and communities also
value, among others, a clean environment, social cohesion, equity across society and
generations, and enhanced resilience to the types of systemic shocks that have the
potential to cause serious damage to various capital assets and through that on our
overall wellbeing.
A public policy that aims to enhance intergenerational wellbeing on a sustained
basis, focuses on ensuring that the wellbeing-generating capacity of capital assets is
sustained or enhanced, and shared, which is to say:
• not eroded by current generations at the expense of future generations (sustain-
ability);
• shared in a manner consistent with sustaining or enhancing the capital base
(equity);
• no particular social group(s) impose their concepts of wellbeing on others, respect-
ing others’ rights to live the kinds of lives they have reason to value (social cohe-
sion);
• comprehensive wealth is protected against major systemic risks (resilience); and
• the material wellbeing-generating potential of capital assets is enhanced (raising
potential economic growth).
These are all outcomes with public good characteristics, with non-rivalry and/or
non-excludability features. They are the sources of significant positive externalities
that public policy can wrap around comprehensive wealth, to enhance the capabilities
and opportunities (i.e., substantive freedoms), of individuals and communities, to
pursue the kinds of lives they have reason to value – i.e., to expand their collective
“wellbeing frontier” (see Sen 2009; McCloskey 2014). These are the domains of
public policy.
Figure 1.2 represents this diagrammatically, with the ultimate sources of wellbeing
(i.e., the capital stocks comprising comprehensive wealth) sitting in the middle,
surrounded by the domains of public policy that collectively define the wellbeing
13 An equivalent classification, using different terminology, distinguishes between the mid-20th
century “entitlement state” and the early-21st century “enablement and self-empowerment state”
(see Cadogan 2013).
Fig. 1.2 Wellbeing frontier
frontier. It is the sustainability domain or dimension of the frontier that brings all the
other dimensions together and links them to intergenerational wellbeing. The main
purpose of public policy is to expand this frontier as much as possible.
Because the returns from investing in the capital assets and institutions that gen-
erate these social goods will not be fully captured privately, this may lead to under-
investment in, and over-use and/or under-protection of, the components of compre-
hensive wealth, relative to a socially optimal equilibrium, in the absence of deliberate
collective action (see Miller 2006). There is no suggestion here that the benefits asso-
ciated with these positive externalities will not be provided at all; however, they may
be under-provided. It is likely that, “decentralised market decisions lead to the under-
provision of public goods such as infrastructure, or the over-provision of public bads
such as pollution” (Sachs 2015, p. 155). Through appropriate and deliberate collec-
tive action, we may be able to make markets, communities, and institutions work
more effectively and efficiently in delivering these public goods. This is a possibility,
and not a certainty.14
What Fig. 1.2 does not capture is the complex interactions between the various
capital stocks (and associated ecosystems) that represent “comprehensive wealth”,
and sustainable wellbeing. We will do so formally in the stylised model presented
in Chap. 2. Meanwhile, in anticipation of that formal presentation, Fig. 1.3 is a dia-
grammatic representation of the key themes.
The Fig. 1.3 attempts to capture two main ideas. First, the interactions between
various types of capital in generating sustained wellbeing. As Costanza et al. (2017),
where we have borrowed this figure from, emphasise, it is the complex interactions
14 In a recent speech, Haldane (2015) provides a wonderful summary of theories and facts about long
term economic growth trends, pulling together threads from Economics, History, Sociology, and
Psychology. We believe that this book, and more specifically the domains of public policy we focus
on, highlight what public policy can do to transform the emerging tensions between “technological
tailwinds” and “sociological headwinds” into a mutually reinforcing dynamic process that leads to
higher sustained wellbeing.
1.4 Domain of Good Public Policy 13
Complementary Ecosystems
Fig. 1.3 Ecosystems
between these various forms of capital assets that affect human wellbeing. Built
capital and human capital are embedded in society, which is in turn embedded in
the rest of nature. Second, “sustainability” refers to the sustainability of human
wellbeing, not necessarily to the sustainability of any particular forms of natural
capital. We should allow for the possibility of substitutability of various types of
natural and other forms of capital in generating human wellbeing.15
1.5 Policy Instruments
It is wellbeing, and not welfare, that is the primary (but not exclusive) focus of good
public policy in our framework. There is a huge literature, and arguments, on the
differences between utility, welfare, and wellbeing. A nice summary can be found
in Sen (2009, 2017).
15 This is where it is appropriate to refer to the distinction between “strong sustainability” and “weak
sustainability. A very useful discussion of this distinction, including relevant empirical evidence, is
provided by Pelenc et al. (2015). Under “weak sustainability”, we allow for a degree of substitutabil-
ity both within different categories of capital stocks (such as different elements of natural capital)
and across them (such as between natural capital and human capital) in generating wellbeing. Under
“strong sustainability”, at the very extreme, all main categories of capital stock are complements
in generating human wellbeing (although a degree of substitutability within each category may be
allowed). Although, in principle, this is an empirical matter, in fact, it is very difficult to resolve
this matter empirically. So, a practical and cautious approach, which is advocated by Pelenc et al.
(2015) as well, is to treat natural, social, and economic ecosystems as complements (i.e., “strong
sustainability”), while allowing substitutability at the margin within different categories of capital
(i.e., “weak sustainability”), in the generation of human wellbeing. Certainly the stylised model of
Chap. 2 can accommodate this precautionary principle.
In a “wellbeing state”, good public policy is targetted at the growth and protection
of, and widespread access to, public capital assets, and the equitable distribution of
private capital assets. This needs to be implemented in a way that does not blunt
incentives for individual and community participation in the creation and productive
application of wealth. It is through these particular mechanisms that public policy can
enhance individuals’ capabilities and opportunities (i.e., substantive freedoms) to live
the kinds of lives they have reason to value. These cover both “internal capabilities”
(such as the skills embodied in individuals) and “external capabilities” (the social
and political institutions that inhibit or promote the expressions of these skills) (see
Heckman and Corbin 2016).
Needless to say, if and when individuals are simply unable to look after themselves,
directly or through family and community support, the government would provide
that support. Hence, almost always, the collective agent who delivers public policy
would have both “welfare state” and “wellbeing state” attributes. Nevertheless, the
primary emphasis of this book is on exploring the formulation, implementation, and
evaluation of public policy for a “wellbeing state”.
Whether actual public policy pursues this purpose or not, and whose wellbeing
is its focus, depends on the institutions that exist in a society. These institutions are
the outcomes of social decisions, made by humans, chosen for their consequences;
they reflect the evolution of the distribution of power in society. The distribution of
power in turn is a reflection of the distribution of political and economic resources
(see Acemoglu et al. 2005; Acemoglu and Robinson 2013).
Creating good institutions is the main vehicle through which a wellbeing state
delivers public policy. Good institutions are those that provide equal access to oppor-
tunities and capabilities to a broad cross-section of society. For such access to gen-
erate good environmental, social, and economic outcomes, institutions need to also
provide security of property rights. Such circumstances are more likely to emerge
when we have political institutions that place checks on those who hold political
power; when political power is broadly distributed, complemented by good invest-
ment opportunities; and when there are only limited rents that power holders can
extract from the rest of society (see Acemoglu et al. 2005; Acemoglu and Robinson
2013).
Good institutions associated with a wellbeing state also ensure that all individuals
are incentivised to do what’s best for themselves as judged by the majority. Thus any
support provided to those who have the capability to look after themselves comes
with some degree of conditionality. By way of example, any transitional support
provided to unemployed individuals comes with the condition that they should do
the best they can to find a job for themselves as soon as possible.
Given the multiplicity and complementarity of the various influences on well-
being, a variety of complementary policy instruments are required. This frames the
desirability of an integrated approach to environmental, social, and economic pol-
icy. This all requires a complementary, and mutually supporting, interaction between
governments (national, regional and local), communities, and markets. The outcomes
of policies should be thought of as ecosystems (including institutions) that facilitate
such effective cooperation.
1.5 Policy Instruments 15
In a complex and evolving world dominated by “radical uncertainty”, and satisfic-

ing (rather than optimising) individuals, the scope and instruments of public policy,
for a wellbeing state, need to be expanded to include the “comprehensive ecosystem”
that surrounds all decision makers. Colander and Kupers (2014), for one, suggest
that, in such circumstances, instead of using direct interventions to achieve certain
goals, the government should rather try to influence the rules of the game in a way
that seems socially desirable. The rules of the game are the institutions that define the
individual agents’ action sets. Instead of solving problems for the agents, the gov-
ernment should create institutions that make it easy for agents to solve problems for
themselves. The government could even go a step further by applying “supernudges”
that have an effect on agents’ preferences and norms. This is in clear contrast to the
mainstream economists’ desire to derive theoretically optimal policies (see also Roos
2015).
New Zealand’s KiwiSaver scheme provides a very good example of all the policy
instruments referred to in the previous paragraph, in action. It is a scheme designed
to encourage individuals to save for their retirement, as well as towards purchasing
their first homes. While it is a voluntary saving scheme for individuals, employers of
individuals have a legal obligation to support, dollar for dollar, individual contribu-
tions to the Scheme, up to a certain percentage of the employee’s salary. In addition,
individuals are “nudged” to join the Scheme by being automatically enrolled when
they take up their employment, but having the option to opt out if they choose to.
A whole ecosystem of private institutions, comprising a variety of fund managers
(“KiwiSaver providers”), offering various types of investment products, has evolved
around the scheme.
1.6 Delivery Mechanisms
Whatever the domain of public policy is, efficient and effective delivery of public
services is essential for sustainable wellbeing – thus support for wider and sustainable
wellbeing is not a vote against efficiency.16
The recognition of radical (or fundamental) uncertainty and adaptive complexity
as essential and unavoidable ingredients of the policy environment adds another
critical dimension to the design and implementation of public policy. Under these
16 The proper domain of public policy remains an area of intense debate. Buchanan and Hartley
(2000) favour a narrower scope for public policy, focused on “efficiency”. For an argument favouring
a possibly broader or different shaped scope for public policy, including interventions such as
“nudging” etc. reflecting behavioural insights, see O’Donnell et al. (2014). As Spence (2014) puts
it, “if countering inequality and promoting intergenerational opportunity introduces some marginal
inefficiencies and blunts some incentives, it is more than worth the price. Public provision of critical
basic services like education or health care may never be as efficient as private-sector alternatives; but
where efficiency entails exclusion and inequality of opportunity, public provision is not a mistake”
(see also Helm 2010). Also very relevant is the evolutionary approach to public policy (see Schmidt
2015), as well as the broader (strategic) scope of public policy advocated by Mazzucato (2015,
2016).
circumstances, the primary focus of policy shifts to enhancing resilience – which, at

the system level, rests on two foundations: protecting comprehensive wealth from
catastrophes, while at the same time nourishing human inventiveness – natural inquiry
through experimentation (Mokyr 2016). There is thus a shift of emphasis away from
a policy setting that is focused on finding optimal solutions to centrally-defined
environmental, social, and economic problems, to building resilient environmental,
social, and economic ecosystems (including institutions).
Haldane (2018), partly based on the work of Broadberry and Wallis (2017), pro-
vides a very powerful and persuasive example of the interface between resilience,
institutions, public policy, and wellbeing over a long period of time covering hundreds
of years. The sustained growth in human prosperity since the Industrial Revolution
turns out to be not only or primarily an economic growth but rather a resilience story.
Growth spurts were ever present for at least a thousand years. It is deliberately cre-
ated social institutions (including schooling, health services, social welfare, and so
on) which protected humanity from the potential negative effects of contractionary
periods owing to war, disease, economic catastrophes, and others, over the period
since the Industrial Revolution, that made the difference.
Thus our proposed policy response to complexity and radical uncertainty is to
shift the focus of policy from direction to nourishment – to building resilience to
potential systemic risks that threaten our way of life, as well as encouraging creative
responses, through cooperative mechanisms. These systemic risks will vary across
countries. Figure 1.4 provides a diagrammatic representation of the need to focus on
resilience, as a platform for sustainability, in a world dominated by radical uncertainty
and complexity.
This revised policy framework also embraces localism, experimentation at com-
munity levels, small steps in implementation, reversibility and fast failures, and plan-
ning on surprises (Ostrom 2009). Collective wellbeing is pursued not by searching
for investment levers that will enhance well-defined social outcomes, but rather by
making it easier for people to pursue the varied lives they value, with no pre-defined
Fig. 1.4 Resilience

1.6 Delivery Mechanisms 17
desirable aggregate social outcomes. Solutions are developed by communities, not

by public servants, although the latter can play a constructive and critical role in
helping communities achieve the outcomes they desire.
1.7 Time Consistency
How do we ensure time-consistency (i.e., ongoing alignment of public policy with

collective wellbeing), as the composition, and with it the preferences, of society
evolve over time?17
The purpose of government policy in our setting is to help us collectively extract
maximum value (in a wider wellbeing sense) from our overall comprehensive wealth
(or stocks of capital) by helping shape, grow, share, and protect these capital assets
for the benefit of current and future generations. In other words, its target is the effec-
tive and efficient accumulation and management of comprehensive wealth towards
expanding our overall “wellbeing frontier”. By way of example, this is the essence of
the investment approach to public policy adopted in New Zealand. In an intergenera-
tional wellbeing context this requires that the evolution of public policy is informed
by and aligned with evolving public preferences across complementary spheres of
wellbeing.
“Time consistent policies are not policies that are never changed, but policies
where any changes required by new circumstances are consistent with maintaining
the original purposes of the policy. They are important for the socially desirable per-
formance of the private and public sectors. This is because they provide stability that
enables individuals and the state to plan for the future. We argue that time consistency
is achievable if intergenerational arrangements between the state and the populace
are treated as relational contracts. A relational contract is quite different from a legal
contract since the latter typically imposes specific constraints on arrangements. A
relational contract does not define specific constraints but rather a process for devel-
oping and changing rules by which all parties agree to abide” (Evans et al. 2013, p. ii).
A very good example of time-consistent policies is provided by the rules, including
voting thresholds, that countries impose for making changes in their constitutions.
This is why democratic institutions are so critical in helping us resolve tensions
through public reasoning and deliberation (see Bertram and Terry 2013; Sen 2009;
Walzer 1983). It is the democratic process, in the form of representative democracy,
supported by appropriate institutions, that provides the forces that push towards an
alignment of collective action (implemented through the government as our agent)
17 Boston (2016) provides a thorough analysis of the role of public policy in bringing the voice of
future generations to decisions made in the present. In that context, he also discusses in detail the
design of governance institutions as “commitment devices” so that public policy is always aligned
with evolving public preferences and interests.
Fig. 1.5 Wellbeing triangle
with evolving private and communal interests (i.e., government action is endoge-
nous). This ensures time consistency.18
By way of example, sticking to the constitution and its rules produces time-
consistent behaviour. Changing the constitution, changes the rules; as a consequence,
the behaviour becomes time-inconsistent.19
A time-consistent policy package for a “wellbeing state”, needs to be strongly
grounded in the history, cultures, and values of the society it represents (see Dalziel
and Saunders 2016), as well as being motivated by a shared vision. Universal access
to basic income,20 as well as health services, housing and education, security, and
freedoms for all, provide some of the key ingredients. A set of environmental, social,
and economic infrastructures (including strong institutions) act as enablers, while
also providing the incentives, to participate productively in social and economic life.
Which particular institutions will work best in different countries will be strongly
influenced by the history and culture of the society concerned (see Liu 2016).
Figure 1.5 provides a diagrammatic representation of this line of argument.
18 This is our interpretation of “time consistency” in the context of this book. There is no suggestion
that the authors of Evans et al. (2013) would agree with this interpretation.
19 As long as an optimal policy is obtained from dynamic programming, it is time consistent. Optimal
control derived from Pontryagin’s maximum principle may be time-inconsistent. For a mathematical
definition of time consistency see e.g., Haurie et al. (2012).
20 This does not suggest that such access to basic income should be unconditional. Conditionality is
critical to ensure that people are incentivised to contribute, to the best of their ability, to their own
wellbeing. Thus, by way of example, temporary unemployment benefits need to be conditional on
the recipient to be actively looking for employment.
1.8 Measuring Progress on Wellbeing 19
1.8 Measuring Progress on Wellbeing
There is now an emerging consensus on the need to complement income with com-
plementary measures of wellbeing in evaluating the success of public policy, since
income is not a good proxy for all indicators of wellbeing. While dissatisfaction with
income as a measure of wellbeing has a long history (see, for example, Coyle 2014,
2015; Nordhaus and Tobin 1972; van den Bergh 2009; Saunders and Dalziel 2017;
Waring and Steinem 1988; Fleurbaey and Blanchet 2013; Boarini and d’Ercole 2013)
the move to expand measures of wellbeing beyond income was given a significant
impetus with the publication of the report of the Stiglitz-Sen-Fitoussi Commission
(see Stiglitz et al. 2010). Useful surveys of this broad literature are provided by the
contributions in Adler and Fleurbaey (2016), Schreyer (2016), in particular, provides
a very valuable discussion of GDP, making the point that “GDP is different from but
not unrelated to theoretically well-founded welfare measures” (ibid., p. 22).
Given the focus of this book on public policy, this literature on measuring progress
on wellbeing has several threads that are worth highlighting.
One thread focuses on the sources (or ingredients) of wellbeing, that is objective
wellbeing measures, such as the components of the OECD’s Better Life Index which
we discussed earlier (which are closely associated with people’s capabilities to pursue
the lives they value). Obviously, income is one of these sources, but there are also
others, such as housing and so on.
In that same vein, the work associated with the Social Progress Index maintained
by the Social Progress Imperative (Imperative 2017 shows that some of the measures
of social progress such as “Basic Human Needs” (including nutrition, basic medical
care, shelter and so on)) are strongly and positively correlated with income, whereas
others such as “Opportunity” (including personal rights, tolerance, and inclusion)
are less so.
In measuring whether a country is or is not making progress, using objective
indicators, some work with multiple indicators and measures, using a dashboard
approach, others construct aggregate or composite indices. The components of the
OECD Better Life Index (BLI) provide a good example of the former.
By way of demonstration, we reproduce below, from two recent OECD reports (see
OECD 2016, 2017b), a diagrammatic and tabular representation of New Zealand’s
progress against the components of the OECD Better Life Index. The chart – see
Fig. 1.6 – shows New Zealand’s relative strengths and weaknesses with respect to
the main influences on wellbeing when compared with other OECD countries and
Australia.21
The table, reproduced as Fig. 1.7, shows comparable information on New
Zealand’s own historic performance over the past ten years.
Prominent among the aggregate or composite indices are the OECD’s Better
Life Index (see OECD 2015, 2017a), the UNDP’s Human Development Index (see
Office 1999), and the UN’s Sustainable Development Goals (see Nations 2017). The
21 We picked the New Zealand chart as an example only because that is where we live; similar charts
are available for all OECD countries.

Fig. 1.6 New Zealand’s wellbeing performance
World Bank uses Genuine Savings, also known as Adjusted Net Savings (see Hamil-
ton 2000), which emphasises sustainability through measures of changes in capital
stocks, while allowing for substitutability between them as sources of wellbeing (i.e.,
“weak sustainability”). Some for-profit and many non-governmental organisations
have developed their own measures, including the Legatum Institute’s Prosperity
Index (Institute 2016) and the New Economics Foundation’s Happy Planet Index
(see Abdallah et al. 2009).
In addition, a Genuine Progress Indicator (GPI) has been proposed since the late
1980s as an improved measure. It includes income inequality, contributions from the
informal economy and households, as well as deductions for unwanted expenses and
the depletion of resources (see Kubiszewski et al. 2013). The UN launched an Inclu-
sive Wealth Index (IWI) focusing on stocks rather than flows (such as GDP). The
IWI measures sustainable economic development to gauge the difference between
produced capital (what the human economy produces through industrial activities)
and human and natural capital (www.inclusivewealthindex.org). The OECD has pro-
posed a number of social progress and wellbeing indicators and implemented them
to better guide policy making (www.oecdbetterlifeindex.org).
As Costanza et al. (2017) emphasise, these measures, while useful, still use the
same linear input-output (I-O) approach of current national accounts. “If we are to
succeed at moving beyond the current model of ‘growth at all costs’ to embrace
the idea of a ‘wellbeing economy’, we need a different approach to data collection
and modelling that is adaptable, evolutionary, and integrated” (ibid., p. 14), and cite
Fioramonti (2017) as an example of work in this direction.
An alternative, complementary, approach focuses directly on measuring out-
comes, in the form of subjective wellbeing. The basic idea here is that, having
attempted to improve objective conditions that are sources of (or preconditions for)
wellbeing, public policy makers would, through carefully structured surveys, test
1.8 Measuring Progress on Wellbeing 21
Fig. 1.7 New Zealand’s wellbeing progress
whether these attempts are having the desired effect on actually increasing wellbe-
ing. Again, the contributions to Adler and Fleurbaey (2016) provide a great entry
point to this rich literature.
1.9 Measuring the Quality of Public Policy
If we go back to Fig. 1.2 as a visual tool for framing wellbeing-focused public policy,
it becomes clear that all the measures proposed above are useful, and should be used
in complementary ways. The data summarised in Figs. 1.6 and 1.7 paint a reasonably
flattering picture of the progress of overall wellbeing in New Zealand.
Nevertheless, this is not a complete picture. These figures do not shine the light
on the quality of public policy as it pertains to the sustainability of wellbeing into
the future – i.e., whether public policy is creating the platform for sustainable inter-
generational wellbeing. To assess that, what we also need are measures of the five
policy domains that represent the “wellbeing frontier” (i.e., equity, social cohesion,
resilience, potential economic growth and, through all these channels, sustainability)
in Fig. 1.2. We need to measure the evolution of both the size and the shape of the
wellbeing frontier over time.
To that end, note from Fig. 1.8 that the wellbeing frontier of Fig. 1.2 can be decom-
posed into three component triangles. Each of these triangles has a common corner
(i.e., sustainability):
• Triangle A (Social Harmony) links up equity and social cohesion as a contribution
to sustainability.
• Triangle B (Inclusive Economic Growth) links up potential economic growth and
equity as a contribution to sustainability.
• Triangle C (Economic Efficiency) is a measure of resilience-adjusted economic
growth, and links up potential economic growth and resilience to systemic risks,
as a contribution to sustainability.
At present, we do not have access to high quality and long-term time series data
on these key policy dimensions of intergenerational wellbeing, to be able to trace
the quality of public policy over a reasonable period of time. Nevertheless, evidence
on the five individual policy domains of wellbeing, summarised in Karacaoglu and
Fig. 1.8 Wellbeing triangles

NZ
1.9 Measuring the Quality of Public Policy 23
Au (2018), suggests that the “wellbeing frontier” has been shrinking over the past
ten years or so. By way of example, the evidence provided by Qasim et al. (2018) is
quite concerning, highlighting pressures arising from strong population growth and
low productivity growth on the sustainability of wellbeing in New Zealand.
1.10 Empirical Evidence
This subsection is substantially based on the material provided by Smith (2015).

The most relevant empirical evidence in this context relates to the critical influ-
ences on wellbeing – or the functionings. We will show that this evidence comes in
three (interrelated) forms: survey-based, experiment/trial based, and econometric.
In principle, there could be wide variance in what matters for wellbeing across
cultures, countries, or other population groups. In fact, we keep discovering the same
general list of critical influences on wellbeing no matter where we look or how we
derive these. There is increasingly robust evidence, from across many countries with
very different histories and cultures, that the eleven key influences on wellbeing
compiled to construct the OECD’s Better Life Index (BLI) identify the common
spheres of wellbeing (see Durand 2015; Van Zanden et al. 2014; Watson et al. 2016).
Smith (2015) has compiled the sources of this evidence into one document. With-
out exception, they all show that the BLI components of wellbeing are pretty univer-
sal.
Alkire (2005) looks across the various different lists developed by philosophers
on the elements of the quality of life. While these reflect a range of perspectives,
they can all be easily mapped onto the OECD’s BLI framework and components.
Any analysis of the determinants of life satisfaction finds that the wellbeing
spheres or influences all have a positive and independent impact on subjective well-
being (see Boarini et al. 2012; Brown et al. 2012; Helliwell et al. 2015).
When we compare an OECD style multi-dimensional approach to wellbeing, with
various one-number mash-up indices dating back to the 1970s, it is clear that multi-
dimensionality is important. Only the UN Human Development Index (HDI) and the
BLI can match GDP as a predictor of life satisfaction, and only the BLI exceeds it
(see Kroll and Delhey 2013).
When different countries run public consultations, they tend to come up with very
similar lists. In some cases (e.g., Israel and the OECD) there is clearly an influence
from one to the other, but in others (the New Zealand Social Report, the OECD, and
the UN Sustainable Development Goals), the lists were developed independently.
Au et al. (2015) and Benjamin et al. (2014) confirm this with very structured
surveys.
Di Tella and MacCulloch (2008) and Ward (2015) provide further confirmation
through their econometric work, that subjective wellbeing indicators are important
in people’s lives.
O’Donnell et al. (2014) provide a useful survey of experiment/trial based evidence.
There is also expert opinion on this matter – e.g., Stiglitz et al. (2010), which is
the proximate source of the OECD framework (see also the contributions in Adler
and Fleurbaey 2016).
Although there are differences in how the outcome spheres are described, the fact
that these are coming from completely different methodological starting points, and
yet have a strong underlying consistency, is quite compelling.
We can thus reasonably confirm that while material sources of wellbeing are
important, so are non-material sources. In addition, we have a very good sense of
the components of these two categories of wellbeing sources. Finally, and quite
critically, while some of the influences on wellbeing (such as nutrition and basic
medical care, and shelter) are strongly positively correlated with income (or material
sources of wellbeing more generally), others (such as personal rights, and tolerance,
and inclusion) are less so (see Porter et al. 2014).
Liu (2016) provides empirical evidence on the impact of culture on the emergence
of specific institutions (original or borrowed) in different countries in pursuit of
similar economic, environmental, social, and economic objectives.
Finally, in the context of a very useful distinction between people’s revealed pref-
erences across “personal-choice scenarios” and “policy-choice scenarios”, Benjamin
et al. (2014) provide survey-based empirical evidence on the relatively high rating
given by respondents to “policy-choice scenarios” that, broadly speaking, provide
opportunities and capabilities for individuals to live the kinds of lives they have
reason to value.
1.11 Methodology and Structure of Presentation
1.11.1 Why Mathematical Models
We present a series of formal, mathematical, models to help with our analysis. Why
do we need them?
While the interactions between environmental, social, and economic influences
on wellbeing can be described verbally, and with the help of pictures and diagrams,
unless we formalise these interactions with the help of mathematical models, it
becomes impossible to assess their collective influence on wellbeing either analyt-
ically or – in particular – quantitatively. Nor is it possible to quantify the comple-
mentarities and trade-offs between various policy interventions. And in any case,
it is impossible to perform policy-informing numerical simulations without such
quantitative models.
The framework and models we have chosen to use reflect our views, preferences,
and background, on modelling the interactions between environmental, social, and
economic influences on wellbeing. Needless to say, there will be various alternatives.
The choice between these will be based on theoretical reasoning and empirical evi-
dence. In any case, there may be no need to choose between the various modelling
1.11 Methodology and Structure of Presentation 25
options. Often, especially in a world of uncertainty and complexity, it is wise to

explore and compare the distillations from various (complementary or competing)
modelling approaches to the policy problems at hand.
1.11.2 What Types of Models
There is a distinct debate in the public policy literature on the types of models that
are best suited to assist with the formulation of public policy.
More specifically, there is an ongoing debate in the literature on whether a model
(such as the stylised model of Chap. 2), using optimising individuals as its unit of
analysis, and tending to converge to equilibrium states, can be of any use as a source
of wellbeing-enhancing policy guidance. The debate intensifies considerably when
it is acknowledged that “radical uncertainty” and “adaptive complexity” are essential
features of the world for which policies are being designed.22
We adopt an agnostic approach to this whole issue. All models are wrong or
incomplete, in the sense that they do not capture the full details of the phenomena
we are interested in analysing. They do however serve as tools for organising our
thoughts, and for analytical purposes. And, as we pointed out above, using a variety
of models to represent policy phenomena may be the wise approach to take to policy-
informing research.
The type of model we build in Chap. 2, which represents just one approach to
modelling, can be used very productively to identify the critical interdependencies
between environmental, social, and economic influences on wellbeing. It can also be
used to show that, unless we adopt an integrated approach to environmental, social,
and economic policies, we cannot achieve sustainable wellbeing. It even takes us
quite far in identifying the kinds of policy packages we should consider if our aim
is to promote intergenerational wellbeing.
The operational simulation model of Chap. 4 can then be used for policy simula-
tions and for “what if” type analysis when these policy packages are altered. We will
suggest this line of analysis in that chapter, to explore the impacts of various policy
packages on environmental, social, and economic influences on wellbeing.
Where such models fall short is in helping us formulate and implement effec-
tive and efficient policy packages that enhance intergenerational wellbeing, in the
presence of radical uncertainty and adaptive complexity. This is where the literature
on complex systems and radical uncertainty provides extremely useful insights. We
explore the implications of these extensions in the second part of the book.
Modelling the behaviour of satisficing (as opposed to optimising) individuals and
businesses in Chap. 6 helps us to deal with “radical uncertainty”, through viability
theory. A viability kernel, whose size will provide a measure of resilience, computed
for uncertain parameters, if nonempty, will be smaller than the kernel computed for
known parameters.
22 See Roos (2015) for an example of this literature.

Under these circumstances, when dealing with complexity and uncertainty, the
focus of policy shifts from the search for socially optimal dynamic solutions, to
building resilience to major catastrophes, and fertilising creativity and innovation.
In Chap. 6 we use viability theory as a tool for policy simulations that are focused
on exploring the implications of various combinations of policy instruments for the
resilience of environmental, social, and economic ecosystems.
In terms of the presentation style, we do not support all statements made through-
out the book with formal proofs, nor do we make an effort to always identify the
initial sources of the main ideas. We refer instead to the literature where these formal
proofs (if they exist) are provided, and the original sources of the main ideas are
cited. Our purpose is to show that a formal model for our wellbeing framework can
be constructed, and this model can then be used to identify the domains of a public
policy that aims to enhance intergenerational wellbeing. We also wish to use the
model to highlight the key complementarities and tradeoffs that we face as a society
in this pursuit, and to explore the policy options and levers available to the policy
makers to relax these tradeoffs and exploit the complementarities to the same end.
Questions for Study and Discussion
1. Assess the following statement: “Economics is a value-free social science.”

2. How convincing do you find the following statement: “The purpose of public
policy is to improve the wellbeing of individuals and communities, by helping
them live the kinds of lives they have reason to value – but without in any way
judging what lives people should value.”
3. What do you understand from the term “wellbeing” – why is income not sufficient
to measure wellbeing? What evidence do you have to support your answer?
4. What are the fundamental cases for public policy – why should the government
be involved with improving people’s lives?
5. How would you distinguish between a “welfare state” and a “wellbeing state”?
What difference does this distinction make for the role, domains, and instruments
of public policy?
6. Suppose we did have data on the policy dimensions of wellbeing in Fig. 1.8, how
would you measure the areas of those triangles over time?
Suggestions for Further Reading

Adler and Fleurbaey (2016) provides a comprehensive entry point, including lists of
references, to the philosophical, technical, as well as empirical, literature on wellbe-
ing and public policy.
Chapter 2
Stylised Model
Abstract This chapter presents a formal “stylised model” in an attempt to pull

together the ideas introduced in the introductory chapter into a coherent whole. This
is intended to facilitate reasoned discussions on how the environmental, social, and
economic influences on individual and community wellbeing interact with each other,
as well as the complementarities and tradeoffs between them. In future chapters, this
will in turn help us formulate policies towards increasing wellbeing on a sustained
basis, while taking into account these complementarities and tradeoffs. Each section
of the chapter has three sub-sections. First, a verbal and diagrammatic presentation of
the core ideas. Second, a mathematical representation of the verbal and diagrammatic
discussion, including the motivations of each of the key actors (such as individual
consumers, producers, and the government). Third, a list of references (accompanied
by a brief discussion) of the empirical literature that attempts to bring evidence to
support or challenge the key assumptions of the theoretical model.
• Explain and discuss the difference between a “framework”, a “theory”, and a
“model”.
• Critically assess the contributions that a “stylised model” makes to the discussion
of policy problems.
• Articulate the respective roles and differences between “generic” and “stylised”
models.
• Explain the respective roles of theory and empirical evidence in the specification
of stylised models.
• Discuss and explore alternative ways in which stylised models can be structured
to capture the interaction of environmental, social, and economic influences on
wellbeing.
• Explain how empirical evidence can assist with identifying complementarities and
substitutabilities between different factors of production, and what difference this
makes to the formulation of policy.
https://doi.org/10.1007/978-981-13-6104-3_2
28 2 Stylised Model
• Critically assess the significance and implications of using different forms of pro-
duction, utility, and other functions in stylised models.
• Formulate and write down policy-informing stylised models, being able to differ-
entiate between useful and not-so-useful ones.
2.1 Introduction
Figure 1.2 in Chap. 1, is an attempt to capture the essence of our framework for public
policy when the objective is to enhance intergenerational wellbeing. In the middle of
the diagram, we have the capital assets that are the key components of “comprehen-
sive wealth”. These are the ultimate sources of the main influences on wellbeing, as
we discussed in Chap. 1. Wrapped around these capital assets is a wellbeing frontier,
framed by five key social outcomes (or public goods) that good public policy aims
to enhance, to help us extract most value (wellbeing) from comprehensive wealth, in
the present and into the future. It is useful to think of them as defining the boundaries
of our collective wellbeing frontier. They are therefore of legitimate interest for a
public policy that aims to push out or extend this frontier, while being cognisant of
the interdependencies (complementarities and tradeoffs) between them. The glue that
binds comprehensive wealth to these public goods is the stewardship of our capital
assets for the benefit of current and future generations.
Our aspiration is to demonstrate the existence of a coherent and unified stylised
model (one possible such model), that is based on this framework. We then use this
model as a platform for designing a wellbeing-focused public policy. In this chapter
we present such a stylised model, and then in Chap. 3 we use it to design public
policy. A distinctive feature of both the general framework which we introduced in
Chap. 1, and the specific stylised model we present in this chapter, is their explicit
recognition of the interactions between environmental, social, and economic factors
in influencing wellbeing. The capital assets (natural, social and cultural, human, and
economic) that serve as the ultimate sources of environmental, social, and economic
benefits, provide the primary channels through which policy interventions influence
wellbeing.1
Figure 2.1 is a diagrammatic summary of the key linkages identified by the math-
ematical model that is presented in the rest of this chapter. The figure highlights:
• the critical link between comprehensive consumption and wellbeing;
• the direct and indirect inputs into comprehensive consumption (through compre-
hensive wealth);
• international influences; and
• the positive and negative (red arrows) externalities associated with poverty, equity,
and social cohesion.
1 See also Galor and Weil (1999, 2000) on the desirability of building unified models.
2.1 Introduction 29
Fig. 2.1 Stylised model
Readers are encouraged to refer back to this diagram as they go through, especially
the mathematical sections of, the material that follows.
The stylised model of this chapter weaves together threads from the existing
literatures on wellbeing, human needs, complex systems, sustainable development,
endogenous economic growth, directed technical change (favouring “clean” technol-
ogy), and credit-based money. In doing so, we also wish to integrate the “subjective
wellbeing“, “capabilities”, “opportunities” and “human needs” approaches to well-
being in one model (see Baujard and Gilardone 2017; Bruni et al. 2008; Ferreira and
Peragine 2015; Gough 2015; O’Donnell et al. 2014; Robeyns 2005).
All the ingredients of the model are already out there, in works usefully sum-
marised, and expanded upon, by Acemoglu et al. (2012, 2016), Arrow et al. (2012,
2013), Atkinson (2015), Chichilnisky (1997), Jakab and Kumhof (2015), Jones and
Vollrath (2018), Krugman (1979), Stiglitz (2015), Turnovsky and Mitra (2013). We
integrate the key insights and methods presented in these works into a coherent
and unified model that underpins our wider wellbeing framework, with the specific
purpose of informing policy advice.2
The main purpose of a stylised model is not to capture all the relevant detail that
may apply to particular circumstances, but rather to provide a structure for thinking,
2 An alternative model can be based on viability theory (see Krawczyk and Judd 2016; Krawczyk
and Kim 2014). In fact, we use such a model to perform numerical policy simulations in Chap. 6
of this book. Yet another potential approach to the problem at hand is provided by overlapping
generation models (see De La Croix and Michel 2002).
30 2 Stylised Model
in a rigorous way, about the matter at hand. In our case, our primary focus is on the role
of public policy in enhancing intergenerational individual and community wellbeing,
and the channels through which it operates to serve this purpose. We use the domains
of public policy as a prism for determining the level of detail that is included in the
model. By way of example, since influencing individual preferences between private
marketable consumption goods is out of scope, no such detail is included in the model
except with reference to the choice between private consumption goods, leisure, and
health services. On the other hand, since the allocation of human, physical, and
financial capital across individuals is critical for potential wellbeing, individuals in
our stylised model are heterogeneous with respect to their ownership of these forms
of capital.3
2.2 Consumption, Wellbeing, and Wealth
2.2.1 Generic Model
Our generic model is a standard dynamic optimisation model that has been expanded
to accommodate “comprehensive consumption” and “comprehensive wealth”. We
use this as a basis for constructing a particular stylised model which eventually
serves as our policy-design tool. This stylised model includes a set of public goods,
such as social cohesion and resilience to systemic risks, which are the sources of
significant positive externalities. These public goods collectively shape the “wellbe-
ing frontier”, and therefore represent the main focus (or “domains”) of public policy.
Policy interventions are targetted at extending the wellbeing frontier, and thereby
intergenerational wellbeing, through appropriate investments in the components of
comprehensive wealth.
2.2.1.1 Motivation and Microfoundations
The generic model is grounded in a framework for thinking about wellbeing-

enhancing environmental, social, and economic policies in an integrated way, and
the distinctive role of government in this context. The main focus of such policies is
to increase intergenerational individual and community wellbeing on a sustainable
basis, by enhancing the capabilities and opportunities (i.e., substantive freedoms) of
individuals and communities to pursue the kinds of lives they have reason to value.
At the core of the model is the interaction between wellbeing, consumption,
wealth, production, and public policy. Wellbeing is a function of comprehensive
consumption, which includes not only standard marketed consumption goods, but
3 One potential criticism is the absence of stochastic, or probabilistic, uncertainty (or risk) in the
model. We return to this point in the second part of the book when we introduce radical uncertainty,
and explore its implications for the design and implementation of public policy.
2.2 Consumption, Wellbeing, and Wealth 31
also (among others) leisure, arts, health services, and environmental services pro-
vided by nature. Comprehensive consumption, in turn, is a function of comprehensive
wealth, which comprises all categories of capital stocks.
Let us now formally define intergenerational wellbeing and comprehensive
wealth.
Definition 2.1 Intergenerational wellbeing is the discounted present value of the
utilities derived by current and future generations from comprehensive consumption,
compare Arrow et al. (2012, 2013), Hamilton and Hepburn (2014).
Definition 2.2 Comprehensive wealth is defined by the stocks of various types of

capital assets, enhanced by infrastructures and ecosystems, as represented by Fig. 1.2
(middle).
The source of comprehensive consumption is comprehensive wealth. Some capital

stocks influence intergenerational wellbeing both indirectly, through the consumable
goods and services they help produce, and also as direct sources of utility. Compo-
nents of natural capital (such as a clean environment) and human capital (such as
good health) are just two examples. (Refer back to Fig. 2.1).
In a social and intergenerational setting, individuals and communities also receive
positive benefits from public goods with potentially significant positive externalities,
such as social cohesion, equity (across society and generations), sustainability of
sources of wellbeing, resilience to major systemic risks, as well as increases in the
growth potential of the economy. (Again refer back to Fig. 2.1). The choice of the
particular public goods and associated externalities we focus on (i.e., the domains
of public policy) is informed by the preferences of the public at large, as revealed or
discovered by various mechanisms (such as surveys). Because the sources of all these
externalities have “public good” attributes, they may be under-provided, and/or their
sources may be under-protected, by private individuals, businesses, and communities
if left to their own devices.
2.2.1.2 Mathematical Model
We now present the skeletons of a mathematical version of the narrative provided in

this section. To serve our aspirations, the model needs to capture all key features of
the wellbeing framework – all capital stocks (including supporting infrastructures)
and all dimensions (i.e., critical public goods, and associated positive externalities)
surrounding the capital assets – and connect them to overall wellbeing.
The generic version of the model is outlined in Arrow et al. (2012), and takes the
following form.4
4 We reproduce some key derivations of that paper here to set the generic platform for the more
specific (stylised) model to be used in the rest of this chapter.
32 2 Stylised Model
Time is continuous and is denoted by s and t (0 ≤ t ≤ s ≤ ∞). The horizon is

taken to be infinite.5
For our purposes what matters is that the “typical individual’s” wellbeing depends
on “comprehensive consumption”. In that particular sense, preferences are identical
and individuals are homogeneous. However, as we will explain when we introduce
our stylised model below, individuals are heterogeneous with respect to how they
generate their market incomes, and their ownership of private human, physical, and
financial capital assets.
It is worth re-emphasising at this juncture that we are constructing a policy-
informing model that will be of use as a guide for policy-advisers to a wellbeing
state. We are not interested in second-guessing or passing judgement on people’s
detailed preferences. Instead, given the wider influences on wellbeing that are com-
mon across societies, our primary focus is on public investments that will enhance
the opportunities and capabilities of individuals and communities to pursue the lives
they have reason to value.
Following Arrow et al. (2012), the wellbeing function for the “typical individual”
(and for society at large) can be written as follows:
∞

WA (t) = U (C(s))e−ρ(s−t) ds, ρ ≥ 0 (2.1)
t
where WA (t) denotes intergenerational wellbeing at t, which depends on C(·) – a

vector of consumption flows that includes not only marketed consumption goods,
but also various others including leisure, arts, health services, and consumption ser-
vices supplied by nature – i.e., it represents comprehensive consumption. U (·) is a
mapping from RmC to R1+ where mC is the number of consumption goods. Hence
U (C(s)), t ≤ s ≤ ∞ represents the collective utility flow at time s. Symbol ρ rep-
resents the discount rate. Thus, intergenerational wellbeing is the discounted utility
flows of current and future generations (compare Definition 2.1).
An assessment of wellbeing at [t – W (t)] – requires a forecast of utility beyond t.
For this purpose, the relevant forecast at t is the pair of vector functions {C(s), K(s)}
for s ≥ t, where K(s) denotes the stocks of a set of capital assets at s (i.e., comprehen-
sive wealth (compare Definition 2.2). K(s) provides the sources of comprehensive
consumption at s. We assume that the integral in expression (2.1) converges for the
forecast.
Current and future wellbeing depends on collective comprehensive wealth. It also
depends on the evolving structure of technology, people’s values and preferences,
and institutions. The stock of assets that comprise comprehensive wealth at any
moment s will be determined by the stocks at the immediately preceding moment.
5 This does not imply that individuals have perfect information. In fact, as we will point out later in
Chaps. 5 and 6, extensions of the stylised model recognise the existence of fundamental uncertainty.
Rather, what is intended is that individuals behave as if they care about current and future generations
into the infinite future.
By proceeding from moment to moment in this way, the entire future course of capital
assets and therefore collective wellbeing will be determined.
Thus, given K(t), K(s), C(s), and U (C(s)) are determined for all future times
s ≥ t. Hence, intergenerational wellbeing W (t) is determined as well. Therefore, we
can write:
W (t) = W (K(t), t) (2.2)
In expression (2.2), W (t) depends directly on t, to reflect the potential impact of

a set of time-varying factors such as population growth, technological change, and
institutions, in addition to capital assets, that affect intergenerational wellbeing. In
the stylised model used below to represent our wider wellbeing framework, we will
also add various public goods which are sources of positive externalities, such as
equity and social cohesion, to this list.
Defining sustainable intergenerational wellbeing as d W (t)/dt ≥ 0 and assuming
W (t) is differentiable in K, a criterion for sustainable intergenerational wellbeing at
t can be specified as follows – where capital assets are indexed by i:

d W (t)/dt = ∂ W/∂t + (∂ W (t)/∂Ki (t)) (dKi (t)/dt) ≥ 0 (2.3)
i
We now define two “shadow prices”, which will in turn help us define “compre-
hensive wealth”. The first, λKi (t), is the (spot) shadow price of the ith capital asset
at time t:
λKi (t) ≡ ∂ W (t)/∂Ki (t), for all i (2.4)
and represents the contribution to W (t) made by Ki (t) both through the goods and
services it helps produce, as well as through direct enjoyment of the stock itself.
A forest, clean water, and health are three examples. Where there are externalities
involved (associated with a “tragedy of commons” for example), an asset’s shadow
price can be negative when its market price is positive. By way of example, this would
be the case if the present value of disposing of a car was higher than its current price.
Here, we share Arrow et al. (2012)’s (pp. 323–324) interpretation of the shadow
prices. “At any date an asset’s shadow price is a function of the stocks of all assets.
Moreover, the price today depends not only on the economy today, but on the entire
future of the economy. So, for example, future scarcities of natural capital are reflected
in current shadow prices of all goods and services. That means that shadow prices
are functions of the degree to which various assets are substitutable for one another,
not only at the date in question, but at subsequent dates as well. Of course, if the
conception of intergenerational wellbeing involves the use of high discount rates
on the wellbeing of future generations (i.e., if ρ is large, the influence on today’s
shadow prices of future scarcities would be attenuated). Intergenerational ethics
plays an important role in the structure of shadow prices” Arrow et al. (2012) (pp.
323–324).
The second shadow price we require, to define comprehensive wealth, is the
shadow price of time which, based on expression (2.2), we can conceptualise as an
additional form of capital asset:
34 2 Stylised Model
λT (t) ≡ ∂ W/∂t (2.5)
We can now use these two shadow prices as weights to construct an aggregate index
of the society’s comprehensive wealth (V ):

V (t) = λT (t)t + iλ
Ki
(t)Ki (t) (2.6)
A society’s comprehensive wealth is the (shadow) value of all its capital assets (see
Arrow et al. 2012, p. 324).
There is an equivalence relationship between changes in comprehensive wealth
at constant prices and changes in intergenerational wellbeing:

W (t) = λT (t)t + iλ
Ki
(t)Ki (t) = V (t) (2.7)
or, letting Ii (t) ≡ Ki (t)/t, we can establish an equivalence relationship between
changes in intergenerational wellbeing and “comprehensive investment” (or “genuine
saving”), which is the expression on the right-hand side of the following expression:

W (t) = λT (t)t + iλ
Ki
(t)Ii (t)t (2.8)
It is very important to note that the term “investment” here is used in a wider
sense than its typical use, which refers to accumulation, when the only type of
capital is “reproducible capital”. When reference is to a broader set of capital assets
comprising comprehensive wealth, including natural, human, and social and cultural
capital, “investment” means “any increase in the flow of services that the asset can
provide over its lifetime. To leave a forest alone so that it can grow is in our extended
sense to invest in the forest. […] To give food to someone hungry […] increases
her future productivity – which means that to prevent hunger is to invest in human
capital” Arrow et al. (2013, p. 513).
A couple of other points are also worth highlighting before we proceed. First, the
ratios of shadow prices are marginal social rates of substitution among the various
capital assets, and when W (t) is maximised these marginal rates of substitution
equal their corresponding marginal rates of transformation. Second, sustainability
and optimality are different concepts. It is quite possible that along an optimum path,
one that maximises W , W (t) may decline for a while (thus violating the sustainability
criterion), and then start increasing again.
2.2.1.3 Empirical Evidence
The key empirical issue here is whether individuals are rational, welfare (or well-
being) maximisers, who base their consumption/saving decisions on present-value
maximisation, or at least they attempt to exploit every opportunity to enhance their
wellbeing, subject to their capabilities.
Greasley et al. (2014) test this hypothesis using long term (130 years) USA data.
They also provide a very useful summary of related (but very limited) previous
empirical work.
Their main conclusion is that when the measures of changes in total capital (or
wealth) are expanded to include human and natural capital (i.e., the measures that
proxy “comprehensive wealth”), and long-term data is used, there is empirical support
for this basic assumption of modern economic growth theory. Empirical support
for the assumption is strengthened when technological progress and the effects of
population growth on wealth dilution are allowed for.
In a similar vein, Greasley et al. (2013, 2017), and Hanley et al. (2016) provide
empirical evidence, for a few countries, in support of the proposition that, when an
expanded version of comprehensive wealth is used, comprehensive investment can
predict future wellbeing reasonably closely in the longer run. Again these results
are strengthened when technological progress is allowed for in the models that are
empirically tested.
2.2.2 Stylised Model
2.2.2.1 Motivation and Microfoundations
As stated in the Introduction to this chapter – Sect. 2.1 – the components of the stylised
model are spread across the existing literature. We simply borrow from this literature,
as summarised and enhanced by Acemoglu et al. (2012, 2016), Arrow et al. (2012,
2013), Atkinson (2015), Chichilnisky (1997), Jakab and Kumhof (2015), Jones and
Vollrath (2018), Krugman (1979), Stiglitz (2015), Turnovsky and Mitra (2013), and
attempt to expand on and integrate key components of the models presented in these
papers, to build a unified stylised model.
There are seven clusters of “actors” in our stylised model: consumers, work-
ers, producers/manufacturers, researchers/scientists, entrepreneurs, financial firms
(banks), and a government through whom individuals and communities implement
collective actions. These actors live and work in a small open economy that is con-
nected to the rest of the world through trade, as well as flows of physical and financial
capital, people, and ideas.
As we work through the stylised model, we provide a high-level specification
of what motivates individual clusters of actors (i.e., the “micro-foundations of the
model”). This is relevant because some policy interjections are targetted at incentivis-
ing private individuals, communities, and firms to adjust their behaviours, towards
generating more socially desirable outcomes. Others involve direct investments by
the government (on its own as our collective agent, or in partnership with individuals,
businesses, and community institutions), in building, shaping, managing, distribut-
ing, and/or protecting the capital assets that are direct and/or indirect sources of
wellbeing for individuals and communities.
36 2 Stylised Model
Individuals (in all their capacities) are intergenerational wellbeing maximisers.

Firms (non-financial as well as financial) are profit maximisers. Government is a
generic term we use to refer to institutions that have been deliberately created to
coordinate and give effect to activities we decide to pursue collectively rather than
individually, with a view to enhancing our individual and community wellbeing.
Since individual wellbeing partly depends on the consumption of private (mar-
keted) consumption goods, individuals need to generate market income to be able to
purchase and consume these goods. They can do so as unskilled or skilled labourers,
or as scientists/researchers/engineers, or as entrepreneurs and business people in our
model.
Income can also be generated from the ownership (as distinct from the production)
of private capital, in the form of physical capital (i.e., the machines used to produce the
final good), financial capital, human capital, and some forms of natural capital (such
as land). In the stylised model we introduce in the next section, individuals differ in
(i.e., are heterogeneous with respect to) their endowments of human, physical, and
financial capital; for simplicity, we do not have land in the model. Using Amartya
Sen’s terminology, in our stylised model individuals’ opportunities and capabilities
(i.e., their substantive freedoms) to generate higher and sustainable wellbeing for
themselves and for future generations are positively affected by their ownership of
various forms of private capital, as well as their access to the services of different
forms of public capital.
All transactions require the use of money (bank deposits), and are costly.
Entrepreneurs have to borrow from banks to fund research (undertaken by scientists
– a subset of households), and then as well to invest in the machinery, manufactured
by the machine producers, imbedding the successful research. Some of the research
is unsuccessful and therefore results in defaults on the loans used to fund them. In
addition, banks also incur monitoring costs.6
A fundamental assumption underpinning the stylised model is that, in addition
to their individual incomes and consumption of private goods (including leisure
and good health), individuals also value (i.e., receive positive benefits from) the
positive externalities associated with, among others, a clean environment, sustainable
comprehensive wealth, social cohesion, equity across society and generations, and
enhanced resilience to the types of shocks that have the potential to cause serious
damage to various capital assets and through them on our overall wellbeing. This is
where the government comes in.
The distinctive role of government is to coordinate and/or undertake activities that
individuals have decided to pursue collectively. Thus, the government is by definition
an agent for collective action. A primary (but by no means the exclusive) purpose of
these activities is to generate positive externalities, and eliminate negative externali-
ties, associated with public goods. This is achieved partly through the establishment
of appropriate institutions (including private property rights, markets, regulations,
taxes, and fees) that incentivise private agents to internalise such externalities. There
6 Fromour perspective, introducing minimum capital requirements and associated costs would be
an unnecessary complication (see Jakab and Kumhof 2015).
are other forms of market failure as well, such as those associated, for example, with
informational asymmetries.
2.2.2.2 Comprehensive Wealth
In the stylised model, there are four broad categories of capital assets, collectively
referred to as comprehensive wealth, which are direct and/or indirect sources of
wellbeing7 : economic (physical and financial) capital, human capital (broadly defined
as we will see shortly), social and cultural capital, and natural capital.8
Especially in a public policy setting, it is extremely useful to think of each
category of capital assets in broad terms, inclusive of supporting and enabling infras-
tructure (or ecosystems). A critical component of this infrastructure is a set of institu-
tions that (North 1994) describes as “the humanly devised constraints that structure
human interaction. They are made up of formal constraints (rules, laws, constitu-
tions), informal constraints (norms of behavio[u]r, conventions, and self imposed
codes of conduct), and their enforcement characteristics. Together they define the
incentive structure of societies and specifically economies” (p. 2).9
Economic capital comprises physical (human-made) capital (such as machines,
buildings, etc.) directly used in production, and financial capital. Part of physical
capital is economic infrastructure, and part of financial capital is financial infras-
tructure. Economic infrastructure comprises physical infrastructure (roads, bridges,
etc.), which has public good attributes, providing and supporting transport, commu-
nication, energy, as well as other networks and services shared by multiple users,
underpinning a variety of production and distribution activities. Financial infrastruc-
ture comprises all the institutions (including the central bank, banks, other financial
intermediaries, money, and the supporting legal infrastructure and regulations) that
underpin an effective, efficient, and resilient financial (including payments) system;
thus it too has public good attributes.
Human (2013) defines a nation’s human capital endowment as, “the skills and
capacities that reside in people and that are put to productive use” (p. 3). It then
proceeds to construct an index for human capital that is based on four ‘pillars’: edu-
cation, health and wellness, workforce and employment, and enabling environment
(p. 4). The stock of knowledge embodied in human beings (i.e., intellectual capital) is
7 Stiglitz(2015) makes a very compelling case for carefully distinguishing between capital (in the
sense of produced assets such as machines) and wealth (including “land or other ownership claims
giving rise to rents”). Our broader definition of both wealth and capital assets should hopefully
address his concerns.
8 Others, such as Gleeson-White (2015), Sachs (2015) refer to six categories of capital but, depending
on how we define them, we are essentially referring to the same types of capital; there are no
substantive differences here.
9 Hodgson Geoffrey (2006) offers a wider definition of institutions as systems of established and
embedded social rules that structure social interactions. Institutions do not only constrain and
influence the choices and actions of individual consumers and businesses, but they also perform a
critical enabling function – and their evolution is endogenous (see also Arvanitidis 2004).
38 2 Stylised Model
also included in human capital. Underpinning these human capital assets is an infras-
tructure, the enabling environment, including schools, hospitals and innovation incu-
bators, with strong public good attributes, that collectively provide a shared platform
for the accumulation and deployment of these human capital assets – i.e., human
development. Our stylised model allows for the dynamics of skill accumulation.
Skills embodied in agents empower people. Enhanced skills increase opportunities
and hence promote capabilities (see Heckman and Corbin 2016; Heckman and Mosso
2014).
While social and cultural capital are distinct, they are very closely related; that
is why we aggregate them into the category of social and cultural capital. For our
purposes, it is very important to emphasise the distinction, as well as the relationship,
between social capital and social cohesion; with the former being the source of
the latter. In this context, we find the definition and dimensions of social capital
provided by Growiec et al. (2017), which is in turn based on the work of Bourdieu
(1986) and Putnam (2001), extremely useful. It is a networks-based definition which
emphasises the aggregate resources available to individuals and communities through
their social networks. It has four distinct dimensions: network degree (number of
social ties), network centrality (position within networks), bridging social capital (ties
with dissimilar others), and bonding social capital (ties with similar others).10 When
we discuss social cohesion later in this chapter, these last two dimensions of social
capital, their relationship with the corresponding dimensions of social cohesion and,
through these, their distinctive (and opposite) impacts on economic and wider social
performance will be clear. We will emphasise again then, that generalised (or social)
trust and willingness to cooperate (which are both dimensions of social cohesion) are
among the immediate outcomes of social capital, while income growth and subjective
wellbeing are two of the ultimate outcomes of social capital (see Growiec et al. 2017,
p. 2).
As we already pointed out, we group cultural capital with social capital, into
a single category labeled social and cultural capital. Culture is often defined as,
“those customary beliefs and values that ethnic, religious, and social groups transmit
fairly unchanged from generation to generation” (see Alesina and Giuliano 2015).
As (Dalziel et al. 2009) emphasise, social capital has a cross-society, current period,
dimension whereas cultural capital has a distinctive intergenerational dimension. The
latter refers to accumulated attributes and wisdom that is transmitted through gener-
ations. However, there is an alternative definition, that does not assume persistence
or vertical intergenerational transmission: “people’s preferences, values, attitudes,
beliefs, and social norms” Gershman (2016, p. 2). The fundamental point is that both
the diversity of cultures within a society, and the strength of social capital (bridging
social capital in particular) have significant impacts on social cohesion and potential
economic growth.
10 See also Hamilton and Hepburn (2014), Jones and Vollrath (2018, Chap. 7), and also Savioli
and Patuelli (2016), who usefully distinguish between three domains (and associated dimensions)
of social capital: cognitive (mental processes, concepts, ideas), relational (trust, norms, identity),
structural (organisations, institutions, leadership).
Natural capital includes underground assets (minerals, fossil fuels), commercial

land, fish stocks, natural land (i.e., both non-renewable and renewable resources),
and ecosystem services that all these resources provide (see Hamilton and Hepburn
2014), underpinned by “green and blue infrastructure” (see Costanza 2015; van den
Belt 2018). A critical ingredient of natural capital is the natural regulatory systems
needed for the reproduction (sustainability, viability) of natural capital stocks; these
regulatory functions protect the climate system and biodiversity (see De Perthuis
and Jouvet 2015). Much like social infrastructure, natural infrastructure (or natural
regulatory functions) too, affect wellbeing both directly, as well as indirectly through
their impact on the potential growth rate of the economy.
2.2.2.3 Mathematical Model
The stylised model that we use in the rest of this book borrows several key features
from the Arrow et al. (2012, 2013) model outlined above. Most importantly, the object
of interest remains intergenerational wellbeing and its sustainability. Wellbeing is
positively influenced by comprehensive consumption, and comprehensive wealth is
the source of comprehensive consumption.
We build on this platform, using the discrete-time model of Acemoglu et al. (2012)
as a benchmark. We then enhance that model through an integrated one that brings
together environmental, social, and economic spheres, and associated externalities,
as essential and complementary ingredients of wellbeing.
The stylised model embodies the point made by Friedman (2006), that “the value
of a rising standard of living lies not just in the concrete improvements it brings to
how individuals live but in how it shapes the social, political and, ultimately, the moral
character of a people. Economic growth – meaning a rising standard of living for the
clear majority of citizens – more often than not fosters greater opportunity, tolerance
of diversity, social mobility, commitment to fairness, and dedication to democracy”
(p. 15). However, “economic progress needs to be broadly based if it is to foster social
and political progress” (ibid., p. 16). Economic growth cannot be sustained unless it
is underpinned by social cohesion and environmental sustainability (see Duraiappah
and Muñoz 2012) – and relying on market forces alone will not achieve all that (see
e.g., Acemoglu et al. 2012).
The model also captures the fundamental notion that the best way to build shared
wellbeing on a sustainable basis is to provide individuals with the opportunities and
capabilities (embodied in components of comprehensive wealth) to participate pro-
ductively in social and economic life, so that they can get their fair share of the
prosperity they help create. Individual wellbeing is enhanced through active contri-
bution, rather than being a passive recipient of welfare payments (see Engelbrecht
2014).
40 2 Stylised Model
2.2.2.4 Mathematical Model - Consumption
Comprehensive consumption refers to the consumption of a set of private and public

goods and services. In addition to deriving utility from the consumption of a marketed
product (the single or composite output Y in the stylised model), individuals also
derive direct utility from the consumption of time (i.e., “leisure”) and good health.
These are the private goods that enter into the individuals’ utility functions.
We assume that individuals also benefit from having access to public goods (and
associated positive externalities) arising from a high-quality (pristine) environment
(E), social cohesion (S), and increased resilience of both (E) and (S) to potential
major systemic shocks – (Γ E ) and (Γ S ) respectively:
∞
1 y
W0 = t u φ(Yt /Yt )Ct , Ct , Ct ; Et , St , Γt , Γt
d x h E S
(2.9)
t=0
(1 + ρ)
Here Y d is the quantity of output obtained through “dirty” technology; C y is the

consumption of the single (or composite) marketed product Y , weighted by a function
φ(Ytd /Yt ) ∈ (0, 1], with φ(Ytd /Yt ) decreasing in (Ytd /Yt ). Thus, the higher the weight
of “clean” technology used in the production of Y , the higher the utility derived from
the consumption of a unit of Y (think of eco-friendly cars, renewable energy use in
heating, organic food).11, 12
Symbol C x denotes the consumption of leisure, C h is the consumption of good
health, and E, S, ΓE , ΓS are all indices ∈ (0, 1), representing the degree of envi-
ronmental quality, social cohesion, and the quality and effectiveness of measures
that enhance environmental and social resilience to potential major shocks to each,
respectively.
Individuals take Et , St , ΓtE , ΓtS as given in any period t, in deciding how to allocate
their income (between the marketed consumption good, financing human capital
formation, and saving), and their time (between work, leisure, education or training,
and health-enhancing activities). As we show below, investment of time and income
in education, training, and health are different forms of human-capital investment.
They may lead to an increase in wellbeing both directly, by providing higher utility,
but also indirectly, by increasing productivity and therefore market-income.
The instantaneous utility function ·u φ(Y d /Y )C y , C x , C h ; E, S, ΓE , ΓS is
twice-differentiable, increasing and jointly concave in (C y , C x , C h , E, S, Γ E , Γ S ).
We assume that:
11 We admit that proposing a function like φ(·) in (2.9) may amount to social engineering. However,
without φ(·), obtaining environment-enhancing policies can be difficult in the stylised model of
this chapter. As we will argue in Chap. 5, this is not the case in models that include complexity and
radical uncertainty.
12 This is an attempt to respond to Gough (2015)’s plea that good policy should be taking into account
environmental, economic, and social considerations in an integrated way; and, in the context of the
environment, consumption as well as production policies should be considered in a complementary
fashion.
∂ W (C y , C x , C h ; E, S, Γ E , Γ S )
lim = ∞; i ∈ {y, x, h} (2.10)
Ci ↓0 ∂Ci
∂ W (C y , C x , C h ; E, S, Γ E , Γ S )
lim = ∞; j ∈ {E, S, ΓE , ΓS } (2.11)
j↓0 ∂j
lim W (C y , C x , C h ; E, S, Γ E , Γ S ) = −∞; j ∈ {E, S, ΓE , ΓS } (2.12)
j↓0
The last two conditions – (2.11) and (2.12) – highlight and emphasise the point
that if the quality of the environment, the degree of social cohesion, and/or the degree
of aggregate resilience to potentially severe shocks to both of these public goods,
were to approach their lower bounds, this would have severe negative wellbeing
consequences.
2.2.2.5 Comprehensive Wealth and Resource Constraints
Comprehensive consumption is sourced from comprehensive wealth which, in the

stylised model, comprises some private and some public capital assets. The private
capital assets are physical capital (in the form of the machines used to manufacture
the single consumable good), human capital (in the forms of skills, education, and
health), and financial (bank) capital. Public capital assets on the other hand comprise
some forms of natural capital (or more broadly, environmental infrastructure), social
infrastructure, and economic infrastructure. From an overall wellbeing perspective,
we are very interested in the evolution of all forms of capital assets, private and
public, widespread access to public assets, and the distribution of private capital
assets. In the formal stylised model, the evolution of capital assets all have the same
structure – investment terms that generate or augment these assets, and depreciation
terms that deplete them.
In the pursuit of higher wellbeing, individuals face three types of resource con-
straints: time, capital assets (or comprehensive wealth), and income (from labour as
well as various types of capital ownership).
In the stylised model, time is the resource that can be partly used to accumulate
human capital (through training or education, and investment in good health), and
income is the resource that can be partly used to accumulate physical capital (through
saving) and/or fund human capital formation. Thus human capital is accumulated by
investing time and income, and physical capital is accumulated by investing income.
In each period, each individual is endowed with one unit of time that can be
allocated to leisure (X ) or non-leisure (NX ) activities. Non-leisure activities comprise
working (L), investing in training (TR) or education (ED) (but not both, because one
requires going to a vocational school and the other to a university, and we cannot be
in two schools at the same time), and investing in one’s health (H̃t ):
1 = Ctx + NXt
(2.13)
NXt = Lt + TRt + EDt + H̃t
42 2 Stylised Model
where, Ctx refers to the consumption of leisure-time; Lt , TRt , EDt , H̃t are each
∈ [0, 1]; and TR = 0 if ED > 0 and vice versa. Thus Eq. (2.13) represent the indi-
vidual’s time-budget constraints.
Time-investment in skills-training is required to convert investment-time (NX −
L) into skilled labour (Ls ), and time-investment in education is required to convert
investment-time into scientists (Sc). In the stylised model, for simplicity, we assume
that it takes one period of investment (with the length of the period to be chosen con-
veniently to suit our purpose) to convert investment-time into skilled labour (through
training) or into a scientist (through education).13 Again, for simplicity, we assume
that investment in health is simply a function of time (say the use of time to exercise).
We assume that these time-conversions take the following forms:
s
Lst+1 = (1 − ξ L )Lst + μtr TRt (2.14)
Sct+1 = (1 − ξ )Sct + μed EDt
Sc
(2.15)
Ht+1 = (1 − ξ )Ht + Υ (Et )μh H̃t
H
(2.16)
where, H ∈ (0, 1) is a “health index” (representing the stock of health as part of our
human capital), ∂Υ/∂E > 0, and μtr , μed , and μh , which are all ≥0, refer to the
returns from the investment of time into training, education, and health activities,
respectively; and ξ j > 0 represents the rate of depreciation of j, for j ∈ {Ls , Sc, H }.
Thus, the evolution of “good health” reflects both private time-investment in health
and the impact of the quality of the natural environment on health.14
In order to be able to consume the marketed product and fund human-capital
formation, the individual has to generate market income, which is spent on consum-
ing the marketed product and funding human-capital investment, or saved (in the
form of physical or financial capital). Individuals can generate market income by
working as (unskilled or skilled) labourers, doing research (as scientists), working
as entrepreneurs, or producing and selling the final product, as well as through their
earnings on their ownership of physical wealth (i.e., the machines that are used to
produce the final output, or bank equity). Thus an individual’s personal income is
measured by the sum of income from physical and financial wealth, and from skilled
or unskilled labour, or from research (as a scientist). We can therefore represent the
individual’s income-budget constraints, all expressed in units of the final output, as
follows:
13 From now on, we will use the term “scientists” to include engineers as well.
14 The possibility of H
t+1 > 1 exists and has not been formally ruled out in our model. To ensure the
health stock remains ∈ (0, 1) is a modelling problem. If we are to propose optimal solutions, they
have first to be feasible. Mathematically, this condition could be achieved via a Lagrange function.
But, it would be messy. This problem concerns other indices as well (e.g., ΓtE ). We recognise it
but, for our purposes in this chapter, we can put it to one side for now.
K K
L
rt b Kb,t + rt p Kp,t
L
+ rtd BdL,t + wt Lt − (1 + Υtc )Cy,t
L
− H̃y,t
L
= SatL
L,Kp L,Kb
= Sat + Sat (2.17)
K Ls K Ls
rt b Kb,t + rt p Kp,t + rtd BdLs,t + wts Ls,t Ls
− (1 + Υtc )Cy,t − TRLy,ts − Ls
H̃y,t = SatLs
L ,K L ,Kb
= Sat s p + Sat s (2.18)
K K
Sc
rt b Kb,t + rt p Kp,t
Sc
+ rtd BdSc,t + wtSc Sct − (1 + Υtc )Cy,t
Sc
− EDy,t
Sc
− Sc
H̃y,t = SatSc
Sc,K Sc,Kb
= Sat p + Sat (2.19)
where:
• r Kp is the (“real”) return to physical capital – K p – using the existing technology;
• r Kb is the (“real”) return to bank equity (capital) – K b ;
• r d is the real interest rate paid by banks on bank deposits (Bd ) (see Sect. 2.5 below);
• ws ≥ w is the market wage of skilled labour (with w being the market wage of
unskilled labour);
• wSc ≥ w is the market wage (or salary) for scientists;
• Υtc represents transaction costs (in real output terms) per unit of consumption
expenditure (see below);
• TRLy,ts is income (part of output) invested in training;
• EDy,tSc
is income (part of output) invested in education;
j
• H̃y,t for j ∈ {L, Ls , Sc} is income (part of output) invested in health services; and
• Sat·,· is saving out of market income, with the first superscript referring to the type
of household making the saving and the second to the form in which saving is
taking place.
The opportunity cost of working (to be able to increase the consumption of mar-
ketable consumer products and fund human-capital investment) is the leisure time
that has to be given up, or the loss of time that can be invested in health. To simplify
the analysis, and minimise notational clutter, we assume that monetary transaction
costs are incurred (by the purchaser) whenever there is an exchange of goods and
services, but not when there is a payment of interest or dividends.
2.2.2.6 Motivation: Utility Maximising Individuals
This section is substantially based on Turnovsky (2013) and Turnovsky and Mitra
(2013).
At the beginning of each period t, we have an existing “mature” stock of unskilled
labour, skilled labour, scientists/researchers and entrepreneurs, as well as a new
generation of “young” people who join them.
The former (“mature”) groups each maximise the following type of intertemporal
wellbeing function:
44 2 Stylised Model
∞ θc
1 y c
θ −1 θc −1 θc −1
W0 = t Ω (Et , St , Γt )h(H̃t )
w ES
φ(Y d /Y )Ct θc + (Ctx ) θc
t=0
(1 + ρ)
(2.20)
subject to their respective income-budget constraints represented by Eqs. (2.18)–
(2.19); the evolution of Ht as specified in Eq. (2.16); the time-budget constraint
below:
1 = Ctx + Lt + H̃t (2.21)
and taking (E, S, Γ ES ) as given, compare (2.13).

We focus on one of these “mature groups”, unskilled labour (L), by way of demon-
stration. The decision (choice, control) variables for their intertemporal optimisation
y,L
problem are the time paths of {Ct , Ctx,L , H̃tL , LLt } for t = {0, 1, 2, 3, . . .}; the state
p
variables are {Ht , Kt , Ktb , Btd }. Note that for each individual household, or cluster of
households, aggregate transaction costs (Υt ) and aggregate velocity (∨t ) are param-
eters.
In addition to the time-budget constraint given by Eq. (2.21) above, each cluster
of households, unskilled labour in this case, also face the following income-budget
constraint for each period t:
+ rtd Btd ,L + wt Lt − (1 + Υtc )Ct

b+p,L Kp+b p+b,L K K p,L y,L
Kt+1 = (1 − ξt )Kt + rt b Ktb,L + rt p Kt
+ rtd Btd ,L + wt (1 − Ctx,L − H̃tL )
Kp+b p+b,L K K p,L
= (1 − ξt )Kt + rt b Ktb,L + rt p Kt
y,L
− (1 + Υtc )Ct (2.22)
Thus, in the first instance, our focus is on total saving (to be allocated between
claims on physical capital (machinery) and banking system net worth), and not on
the allocation of saving between the two, which is of second order of importance
for us. The second point to re-emphasise, is that bank deposits (BdL,t ) are held for
transaction purposes, as a means of payment; they are not a saving vehicle:
d ,L
= Btd ,L + ∨ct (Ct+1 − Ct )
y y
Bt+1 (2.23)
and the evolution of Ht is described by Eq. (2.16).

On the optimal time path for these choice (or control) variables, the optimality
conditions comprise a set of efficient allocation conditions. These essentially state
that, under an optimal consumption plan, the marginal wellbeing derived from all
forms of consumption in period t is equal to the marginal wellbeing derived from
consumption in all future periods discounted to the present at the rate β ≡ 1/(1 + ρ).
There is an equality between the optimum rate of intertemporal marginal rate of
substitution in consumption and the marginal rate of transformation in production:
∂u(.) K
Ω w (.)h(.) y,L
= β λt+1
p+b
(1 + Υtc ) + λBt+1
d
∨ct (2.24)
∂Ct
∂u(.) K
Ω w (.)h(.) = βλt+1
p+b
wt = λH̃
t+1 μh (2.25)
∂Ctx,L
and arbitrage conditions:

K Kp+b Kp+b K
βλt+1
p+b
(1 − ξt ) + rt = λt p+b (2.26)
∂h(.)
Ω w (.)u(.) + βλH̃
t+1 (1 − ξ ) = λt
H̃ H̃
(2.27)
∂ H̃t

Kp+b d
β λt+1 rt + λBt+1
d
= λBt d (2.28)
and the income-budget, time-budget and money-holding constraints already given

above. These optimality conditions, written for L, also hold for Ls and Sc, with
suitable adjustments made in Eqs. (2.18)–(2.19).
Combining first-order (necessary) optimality conditions (2.24) and (2.25) yields:
∂u(.) Kp+b
Ω w (.)h(.) y,L
= βλt+1 (1 + Υtc ) − rtd ∨ct + λBt d ∨ct (2.29)
∂Ct
which shows the wedge that transaction costs, adjusted for velocity of money (bank
deposits), puts between the equality of the intertemporal marginal rate of substitution
in consumption with the intertemporal rate of transformation
of consumption. At an
aggregate level, this wedge is equal to 1 − Υtc (∨t )2 /rtd (see Jakab and Kumhof
2015, p. 23).
p,L
The allocation of saving between Ktb,L and Kt is of second-order importance for
our purposes, and will reflect the risk-adjusted expected returns from each option,
and is driven by the usual arbitrage conditions that are a subset of the first-order
conditions of the individuals’ intertemporal utility-maximisation problem (refer back
to Eq. (2.28)).
The new generation of “young” people who are added to the population have an
additional choice to make, namely to join the work force immediately as unskilled
labour L, or take some time to train themselves as skilled labour, or educate them-
selves to be scientists. This new generation tries to maximise the same type of
objective function as in Eq. (2.20), subject to an income-budget constraint given
by Eq. (2.18), and time-budget constraints that replace Eq. (2.21) with Eqs. (2.13)–
(2.15), but also retaining the evolution of H̃t as specified in Eq. (2.16). If they choose
to invest in skilling or education, they need to work a sufficient length of time (as
unskilled labour) to be able to afford initial-period consumption.
This choice reflects the risk-adjusted expected returns from each option, and is
driven by the usual arbitrage conditions that are a subset of the first-order conditions
of the individuals’ intertemporal utility-maximisation problem. The opportunity cost
of the investment in building human capital as skilled labourers or scientists, is
46 2 Stylised Model
the wages, or leisure, or health-investment lost. The return from the investment is
the higher wages earned from skilled labour or the profit earned from converting
science education into profitable investment. These would be captured as variations
to Eqs. (2.24)–(2.28); they affect the supply of skilled and unskilled labour, and of
scientists, from period t + 1 onwards.
It is evident, based on day-to-day observations, that individuals value (they spend

their time and income on) material consumer goods, leisure, good health, education,
and training. This is also confirmed by the domains of wellbeing identified by the
empirical studies reviewed by Smith (2015).
Krause (2014) surveys the evidence that supports the hypothesis that work and
employment make a significant direct contribution to individuals’ wellbeing, in addi-
tion to contributing to wellbeing indirectly by generating income. This direct effect
of work on wellbeing is not captured in our stylised model.
2.3 Production and Wellbeing
2.3.1 General Introduction
We assume that there is a single final (produced) output that is either consumed, or
exported, or saved (and partly invested in machines – i.e., converted into the machines
used to manufacture the final output). Such conversion can take the form of domestic
manufacture, or imports, of the required machinery.
Two types of technology are used in the production of the (single) final output:
(relatively) “clean” (new) technology and (relatively) “dirty” (old) technology. This
language (“clean” vs “dirty” technology) is used for convenience only, to differentiate
between technologies that are more or less skill- and natural-resource intensive. These
two technologies are assumed to be “gross substitutes” in the production of the final
output.
Firms use physical capital (human-made machines), unskilled labour, human cap-
ital (in the form of skilled labour), and natural capital (in the form of exhaustible natu-
ral resources) in the production of the final output.15 All economic activity (including
costs of economic activity) is measured in units of this single output.
The machines (physical capital) used in the production of the final output can
be imported, or manufactured at home. In both cases, they are “funded” through
15 Although the generic production function introduced below does include renewable resources as
well, this additional complication would not add value to our analysis. We implicitly capture the
presence of renewable resources through allowing for “clean” (new) technology in production, in
our stylised model.
2.3 Production and Wellbeing 47
saving (the portion of the final output that is not consumed or exported). When
these machines are manufactured at home, the transformation of the output that is
saved into machines occurs within a single period of time, through a very simple
production process (another instance of gross simplification, to be able to focus on
our core purpose).
Private inputs used in production are deemed to be “gross substitutes”; and so are
the two categories of private goods used in consumption (marketed consumer goods
and leisure), conditional on “good health”. On the other hand, the sources of positive
externalities (i.e., public or social goods) are assumed to be “gross complements”.
2.3.2 Mathematical Model
Comprehensive wealth affects wellbeing both through the direct influences of capital
assets (such as a pristine environment and good health) on individuals’ utility, and
also indirectly through the production of consumable goods and services, which uses
capital assets as inputs, which again enter individuals’ utility functions.
2.3.2.1 Generic Model
Voosholz (2014) provides a very useful summary of alternative specifications for

production functions that link comprehensive wealth to production: Cobb–Douglas
(CD), Constant Elasticity of Substitution (CES), Variable Elasticity of Substitution
(VES), and Leontief. Applications vary not only in terms of the version of production
function that is used, but also viz the choice of capital assets that are represented in
comprehensive wealth. These choices are not inconsequential; they do influence the
outcomes of the analysis.
The general version of the production function used in the endogenous growth
literature takes the following form16 :
Y = F(K p , L, K h , A, R̃, Z̃, K s ) (2.30)
where:
• we have dropped the time index t from the symbols on each side of the equation;
• K p is physical (human-made) capital (or machines) used in production;
• L is (unskilled) labour;
• K h is human capital;
• A represents technological change;
16 Note that E does not enter the production function directly. Below, in our stylised model, E affects
production, through its effect on the health and therefore the productivity of the labour force. L will
be less productive if people’s health is adversely affected because of dirty air (hence our reference
to health-adjusted labour).
48 2 Stylised Model
• R̃ represents the use of the non-renewable (exhaustible) resource (R) and Z̃ the use
of the renewable resource (Z) in production; and
• K s is social capital.
The inclusion of social capital (K s ) in the production function can be motivated
in various ways. From an economic perspective, we have conceptualised social cap-
ital as distinctly networks-based, and as such as a source of various dimensions of
social cohesion (including generalised trust). This in turn reduces transaction costs
by enabling cooperation and information sharing between innovators. Thus one way
in which social capital can have an impact on economic growth is by reducing trans-
action costs and increasing innovation capacity and, through these channels, produc-
tivity growth (see Agénor and Neanidis 2015; Algan and Cahuc 2013; Thompson
2015).
The model is then closed by formulating a set of equations of motion that specify
how each component of comprehensive wealth evolves over time through a set of
influences that lead to their depletion or depreciation, and regeneration.
In what follows, we use the CES specification primarily because of our focus on
emphasising whether key inputs into our production, utility, wellbeing, and exter-
nality functions are complements or substitutes. The CES function provides a very
convenient way of making this differentiation, between complementarity and substi-
tutability.
2.3.2.2 Stylised Model
The stylised model is primarily based on the one used by Acemoglu et al. (2012),
but also influenced by the discussion in Acemoglu et al. (2016). One of the main
differences is that our stylised model represents a small open economy, whereas the
Acemoglu et al. papers work with a closed economy.
There is a single, consumable and internationally tradeable, final good (Y ), pro-
duced competitively using two types of technologies, “clean” (or new) and “dirty”
(or old) according to the following aggregate production function17 :
⎡ ⎤ θ θ−1
y
1 θyθ−1
y
1 θyθ−1
y
y
Yt = ⎣ (Ls H̃ )1−α
t A1−α α
cit mcit di + R̃αt 2 (LH̃ )1−α
t
1
A1−α 1 α1
dit mdit di
⎦ (2.31)
0 0
These different technologies are imbedded in the two types of machines used in
producing the single output. The symbols have the following meanings:
• θy ∈ (0, +∞) in Eq. (2.31) represents the elasticity of substitution (in production)
between the two technologies;
17 To repeat, the term “dirty technology” is not used in a derogatory sense, but simply as a convenient
means of differentiating between two types of technology with significantly different impacts on
the accumulation of human capital and the preservation of natural capital.
• R̃ refers to the quantity of the exhaustible natural resources R used with the “dirty
technology”;
• mdi and mci refer to the continuum (indexed by i) of machines (physical capital)
using the “dirty” and “clean” technologies, respectively; and
• Aci and Adi correspond to the productivity (or “quality”) of these machines.
It is assumed that the (health-adjusted) labour required to operate the machines
using the “dirty technology” (LH ) is “unskilled” (or comparatively less skilled),
whereas the (health-adjusted) labour required to operate the “clean technology”
(Ls H ) is “skilled” (or comparatively more skilled). Note that labour is not indexed
by i; Ls and L can operate all versions of mc and md , respectively, equally efficiently.
The two technologies are referred as (gross) substitutes when θy > 1 and (gross)
complements when θy < 1. Reflecting the assumption, and expectation, that, over
time, successful “clean technologies” will substitute for “dirty technologies”, in our
stylised model we assume that θy > 1. Throughout we ignore the Cobb–Douglas
case of θy = 1. α, α1 , α2 ∈ (0, 1), α1 + α2 = α.
The evolution of the exhaustible resource (R) is given by:
Rt+1 = Rt − R̃t (2.32)
The per unit extraction cost for the exhaustible resource, expressed in units of the
final output (Yt ), is c(Rt ), where c is a non-increasing function of R.
2.3.2.3 Motivation: Producers of the Final Product
The final-goods sector of the economy consists of a large number of competitive firms
that use the production technology given by Eq. (2.31) to produce the final good –
Y . These firms are price-takers in world markets. The price of the product they sell
is governed by Eq. (2.59); in other words, the higher the “quality” of the product
(reflecting the proportion of clean technology used in producing it), the higher the
price-premium these producers receive in world markets.
Firms in this sector have to decide how much of each type of labour (skilled and
unskilled), and of each type of capital or machine (“clean” and “dirty”), as well as
the exhaustible resource (R̃) they use in producing the output.
They do this by solving the following profit-maximisation problem,18 taking the
prices/costs of each input as given:

y y m y
max Πt = Pt Yt − wts Lst + wt Lt + ptmc mct + pt d mdt (1 + Υt + rtl ) − c(R)R̃t
{Lst ,Lt ,mct ,mdt ,R̃t }
t = 0, 1, 2 . . . (2.33)
where Yt is given by Eq. (2.31). The remaining terms have the following meanings:
18 The
labor-market-clearing conditions (2.67) and (2.68) in Sect. 2.7.2 will ensure that the labour
maximisants of (2.33) are not greater than the available stocks, governed by (2.13) and (2.14).
50 2 Stylised Model
y
Πt is the profit level from the production of the final product in period t;
y
Pt refers to the exogenous market price of the final product, determined in
international markets;
y
Υt is an average measure of monetary transactions costs associated with the
purchase of labour services and machinery in the production of the final
output;
rtl is the interest rate producers need to pay on the loans they obtain from banks
to pre-pay for labour services and machinery at the beginning of each period,
before they can produce and sell their produce at the end of the period; and,
as a reminder,
c(Rt ) is the per unit extraction cost for the exhaustible resource, expressed in units
of the final output (Yt ), where
c(·) is a non-increasing function of R, reflecting economies of scale.
Note that we have suppressed the index i, as that additional level of detail does not
add any new insights to the analysis that follows in this section. In other words, we
are working with average productivities (Ac , Ad ) and a “typical” or “representative”
machine within each sector (“clean” and “dirty”).
Under suitable concavity assumptions for the profit function, the necessary and
sufficient conditions characterising the solution of this problem state that, on the
optimal (profit-maximising) time path, these firms hire or rent all inputs until the
value of the marginal product of the input equals the price or rental price of the input,
for all inputs (Lst , Lt , mct , mdt , R̃t ):
− θ1 α
∂Yt wts y (1−α) Ytc y mct
s = y (1 + Υt + rt ) = (1 − α)Act
l
(2.34)
∂Lt Pt Yt Lct
d − θ1y d α1
∂Yt wt Yt mt
= y (1 + Υt + rtl ) = (1 − α1 )A(1−α 1 ) α2
y
R̃ t (2.35)
∂Lt Pt dt
Yt Lt
mc c − θ1 s (1−α)
∂Yt pt y (1−α) Yt
y Lt
c = y (1 + Υt + rt ) = αAct
l
(2.36)
∂mt Pt Yt mct
d − θ1y (1−α1 )
∂Yt ptmd y (1−α1 ) α2 Yt Lt
= y (1 + Υ t + r l
t ) = α A
1 dt R̃ t (2.37)
∂mdt Pt Yt mdt
d − θ1y
∂Yt c(Rt ) (1−α1 ) α2 −1 Yt
= y = α2 Adt R̃t L1−α
t
1
mαdt1 (2.38)
∂ R̃t P t Yt
where, in all cases for Eqs. (2.34)–(2.38), t = 0, 1, 2, . . . , the five equations (first-
order conditions) represent the demand functions for the five inputs into the produc-
tion of the final good (Lst , Lt , mct , mdt , R̃t ), respectively. In what follows, we use the
notation: j − θ1
y
j Yt
yt ≡ ; j ∈ {c, d } . (2.39)
Yt
2.3.2.4 Motivation: Machine Producers or Importers
The machines using both technologies are supplied (either manufactured or imported)
by monopolistically competitive firms. These represent the intermediate-goods sec-
tors. They produce (or import) the machines (the capital goods) that are sold to the
final-goods sector. These firms gain their monopoly power by purchasing the design
for a specific capital good from the research sector (the entrepreneurs funding the
scientists). Owing to patent protection, only one firm manufactures (or imports) each
machine.
Like the final-goods producers, these machine producers are also profit-
maximisers. The profit-maximisation problem of the “representative” producer of
machines, at time t (t = 0, 1, 2, . . .), working with technology j ∈ {c, d }, can be
written as:
m m j
max
m
πt j = (pt j − ψ)mt (2.40)
j
{pt }
subject to the demand functions for machines derived from Eqs. (2.36) and (2.37):
j
1−α
1
j αyt j j
mt = mj Lt At (2.41)
pt
y
with Pt = 1. The unit cost, inclusive of transaction costs (measured in units of the
final output), of producing (or importing) any machine is ψ.
It can be shown (see Acemoglu et al. 2012 or Jones and Vollrath 2018) that the
profit-maximising price is a constant mark-up over marginal cost:
m ψ
pt j = (2.42)
α
so that all machines sell at the same price. Substituting back into Eq. (2.41) yields
the demand for the “representative” machine using technology j ∈ {c, d }:
j
1−α
1
j α 2 yt j j
mt = Lt At (2.43)
ψ
Thus the (temporary) equilibrium profits of the “representative” machine produc-

ers of machine type j ∈ {c, d } can be written as follows:
2 j 1−α
1
m 1 α yt j j
πt j =ψ −1 Lt At (2.44)
α ψ
Combining the profit-maximisation solution of the final-good producers with the

profit-maximisation problem of the machine producers yields the following important
52 2 Stylised Model
result for our purposes, regarding relative equilibrium employment of the two types
of labour working with the “clean” and “dirty” technologies (see Acemoglu et al.
2012):
(θy −1)
−ϕ
c(Rt )α2 α 2α (1 + Υt + rtl )
y
Lct Act
= −ϕ (2.45)
Lt ψ α2 α12α1 α2α2 Adt 1
where ϕ1 ≡ (1 − α1 )(1 − θy ). The higher the extraction cost c(Rt ), the higher the
amount of labour allocated to “clean” technology when θy > 1. The relative costs
of funding for clean and dirty technology will also have a significant bearing on the
allocation of labour between “clean” and “dirty” technology processes.
The key empirical issue here is the degree of substitutability of “clean” and “dirty”
technologies in the production of goods (in our model, the single product). This is
measured by the size of the elasticity of substitution, with the critical criterion being
θy > 1.
For a review of the endogenous growth literature in general, with emphasis on
directed technical change towards “clean” technologies, including supporting empir-
ical literature, see Popp et al. (2010).
See Saam et al. (2014), in particular, on supporting evidence for θy > 1.
As already stated, the policy implication of this is that, using an appropriate mix
of policies, to be discussed in detail in Chap. 3, with θy > 1, the economy can be
steered to a sustainable growth path.
Referring back to the weak versus strong sustainability discussion of Chap. 1,
Drupp (2016) empirically explores the elasticity of substitution between categories
of capital assets (ecosystem services and manufactured goods in particular). He finds
that, while in the short- to medium term this elasticity is quite high, over the very long
term it diminishes quite significantly, with ecosystem services declining towards their
subsistence levels. This reinforces the wisdom of the practical and cautious approach
promoted in Chap. 1, towards treating natural, social, and economic ecosystems as
complements (i.e., “strong sustainability”), while allowing substitutability at the
margin within different categories of capital (i.e., “weak sustainability”), in the gen-
eration of human wellbeing.
2.4 Ideas and Technology
Ideas and technology are a key input into production, through physical and human
capital. (Refer to Fig. 2.1.)
2.4 Ideas and Technology 53
The ultimate source of progress in our individual and communal wellbeing is the
generation or importation, as well as liberation and conversion, of ideas and knowl-
edge into goods and services that individuals value, supported by institutions and
ethical values. The very significant progress achieved over the last few centuries has
been based on the widening adoption of two ideas, the new economic idea of liberty
for ordinary people and the new social idea of dignity for them (a dignity that is not
based on social status) (see McCloskey 2014, p. 7; McCloskey 2016).19
A very similar thesis is presented by Phelps (2013), who emphasises the critical
role of values, attitudes, and beliefs (i.e., culture) in encouraging new ideas, the role
of institutions in promoting them, and the role of population density in underpinning
their impact: “The modern economies emerging in the 19th century were a stunning
success in both non-material and material dimensions: intellectual engagement and
personal development as well as sustained economic growth and an inbuilt tendency
towards inclusion. This depended on the rise of a new force: economic dynamism.
And what sparked this dynamism was a new economic culture. Its necessary nutrients
were representative democracy and a cultural revolution originating in Renaissance
humanism, Baroque vitalism, and Enlightenment modernism. Representative democ-
racy ensured property rights but also stimulated self-reliance and social engagement.
Altruism, vitalism, and modernism caused people to reach out to the world and
find meaning through innovative activity. The resulting culture and the economic
institutions it led to provided people with the urge and capabilities to innovate. Ade-
quate numbers of people was the last of the necessary, but not sufficient, conditions”
(p. 109).
In the stylised model, economic growth is a source of improvements in general
material wellbeing. A very important source of the growth potential of an economy is
knowledge-based innovation – knowledge generated at home or abroad. This knowl-
edge is embodied in human and physical capital assets in the form of productivity
improvements and converted, through the production process and market testing,
into goods that consumers want.
The growth in the number of (domestically grown or immigrant) scientists and
engineers, their success rate in innovating, and the productivity impact of that inno-
vation are the key drivers of economic growth in our model.
The research sector, which is comprised of scientists and engineers, generates new
ideas and knowledge. Some of these, if adopted because they are commercially lucra-
tive, lead to productivity improvements in machines using the “dirty” technology,
whereas others lead to further improvements in the “clean” technology. Entrepreneurs
and engineers work collaboratively to give effect to these improvements, comple-
mented and supported by economic infrastructure provided or coordinated through
19 “The modern world was made by a revolution in ethical judgments about commercial virtues and
vices, in particular by an up-valuation of market-tested betterment – […] the enrichment of ordinary

people, by ordinary people, for ordinary people” (McCloskey 2014, pp. 5–6).
54 2 Stylised Model
collective action (i.e., the government). This is one channel through which new ideas
and knowledge may lead to productivity improvements.
Other ideas and knowledge lead to productivity improvements through enhance-
ments in human capital. Scientists and engineers are part of our human capital.
A country’s scientific base (and stock of ideas and knowledge) can be enhanced
through investment in human capital or augmented by inward migration. Thus ideas
and knowledge (or technology) can be generated within the country or imported. The
vehicles for importing these can be in the form of human capital (e.g., scientists) or
physical capital (e.g., machines).
Ideas and knowledge need not take the form of inventions. They may also take the
form of smart applications of existing or new technology, developed domestically or
imported from overseas, combined and adapted to suit domestic circumstances (see
Bruton 1998).
Ideas and knowledge are necessary but not sufficient for promoting economic
growth and overall wellbeing on a sustained and equitable basis. What is required is
an institutional framework that incentivises the generation of ideas, as well as their
connection to entrepreneurial effort. The consequent “creative destruction”, fuelled
by competition, is what provides the dynamics of some of the endogenous growth
models.20 This is especially effective in promoting growth in economies operating
with frontier technologies (see Aghion et al. 2018).
Three complementary pillars are required:
• the accumulation, investment, and upgrading of ideas and knowledge; comple-
mented by,
• implementation mechanisms and incentive structures that enable and encourage
knowledge to be exploited such that growth and societal prosperity are achieved;
in turn enabled and supported by,
• the relevant economic infrastructure (see Braunerhjelm and Henrekson 2016).
Skilled labour is required to operate the machines that embody the new technology
and, through upskilling, individual labourers can increase their material rewards.
Further material reward is provided to the producers of the final good using cleaner
technology, through the price premium received in international markets.
In line with the literature on endogenous technical change (see Jones and Vollrath
2018), the production side of the economy consists of three sectors: a final-goods
sector, an intermediate-goods sector, and a research sector.
20 What is potentially ignored here is what Antonelli (2016) describes as the “evolutionary complex-
ity approach” to endogenous innovation – based on the interplay between evolution, complexity,
and Schumpeter (1947)’s idea of creative responses to significant gaps between actual and expected
outcomes. We will return to the implications of complexity and radical uncertainty for public policy
in Chap. 5.
The single final output (Y ) is consumable and internationally tradeable. It is pro-

duced using the process described by the production function in Eq. (2.31). Although
our small-open economy is a price-taker in the global market for this good (P̄), there
is an increasing price premium on this product as the weight of “clean technology”
increases in its production (see Sect. 2.6 below).
The machines using both technologies are supplied by monopolistically compet-
itive firms. These machines may be manufactured domestically or imported from
overseas. This represents the intermediate-goods sectors.
The research sector, which is comprised of scientists, generates new ideas. These
scientists may be domestically-educated or may be immigrants from overseas. Some
of these ideas lead to productivity improvements in machines using the “dirty tech-
nology”, whereas others lead to further productivity improvements in the application
of “clean technology”.
This productivity term is referred to as A in Eq. (2.31), and we define the average
productivity of process, or machine type, j ∈ {c, d } as:
1
j j
At ≡ Ait di (2.46)
0
which implies that Adt corresponds to “dirty technologies” whereas Act relates to
“clean technologies”. For concreteness, “clean innovation”, that is increasing Act , can
be thought of as reducing the pollution, or the exhaustible natural resource intensity,
of the overall production process.
At the beginning of each period, each scientist decides whether to direct his/her
research to improving the quality of machines using “clean” or “dirty” technology.
She/he is then randomly allocated to at most one machine (without any congestion;
so that each machine is also allocated to at most one scientist). A successful sci-
entist, who has invented a better version of machine i using technology j ∈ {c, d },
obtains a one-period patent and becomes the entrepreneur for the current period in
the production or importation of machine i. In cases where innovation is not suc-
cessful, monopoly rights are allocated randomly to an entrepreneur drawn from the
pool of potential entrepreneurs, who then uses the old technology. This “innova-
tion possibilities frontier” where scientists can target only a technology or machine
type (rather than a specific machine) ensures that scientists are allocated across the
different machines using a particular technology.
j
At evolves over time according to the following difference equation (see
Acemoglu et al. 2012; Sunaga et al. 2015)21 :
j j j j
At = At−1 + μηj (Sct )Āt (2.47)
where:
21 Sunaga et al. (2015) in turn acknowledges his indebtedness to Acemoglu et al. (2006), Aghion
et al. (2005), Michalopoulos et al. (2009).
56 2 Stylised Model
• Scj references the group of scientists (or engineers) working with technology j,
• ηj ∈ (0, 1) is the probability function ηj : R+ → [0, 1] (which is a concave and
increasing function of Scj ) which represents the likelihood of successful innova-
tion,
• (1 + μ) (where μ > 0) is the factor by which innovation increases the quality (or
productivity) of a domestically-used machine when the scientists are successful
in innovating; and
j
• Āt represents the world technology frontier.
It is assumed that the world technology frontier is expanding at the exogenous
growth rate (ḡtech ):
j j
Āt = (1 + ḡtech )Āt−1 (2.48)
This does not preclude domestic scientists, or migrant innovators to the small
open economy, from innovating. Successful innovation, wherever it happens and
whomever it is done by, pushes out the world-technology frontier through the machin-
ery in which it is embedded. The efficiency and effectiveness with which such inno-
vation is reflected in domestic production is captured by the parameter μ.
It follows from Eqs. (2.47) and (2.48) that the number of scientists (and engineers)
working on innovation in the small open economy, which increases the probability
of innovation, and the effectiveness and efficiency with which such innovation is
incorporated into domestic production processes (μ), are the two levers through
which the production processes used in the small open economy can manage to stay
close to the world technology frontier.
2.4.2.1 Motivation: Entrepreneurs, Scientists, Technology,

and Productivity Growth
As we explained in Sect. 2.4 above, the research sector, which is comprised of scien-
tists, generates new ideas. Some of these ideas lead to productivity improvements in
machines using the “dirty” technology, whereas others lead to further improvements
in the use of “clean” technology. One of the possible outcomes of the typical indi-
vidual’s intertemporal utility maximisation problem is the investment of time that
the individual makes in education, for acquiring skills towards becoming a scien-
tist/researcher – EDt (this is the income invested in education as seen in Eq. (2.19)).
In the stylised model, economic growth is driven by productivity growth. The
evolution of productivity growth is described by Eqs. (2.47) and (2.48). It is clear
from those equations that the growth in the number of scientists, their success rate
in innovating, and the productivity impact of that innovation are the key drivers of
productivity growth. The growth in population (assisted by the growth of inward
migration), the composition of migrants, and the decision of individuals on whether
to pursue science as a career or not, are the key influences on the growth of scientists.
Knowledge provides the engine of growth, and the application of knowledge as
between “clean” and “dirty” technologies determines the “quality” of growth. The
allocation of scientists and engineers between the “dirty” and “clean” technologies, or
equivalently the choice by a scientist or engineer as to whether they apply their skills
towards improving the quality of the “dirty” or “clean” technologies or machines,
reflects the relative expected profit of this choice.
Entrepreneurs intermediate between financiers (banks in our model) and scientists.
They borrow from the banks to fund the scientists before the research is undertaken
and yields results, carry the risks of scientific failure, and profit from scientific success
that results in innovation.
To determine the profitability of research, we need to start with the profitability
of the machine producers for each of the two technologies, as given by Eq. (2.44).
We then note that the profits of these machine producers are completely extracted
by the entrepreneurs, whose research funding leads to a new-design machine using
each technology – i.e., to an actual invention. Since the probability of a successful
innovation in technology j ∈ {c, d } is ηj ∈ (0, 1), and innovation increases the quality
(or productivity) of a machine by a factor (1 + μ), it follows that the expected profit
j
Πt defined in Eq. (2.33) for an entrepreneur funding research with technology j at
time t is:
2 j 1−α
1
j 1 α yt j j j
Πt = ψηj (1 + μ) −1 Lt At − (1 + rtl )wtSc Sct (2.49)
α ψ + rtl
The entrepreneur needs to hire scientists to do research and borrow the funds
required to pay their wages (or salaries), hence the inclusion of the last term on the
right hand side.
It is the relative profitability of doing research with the two technologies that will
drive where innovation is directed. This in turn will have a significant impact on the
“quality” and sustainability of economic growth.
It can be shown (see Acemoglu et al. 2012) that, in equilibrium, the relative
profitability of undertaking research in sector c relative to sector d is given by:
Πtc
= (2.50)
Πtd

ηc c(Rt )(1 + Υt + rtl )α2 (θy −1) (1 + μηc Scct )−ϕ−1 (Act−1 )−ϕ − (1 + rtl )wtSc Sctc
y
κ
ηd (1 + μηd Scdt )−ϕ1 −1 (Adt−1 )−ϕ1 − (1 + rtl )wtSc Sctd
(θy −1)
(1 − α)α α 2α
where κ =
(1 − α1 )α1
(1+α2 −α1 )/(1−α1 )
(ψ)α2 α12α1 α2α2
What Eq. (2.50) says is that, so long as c(Rt ) is decreasing in Rt , and the two
technologies are gross substitutes (i.e., θy > 1), as the exhaustible resource gets
depleted the incentives to direct innovations towards the clean technology increases.
58 2 Stylised Model
Again, also note the effect of funding costs – rtl . So the market mechanism is working
in the stylised model; whether it needs support to work more effectively and efficiently
is a matter we address later in Sect. 2.7 below.
For a detailed theoretical and empirical (historical) presentation of the critical role of
institutions in promoting sustained prosperity, see Acemoglu and Robinson (2013).
See Bitzer et al. (2015), Bosetti et al. (2015), Czaika et al. (2015), and d’Albis et al.
(2016) for recent work on the links between immigration, innovation, and economic
growth. These works also have very useful summaries of related previous literature.
For evidence supporting the positive impacts of smart applications of existing or
new technology, developed domestically or imported from overseas, combined and
adapted to suit domestic circumstances, see Porter and Heppelmann (2014).
2.5 Finance - Money, Credit, and Banking
Ideas, technology, and machinery need to be funded before (the successful ones) can
be put into use profitably.
As Blanchard (2015) notes, on reflecting about what we learned from the Global
Financial Crisis, “mainstream macroeconomics had taken the role of finance for
granted” (p. 2). So do mainstream economic growth theories. The (hidden) financial
system does what it needs to do in the background, ensuring that the economy is
always functioning on the production possibilities frontier – or, in this case, the
technology frontier.
Surprisingly, the same criticism can be extended to endogenous economic growth
theories, some of which are based on Joseph Schumpeter’s economic development
framework where “creative destruction” plays a critical role (see Schumpeter 1982).
In these models, innovation is a key influence on long-run economic growth and
development. Innovation must be financed before it can generate revenue. Yet,
in much of this literature, money and credit are in the background. “The banker,
Schumpeter’s ephor of capitalism, plays an important role in helping to get innova-
tions financed. The idea is that innovations cannot be self-financing because they do
not yet exist. The banking ephors need to provide finance first, to let the innovation
proceed. The [successful] innovations then are rewarded with profits (allowing the
finance to be repaid) even as they destroy competitors” Mazzucato and Wray (2015,
p. 12).
2.5 Finance - Money, Credit, and Banking 59
But what exactly is it that the financial system needs to do to ensure these out-
comes – and indeed (in our broader wellbeing context) support the expansion of
the wellbeing frontier? And is there a role for public policy to achieve it, instead of
assuming that it will automatically happen?
As we emphasised before, the key to sustaining and sharing wider intergener-
ational wellbeing is the accumulation, quality, allocation (across activities), distri-
bution (across society and generations), and preservation of the capital assets that
comprise comprehensive wealth.
The critical functions performed by the financial system in underpinning and
promoting this are (see Merton and Bodie 1995; Laeven et al. 2015; Levine 2005;
Jakab and Kumhof 2015):
• the provision of a common medium of exchange,
• effective and efficient payments infrastructure and services (broadly defined),
• effective and efficient financial intermediation (efficiency to be defined in terms
of the wider intergenerational wellbeing function),
• effective and efficient risk sharing, and
• credit creation (as distinct from, and in addition to, the allocation of existing funds)
through the banking system.
A complementary perspective and classification on the interaction of financial
intermediation and economic growth is provided by Yusifzada and Mammadova
(2015) where four aspects of financial development and intermediation are empha-
sised as critical for economic growth and development: depth, widespread access to
finance, efficiency of financial intermediation, and stability of the financial interme-
diaries.
In the traditional literature, the two clusters of “frictions” that frustrate the achieve-
ment of these outcomes automatically, through market forces left to their own devices,
are asymmetric information, and enforcement-related and other transaction costs.
These lead to a whole series of imperfections, including financing constraints, incom-
plete risk-sharing, liquidity shortages, and poor market discipline through which the
extension of the wellbeing frontier is constrained.
In this context, the existence and well-functioning of banks, as both financial
intermediaries and money creators, is critical.22
We are dealing with a small and open “credit economy”, where money and credit
creation are endogenous through the banking system, and both the creation and
allocation of credit to fund wellbeing-enhancing investments is critical (see Fontana
and Sawyer 2015; Hermele et al. 2015; Jakab and Kumhof 2015; Kratena 2015;
Stiglitz 2015). Thus the aggregate growth of credit, as well as its allocation between
22 In our model, “banks” are both financial intermediaries between savers and investors, and
financiers of loans to investors: “In the real world, the key function of banks is the provision
of financing, or the creation of new monetary purchasing power through loans […] The bank there-
fore creates its own funding, deposits, in the act of lending” (Jakab and Kumhof 2015, p. 3). While,
for simplicity, we conceptualise “banks” as playing both roles, in fact the latter is the distinctive
role of banks among all financial intermediaries.
60 2 Stylised Model
“clean” and “dirty” technologies, affect both the quality and the sustainability of the
potential economic growth rate.
Although we are dealing with a small open credit economy, where credit-created
money serves as the medium of exchange, we do not model inflation or changes
in the nominal exchange rate directly but work with the real exchange rate instead.
The distortionary effects of inflation on wellbeing work through misallocations of
(consumption and production) resources across industries and time, as well as through
the distribution of wealth. The stylised model has channels through which all these
effects can be captured and transmitted.
The lending capacity of banks is positively affected by their revenues, and nega-
tively by their bad loans, monitoring costs, and required capital ratios.
There is however, a critical additional nexus between finance and sustained devel-
opment, which is not captured by any of the dimensions mentioned above, and yet
provides not only a critical bridge between long-term development and finance, but
also a distinctive role for public policy. Innovation-supporting investment needs a
long-term commitment, supported by “patient finance”, as well as an appetite for
risk-taking in the face of “radical uncertainty”. This is especially the case when we
are talking about transformative infrastructure investments, such as those supporting
“green growth”, nanotechnology, biotechnology, and so on. These are the circum-
stances where the state has to lead (and has historically done so), and only then does
the private sector (including venture capital) follow (see Mazzucato and Wray 2015;
Mazzucato and Perez 2014; Mazzucato 2013a, b).
In a similar vein, Campiglio (2016) adds, “In order to carry out their activities,
low-carbon firms necessitate credit. Under certain economic conditions, of which
the post financial crisis period represents the most recent realization, banks may lack
the confidence to create new credit even in the presence of right prices and prof-
itable investments. This credit market failure, together with the deep uncertainties
surrounding the future implementation of a carbon price, makes the case for consid-
ering a wider portfolio of policies. Examples include green differentiated reserve and
capital requirements, modifying the risk weights for computing capital requirements
in favour of low- carbon assets and other quantitative macroprudential policies aimed
at easing lending conditions for low-carbon firms” (p. 227).
In passing, we should also emphasise that there are other ways of modelling
the interface between economic growth, the environment, and finance. An emerg-
ing complementary literature, covering the same ground, can be generically labeled
“post-Keynesian ecological macro models” (see Naqvi and Engelbert 2017; Hardt
and O’Neill 2017 for very recent contributions and reviews of the related literature).
Among others, one of the more distinctive aspects of this particular literature is the
emphasis on the demand-side of the economy, and its linkages with finance and its
influence on economic growth. What is comforting, from our perspective, is that the
policy implications of this strand of literature is qualitatively not that different from
what we have outlined above in this section.
In our stylised model, the government funds all its expenditure, operational as
well as investment, with a tax on production. Of course the government, just like the
private sector, can pre-fund its investment expenditure through borrowing and (in
the case of the government) through “printing money” as well. Since our focus is on
the distinctive role of government in wellbeing-enhancing infrastructure investment,
rather than how it funds it, we deliberately keep the model as simple as possible.
2.5.2.1 The Financial Sector
A stylised model cannot capture all the richness of the qualitative discussion pre-
sented in the previous section. Our primary focus is on the interface between “banks”
(to be interpreted as the financial sector at large) and entrepreneurs. The role of gov-
ernment in supporting infrastructure investments (broadly defined) is treated sepa-
rately in a separate section below in this chapter.
Money (as the common medium of exchange)23 and the banking system exist as
essential components of financial infrastructure, supported by appropriate govern-
ment policies which are funded through a tax on final-good production.
In the real world, by far the main component of money (used as the means of
payment in all transactions) is the deposits of the banking system. Banks create new
deposits through lending.
Like the producers of intermediate goods (machines in our model), banks too
are profit-maximising, monopolistically competitive firms. Their partial-monopoly
status reflects the regulatory requirements (imposed by government – which includes
the central bank) under which they are allowed to register and operate.
One of these regulatory conditions is the capital requirements imposed by the
government. Banks have their own balance sheets and net worth. Loans comprise
their assets, and demand deposits and issued capital comprise their liabilities.
In our model, banks make loans to the entrepreneur-scientists. Then, those who
succeed in innovating either manufacture or import the machines in which the suc-
cessful research is embodied. Because only a certain percentage (ηj ∈ (0, 1), see
Eq. (2.47)) of the research is successful (i.e., leads to innovation), banks are exposed
to credit (or default) risk. In addition, banks incur monitoring costs towards the
lowering of credit risk.
While all agents use money for transaction purposes, the ultimate (willing) holders
of money are the households. Their underlying motive for doing so is to save on
transaction costs; bank deposits are not a saving vehicle for households. Bank capital
or equity is one of the saving vehicles for the household sector, the other one being
claims on private physical capital – or machines in our model.
Banks lend to businesses (entrepreneurs) at a nominal interest rate (il ) and pay
a nominal deposit rate (id ) to depositors. These rates are set at a margin above the
international wholesale nominal interest rate (ı̄) (which can be conceptualised as a
23 Thissection is substantially based on Jakab and Kumhof (2015), and represents a simplified
version of their model to suit our purposes. It also borrows ideas from Sunaga et al. (2015).
62 2 Stylised Model
mark up over cost, as in any monopolistically competitive industry). The lending rate
also includes a lending-risk premium:
itd = (1 + ϕ)ı̄ t (2.51)

itl = itd + ltrisk (2.52)
where, 0 < ϕ < 1 and l risk > 0 (and may vary across sectors).
The corresponding “real” (inflation-adjusted) interest rates are denoted by r̄, r d , r l
where in each case the adjustment factor is πt = Pt /Pt−1, and P is the output (GDP)
deflator. Thus, r̄t = ı̄ t−1 /πt , rtd = it−1
d
/πt , rtl = it−1
l
/πt .
The aggregate banking system balance sheet, in nominal terms, is given by:
BtRFA + Btl = Btd + Ktb (2.53)
where:
• BtRFA is bank holdings of foreign risk-free assets (held for liquidity purposes),
• Bl (= Bcl + Bdl ) is bank loans to domestic entrepreneurs (for investing in “clean”
and “dirty” technology),
• Bd is bank deposits held by domestic households, and
• K b is bank capital (or net worth) held by domestic households or overseas investors.
The evolution of banks’ aggregate net worth (Ktb ) is given by:
Ktb = Kt−1
b
+ (2.54)

l,c
(1 − τ ) (itl (ηc Bt−1 l,d
+ ηd Bt−1 ) + ı̄ t BtRFA − itd Btd − monc(Btl,c + Btl,d ) − rtKb Ktb
where:
• τ is the uniform tax rate on gross (before tax) income,
• monc refers to the monitoring costs incurred by banks,24 and
• rtKb Ktb are bank dividends paid out to investors in K b .
K b is one of the two savings vehicles available for households, the second
being K p . The relative demands for these savings vehicles will be determined by
the arbitrage conditions that represent a subset of the first-order conditions of an
intertemporal optimisation model for consumer-households (see e.g., Turnovsky
2013; Turnovsky and Mitra 2013).
24 There is also another cost that we should be allowing for, namely the penalty banks have to pay if
they do not meet their minimum capital requirements (see Jakab and Kumhof 2015), but that level
of detail is not required for our purposes.
2.5.2.2 Motivation: Banks and Households
Individual banks choose the volume of loans made to the entrepreneurs investing in
“clean” and “dirty” technology to maximise:
∞
1
max K0b ≡ · (2.55)
{Btc,l ,Btd ,l } t=0
(1 + ı̄)t

(1 − τ ) itl (ηc Btc,l + ηd Btd ,l ) + ı̄ t BtRFA − itd Btd − monc(Btc,l + Btd ,l )
Under suitable concavity assumptions for the profit function, the necessary and
sufficient conditions characterising the solution of this problem state that, on the
optimal (net worth maximising) time path, the allocation of loans will reflect the
following condition:
itl ηc (Scct ) = monc = itl ηdl (Scdt ) (2.56)
Thus, the key determinant of the allocation of Btl between Btc,l and Btd ,l will be the
relative probability of successful innovation in the two research sectors using the
“clean” and “dirty” technologies.
The demand (to hold) bank deposits, by households, is underpinned by the fol-
lowing transaction-cost function:
U2
Υ t = U 1 ∨t + − 2 U1 U2 (2.57)
∨t
where Υ is transaction cost (in units of the final output) per unit of expenditure, U1 and
U2 are parameters, and ∨ is transaction velocity (= total transaction value/Btd ).25
The effective market price of any good that is purchased using money (i.e., deposits)
is (1 + Υt + Υt ∨t )P, for all prices (P), where Υt (the partial derivative of Υt with
respect to ∨t ) < 0 and represents the impact of the effect of an increase in the volume
of monetary purchasing power in circulation on transaction costs.
Our primary concern is the aggregate growth and allocation (i.e., “quality”) of credit.
In the latter context, we are particularly interested in the allocation of credit in funding
investments in “clean” and “dirty” technologies, which in turn affects both the quality
and the sustainability of the potential economic growth rate.
There is a substantive theoretical and empirical literature on the relationship
between finance and “financialisation” on the one hand, and economic development
and economic growth on the other. Recent surveys of this literature have been pro-
vided by, among others, Popov (2017), Benczur et al. (2017), Panizza et al. (2017),
25 See Jakab and Kumhof (2015) for an articulation of the micro-foundations of Eq. (2.57).
64 2 Stylised Model
and Yusifzada and Mammadova (2015). As Benczur et al. (2017) state, “There is
growing evidence that the impact of financial development [especially “financiali-
sation” or financial deepening] on economic growth might be non-linear and hump-
shaped, exhibiting a turning point” (p. 3). This latter result is partly associated with
the impact of excessive financial deepening on the misallocation of resources and
creation of financial fragility, a point strongly emphasised by Hyman Minsky (see,
for example, Hyman 2008; Mazzucato and Wray 2015).
Broadening this discussion, Popov (2017) summarises the key findings of his
survey as follows: “First, the bulk of the historical evidence indicates that on average,
financial development has a positive, monotonic effect on economic growth. […]
Second, more recent studies have cast doubt on the proposition that the effect of
financial development on economic growth is not just positive, but also monotonic.
A number of papers have pointed to non-linearities in the finance-and-growth nexus.
[…] Third, while much of the earlier literature suggested that a country’s financial
structure – i.e., the mix of financial intermediaries operating in the economy – has no
independent effect on economic growth, more recent evidence has challenged this
view. As per capita income rises, countries’ financial structures tend to move towards
non-bank financing. […] Fourth, various aspects of financial development can have
a non-negligible social impact, beyond their first-order effect on economic growth.
For example, recent evidence has suggested that in particular banking competition is
associated with a decline in both income and of gender inequality, as the increase in
economic activity associated with the deregulation of previously parochial banking
markets creates economic opportunities for agents that in the past found it difficult
to join the formal labor market” (pp. 2–3).
On evidence relating to both under-investment in long-term-horizon infrastructure
investments, as well as the leadership and directive roles played by government
in transformative infrastructure investments in biotechnology and other industries
over the centuries, see Mazzucato and Semieniuk (2017), Mazzucato (2013a, 2016),
Mazzucato and Wray (2015).
2.6 International Connections
The small open economy is connected to the rest of the world through flows of trade,
finance, machinery, people, and ideas.
The small open economy of the stylised model is connected to the rest of the world
through trade in goods and services, as well as flows of physical and financial cap-
ital, people, and ideas or new technology. The vehicles for importing new ideas,
2.6 International Connections 65
knowledge, and technology from overseas can come in the form of human capital
(e.g., scientists or engineers) or physical capital (e.g., machines).
The final good (the single representative of “goods and services” in the stylised
model) can be exported, consumed at home, or saved (= invested). Physical capital
(i.e., machines used in the production of the final output), can be manufactured
at home or imported. Financial (equity) capital can be obtained from domestic or
overseas saving. In addition, part of domestic lending by banks can be funded from
overseas borrowing. People can emigrate or immigrate; immigrants can be unskilled
labourers, skilled labourers, engineers or scientists.
Imports of machinery are associated with local or overseas entrepreneurs bid-
ding for ideas generated in the small open economy or abroad and, if successful,
obtaining the monopoly rights for producing or importing the machines embedding
the new ideas. To the extent that the small open economy operates at the frontier of
“clean” technology, it is able to exercise influence on the prices of the machines it
imports. This reflects the (temporary) monopsony advantage it enjoys for leading the
implementation of frontier technology (e.g., in green agriculture).
People immigrate to the small open economy from their own countries because
they are attracted by “the quality of life” there, and/or they obtain higher economic
rewards (real wages or return to research) by doing so. In the stylised model we
present below, “quality of life” is captured by social cohesion, and the quality of the
natural environment.
The single, internationally tradeable and consumable, final good can be produced
with different mixes of “clean” and “dirty” technology. The greater the weight of
“clean” technology in producing the final good, the higher the price-premium (quasi-
rent) the producers earn from the sale of the good on both the domestic and interna-
tional markets, because people are willing to pay a higher price for goods produced
with cleaner technology (think of electric cars or organic foods). This in turn has
a positive influence on both the real income of skilled labour used to operate the
“clean” technology, and the return earned on the machines (physical capital) that
embody the “clean” technology.
Accommodating the inflow of skilled labour, scientists, and engineers from other
countries in the stylised model, also helps highlight the (hopefully temporary) ten-
sions between economic growth and social cohesion. The inward migration of skilled
labour, or scientists and engineers, while welcome by business because it assists with
innovation and productivity growth, at the same time potentially creates social ten-
sions, and puts pressure on social and economic infrastructure.
The primary insights for this section come from Krugman (1979). We capture these
insights through the following equations:
66 2 Stylised Model
j j j
At = At−1 + f A (Āt ); j ∈ {c, d } (2.58)
y
Pt /P̄ =p y
(Ytd /Yt ) (2.59)
Mc F Mc y
Ft+1 = (1 − ξ )FtMc + f Mc
(Pt /P̄) (2.60)
L
L
Ft+1 = (1 − ξ )FtL + f L (St+1 /S̄, Et+1 /Ē, wt /w̄)
F
(2.61)
Ls Ls
Ft+1 = (1 − ξ F )FtLs + f Ls (St+1 /S̄, Et+1 /Ē, wts /w̄s ) (2.62)
Sc
Sc
Ft+1 = (1 − ξ F )FtSc + f Sc (St+1 /S̄, Et+1 /Ē, wtSc /w̄Sc ) (2.63)
itd = (1 + ϕ)ı̄ t (2.64)
p,F
netasstF = Ktb,F + Kt − BtRFA (2.65)
where:
• F i , i ∈ {A, L, Ls , Sc, M c } refer to the stocks of foreign unskilled and skilled labour,
scientists, and “clean” machines respectively;
• ξ i ∈ (0, 1) represents the rate of depreciation of i, for i ∈ {L, Ls , Sc, M c };
• a bar over a variable denotes its exogenous “world” counterpart;
• f i (.), i ∈ {A, L, Ls , Sc, M c } are all twice-differentiable and increasing in all their
arguments;
• py (Ytd /Yt ) is also twice-differentiable but decreasing in its argument (Ytd /Yt ); and
• F stocks will enter the country’s equations, through the market-clearing conditions,
in Sect. 2.7.
In addition:
• Equation (2.58), which we have elaborated on in Sect. 2.4 above, captures the
impact of international technological progress on domestic technology and pro-
ductivity.
• Equation (2.59) captures the positive impact of using cleaner technology in the
production of the final output, on the price-premium that can be earned on this
product in international markets where the small open economy is a price-taker.
• Equation (2.60) represents the equation of motion for overseas-produced machin-
ery (assumed to be the machinery using clean technology), relating it to the price
premium that can be earned on the “clean technology” component of the produc-
tion process.
• Equations (2.61)–(2.63) represent the equations of motion for various types of
overseas labour, highlighting the common attractors of the small open economy
for them, being relative social cohesion, relative environmental quality, and relative
material rewards.
• Equations (2.64)–(2.65) represent the financial linkages between the small open
economy and the rest of the world, which we have elaborated on, in Sect. 2.5.2
above.
• Equation (2.64) links the domestic (deposit) rate to the international wholesale
benchmark rate, plus a margin.
2.6 International Connections 67
• Equation (2.65) on the other hand states that the net overseas financial asset position
of the small open economy is the gap between bank equity and claims on physical
assets by overseas investors and the holdings of overseas (risk free) assets by the
banking system (see Sect. 2.5).
We note that inward migration is a potential vehicle for changing the composition
of our human capital through the inflow of skilled labour, and/or scientists. In the
stylised model, to lift the rate of economic growth we must lift productivity growth.
One avenue for this is to increase the number of skilled workers, scientists and
engineers, which can partially be achieved by building connections with the rest
of the global economy. The return from inward migration of human capital is the
positive effect of the quantum and skill-base of migrants on the rate and composition
of our potential economic growth. The cost of inward migration is the (hopefully
temporary) pressure it places on social as well as economic infrastructure. This is
modelled in Sect. 2.8.
2.6.3 Empirical Evidence
Collier (2013) provides evidence on the tensions between the positive economic
effects of the immigration of skilled labour, and the pressures on social and economic
infrastructure associated with immigration.
For a review of the literature, including empirical evidence, on the effects of
ethnic diversity for economic development, see Watkins and Ferrara (2005). The
focus has been on the tradeoff between the benefits of diversity and the costs of
heterogeneity. On the positive side, ethnic diversity can be beneficial by increasing
productivity through innovation, skill complementarities, as well as more creativity,
trade, and product variety. On the negative side, it can create tensions through, among
other channels, competing demands on public goods. García-Montalvo et al. (2017)
provide an updated survey, as well as presenting their own work, showing that in
smaller geographic areas, such as cities, there is a positive correlation between ethnic
diversity on the one hand and wages and/or productivity on the other. This relationship
dissipates as the areas get larger.
Ravlik (2014) provides evidence of both economic opportunities and quality of
life (broadly defined) in the choice of destination-countries for migrants.
For a very useful survey of the interface between innovation that delivers high
value (say through the use of environmentally friendly technology) and the price
premium (and therefore higher returns) earned by producers of small open economies
in international markets, see Dalziel et al. (2017). This provides evidence supporting
our assumption that small open economies that use “clean” technology and operate
at the technological frontier are rewarded by earning price premiums on international
markets.
For evidence on the positive influence of being an innovative economy on attract-
ing high quality human capital (in the form of international skilled labour, scientists
and engineers), as well as overseas investment supporting the import and use of
“clean” machines, see Buera et al. (2015) and Casey and Galor (2014). The chan-
68 2 Stylised Model
nels through which this attraction operates are higher returns to human and physical
capital, as well as a cleaner environment.
2.7 Pulling the Pieces Together - Market Clearing
In their capacity as economic agents (as consumers, producers, employees,

entrepreneurs, engineers and scientists), individuals interact with each other through
various types of markets, with a view to improving their material wellbeing. The
purpose of this section is to pull together and summarise these market interactions
with the help of a set of market-clearing conditions.
We acknowledge that spheres of wellbeing extend well beyond the material dimen-
sion. Individuals who strive to improve their wellbeing take this into account in
different ways in making their decisions – e.g., in allocating their time between
leisure and income-generating work. Individuals, again in pursuit of higher wellbe-
ing, deliberately interact with others in all types of non-market social settings as well
(e.g., churches, schools, charitable and civic institutions).
In addition, there are spheres of individual wellbeing that can be most effectively
and efficiently enhanced through deliberate collective action by an appointed or
elected collective agent (i.e., the government). The domain of this activity will be
different for different societies, and will vary over time (see Cadogan 2013; Tanzi
2011).
2.7.2 Mathematical Model - Market-Clearing Conditions
There are seven market-clearing conditions – for the final product, unskilled labour,
skilled labour, scientists, machines, credit, and international exchange of goods,
machines, and financial assets.
2.7.2.1 Final Good
Market clearing for the final good implies, for t = 0, 1, 2, . . .:

1 1
y y
Ct + χZt = Yt − Xt − ψ mcit di + mdit di − c(Rt )R̃t − τ Yt − f inancecosts (2.66)
0 0
where:
• all costs and taxes are measured in units of the final output;
2.7 Pulling the Pieces Together - Market Clearing 69
y
• Ct is aggregate domestic consumer demand for the final product;
• Zt χ represents the wedge between output and consumption arising from the poten-
tial distortionary effects of government activities, as captured by the government
budget constraint (2.103)26 ;
• ψ is as in (2.40);
y
• Xt is the export of the final product; and
• ψ is the unit cost of producing (or importing) any machine (and the term in
brackets represents the total quantity of machines used in production in period
t, i.e., Mt ), the fourth term on the right hand side measures the total cost of the
exhaustible resources used in producing the final output in period t, and the final
term represents the total tax collected in period t to pay for all the services provided
by the government.
2.7.2.2 Labour
Market clearing for unskilled labour and skilled labour can be represented as follows:
sup
‘efficient’Ldem
t = min(Ldem
t , Lt ) (2.67)
sup
‘efficient’Ldem
st = st , Lst )
min(Ldem (2.68)
where the superscripts dem and sup refer to demand and supply respectively.
The demand for unskilled labour and skilled labour for t = 0, 1, 2, . . . are derived
from the profit-maximisation conditions of the final-output producers, as represented
by Eqs. (2.34)–(2.38).
The supply of both types of labour for t = 0, 1, 2, . . . has two sources: the solution
of the optimisation problem of the consumers, as represented by Eqs. (2.24)–(2.28),
and the inflow of labour from overseas:
sup
Lt = Lt + (FtL − Ft−1
L
) (2.69)
sup
Lst = Lst + (FtLs − Ls
Ft−1 ) (2.70)
2.7.2.3 Scientists
Market clearing for scientists takes the following form:
Sctc + Sctd = Sct (2.71)
where Sct refers to the total number of scientists available at the beginning of period
t, who choose to target working with the “clean” (Scct ) or “dirty” (Scdt ) technologies
for the period ahead.
26 Z
t ≡ cΩw (Ωw ) + cΩy (Ωy ) + cB (B) as in Eq. (2.103) (see Acemoglu et al. 2016).
70 2 Stylised Model
The supply of both types of scientists for t = 0, 1, 2, . . . has two sources: the
solution of the optimisation problem of the consumers as represented by Eqs. (2.24)–
(2.28), and the inflow of scientists from overseas:
Scts = Sct + (FtSc − Ft−1

Sc
) (2.72)
2.7.2.4 Machines
Market clearing for machines in aggregate takes the following form:
j,dem j,sup
Mt = Mt (2.73)
for j ∈ {c, d }, where the aggregate domestic demands for the two classes of machines
(the left-hand side of the equalities), for t = 0, 1, 2, . . . , are derived from the profit-
maximisation conditions for final-output producers, as represented by Eqs. (2.34)–
(2.38).
The supply of new machines to each sector reflects the successful innovations
introduced by scientists, which in turn determines the total domestic production
and/or imports of machines which embody these innovations:
d ,sup
Mt = Mtd (2.74)
c,sup
Mt = Mtc + (FtMc − Mc
Ft−1 ) (2.75)
2.7.2.5 Exhaustibe Natural Resources
In the version of the model we are working with, there is no private property rights
on the exhaustible natural resource, so it is inappropriate to refer to “market clearing
conditions”.
An alternative condition is that the demand for the resource in production is less
than the available stock:
R̃t ≤ Rt (2.76)
where the left-hand side of the inequality (the demand for the natural resource),
for t = 0, 1, 2, . . . , is derived from the profit-maximisation conditions for the final-
goods producers, as represented by Eqs. (2.34)–(2.38).
The stock of the natural resource (the right-hand side of the inequality) is given
by the equation of motion (2.32).
One of the possible policy levers available to the government is to assign private
property rights to natural resources; this is an option we return to in the next Chap. 3
when we consider various policy options.
2.7 Pulling the Pieces Together - Market Clearing 71
2.7.2.6 Finance
The relevant market is the market for loans – the credit market:
Btl − Bt−1
l
= wts Lst + wt Lt + ptmc mct + ptmd mdt + wtSc Sctc (2.77)
In addition, the households need to be willing to hold the outstanding volume of

bank deposits:
sup
Blt = Bltdem (2.78)
sup
Bdt = dem
Bdt (2.79)
2.7.2.7 International Payments
The stylised small open economy can export the single final (consumable) product
and can import “clean” machines.27 In the short run, it is a price-taker for both
products in international markets. However, in the longer run, by increasing the
weight of the use of “clean” technology in the production of the final good, it can
increase the premium it earns on the production and sale of the final good.
Without loss of generality, we assume that the nominal exchange rate is fixed at
a value of one, and specify the condition for the international payments balance of
this small open economy as follows:
y y
Pt (Yt − Ct − Sat ) = ptmc (Ft+1
Mc
− FtMc ) + (netasst+1
F
− netasstF ) (2.80)
y y y
where, Xt = (Yt − Ct − Sat ). Thus, the relative price of the final product (Pt ) viz
the price of the “clean” machine (ptmc ) is also the small open economy’s terms of
trade (and “real” exchange rate).
2.8 Environmental, Social, Economic Externalities
A fundamental assumption underpinning our wider concept of wellbeing is that, in

addition to their individual incomes and consumption of private goods (including
leisure and good health), individuals and communities also value, among others,
a clean environment, social cohesion, equity across society and generations, and
enhanced resilience to the types of systemic shocks that have the potential to cause
serious damage to various capital assets and through that on our overall wellbeing.
27 Given the purpose of this chapter, complicating the model by adding final products that can also
be imported, or switching to importing the final product but exporting the machines, does not add
anything of value to our analysis.
72 2 Stylised Model
Because the returns from investing in the capital assets and institutions that
generate these social goods will not be fully captured privately, this may lead to
under-investment in, and over-use and/or under-protection of, the components of
comprehensive wealth, relative to a socially optimal equilibrium, in the absence of
deliberate collective action (see Miller 2006). There is no suggestion here that the ben-
efits associated with these positive externalities will not be provided at all; however,
they may be under-provided. It is likely that, “decentralised market decisions lead
to the under-provision of public goods such as infrastructure, or the over-provision
of public bads such as pollution” Sachs (2015, p. 155). Through appropriate and
deliberate collective action, we may be able to make markets, communities, and
institutions work more effectively and efficiently in delivering these public goods.
This is a possibility and not a certainty.
The dimensions of the wellbeing frontier, seen from an environmental, social, and
economic policy perspective, are public goods (potential sources of significant posi-
tive externalities) wrapped around the capital assets that are the sources of wellbeing.
These are the main social outcomes that good public policy pursues, through its influ-
ence on the growth, allocation, and preservation of components of comprehensive
wealth, towards enhancing sustainable and shared intergenerational wellbeing. They
are delivered through (public and/or public/private) investments in various forms of
complementary infrastructure, including institutions.
Resilience refers to the capability of the economy and society to respond to,
and continue to operate reasonably effectively and efficiently in the aftermath of,
significant systemic shocks to comprehensive wealth – measured, in principle, by
the speed and cost of adaptation.28
It is achieved through appropriate governance and management towards reducing
the impact, on our wellbeing, of the major (systemic) shocks that we are collectively
most likely to be exposed to. We first identify such systemic risks, and then decide
how much we are willing to collectively invest towards building resilience to them
(see Guild et al. 2013).29
To use and adapt the language of Hansen (2014), here we are focused on “uncer-
tainty outside models” but stop short of assigning probabilities to the occurrence of
the types of systemic shocks we are most concerned about – because there is no way
of knowing these probabilities.
28 For a recent empirical study of the determinants, and population-distribution and heterogeneity,
of resilience (in terms of total psychological response) at the individual level, to ten major adverse
life events, as well as a useful survey of this literature, see Etilé et al. (2017).
29 The closely related critical question, faced with numerous potential catastrophes, with uncer-
tainties surrounding occurrences and timings, which should society attempt to avert, is carefully
analysed by Martin and Pindyck (2015).
2.8 Environmental, Social, Economic Externalities 73
A system may show resilience to major systemic risks not necessarily by return-
ing exactly to its previous state following a shock, but instead by finding different
ways to carry out essential functions; that is, by adapting. We therefore think of
resilient environmental, social, and economic systems as ones that have the capabil-
ity to: withstand sudden shocks, adapt to changing contexts, and recover to a desired
state (either the previous one or a new one), while preserving the continuity of its
operations. Thus resilience encompasses both recoverability (the capacity for speedy
recovery after a crisis) and adaptability (timely adaptation in response to a changing
environment).
To paraphrase (North 1994), it is adaptive rather than allocative efficiency that
is the key to long run wellbeing. Successful economic, political, and social systems
have evolved flexible institutional structures that can survive the shocks and changes
that are part of successful evolution.
We classify resilience-enhancing measures against systemic shocks into three
clusters – resilience against systemic natural, social, and economic risks. By way of
examples, regulations and investments targeted at reducing the impact of earthquakes,
the sources of climate warming, and the likelihood of imports of diseases from the rest
of the world, increase resilience to systemic natural risks; ensuring universal access
to basic income, health services, education, and housing, increases the resilience of
our social infrastructure; and investments (including regulations) towards enhancing
financial and macroeconomic stability increase resilience to potential economic risks.
We need to also recognise that the presence of such resilience-increasing measures
against systemic risks allows individuals to be prepared to take more risks, thus
potentially enhancing the economic-growth potential of the economy in aggregate –
and possibly the wider wellbeing potential of our collective comprehensive wealth.30
Social cohesion may be conceptualised as “a phenomenon of togetherness in
a society” (represented by the cohesiveness of communities, neighborhoods, and
societies) Pervaiz and chaudhary (2015, p. 369). It refers to our ability and willingness
to live and work together in social harmony, comprising tolerance, trust, and mutual
respect. It reflects a state of social being where our differences in all its dimensions
are respected, embraced, and celebrated. It does not refer to ethnicity only; it also
covers cohesion across genders as well as different generations. In this broad context,
Sachs (2015) refers to generalised trust, social support networks, generosity and
voluntarism, and perceptions of low corruption in business and government (p. 154).
We can enhance social cohesion, as well as the growth potential of the economy and
overall resilience, by deliberately investing in social infrastructure.
The literature makes a distinction between different dimensions of social cohe-
sion, and their impacts on economic and social outcomes. Strong within-group social
cohesion, associated with bonding capital, to the extent that it is divisive, discrim-
inatory, and exclusive can have negative effects on trust, democratic institutions,
political and social stability, and transaction costs, thus having negative implications
30 We want to specifically thank Ken Warren at the New Zealand Treasury for his insights on risk
management as resilience-enhancement, particularly in the context of dealing with systemic risks

from a policy perspective.
74 2 Stylised Model
for social and economic outcomes. On the other hand, strong across-group social
cohesion, associated with bridging capital, by underpinning and supporting the for-
mation of strong institutions, generalised trust, collaboration, lower transaction costs,
and innovation, can have a very positive impact on social and economic outcomes
(see, for example, Algan and Cahuc 2013; Varvarigos et al. 2015).
The promotion of social and intergenerational equity in our framework is not a
pursuit of equality of outcomes or of opportunities. The pursuit of equality as such,
as a policy aim, would be fruitless (see Buchanan and Hartley 2000; Walzer 1983),
and possibly counterproductive as well (see Breunig and Majeed 2016). McCloskey
(2014) distinguishes between the French Enlightenment focus on the pursuit of equal-
ity of material outcomes, versus the Scottish Enlightenment version which aspires
to give everyone the opportunity “to have a go” at living the kinds of lives they have
reason to value. As Atkinson (2015) emphasises, however, in an intergenerational
context this distinction may not be that useful since today’s outcomes determine
tomorrow’s opportunities: “Inequality of outcome among today’s generation is the
source of the unfair advantage received by the next generation. If we are concerned
about equality of opportunity tomorrow, we need to be concerned about inequality
of outcome today” (p. 12).
It is the distribution of opportunities and capabilities (i.e., substantive freedoms)
among individuals and communities, in a way that gives them a fair chance of pur-
suing the kinds of lives they have reason to value, today and into the future, that is
at the core of our concept of equity. The allocation of private wealth (in the form of
the ownership of physical capital, human capital, and some forms of natural capi-
tal such as land), the evolution of the production technology used in producing the
final output, and public policy that enhances and provides widespread access to pub-
lic wealth, all have a major bearing on equity defined in this way. In our stylised
model, increasing equity has a direct positive benefit on wellbeing through enhanced
social cohesion. In addition, policies that incentivise skilling, innovation-supporting
education and R&D, and the switch of production technology towards “clean tech-
nology”, all have positive benefits for both equity, and the quality and sustainability
of potential economic growth.31
There is a critical difference between the pursuit of equity and the pursuit of
poverty-reduction. The first (equity) is a relative concept centred on opportunities
and capabilities; the second (poverty) is an absolute concept centred on deprivation
– not only income deprivation per se but (multi-dimensional) wellbeing deprivation
(see Alkire 2016; Farina et al. 2015; Toledo et al. 2016).32 They have separate,
31 Ina similar vein, Acemoglu and Robinson (2015) emphasise that, “[…] inequality should not be
thought of as always summarized by a single index, such as the Gini index or the top 1% share.
Rather, the economic and political factors stressed here determine the distribution of resources more
generally […]” (p. 16).
32 Toledo et al. (2016) provides a very useful summary of the literature on multi-dimensional poverty,
as well as alternative measures to capture poverty as exclusion and deprivation. Eiffe (2010) sum-
marises Smith’s and Sen’s work on poverty and notes that they both have emphasised both the
multi-dimensional and absolute nature of poverty, and therefore poverty as exclusion and a source
of shame has to be conceptualised and measured relative to the society in which it occurs.
and distinctive, influences on social cohesion in our stylised model. Complementary

policies are required to reduce poverty while promoting equity. Both concepts (equity
and poverty) are ultimately about wellbeing and expressed in terms of wellbeing in
the formal model that is presented in the next section.
The concept of sustainability used in this book is sourced from the literature on
“sustainable development”. It is defined in terms of our capacity to bequeath capital
assets (comprehensive wealth) to future generations so that they can generate a level
of wellbeing (in its broadest sense) that is at least as high as our own – critically,
however, recognising the substitutability of these capital assets (often aided by tech-
nological advances) as sources of wellbeing. Following Arrow et al. (2012, 2013),
“we take intergenerational wellbeing to be the object of interest in sustainability
analysis” (p. 319). In operationalising this, Arrow et al. note that since the source of
potential wellbeing is comprehensive wealth, sustainability is achieved at any point
in time if comprehensive wealth at constant shadow prices is non-decreasing at that
time. Ultimately, however, sustainability (just like equity and poverty) is defined
in terms of wellbeing – maintaining or enhancing overall wellbeing across time.
The emphasis on the sustainability of wellbeing as such allows for substitutability
between the various components of comprehensive wealth as sources of wellbeing
(see Greasley et al. 2013, 2014, 2017). To refer back to our earlier discussion in
Chap. 1, this is “weak sustainability”.
Material wellbeing is critical for overall wellbeing. Our focus is on long-term
intergenerational wellbeing. Thus it seems appropriate to focus on lifting potential
economic growth as the proper domain of public policy so far as material wellbeing
is concerned. To repeat an earlier quote from Arrow et al. (2012), “When scholars
adopt intergenerational wellbeing as the object of interest, their presumption is that
at any given date social wellbeing is not only the wellbeing of the current generation,
but also the potential welfare of the generations that are to follow. The point is to ask
whether the society under study is functioning sufficiently well to ensure that some
measure of intergenerational wellbeing does not decline” (p. 318).
The formal model presented in the section that follows captures the strong interde-
pendencies between the public goods that are the sources of significant environmen-
tal, social, and economic externalities. By way of example, both higher resilience to
systemic (environmental, social, and economic) risks and stronger social cohesion
have positive effects on sustainability. Similarly, increased resilience has a positive
effect on the growth potential of the economy. Furthermore, policies that are able
to increase (cross-society and/or intergenerational) equity and reduce poverty, sepa-
rately and distinctly, contribute to both resilience and social cohesion.
76 2 Stylised Model
2.8.2.1 Consumption
In terms of consumers, policies that enhance environmental quality, social cohe-

sion, and environmental and social resilience represent significant sources of positive
externalities. We can capture these by re-writing Eq. (2.9) as follows33 :
∞
1 y
W0 = t Ωt (Et , St , Γt , Γt ) · u φ(Y /Y )Ct , Ct , Ct
w E S d x h
(2.81)
t=0
(1 + ρ)
y
where, we assume that u φ(Y d /Y )Ct , Ctx , Cth takes the following specific form:
θc
y y c
θ −1 θc −1 θc −1
u φ(Y d /Y )Ct , Ctx , Cth = h(H̃t ) φ(Y d /Y )Ct θc + Ctx θc (2.82)
where, H̃t ≡ Cth , ∂h/∂ H̃ > 0, h(H̃t ) ∈ (0, 1), and Ωtw (Et , St , ΓtE , ΓtS ) ∈ (0, 1) rep-
resents the wellbeing-externality function. θc is the elasticity of substitution, in the
provision of utility, between the marketed consumer product and leisure; and we
assume these to be gross substitutes (i.e., θc > 1).
To re-emphasise a point we made earlier in relation to Eqs. (2.11) and (2.12),
what Eqs. (2.81)–(2.82) say is that we enjoy life, including consumption of marketed
products, leisure, and our good health, more when we live in a country that offers
relatively higher levels of environmental quality, social cohesion, and resilience to
potential threats to both. Furthermore, as our health deteriorates, we derive less and
less utility from consuming marketed consumer products and/or leisure. We assume
that Ω w (E, S, Γ E , Γ S ) takes the following specific (CES) form:
θΩw −1 θΩw −1 θΩw −1 θΩw −1

θ θΩw−1
Ωw
Ω (Et , St , Γt , Γt ) = (Et ) w + (St ) w + (Γt ) w + (Γt ) w
w E S θΩ θΩ E θΩ S θΩ
(2.83)
The arguments of Ω w (·, · · · ) are represented by index numbers in the interval
[0, 1].34 The parameter θΩw represents the elasticity of substitution, in the provi-
sion of wellbeing-externalities, between environmental quality, social cohesion, and
33 Strictly speaking u(·, · · · ) here is different from that in (2.9) and from the subsequent utility
functions in the next sections. However, for simplicity, we call all such functions u(·, · · · ).
34 So are the arguments of Ω y (·, · · · ) in Eq. (2.90). In general, it is difficult to assure that these index
numbers, which satisfy the respective stock Eqs. (2.84)–(2.87) (and (2.91)–(2.93)) remain in [0, 1].
An appropriate calibration of the parameters ξ ·,· , δt· , γ · , etc. can assure this condition for a range of
values of Et , St , etc. Another possibility would be to model the stock equations using the function
arctan(·). E.g., Eq. (2.86) could be modelled as follows:
2
Γt+1
E
= arctan (1 + δ ΓE )ΓtE + γ ΓE Et
π
the infrastructure that supports the resilience of both to systemic shocks. We assume
these to be gross complements (i.e., θΩw < 1).
The equations of motion describing the evolution of E, S, Γ E and Γ S are assumed
to take the following very similar forms in the stylised model:
Et+1 = −ξ E,Y (Ytd /Yt )Yt + (1 + δ E )Et − ξ E,POV POVt (2.84)

St+1 = −ξ S,F
Ft − ξ S,Q
Qt − ξ S,POV
POVt + (1 + δ Ks
)Kts + (1 + δ )St
S
(2.85)
ΓE ΓE
Γt+1
E
= (1 + δ )Γt + γ E
Et (2.86)
ΓS ΓS
Γt+1
S
= (1 + δ )Γt + γ St
S
(2.87)
where ξ E,Y > 0 represents the rate of environmental degradation resulting from the
production of the single output (Y ) (which depletes the stock of our exhaustible
natural resources), weighted by (Ytd /Yt ) that is a measure of the extent to which
overall production uses “dirty” technology (imbedded in “dirty” machines); and
δ E ≥ 0 is the rate of “environmental regeneration”. This equation introduces the
environmental externality associated with production. Innovations or investments in
both ξ E,Y and δ E are public goods, and therefore potential targets for public policy.
ξ E,POV measures the impact of poverty on environmental quality. Poor people do
not have the luxury of enjoying an environmentally-friendly life style (think of the
affordability of electrical cars or solar-powered heating devices).
Similarly, ξ S,i > 0, i ∈ {F, Q, POV }, measure the rates of degradation of our
social cohesion, arising, respectively, from:
immigration, where Ft is measuring the number (stock) of migrants at the beginning
of period t,
inequity – Qt – which is measured by the variance of wealth ownership across
different clusters of individuals with different skills (as in (2.100)), and
poverty – POVt – measured by the percentage of people whose ownership of “com-
prehensive wealth” is below a certain threshold.
The parameter δ Ks ≥ 0 is the rate of enhancement of social capital. γ ΓE and γ ΓS
are both > 0; thus, the higher the levels of environmental quality and social cohe-
sion, the greater their resilience to systemic shocks. δ ΓE ≥ 0 and δ ΓS ≥ 0 refer to the
rates of regeneration or enhancement in environmental and social resilience respec-
tively. These equations collectively introduce the wellbeing-externalities associated
with investments in environmental quality and social cohesion, as well as enhanced
resilience of both to systemic risks.
Innovations in ξ S,F , ξ S,Q , ξ S,POV , δ Ks , γ ΓE , δ ΓE , γ ΓS , δ ΓS are public goods, and
therefore potential targets for public policy.
POV directly and negatively affects social cohesion (Eq. (2.85)) and indirectly,
through its negative effect on social cohesion, social resilience (Eq. (2.87)). In addi-
tion, it directly and negatively affects environmental quality (Eq. (2.84)) and indi-
The right hand side of this equation has the property of decreasing returns and remains between 0
and 1 for positive arguments.
78 2 Stylised Model
rectly, through its negative effect on environmental quality, environmental resilience

(Eq. (2.86)). Thus, reducing poverty improves social cohesion and environmental
quality, as well as enhancing social resilience (partly by providing a “safety net” for
the most vulnerable) and environmental resilience. It is wellbeing-poverty (and not
just income-poverty) that matters for social cohesion. We specify the equation of
motion for poverty as follows:
POVt+1 = (1 − δ POV )POVt − γ V,POV Vt . (2.88)
Here, V is the society’s comprehensive wealth – a discrete-time equivalent of V (t)

in (2.6) and (2.7). The coefficients δ POV ∈ (0, 1) and γ V,POV > 0 represent, respec-
tively, the rate of change of poverty – a potential target for public policy, and the
positive effect of an increase in overall wealth on poverty reduction.
2.8.2.2 Production
On the production side of the stylised model, policies that improve economic infras-
tructure, enhance broader economic resilience, and build social capital represent
significant sources of positive externalities. We attempt to capture these particular
positive externalities by expanding Eq. (2.31) as follows35 :
y y
Yt = Ω y (Λt , Γt , Kts , ΓtE )
⎧ ⎫
⎪⎡ θy −1 ⎤ θy −1 ⎪
θy
⎪
⎨ 1 θyθ−1 1 θy
⎪
⎬
y
⎣ (Ls H )1−α Aαcit mαcit di + R̃αt 2 (LH )1−α1
A1−α 1 α1
mdit di ⎦ (2.89)
⎪
⎪
t t dit ⎪
⎪
⎩ 0 0 ⎭
where, Ω y (Λy , Γ y , K s , ΓtE ) ∈ (0, 1) represents the economic-externality

function, with economic (or production) infrastructure (Λy ), economic (or produc-
tion) resilience (Γ y ), social capital (K s ), and the resilience of environmental ser-
vices being its four main arguments. It is best to think of all four arguments of
Ω y (Λy , Γ y , K s , Γ E ) as index numbers ∈ (0, 1). Ω y (Λy , Γ y , K s , Γ E ) is increasing
in all of Λy , Γ y , K s , Γ E , twice-differentiable and jointly concave in (Λy , Γ y , K s ,
Γ E ). The economy approaches its productive-potential when Λy , Γ y , K s , Γ E are all
close to 1, and operates significantly below potential when they are close to 0, albeit
production remains positive even then. The growth in the economy’s productive
potential in turn is primarily driven by the growth in productivity (A).
The inclusion of Γ E in the economic-externality function can be motivated by
noting that the deterioration of environmental or natural infrastructure “may lead
to the depletion or even disappearance of goods previously viewed as ‘free’, that is,
35 Strictly speaking, the same symbol suggests, if not implies, that the expressions are identical. So,
by way of example, we should be careful to note, when we use Y in Eq. (2.89) and then again in
(2.31), that we are not referring to “the same” Y . However, for our purposes, in both cases we are
referring to the single final output produced in our small economy.
available in unlimited quantities, such as water, air, climate stability, biodiversity, and
so on. […] In terms of regulatory services (provided by nature), there is air quality,
erosion, water purification, climate regulation, and the cushioning of the effects
of natural hazards such as tornadoes and tsunamis. Primary services comprise soil
formation, photosynthesis, the water cycle, and the cycle of nutrients essential to life.
[…] the deterioration of these different services may negatively impact, directly or
indirectly, the planet’s production capacities” De Perthuis and Jouvet (2015) (p. 83).
We assume that Ω y (Λy , Γ y , K s , ΓtE ) takes the following specific form:
$ θΩy −1 θΩy −1 θΩy −1 θΩy −1

% θ θΩy−1
y y
y θ y θ θ θ
Ωy
Ω y (Λt , Γt , Kts , ΓtE ) = Λt Ωy + (Γt ) Ωy + (Kts ) Ωy + (ΓtE ) Ωy
(2.90)
where θΩy represents the elasticity of substitution, in the provision of production-
externalities, between economic infrastructure, economic resilience, social capital,
and the resilience of the environmental or natural infrastructure. We assume these
to be gross complements (i.e., θΩy < 1). The equations of motion describing the
evolution of Λy , Γ y , and K s can be written as follows36 :
Λ
Λt+1 = (1 + δt y )Λt − ξ Λy ,Y Yt − ξ Λy ,F Ft
y y
(2.91)
Γ
δt y )Γt + γ Γy Λt+1
y y y
Γt+1 = (1 + (2.92)
s
Kt+1 = (1 + δtKs )Kts + γ Ks Πt (2.93)
where:
• ξ Λy ,Y > 0 measures the rate of degradation of economic infrastructure, resulting
from production activities in this economy.
• ξ Λy ,F > 0 represents the pressure on economic infrastructure arising from an
increase in the number of migrants (or population growth more generally).
• δ i ≥ 0, i ∈ {Λy , Γ y , K s , Γ E }, are the rates of enhancements to economic infras-
tructure, economic resilience, social capital, and resilience of environmental or
natural infrastructure respectively.
• γ Γy > 0; thus, the stronger economic infrastructure is, the greater its resilience to
major systemic shocks.
• γ Ks > 0; thus, the greater the profitability of scientific/entrepreneurial activities,
the greater the incentive to invest in social capital (in the form of building clusters
and networks).
• And so is γ ΓE > 0; thus, the stronger environmental or natural infrastructure is,
the greater its resilience.
These equations introduce the economic externalities associated with investments
in economic infrastructure, economic resilience, social capital and the resilience of
36 See comments in footnote 34, Sect. 2.8.2.

80 2 Stylised Model
natural or environmental infrastructure. Innovations in ξ Λy ,Y , ξ Λy ,F , δ Λy , δ Γy , δ Ks ,

γ Λy , γ Ks and γ ΓE are public goods, and therefore potential targets for public policy.
2.8.2.3 Distribution and Equity37
It is the distribution of opportunities and capabilities (i.e., substantive freedoms)

among individuals, in a way that gives them a fair go at pursuing the kinds of lives
they have reason to value, that is at the core of our concept of equity. Thus, access to
public forms of capital (such as social and economic infrastructure, and some forms
of natural capital), and public policy that has an influence on both the quantum of
and access to public goods that generate positive externalities, have a major bearing
on equity defined in this way. So do the allocation of private wealth (in the form
of the ownership of physical, financial (bank), and human capital), and the evolu-
tion of the production technology used in producing the final output. In our stylised
model, increasing equity has a direct positive benefit on wellbeing through enhanced
social cohesion. In addition, policies that incentivise skilling, innovation-supporting
education and R&D, as well as the switch of production technology towards “clean
technology”, all have positive benefits for both equity, and the quality and sustain-
ability of potential economic growth. (Refer back to Fig. 2.1.)
In order to be able to discuss and present distributional aspects or equity across
society, we need to differentiate between individuals. In our stylised model, we differ-
entiate between three groups of individuals – unskilled labourers, skilled labourers,
and scientists/entrepreneurs.38
In addition to using their non-leisure time in different ways, to generate income
or to invest in building human capital, these groups of individuals also differ in
Ls
their initial (period 0) endowments of physical capital Kp,0L
, Kp,0 , Kp,0
Sc
(comprising
Ls
L
the machines referred to earlier), bank capital Kb,0 , Kb,0 , Kb,0
Sc
, and human capital
Ls
L
Kh,0 , Kh,0 , Kh,0
Sc
:
1 1
p p,L p,Ls p,Sc
K0 = mci,0 di + mdi,0 di = K0 + K0 + K0 (2.94)
0 0
K0b = K0b,L + K0b,Ls + K0b,Sc (2.95)
λK0 h K0h = λL0 s Ls0 + λSc
0 Sc0 (2.96)
where the symbols mc(·) and md(·) were introduced after Eq. (2.31), and λj , j ∈
h s
K0 , L0 , Sc0 represent the respective spot shadow prices of these various forms of
human capital (or human wealth). These three components of private capital represent
37 This section draws heavily on Turnovsky (2013) and Turnovsky and Mitra (2013).
38 In the stylised model, successful scientists become entrepreneurs; and unsuccessful ones generate
income as skilled labour.

the accumulated, initial or period 0, collective private capabilities and opportunities

of these groups of individuals.
Ultimately, we are interested in the distribution of wellbeing. Wellbeing is a func-
tion of comprehensive consumption, and comprehensive consumption is a function
of comprehensive wealth. Comprehensive wealth comprises public capital (compris-
ing components of natural capital and social and cultural capital) and private capital
(in the form of physical, financial, and human capital). The distribution of compre-
hensive wealth, which is a primary determinant of the distribution of wellbeing, is
thus a function of access to public capital assets and ownership of private capital
assets.
With that context and background in mind, let us start with the initial distribution
of total private human capital K0h (which comprises the skills, education, and health
p
of individuals), total private physical capital K0 (comprising either type of machine),
b
and total private bank capital K0 . In a growing economy, we are interested in the
evolving ownership shares of these three groups of individuals viz the accumulating
h,j h,j p,j
total stocks of human capital, kt ≡ Kt /K̃th , j ∈ {Ls , Sc}, physical capital, kt ≡
p,j p b,j b,j
Kt /K̃t , j ∈ {L, Ls , Sc}, and financial (bank) capital, kt ≡ Kt /K̃tb , j ∈ {L, Ls , Sc},
p
where K̃th , K̃t and K̃tb denote the corresponding economy-wide average quantities.
The initial relative endowments across these three groups of individuals have
k
mean one, standard deviations σ0kh , σ0 p , and σ0kb measured around the economy-wide
k k
averages, and covariances (or in this case, coefficients of variation) σ0 h p , σ0kh kb and
k k
σ0 p b (possibly zero).
At the beginning of each time period t, the total private wealth (V ) of the three
groups of individuals comprises their physical capital, financial (bank) capital, plus
the value of their human capital, and is defined by:
K p,L
VtL = λt p Kt + λKt b Ktb,L (2.97)
K p,L
VtLs = λt p Kt s + λKt b Ktb,Ls + λLt s Lst Ht (2.98)
K p,Sc
VtSc = λt p Kt + λKt b Ktb,Sc + λSc
t Sct Ht (2.99)
where λi , i ∈ {K b , K p , Ls , Sc} are the shadow prices of physical capital, skilled

labour, and scientists.
Wealth accumulation at constant prices is the sum of physical capital accumulation
(a function of saving and investment out of market income), financial (bank) capital
accumulation (a function of net bank profits), and human capital accumulation using
time (in the form of investment in skilling, education, and health) (see Eqs. (2.18)–
(2.19)).
We define the share of total private wealth owned by each of the three clusters of
individuals as vj ≡ V j /V, j ∈ {L, Ls , Sc}.
It is quite obvious from Eqs. (2.89), (2.98)–(2.99) that the evolution of wealth
inequality over time will reflect the evolution of the structure of the economy (i.e., the
82 2 Stylised Model
composition of production), and the values of the various types of physical, financial
(bank), and human capital.
Using a similar, but not identical, model (Turnovsky and Mitra 2013) show that an
individual who has above average long run (equilibrium) wealth also enjoys above
average long run consumption and thus wellbeing.
We can then define the degree of inequality in society by the degree of inequality
of opportunities and capabilities (and thereby of wellbeing) across the three clusters
of individuals, which in turn reflects the standard deviation of private wealth around
mean wealth (σtv ), measured across all capitals:
Qt+1 ≡ σtv (2.100)
This standard deviation is the standard deviation of total private wealth (hence the
superscript v) across the three clusters of households indexed by j as before (i.e., vj ≡
V j /V, j ∈ {L, Ls , Sc}).39
An important result of the analysis of Turnovsky and Mitra (2013) is that, in
y
the steady state, σ̃tv > σ̃t > σ̃tw (where ˜ above a variable indicates its steady state
value). That is, wealth inequality tends to overstate income inequality, which in turn
tends to overstate wellbeing (or welfare) inequality.
Q can converge towards zero around high or low aggregate average wealth (and
thus wellbeing) levels. In our stylised model, everything else being equal, it is conver-
gence of inequity to zero around high levels of wealth that would be consistent with
higher levels of overall wellbeing. The aim of policy is to ensure a greater degree of
equality of opportunities and capabilities (and thereby of wellbeing) across groups
of individuals, without compromising overall wellbeing.
As we will argue later in this book, this can be achieved with the help of policies
that simultaneously support the switch of production towards the use of “clean”
technology (through a combination of subsidies and taxes), while concurrently
(through R&D support and related measures) encouraging investment in (and thereby
increased supply of) human capital in the form of skilled labour and scientists, as well
as “clean” machines. Subsidies for education and skilling can help reduce the inequal-
ity in the distribution of human capital. Since this would decrease the inequality in
the distribution of market income, it also decreases the inequality in the ownership of
physical and financial (bank) capital (through the convergence of saving rates across
these clusters of individuals). Overall we would have lower σ v , and thus Q. We thus
potentially have a combination of policies that can simultaneously increase both the
quality and sustainability of the economic growth (i.e., material wellbeing) potential
of the economy, while also increasing equity and reducing poverty – thus increasing
social cohesion and overall wellbeing.
39 In fact overall wellbeing reflects access to both the private and public components of comprehen-
sive wealth. We are implicitly assuming here that there is equity of access to public components of
comprehensive wealth.
Sachs (2015) notes that, “societies with high social capital outperform those with low
social capital in terms of subjective wellbeing (SWB) and economic development”
(p. 152). In this context, empirical evidence, which also has implications for public
policy, is more nuanced. Pervaiz and Chaudhary (2015) provide empirical evidence
that supports the proposition that inter-group versus intra-group social cohesion,
supported by bridging social capital and bonding social capital, respectively, have
different (in fact opposite) relationships with economic growth – the former a positive
one and the latter a negative one.
Pervaiz and Chaudhary (2015) argue, and provide both a theoretical rationale and
empirical evidence to support their argument, that different dimensions of social
cohesion have different impacts on economic and social outcomes. Strong within-
group social cohesion, associated with bonding capital, to the extent that it is divisive,
discriminatory, and exclusive can have negative effects on trust, democratic institu-
tions, political and social stability, and transaction costs, thus having negative impli-
cations for social and economic outcomes. On the other hand, strong across-group
social cohesion, associated with bridging capital, by underpinning and supporting
the formation of strong institutions, generalised trust, collaboration, lower transac-
tion costs, and innovation, can have a very positive impact on social and economic
outcomes.
Another study that finds strong, positive, and lasting correlations between diver-
sity and economic development is Rodríguez-Pose and von Berlepsch (2017). In the
context of the USA, and comparing the experiences of states and counties across
the country, they “identify the presence of a strong and very long lasting impact of
diversity on county-level economic development. Counties that attracted migrants
from very diverse national and international origins over a century ago are signifi-
cantly richer today than those that were marked by a more homogeneous population
at the time. […] Yet the benefits of diversity came with a strong caveat: the gains of
having a large number of groups from different origins within a territory (fraction-
alisation) only materialise if the diverse groups are able to communicate with one
another (low polarisation). Deep cut lines separating the groups (high polarisation)
emerge as an important barrier for economic development. Hence, diversity becomes
a double edged sword: it works only if the different groups can interact, that is, if the
‘melting pot’ really happens. Where it is not possible to build a dialogue between the
different groups, where bridging does not occur, groups and communities remain in
their own physical or mental ghettoes, undermining any economic benefits from a
diverse environment” (pp. 24–25).
When it comes to subjective wellbeing (or life satisfaction, or happiness), on the
other hand, Growiec et al. (2017) cite a number of studies “confirming the importance
of maintaining frequent social interactions, both with similar and dissimilar others,
for individuals’ life satisfaction and happiness” (p. 11).
A very useful survey of the literature on poverty, inequality, economic growth,
and social cohesion is provided by Breunig and Majeed (2016). The same paper
84 2 Stylised Model
also provides empirical evidence on the interface between inequality, poverty, and
economic growth, and concludes that: when poverty is low, the relationship between
inequality and economic growth is statistically insignificant. As the level of poverty
increases, so does the negative effect of inequality on economic growth. It is also
emphasised however that both poverty and inequality can and do negatively impact
wellbeing through other channels as well, including a decrease in social cohesion
and a deterioration in the quality of institutions.
Welsch and Biermann (2016) provide empirical evidence in support of the hypoth-
esis that poverty, like unemployment, is not only a “private bad”, but also a “public
bad”.40 Not only personally being poor (unemployed), but also the aggregate level of
poverty (unemployment) negatively affects subjective wellbeing, especially of those
who are not poor themselves. In both cases, “people may be unhappy about poverty
(unemployment) even if they are not poor (unemployed) themselves. They may feel
bad about the unfortunate fate of the poor (unemployed) and they may worry about
the possibility of becoming poor (unemployed) themselves in the future. They may
also feel repercussions on the economy and society as a whole. They may dislike
the increase in taxes likely to happen in the future. They may fear that crime and
social tensions increase, and they may even see the threat of violent protests and
uprisings”(ibid., p. 2).41
The latter reference refers to concerns about the potential negative link between
poverty and social cohesion. Acemoglu and Robinson (2000) and Yamamura (2016)
provide theoretical arguments and empirical evidence in support of the proposition
that high-income earners’ support and stated preference for income redistribution
relates to their concern about the potential conflict between the rich and the poor
– i.e., the erosion of social cohesion. In addition, as Welsch and Biermann (2016)
point out, “From a policy point of view, the finding that poverty is a public bad and/or
creates negative externalities suggests that poverty implies market failure. This, in
turn, suggests that poverty alleviation is a matter not only of equity, but of efficiency”
(p. 9).
Angelopoulos et al. (2017) provide a very useful survey of the literature, particu-
larly in the context of the UK, on whether job-related training targeted at low-skilled
workers helps increase income, and reduce the inequality of income between skilled
and unskilled workers, as well as conducting their own independent empirical inves-
tigation.
They find that while subsidising job-related training for unskilled workers does
lead to an increase in their income, it has only a modest impact on reducing income
inequality between unskilled and skilled labour, because it does have a positive
spillover effect on skilled labour as well, thus increasing overall income by increas-
ing the earnings of both unskilled and skilled labour. The spillover effects occur
primarily through the benefits that skilled labour receive from the higher productiv-
ity of unskilled labour on their own (i.e., the skilled labourers’) wages.
40 The corresponding evidence regarding unemployment is presented in Di Tella et al. (2001).

41 The parallel references to unemployment are from Frey and Stutzer (2002).
2.8.4 Motivation: Government
The key actor, our collective agent, in delivering or facilitating the delivery of public
goods is the government.
In the context of our wider wellbeing framework, and to motivate a role for public
policy and for the government, it is very important to focus on Eqs. (2.81) and (2.89)
jointly:
• the four capital stocks that sit in the middle of the wellbeing frontier in Fig. 1.2, are
imbedded in the production or wellbeing functions, yielding services and wellbeing
either directly, or indirectly through production;
• the dimensions of the wellbeing frontier surround (or are wrapped around) the
production and utility functions as public goods that are potential sources of sig-
nificant positive externalities; and
• if the values of these externality functions tend towards zero, reflecting a degrada-
tion in economic and/or social infrastructure, and/or the quality of the environment
(environmental infrastructure), overall wellbeing also tends towards zero.
The government can undertake collectively-funded investments (on its own or
through private-public partnerships) towards generating positive externalities or
eliminating negative ones.
The government can also use scale (e.g., the size of its balance sheet) to
generate positive externalities through its contribution to intergenerational equity
(e.g., through retirement income policy).
The wellbeing framework identifies the domains of public policy aimed at dealing
with market failures, by complementing, supporting, and/or correcting (redirecting)
market forces where relevant and appropriate. The Crown balance sheet, regulation,
market creation, incentives, and taxes, are possible levers to give effect to such
policies.
The government’s reason for being is not exclusively to deal with market failures.
Although this is contentious, the government may also be assigned the task of
deliberately pursuing actions, taking a longer-term view than private individuals may
choose to do, that will promote the wellbeing of future generations yet to be born,
however far into the future that may be. This is where the selection of appropriate
social discount rates in different contexts becomes critical.
In addition, governments (both central and regional/local) also have a role to play
in building a consensus around a shared vision, within which various futures can be
accommodated. Ethics and values are critical stakes in the ground (see Mazzucato
2016).
All such government (i.e., collective) activities are intended to enhance wellbeing
either directly (say through improved social cohesion) or indirectly by increasing the
production of material goods, and they are funded by taxing the income generated
from the production of the material goods.
The broader role of a representative government is embedded in the wider inter-
generational wellbeing function represented by Eqs. (2.81)–(2.82) that we have been
86 2 Stylised Model
working with. The essence of this role is well captured by Mazzucato (2015), who
argues that when governed effectively and efficiently, “the State’s visible hand is firm
but not heavy, providing the vision and the dynamic push (as well as some ‘nudges’)
to make things happen that otherwise would not have.42 Such actions are meant to
increase the courage of private business. This requires understanding the State as
neither a ‘meddler’ nor a simple ‘facilitator’ of economic growth. It is a key partner
of the private sector – and often a more daring one, willing to take the risks that
business won’t” (p. 31).
We observe from Eqs. (2.81) and (2.89) that the impact of the wellbeing and eco-
nomic externality functions are of the Hicks-neutral type; they are multiplicative
and wrap around the utility and production functions, respectively. They affect the
potential aggregate wellbeing and economic growth that can be achieved, and their
evolution is endogenous. A “government” that is interested in enhancing intergenera-
tional wellbeing (either towards optimising intergenerational wellbeing in Eq. (2.81)
or at least maintaining d Wt /dt ≥ 0) would have to respond to the evolution of the
factors that affect these externalities.
There is another potential distinctive role for the government, that also provides
a positive externality for future generations, namely that it could adopt a longer time
horizon than a typical individual or business. To capture this additional role of gov-
ernment in our model, we introduce the following (expanded) version of Eq. (2.81),
which we refer to as the Chichilnisky wellbeing function, in reference to Chichilnisky
(1997)43 :
∞
1 y
W0 = ω t Ω (Et , St , Γt , Γt ) · u φ(Y /Y )Ct , Ct , Ct +
w E S d x h
t=0
(1 + ρ)
y
(1 − ω) lim Ω w (Et , St , ΓtE , ΓtS ) · u φ(Y d /Y )Ct , Ctx , Cth (2.101)
t→∞
where, ω ∈ [0, 1] represents the weight we put on the welfare of future generations
versus current generations, and:
y
lim Ω w (Et , St , ΓtE , ΓtS ) · u φ(Y d /Y )Ct , Ctx , Cth > 0 (2.102)
t→∞
Chichilnisky refers to ω as a measure of the tyranny (or dictatorship) of present

generations against future generations. The case where (1 − ω) > 0 is referred as
as representing “sustainable preferences”, where the preferences of both current and
future generations are reflected in the aggregate wellbeing function. Thus setting
ω ∈ (0, 1) is in effect a way of reducing ρ.
42 “Most of the radical, revolutionary innovations that have fuelled the dynamics of capitalism –
from railroads to the Internet, to modern-day nanotechnology and pharmaceuticals – trace the most
courageous, early and capital-intensive ‘entrepreneurial’ investments back to the State.” Mazzucato
(2015, p. 2).
43 See also Cowen (2007).
Generically speaking, the solution to the optimisation problem with sustainable

preferences will lead to a lower production and consumption path but higher well-
being than one where ω = 1 (see Chichilnisky 1997).
The government has two sets of policy instruments it can use to affect the aggregate
outcomes of economic and social activity: regeneration investments (often in cooper-
ation with individuals, businesses, and communities) through which it influences the
“equations
of motion” for public goods by adjusting the parameters or instruments
δ j , j ∈ E, S, Γ S , Γ E , Γ y , Λy , POV, K s ; and behaviour-affecting instruments such
as subsidies or penalties (operating through the “equations of motion” of private cap-
ital assets), through which it influences, among others, the mix of production, mix
of human capital through skilling, education and immigration, mix of research and
science, and the quantum and mix of physical capital investments. It pays for all this
through a tax on the production of the final output.
We write the government’s budget constraint as follows:
y
cΩw (Ωtw ) + cΩy (Ωt ) + cB (Bt ) (1 + χ ) + f inancecosts = τ Yt (2.103)
for t = 0, 1, 2, . . . . The first two terms on the left-hand side are the cost functions for
the delivery of positive externalities, or reduction of negative externalities, to individ-
uals and businesses, respectively. B refers to all government activities (subsidies and
penalties) directed at influencing the behaviour of private people, institutions, and
communities – all measured
in units of the final output (Y ). All three cost functions
cj (.), j ∈ Ωw , Ωy , B are assumed to be increasing in their respective arguments. As
in Acemoglu et al. (2016), the parameter χ > 0 represents the potential distortionary
effects, in terms of lost output, of all three types of government interference with
economic and social life, in recognition of the possibility that government activi-
ties may lead to wastage. The government pays for its activities through the taxes it
imposes on production, τ Yt .
Equation (2.103) highlights an obvious, but critical, point, namely that we need
production (income) to generate the revenue with which we fund the positive exter-
nalities and behavioural adjustments that deliberate collective action may lead to,
towards improving our overall wellbeing. There are potential costs if we do not get
it right. If, in trying to generate and fund these potential benefits, we blunt the desire
to work, invest and produce, we risk ending up with lower aggregate wellbeing than
we would have achieved with no collective action. Thus the stylised model captures
both the direct (τ Yt ) and the potential indirect (χ ) costs of government interference
on wellbeing through its effects on production, Yt .
Non-technical
1. Go back to Chap. 1 and then ask yourself what value this chapter adds to the
analysis - why do we need a mathematical model at all?
2. Consider the structure of the model developed in this chapter – what are the key
structural components of the model?
88 2 Stylised Model
3. What do you understand from the description “stylised model” - in what sense is
the model we develop in this chapter “stylised”?
4. Do you find the model developed in this chapter convincing - how did you arrive
at your conclusion?
5. If you were to change or extend this model, what would be the three aspects of
the model you would like to change - how?
Technical
1. What do we lose by not including a renewable natural resource in the stylised
model?
2. Write the generic production function (2.30) in the form of a Cobb–Douglas func-
tion, a Constant Elasticity of Substitution (CES) function, a Variable Elasticity
of Substitution (VES) function, and a Leontief function.
3. Derive Eqs. (2.24)–(2.28).
4. Derive Eqs. (2.34)–(2.38).
5. Derive Eq. (2.44).
Readers who wish to pursue the technical aspects of this literature are strongly urged
to study Acemoglu et al. (2012) and Arrow et al. (2012).
Chapter 3
Wellbeing-Enhancing Public Policy
Abstract The primary purpose of this chapter is to explore the policy implications of
the stylised model presented in the previous chapter. We first explore the steady state
properties of three versions of the model, including one that reflects the outcomes of
socially-optimal policies. Then, by way of demonstration of how the dynamics of the
model works, we trace some of the interactions between wealth, poverty-reduction,
social cohesion, environmental quality, output, technology, and overall wellbeing,
through a few simple numerical simulations. Both exercises highlight a key theme
of this book, namely, that when we are dealing with a multi-dimensional wellbeing
function, policy makers will always be looking for policy interventions that have
multiple and mutually reinforcing benefits. This will be a common theme across the
examples that we present in Chaps. 4 and (in the form of numerical simulations) 6.
Throughout these three chapters (i.e., Chaps. 3, 4, and 6), think of the policy question
in terms of Fig. 1.2 in Chap. 1. The question is, which of the capital stocks that sit
in the middle of the figure should we invest in, and how, towards expanding the
“wellbeing frontier”; this is the primary policy concern of a “wellbeing state”. Each
section of the current chapter has three sub-sections. First, a verbal and diagrammatic
presentation of the core ideas. Second, a mathematical representation of the verbal
and diagrammatic reasoning. Finally, a list of references (accompanied by a brief
discussion) of the empirical literature that attempts to bring evidence to support or
challenge the key assumptions of the theoretical model.
• Derive optimisation conditions for policy models and be able to interpret them.
• Explain what the “steady state properties” of a model mean - and how they are
different from the “equilibrium properties” of a model.
• Explain why the “steady state properties” of the three models presented in this
chapter are different. What are the fundamental differences?
• Explain why we would construct different types of models. How would we choose
which model to use?
• Critically assess what role evidence plays in the choice of models. What types of
evidence are used?

https://doi.org/10.1007/978-981-13-6104-3_3
90 3 Wellbeing-Enhancing Public Policy
• Combine optimisation conditions with market-clearing conditions to derive policy-

guiding solutions to stylised models.
• Assess and explain the policy implications of the three different models used in
this chapter.
• Assess and explain the implications of the three models for the choice of policy
instruments.
• Explain, using the full stylised model, how we would prioritise the choice of
policies and policy instruments.
• Use the models to discuss the complementarities between environmental, social,
and economic factors on wellbeing, and identify policy interventions and choice
of instruments that would be able to leverage these complementarities towards
increasing wellbeing on a sustained basis.
3.1 Introduction
Our focus in this chapter is on the analysis of the impact of public policy on inter-
generational wellbeing, and the channels through which policy operates.
As we pointed out in the previous Chap. 2, in the absence of deliberate public
policies there is a danger that, due to under-investment in various forms of com-
prehensive wealth, overall wellbeing will not be improving as much as it could.
Furthermore, whatever level it does reach may not be sustainable. To repeat, this is
not to deny that, even in the absence of government interventions (i.e., deliberate
collective actions funded through a tax on production), faced by pressures on social
cohesion, and economic and social resilience, the community would find ways of
gathering resources and investing towards addressing these issues. However, these
may be inadequate because the benefits of such actions may not be fully internalised.
Thus, through appropriate public policies, we may be able to enhance overall well-
being by improving environmental, social, and economic resilience to the systemic
risks that our comprehensive wealth is exposed to.
We first derive and compare the steady state properties of three versions of the
stylised model introduced in the previous Chap. 2, as a platform for the policy dis-
cussion that follows.
• The first version (see Sect. 3.2.1), which serves as a very useful benchmark, is a
closed-economy model with no banking sector and no government intervention
towards influencing environmental, social, and/or economic outcomes; the only
externalities present relate to natural resources and the environment. This is essen-
tially the laissez-faire version of the model presented and analysed by Acemoglu
et al. (2012).
• The second (see Sect. 3.2.2) is the stylised model presented in Chap. 2, but in the
absence of deliberate public policy (i.e., the laissez-faire version of the stylised
model).
3.1 Introduction 91
• The final version (see Sect. 3.2.3) is the full stylised model presented in Chap. 2,
with an explicit role for public policy towards enhancing intergenerational well-
being. We use it as a basis for policy analysis.
The model solutions provide time-paths for outcomes of economic and non-
economic dimensions of wellbeing, as well as overall wellbeing.
We then compare wellbeing outcomes, across model solutions that do and do not
incorporate deliberate collective actions to manage various externalities.
We show that, from an overall intergenerational wellbeing perspective, a laissez-
faire steady state is socially sub-optimal (see Acemoglu et al. 2012), and so are
policy interjections that are solely targeted at increasing economic growth. Material
wellbeing is essential for overall wellbeing, and as a basis for social progress, but
economic growth cannot continue unless it is socially and environmentally sustain-
able (see Duraiappah and Muñoz 2012). Furthermore, economic progress needs to
be broadly based if it is to foster social progress (see Friedman 2006).
There is no presumption that the government necessarily knows better or can do
better. Our purpose rather is to raise awareness of the wider spheres of wellbeing
(including material wellbeing), and highlight their relevance and complementarity
for a policy aimed at enhancing intergenerational wellbeing. Policy interjections that
ignore these interdependencies will lead to sub-optimal outcomes.
We start with the mathematical models that underpin these three steady state
outcomes.
3.2 Three Steady State Mathematical Models
3.2.1 Mathematical Model: Laissez Faire - A Benchmark

Model
The model here is essentially the closed-economy version of the stylised model intro-
duced in Chap. 2, with no banking sector. It excludes references to social cohesion,
equity, resilience, or the Chichilnisky extension. Utility is both a function of the
consumption of the single output (Y ) and the quality of the environment (leisure and
good health are excluded), with both directly entering the individual’s utility func-
tion. Thus on the consumption side, the equivalents of Eqs. (2.9)–(2.12) hold (with
the exclusions just referred to) and the production side of the economy is identical
to the one used in the stylised model, except for no differentiation between “skilled”
and “unskilled” labour – labour is perfectly substitutable between “clean” and “dirty”
technologies – and where “good health” affects neither the individual’s utility nor
the productivity of labour. As stated above, this is essentially the laissez-faire version
of the model presented and analysed by Acemoglu et al. (2012), and it serves as a
benchmark model, against which we compare the steady state properties of the two
models that follow.
The steady state of this model is represented by the vector of: final-good output
j j
(Yss ), unskilled wages (wss ), prices for machines (pi,ss ), demands for machines (mi,ss ),
labour demands (Ldj,ss ) for use with the two technologies j ∈ {c, d }, allocations of
scientists (or research) to “dirty” and “clean” technologies (Scd ,ss , Scc,ss ) respectively,
use of the non-renewable natural resource (R̃ss ), and the quality of the environment
(Ess ), such that:
• (Yss , R̃ss ) maximise the profits of the final-good producers;
• (pss
d
, Scss
c
) maximise the expected profits of researchers choosing to work with
“dirty” and “clean” technologies respectively;
• wss clears the labour market;
• the final-goods market clears as per Eq. (2.66) – setting τ = 0, financecosts = 0;
and
• the evolution of the exhaustible resource is given by Eq. (2.32), and that of the
quality of the environment by Eq. (2.84).
As Acemoglu et al. (2012) demonstrate, both the dynamics and the steady state
properties of this model critically depend on the degree of substitution between the
two technologies (“clean” and “dirty”) in the production of the final output.
The exhaustible resource is a key ingredient in the use of the “dirty” technology.
Going back to Eq. (2.50) which shows the relative expected profitability from research
with the two technologies, and assuming θy > 1 and c(Rt ) is non-increasing in Rt , as
the non-renewable (exhaustible) resource stock gets depleted, the market generates
the incentives for research to direct innovation towards the “clean” technology, and
leads to a rise in the demand for Lc relative to L (see Eq. (2.45)).
Acemoglu et al. (2012) prove that, under these circumstances, innovation will
eventually be directed to the clean technology only, and the long-run steady state
growth rate of the economy will be μηc . Provided the initial quality of the environ-
ment and the value of θy are sufficiently high, the switch to clean technology occurs
faster and an environmental disaster is avoided. We thereby achieve a positive and
sustainable economic growth rate, with a positive sustainable level of wellbeing.
However, even under its own terms, this steady state is not socially optimal (i.e.,
it does not maximise W (t) in Eq. (2.2), or its discrete equivalent). The fundamental
justifications for the latter conclusion (i.e., the sub-optimality of the laissez-faire
steady state) are four-fold (see Jones and Vollrath 2018, Chap. 5 and Acemoglu et al.
2012), all related to externalities or market structures:
• First, scientists are not able to internalise the positive externalities associated with
their research (researchers are not compensated for their contribution towards
improving the productivity of future researchers, so there is too little research).
• Second, the negative environmental externality created by the use of “dirty” tech-
nology is not fully internalised.
• Third, the private cost of extraction c(Rt ) does not reflect the total social (scarcity)
value of the exhaustible resource.
• Finally, there is the standard (static) monopoly distortion (the “consumer-surplus
effect”) – an inventor of a new design (for a machine) captures the monopoly profit
3.2 Three Steady State Mathematical Models 93
but not the entire potential gain to society of the invention, which results in too
little innovation.
3.2.2 Mathematical Model: Laissez Faire - Stylised Model
We now consider the laissez-faire steady state properties of our expanded stylised
small-open-economy model where, in addition to exhaustible resources and the envi-
ronment, we also include a banking sector which provides credit, as well as exter-
nalities associated with social cohesion, resilience to systemic risks, and equity. The
Chichilnisky extension is included but then suppressed (see below), because there is
no role for government in the laissez-faire model. The individual’s utility function
now includes, in addition to the consumption of the single output (Y ), also the con-
sumption of time (or “leisure” in its most general sense) and “good health”. Finally,
we now conceptualise the specialised skills embedded in “skilled” labour and scien-
tists as part of “human capital”, which requires investment to convert investment-time
into skilled labour or scientists.
The steady state of this model is represented by the vector of:
Yss – final-good output;
y
Pss – final-good price;
y
Css , Cssx , Cssh – consumption of products, time, and good health, respectively;
y
Xss – final-good export;
wss – unskilled wages;
s
wss – skilled wages;
Sc
wss – scientist’s wages or salaries;
j
pss – prices for machines; and
j
mss , j ∈ {c, d } – demands for machines;
j
mss – machine productions using technologies j ∈ {c, d };
Mc
Fss – stock of foreign “clean” machines imported;
Ajt – qualities of machines using technologies j ∈ {c, d };
Lds,ss – skilled-labour demand;
Ldss – unskilled-labour demand;
Scd ,ss , Scc,ss – allocations of scientists (or research) to “dirty” and “clean” tech-
nologies respectively;
Lss,ss – human capital formation through skilling (i.e., supply of skilled labour);
s
Scss – education (i.e., supply of scientists);
FssL , FssLs , FssSc – stocks of foreign skilled and unskilled labour and scientists;
l
iss , rssl , iss
d
, rssd – nominal as well as “real” loan, and deposit and interest rates;
l
Bss – bank loans;
d
Bss – bank deposits;
RFA
Bss – bank holdings of foreign risk-free assets;
R̃ss – use of the non-renewable natural resource;
Ess – quality of the environment;
Ksss – social capital;

p
Kss – physical capital; and
Kssb – bank capital;
Sss – degree of social cohesion;
y
Γss , ΓssS , ΓssE – degree of resilience to potential systemic risks to economic and
social infrastructure, and the environment, respectively;
Qss – degree of inequity;
POVss – degree of poverty; and
y
ss – economic infrastructure,
such that, these conditions are all satisfied at a steady state ss :
• (Yss , R̃ss ) maximise the profits of the final-good producers;
j j
• (pss , mss ) maximise profits of the producers of machines using each technology j,
j ∈ {c, d };
• (Scss d
, Scssc
) maximise the expected profits of entrepreneurs choosing to invest in
“dirty” and “clean” technologies, respectively;
• (wss , wss s
, wss
Sc
) clear the unskilled and skilled labour, and scientist markets, respec-
tively;
y y
• (Xss , Pss , FssMc , Bss RFA
) clear the international good and capital balance;
• Bss clears the bank-loan (credit) market for given nominal and “real” loan rates;
l
• all other market-clearing conditions set out in Sect. 2.7.2 hold; and
• the evolution of the exhaustible resource is described by Eq. (2.32); technology or
productivity by (2.47)–(2.48); the quality of the environment by (2.84); social cohe-
sion by (2.85); environmental and social resilience by (2.86) and (2.87), respec-
tively; poverty by (2.88); economic infrastructure by (2.91); economic resilience
by (2.92); social capital by (2.93); and inequity by (2.100).
The absence of government intervention inthe laissez-faire scenario is captured
by setting τ = 0 in Eq. (2.103) and δ j = 0, j ∈ E, S, Γ S , Γ E , Γ y , y , POV, K s in
Eqs. (2.84)–(2.88) and (2.91)–(2.93)1 ; as well as setting ω = 1 in Eq. (2.101) (this
automatically eliminates the Chichilnisky extension).
The differences between the two laissez-faire versions of the model, presented in
the previous section and the current section, can be attributed to the way the wellbeing
function is specified, including additional public goods through which wellbeing are
affected: social cohesion, equity, and resilience to systemic risks.
The fundamental economic-dynamic properties of the benchmark laissez-faire
model of the previous section are not affected by the extensions introduced in this
section, in the specific sense that economic growth is still driven by the growth and
composition of technology, in turn reflecting where innovations occur, as well as
the composition of the labour force. The introduction of immigration, and possible
imports of machines and exports of the single final good, do not affect this basic
1 Whether we set these policy-parameter values to zero or to an arbitrary minimum positive value
does not affect the qualitative results presented here. δ j , ξj , and γj , for all j, are treated as fixed
parameters for convenience.
3.2 Three Steady State Mathematical Models 95
outcome, although they do add other avenues through which the evolution of the size
and composition of the labour force, scientists, and machines are affected.
We now care about both the rate and the composition of economic growth, not
only because this may affect the potential sustainable growth rate of the economy,
but also because it affects equity, which in turn has an effect on social cohesion.
Another channel that is opened in this model, immigration, has an effect on both
potential economic growth and equity, as well as on social cohesion.
These additional externalities open up further channels through which the
expanded laissez-faire equilibrium of this section may not be socially optimal. It
is not only the externalities associated with environmental influences that may not be
fully internalised, but also those associated with the additional spheres of wellbeing
the stylised model introduces.
Market- and/or community-driven mechanisms may not be sufficient to fully
internalise the potential negative externalities resulting from the degradation of envi-
ronmental or economic-infrastructure caused by production (Eqs. (2.84) and (2.91)),
and/or the potential negative impact on social cohesion of immigration, or rising
inequity, or rising poverty (Eq. (2.85)). Based on the specification of the individ-
ual utility and aggregate wellbeing functions (Eqs. (2.9)–(2.12), (2.81)–(2.83), and
(2.101)–(2.102)), this leads to a potential failure to achieve the full wellbeing-
potential of the society and economy.
In the absence of deliberate regeneration policies directed to economic infrastruc-
ture, the environment, and social cohesion, ongoing economic growth, decreasing
equity, and deteriorating social cohesion will push overall wellbeing to a level that is
below potential. To repeat, this is not to deny that, even in the absence of government
intervention (i.e., deliberate collective action funded through a tax on production),
faced by pressures on social cohesion, and economic and social resilience, the com-
munity would find ways of gathering resources and investing towards addressing
these issues. However, these may be inadequate because the benefits of such actions
may not be fully internalised. Thus the scope and breadth of possible environmen-
tal, social, and economic policies need to be expanded beyond environmental and
economic growth considerations to help us get towards a socially optimum solution.
Furthermore, in this expanded model, we can also enhance overall wellbeing by
improving social and economic resilience to the systemic risks that our social and
infrastructures are exposed to.
3.2.3 Mathematical Model - Stylised
Now consider the full (stylised) model presented in this book, in Chap. 2.
The socially optimal (wellbeing maximising) steady state of this model is repre-
sented by the vector of:
• final-good output (Yss );
y
• final-good price (Pss );
• consumption of products, time, and good health respectively (Cy,ss , Cx,ss , Ch,ss );
y
• final-good export (Xss );
• unskilled wages (wss ) and skilled wages (ws,ss );
• prices for machines (pj,ss ) and demands for machines (mj,ss ), j ∈ {c, d };
• machine productions (mj,ss ) using technologies j ∈ {c, d };
• stock of foreign “clean” machines imported (FssMc );
• qualities of machines (Aj,ss ) using technologies j ∈ {c, d };
• skilled-labour demand (Lds,ss ) and unskilled-labour demand (Ldss );
• human capital formation through skilling (i.e., supply of skilled labour) (Lss,ss ) or
s
education (i.e., supply of scientists) (Scss );
• allocations of scientists (or research) to “dirty” and “clean” technologies (Scd ,ss ,
Scc,ss ), respectively;
• stocks of foreign skilled and unskilled labour and scientists (FssL , FssLs , FssSc );
• use of the non-renewable natural resource (R̃ss );
• quality of the environment (Ess );
• social capital (Ks,ss );
• degree of social cohesion (Sss );
• degree of resilience against potential systemic risks to economic and social infras-
tructure and the environment (Γy,ss , ΓS,ss , ΓE,ss ), respectively;
• degree of inequity (Qss ) and degree of poverty (POVss ); and
• economic infrastructure (y,ss );
that maximises the intertemporal wellbeing function (2.101)–(2.102), subject to the:
• parameter ω set through the political process;
• utility function (2.81), (2.82), (2.83);
• production function (2.89)–(2.90);
• the evolutions of: the exhaustible resource (2.32), the quality of the environment
(2.84), social cohesion (2.85), inequity (2.100), poverty (2.88), economic infras-
tructure (2.91), economic resilience (2.92), social capital (2.93), environmental and
social resilience (2.86) and (2.87), the quality (or productivity) of the machines
(2.58)–(2.48), immigration of machines and people (2.60)–(2.63), total physical
capital (or machines) (2.74)–(2.75), unskilled labour (2.69), skilled labour (2.14)
and (2.70), scientists (2.15) and (2.72), banks’ aggregate net worth (2.54), individ-
uals’ income-budget constraints (2.17)–(2.19), and the quality of health (2.16);
• market clearings for the final good (2.66), unskilled labour (2.67), skilled labour
(2.68), scientists (2.71), machines (2.73), exhaustible natural resources (2.76),
loans or credit (2.77);
• international payments balance (2.80); and
• the government budget constraint (2.103).
3.3 Policy Analysis 97
3.3 Policy Analysis
3.3.1 Policy in the Benchmark Model
In the benchmark model outlined in Sect. 3.2.1, the government, through appropriate
policies, can incentivise private actors to move the economy towards the “socially
optimal” solution. Acemoglu et al. (2012) show that this can be achieved through
the combination of the following policies:
• a “carbon tax” (i.e., a tax on the use of the “dirty” technology, or tasks);
• a Research and Development (R & D) subsidy for “clean” research;
• a subsidy on the use of all machines (using “dirty” or “clean” technology); and
• a permanent resource tax (all proceeds from taxes/subsidies being redistributed/
financed lump sum).
The intended collective impact of this package of policies is to switch produc-
tion towards the use of clean technology, with a view to increasing the long-term
sustainable growth rate of the economy.2
By way of demonstration, a subsidy for “clean” research, represented through an
adjustment to Eq. (2.49), would take the following form:
⎛ ⎞
2 j 1
α yt
1−α
1
= (1 + qt ) ⎝ψηj (1 + μ) Lt At−1 − (1 + rtl )wtSc Sct ⎠
j j j j
t −1
α ψ + rtl
(3.1)
where qt represents the subsidy rate.
It is worth highlighting that the subsidy in support of the use of machines is to cor-
rect for the under-utilisation of machines due to monopoly pricing in the laissez-faire
equilibrium. In addition, the “user-cost” of the exhaustible resource is determined
by the cost of extraction and does not reflect its scarcity value.
3.3.2 Policy in the Stylised Model
When the object of interest is material wellbeing, delivered through economic growth,
it might at first be intuitive to think of the other policy domains, represented by the
dimensions of the Wellbeing Frontier in Fig. 1.2, as constraints on the growth poten-
tial of an economy. If, instead, as we do in this book, one defines the object of interest
2 An important result highlighted in the Acemoglu et al. (2012) paper relates to whether government
policy intervention needs to be temporary or permanent in nature, to incentivise the switch of

production towards the use of clean technology, thus generating socially optimal outcomes. They
show that a critical influence on this choice is the degree of substitutability between the clean and
dirty technologies in the production of the final output.
as the enhancement of intergenerational wellbeing, then these dimensions will nat-

urally be seen as complements towards increasing our individual and community
wellbeing on a sustainable basis – they are all essential, complementary, ingredients
of wellbeing (see van den Bergh et al. 2015).
It is interesting, and informative, to note in this context that, “In the early days
of the European welfare state, social and economic policies were seen as work-
ing in the same direction. This view persisted for several decades. When, in the
United Kingdom, Beveridge drew up his 1942 plan for postwar social security, he
collaborated with Keynes to ensure that macroeconomic and social policy worked
together, notably via the role of social transfers in providing automatic stabilisers.
In the United States, Moses Abramovitz argued that ‘the support of income minima,
health care, social insurance, and other elements of the welfare state, was [...] a part
of the productivity growth process itself’. Only later in the 1980s and 1990s, did the
predominant view shift and come to see social protection as an impediment, rather
than as a complement, to economic performance” (Atkinson 2015, p. 265).
In this broader context, the wider wellbeing framework that we have been pro-
moting in this book, as a guide to good policy, following the many scholars we cited
earlier, can serve at least three separate but possibly complementary purposes.3
• First, as a a reminder of the wider dimensions of wellbeing that policy-advisers
should take into account in formulating policy advice.
• Second, hopefully in time as appropriate models are built and the public’s pref-
erences are able to be more rigorously ascertained and weighted, as a a guide for
the quantitative assessment of the tradeoffs between alternative policy options and
outcomes (see Au et al. 2015; Benjamin et al. 2014).
• Third, conditional on having a shared vision of how we wish to live as a society,
as a a framework and a guide for designing policies that may get us there.
It is the third purpose that is the most exciting and that provides the context for
our reference to the late, great Sir Paul Callaghan’s vision of New Zealand as a place
where talent wants to live. What may that vision look like?
The generic version of that vision is an expanding wellbeing frontier over time.
The specific features of that vision will reflect the peculiarities of each country (its
history, cultures, values, etc.). The ones that are particularly suited for a country like
New Zealand comprise a clean, blue and green, safe and prosperous, country where
people of different ethnicities, backgrounds, and beliefs live in harmony, and are able
to apply their individual talents towards improving their wellbeing. Talent (in all its
dimensions) is attracted to this country because it offers the potential to reap high
material rewards, as well as a great quality of life in its broadest sense.
In the absence of deliberate collective action, several tensions may arise in the
stylised model. Examples are possible tensions between: poverty and inequity, and
social cohesion; immigration and economic growth, and social cohesion; economic
3 Conveniently, the dimensions of the Wellbeing Frontier in Fig. 1.2 can also be used to measure the
success of policies, across time and countries, in expanding the frontier. The New Zealand Treasury
has done some work in this area (see Sadetskaya 2014; Thomson 2013).
3.3 Policy Analysis 99
growth, and environmental degradation and resource depletion; and economic growth
and equity.
A deliberate multi-dimensional strategy that aims to deliver our vision would
be one targeted at expanding the wellbeing frontier, the key dimensions of which
are resilience to systemic risks, potential economic growth, equity, social cohesion,
and sustainability, by focussing primarily on the complementarities between pol-
icy instruments and outcomes. A set of interconnected and complementary policy
interventions would be jointly targeted at growing, shaping, managing, appropriately
distributing capital stocks across society and across generations, and protecting them
against systemic risks. They would be effected by ensuring that the required eco-
nomic (including financial), environmental, and social infrastructures are in place.
The policies would operate through their influence on the “equations of motion”
that define the evolution of capital assets, and on the public goods that are wrapped
around these capital assets.
The policy measures would be underpinned by technology, by institutions (“rules
of the game”), and by cultures embodied in the country’s social infrastructure (e.g.,
the rule of law, well defined and secure private property rights, well functioning
financial markets and institutions). Although these measures offer the potential to
enhance our collective wellbeing, whether that potential will be realised or not will
depend on the effectiveness and efficiency of the choice of policy instruments and
how they are implemented – including their effects on private individuals’ incentives
to do the best they can for themselves and their businesses.
The initial values of the dimensions of the wellbeing frontier, as well as the initial
distribution of resources (such as physical and human capital), are “state variables”
(i.e., inherited initial conditions for the dynamics of the future). They all evolve
endogenously in response to economic, social, and political forces in pursuit of
higher individual and community wellbeing.
To give a flavour of the endogenous dynamic process at work, note that social
infrastructure includes the social and political institutions we have inherited. Inequal-
ity of access to resources creates inequality of opportunities and capabilities, and
negatively affects social cohesion. Policy responds by leaning against such inequal-
ity in an attempt to enhance social cohesion – and hence overall wellbeing. Policy
affects inequality through prices and incentives (via technology) – thus generating
“good growth” through effects on technology. This in turn enhances social cohesion
and thereby overall wellbeing. Thus we generate a virtual upward wellbeing spiral.
This endogeneity is an outcome of the responsiveness of a representative government
and the democratic process.
The government funds all its interventions through a tax on the production of
the single output. In doing so, it is aware of the complementarities and tradeoffs
between the costs and benefits of its interventions on overall wellbeing. Its purpose
is to enhance overall intergenerational wellbeing.
Table 3.1 Summary of equilibrium properties of alternative models

Benchmark Stylised W-Max
1 W̃ >0 −→ 0 W̃W −max > W̃LF2
2 gỹ μηc −→ 0 gỹ,W −max < μηc
3 ω̃ − 1 ∈ (0, 1)
4 Ẽ ∈ (0, 1) −→ 0 ẼW −max > ẼLF2
5 Q̃ − max Q̃W −max < Q̃LF2
6 S̃ − −→ 0 S̃W −max > S̃LF2
7 Γ̃ − −→ 0 Γ˜W −max > Γ˜LF2
3.4 Comparative Dynamics
We summarise the qualitative (comparative) equilibrium outcomes of the three mod-

els outlined in Sect. 3.2 above in the following Table 3.1.
The first substantive column of the Table, labelled Benchmark represents the key
steady state features of the model outlined in Sect. 3.2.1. The second column, labelled
Stylised relates to the steady state features of the model outlined in Sect. 3.2.2. Finally,
the last column labelled W-Max relates to the model of Sect. 3.2.3; this is the full
stylised model used as a platform for designing wellbeing-enhancing policies.
As we stated before, in the Benchmark model of substantive column one, inno-
vation will eventually be directed to the clean technology only, and the long-run
steady state growth rate of the economy will be μηc . This result is conditional on
suitable assumptions about the degree of substitutability between “dirty” and “clean”
technology in the production of the final output. Provided the initial quality of the
environment and the value of θy are sufficiently high, the switch to clean technology
occurs faster and an environmental disaster is avoided. Thus we achieve a positive
and sustainable economic growth rate, with a positive sustainable level of wellbeing.4
However, in the context of this book, this result is misleading because it does
not allow for the full set of externalities included in the Stylised model, with no
government (second substantive column). Once these additional externalities are
allowed for, there is no longer a guarantee that the model will converge to a sustainable
steady state with a positive output-growth rate and a positive level of wellbeing, even
under the favourable assumptions of the Benchmark model.
In any case, for reasons similar to the ones given for the Benchmark model, even if
a sustainable steady state with a positive output-growth rate and positive sustainable
wellbeing exists, it is most unlikely to be socially optimal for similar reasons to those
given for the sub-optimality of the model’s steady state outcomes. The reasons for
this result were outlined in Sect. 3.2.1 above, and are elaborated in Acemoglu et al.
(2012). What the last column shows are the comparative outcomes of the Wellbeing-
Maximisation model, with a direct role for government (or collective) action.
4 See Acemoglu et al. (2012) for a rigorous derivation of this result.

3.4 Comparative Dynamics 101
The key results can be summarised as follows:

• First, the steady state growth rate of the wellbeing-optimising model, should a solu-
tion exist, will be less than μηc [row marked 2 in Table 3.1]. To see this all we need
to do is to observe, by way of demonstration, that to the extent that the implications
of a positive growth rate on economic infrastructure [Eq. (2.91)], the implications
of the structure of production and of employment on equity [Eqs. (2.97)–(2.100)],
and the implications of the evolution of equity itself on social cohesion [Eq. (2.85)]
are excluded, and no policies are in place to correct for their social effects, both
externality functions and therefore the wellbeing function will tend towards zero
(or to a level below potential) [row marked 1].
• Second, whether such a sustainable wellbeing-optimising solution exists, and if it
does what the corresponding level of the economic growth rate will be, will depend
on whether the government can effectively and efficiently provide the external-
y y
ities summarised by the functions Ω w (Et , St , ΓtE , ΓtS ) and Ω y (t , Γt , Kts , ΓtE )
in Eqs. (2.83) and (2.90), respectively. These solutions will also depend on how
much weight we put on the wellbeing of generations in the very distant future [the
value of parameter ω in (2.101); row marked 3].
• Third, should we be able to switch production effectively and efficiently to the use
of “clean” technology [row marked 4], and also provide the appropriate environ-
mental, social, and economic infrastructures, we will be significantly increasing
our chances of enjoying positive sustainable wellbeing accompanied and supported
by a positive sustainable economic growth rate, high degree of social cohesion
(supported by low inequity and low poverty) [rows marked 5, 6], a clean environ-
ment, and good health [row marked 4], and enhanced resilience to major systemic
shocks [row marked 7]. That would indeed be a country (in our case, New Zealand)
where talent would want to live. Column 3 [row marked 1] in Table 3.1 captures
the outcomes of such a complementary collection of effectively and efficiently
implemented wellbeing-enhancing policies.
The critical additional tradeoffs that the optimisation conditions of the expanded
wellbeing-optimisation problem introduce, for consideration in formulating public
policy, include:
• the positive and negative impacts of immigration on skills and technological
progress, versus social cohesion, respectively;
• the direct and indirect wellbeing costs versus benefits, of investments in enhancing
environmental, economic, and social resilience to potential major systemic shocks;
• the direct and indirect impacts of the change of the structure of production and
employment on equity, and through that on social cohesion;
• the impact on overall wellbeing of the weight we put on the wellbeing of genera-
tions way into the future; and
• the costs of taxation, versus the benefits of the externalities generated by the use
of the funds collected through taxation.
What should be the focus, and critical ingredients, of an expanded policy pack-
age that would complement the policy recommendations of Acemoglu et al. (2012)
(targeted at enhancing the Benchmark model, as presented above)? The purpose of

this expansion would be to enhance the public goods (the key dimensions of the
wellbeing frontier) that are wrapped around comprehensive wealth in the stylised
model, while also capturing the complementarities and tradeoffs highlighted in the
preceding paragraph.
Note as well the broader context of the search for policies for a “wellbeing state”,
where the focus is on enhancing the capabilities and opportunities of people and
communities to pursue the lives they have reason to value, without judging what
those lives should be.
There is a very strong suggestion here that the primary focus of a wellbeing state
should be on building resilience to the systemic shocks that represent risks to the
way we live as individuals and communities.
It is evident from Eqs. (2.81)–(2.87) and (2.89)–(2.93) that an appropriate suite of
policies that target building resilience will enhance all five public goods dimensions
that surround the capital stocks which are the ultimate sources of wellbeing – directly
(by enhancing potential economic growth, equity, and social cohesion), as well as
by increasing economic, environmental, and social resilience to systemic shocks –
thus contributing to sustainability overall.
The common thread that runs across all the ingredients of this whole policy pack-
age (an expanded version of the one proposed by Acemoglu et al. 2012) is that they
all positively contribute to individual and community wellbeing by influencing the
growth, distribution, and protection of the capital assets that comprise comprehensive
wealth.
Start with the equations of motion (2.84)–(2.87) and (2.91)–(2.93), that drive the
externality functions (2.83) and (2.90) respectively. Then focus specifically on the
subset of these equations [Eqs. (2.86), (2.87), (2.92)] that describe the evolution of
the resilience functions. Policy levers are δ ΓS , δ ΓE , and δ Γy . Substituting the others
into those three, guide us towards direct investments, regulations, subsidies, and taxes
that:
• enhances the quality of the environment through encouraging a switch in the
composition of production towards the use of clean technology, which is the main
focus of the work of Acemoglu et al. (2012) (i.e., environmental ecosystem and
natural capital);
• enhances economic and financial infrastructure;
• encourages the immigration of skilled labour and scientists;
• enhances the wider health of the population;
• increases the domestic supply of skilled labour and scientists (i.e., various com-
ponents of human capital); and
• focuses on enhancing equity and eradicating poverty in its broadest sense of “exclu-
sion” (i.e., enhancing social capital).
In the context of financial infrastructure, the existence of money and a well func-
tioning banking and broader financial intermediation system are critical ingredients
of the required financial infrastructure (see Jakab and Kumhof 2015). Both the quan-
tum of credit, and allocation of credit through bank lending (between “clean” and
3.4 Comparative Dynamics 103
“dirty” technology firms), influences the potential growth rate of the economy and
its sustainability. Public policy can enhance the chances of finance playing its part
in the extension of the wellbeing frontier by:
• setting up institutions that ensure well defined property rights;
• enforcing contracts effectively and efficiently;
• achieving monetary, financial, and macroeconomic stability;
• enforcing regulation, competition policy, and financial openness; and
• enhancing social norms and other informal institutions that support trust and will-
ingness to take risks (see Fernández and Tamayo 2015; Stiglitz 2015).
3.5 Policy Examples
In this final section of the chapter, we provide some policy examples, and a few
numerical simulations, that are consistent with the policy recommendations outlined
above.
By way of introduction to the example and the numerical simulations that follow,
we provide some empirical evidence that supports the priority of poverty-reducing
policies as a platform for enhancing wellbeing on a sustained basis.
Breunig and Majeed (2016) provide empirical evidence in support of their pol-
icy advice to, “reduce inequality by attacking poverty rather than by redistributing
income,” on the grounds that this has a wider positive impact on wellbeing through its
positive impact on potential economic growth, the quality of institutions, and social
cohesion.
A useful summary of the measures used to capture multi-dimensional poverty can
be found in Toledo et al. (2016).
There is also a growing empirical literature on the relationship between the income
and socio-economic characteristics of individual households and the carbon content
of their consumption. According to the literature survey conducted by Berthe and
Elie (2015), there does not appear to be a systematic relationship between inequality
of income and pollution. Sager et al. (2017) attributes this partly to the limitations
of drawing such inferences from cross-country studies. When he then attempts to
quantify the ‘equity-pollution dilemma’ using micro-data on household consumption
within a single country (i.e., the USA), controlling for the scale (i.e., income and
level of consumption), composition (i.e., preferences), and technology effects on
consumption baskets, he finds, “Environmental Engel Curves (EECs) for CO2 to be
upward-sloping, concave, and shifting downwards over time” (ibid., p. 1). And then,
“Given the higher pollution intensity of consumption per expenditure by poorer
households, progressive redistribution [of income] may result in higher aggregate

pollution from consumption” (ibid., p. 34).
A set of policy recommendations to reduce multi-dimensional poverty, that are
fully aligned with the policy implications of our stylised model, are provided by
Atkinson (2015) and Boston and Chapple (2014).
We argued throughout the book so far, that the purpose of a wellbeing-focused
public policy should be to provide individuals and communinities with the oppor-
tunities and capabilities to live the kinds of lives they value. Access to “compre-
hensive wealth” provides a source of opportunities and capabilities. In this context,
the proposal by Smith (2017) is particularly relevant and interesting. He argues for
asset-based assistance for high-risk children. Having identified children who are at
high risk of future poverty (“wards of the state”), we would provide them with a rea-
sonably generous cash endowment at the age of 18, so that they have the foundation
for a positive start to their adult lives.
What follows is a real-world example of the anti-poverty policy process at work.
3.5.2 Example: Familias en Acción Urbano
This example is sourced from Galama et al. (2017). They start by acknowledging,
with references to the literature, that, “it is unclear whether anti-poverty programs
can have lasting impacts on economic wellbeing, subjective wellbeing and self-
sufficiency, while mitigating the costs to taxpayers. Critics argue, for example, that
anti-poverty programs are costly redistribution programs that promote government
dependency through reduced work incentives [...] Alternatively, these programs may
expand participants’ choice sets (e.g., through improved nutrition) enabling them
to take advantage of labor-market opportunities, and by increasing self-sufficiency,
reduce the burden on taxpayers” (ibid., p. 2).
They then go on to report: “we empirically address this question in the context of
Familias en Acción Urbano (FAU), a conditional cash transfer program implemented
at scale in the country of Colombia. FAU targets socio-economically disadvantaged
households in Colombia’s fourteen largest cities, which concentrate over 80% of the
country’s population. At full scale, Familias en Acción (rural plus urban expansion)
benefits over two million households nationwide. For eligible households, FAU pro-
vides cash transfers to families with children under the age of 18 conditional on
medical visits for younger children and school enrolment and continued attendance
for secondary school students. We take advantage of the fact that a means-test score
determines FAU eligibility. As we demonstrate with the aid of administrative baseline
data, the means-test score provides exogenous variation in program participation near
the program eligibility cutoff. We then reproduce the means-test score assignment
rule in a nationally representative Colombian living standards household survey
collected three years into the program. The survey measures multiple dimensions
of evaluative wellbeing as well as more objective measures that include income,
consumption, health and education, and that lacks the incentives for strategic report-
3.5 Policy Examples 105
ing because it was not designed to explicitly target program beneficiaries or track
program-related outcomes” (ibid., p. 2).
“Three years into the program, beneficiary households at the margin of eligibility
report greater income, consumption and formal employment participation for both
household head and partner. Impacts on household income at the margin are a factor
of ten times greater than would be expected solely on the basis of the cash transfers,
likely because of gains in formal employment. This finding casts doubt on the notion
that anti-poverty programs always generate work disincentives and is consistent with
recent evidence from other cash transfer programs. [...] Increases in consumption that
result from program participation map almost one to one to increases in income, and
there is a substantial increase on food expenditures among beneficiary households at
the margin, which may have influenced the quantity and quality of nutritional intakes.
Through enhanced choice sets, anti-poverty programs like Colombia’s conditional
cash transfer program may open new avenues for increased market opportunities
among beneficiary households” (ibid., p. 3).
“To summarize, our main contribution to this nascent literature is threefold. We
document how a conditional cash transfer anti-poverty program can simultaneously
improve material and subjective conditions for participants, and increase their self-
sufficiency. By simultaneously measuring objective and subjective measures of well-
being our results may shed light on potential pathways by which anti-poverty pro-
grams can improve wellbeing among households with unmet needs. Second, our
findings suggest that those goals can sometimes be achieved at a reduced cost to tax-
payers, to the extent that anti-poverty programs help enhance participants’ market
opportunities” (ibid., p. 5).
In an attempt to illustrate how our stylised model works we calibrate it in Sect. 3.5.3
below. We then examine if this model can reproduce the above findings.
3.5.3 Numerical Simulation
In what follows, we provide a brief summary of some numerical simulations, using

a simplified version of the stylised model of Chap. 2.5 All simulations are calcu-
lated over a twenty-year period. The purpose of these simulations is to explore and
demonstrate the interactions of wealth, poverty reduction, social cohesion, output,
technology, and environmental quality, in affecting wellbeing. These are some of the
linkages highlighted by the empirical literature cited in the previous subsection.
Figure 3.1 represents the outcomes of these simulations diagrammatically. It high-
lights the channels through which these various interactions influence overall well-
being.
Moving from the upper left-hand side diagram in Fig. 3.1, the first diagram rep-
resents the interaction between average per capita real wealth and poverty, based
5 We want to acknowledge the assistance provided by Ms Thakshila Gunaratna in the calculation

and diagrammatic representation of these numerical simulations.
Fig. 3.1 Numerical simulations
on Eq. (2.88). In these simulations, “poverty” is measured (or represented) by the

percentage of people whose ownership of comprehensive wealth is below a certain
threshold. The source of our long-run data on “comprehensive wealth” for New
Zealand is Qasim et al. (2018). Using the data presented there as a general guide,
we calibrate this particular equation by setting the values of δtPOV = −0.0009 and
γ V,POV = −0.000000006. The latter value is intended to highlight and capture the
very slow impact of an overall (average) increase in wealth on reducing poverty,
as suggested by the evidence provided in Qasim et al. (2018). The solid poverty
line is obtained with no impact of comprehensive wealth, so γ V,POV = 0 on this

line. The dashed line captures the impact of comprehensive wealth on poverty. We
note that poverty reduces from 20% to 19.2% over the twenty-year period due to
policy-independent factors.
We then assume that average per capita “comprehensive wealth” follows the long-
term improving trend shown in Qasim et al. (2018). We represent this trend numeri-
cally in this simulation as growth from $5,000 to $11,000 in 20 years, which assists
a further decline in poverty to 16.4%; see the dashed line. So, in our model, a 6% p.a.
growth of “comprehensive wealth” can reduce poverty by about 4 percentage points,
after 20 years (i.e., from 20% to 16.4%). This is consistent with our expectation
that an improvement in average per capita wealth alone will not suffice to make a
significant dent in poverty levels.
The second diagram represents the consequential positive impact, on social cohe-
sion, of the positive influence of increasing average wealth on poverty. This is based
on Eq. (2.85). To keep it simple and focus on some of the key relationships, we
calibrate the equation by setting ξ S,POV = 0.001 and δ S = 0.00001. We suppress all
other variables in the equation. These calibrations are informed by Green et al. (2011)
and Larsen (2014), who confirm that “social cohesion” (measured by survey-based
proxies for “generalised trust”) is very “sticky”, especially when we control for the
main influences (such as inequality of various types and corruption) on it. Thus, in the
diagram, social cohesion declines by only 0.9 percentage points over twenty years,
with the consequential improvement arising from the positive influence of increasing
wealth on poverty being very marginal as well (the dashed line in the diagram).
The third diagram in Fig. 3.1 (i.e., the first diagram in the second row) depicts
the consequential impact of an increase in “comprehensive wealth” on environmen-
tal quality. This is based on Eq. (2.84), when the effect of the output term is sup-
pressed (i.e., ξ E,POV = 0; otherwise, ξ E,POV = −0.003). The rate of environmental
improvement due to self-cleaning is represented by δ E = 0.001. We note that, in
this calibration, both the “natural” improvement in environmental quality (with no
production – the solid line), and the consequential positive impact on environmental
quality (through reduction in poverty) of an increase in average wealth (the dashed
line), are small. The parameter estimates are informed by the projections provided in
Conti et al. (2016) and related USA Energy Information Agency reports. We observe
that the estimates characterise an economy with little environmental self-cleaning
capacity and whose environmental quality is positively, albeit only mildly, related to
poverty reduction.
In the next (fourth) diagram, we do allow for the impact of production on environ-
mental quality – i.e., we “turn on” the output (Yt ) impact in Eq. (2.84). We assume
that per capita output (or real income) grows from $32,000 to $100,000 over the
20-year period. We proceed to study the impact, over a twenty-year period, of pro-
duction with “dirty” and relatively “clean” technologies as measured by the function
ξ E,Y (Ytd /Yt ) that varies with the share of output Ytd obtained via dirty technology,
in total output Yt . All lines start at ξ E,Y (Ytd /Yt ) = 0.000000016. On the dashed line,
the function ξ E,Y (Ytd /Yt ) reduces to 0.00000001 at t = 20. Function ξ E,Y (Ytd /Yt )
remains constant on the solid line over twenty years. Overall, we see that environ-
mental degradation can be reversed but only as a result of rather strong policies –
like one that would result in a 16-fold decrease in the use of dirty technology (see
the dashed line turning upwards).
Finally, in the last diagram of Fig. 3.1, we bring all these influences together
through Eq. (2.83), where we set the value of θΩw (the elasticity of substitution in
production) to 0.9. We have selected θΩw < 1 to reflect the assumption that all the
arguments of the function are complementary in nature in collectively generating
the externalities on wellbeing. To keep it simple, we set both ΓtE and ΓtS = 0 in
these numerical simulations. In this figure, we see five lines, each corresponding
to a policy scenario that results in a different time profile of comprehensive wealth,
output, and technology improvement. The fundamental point to note is that the largest
positive impact on the “externality multiplier” on wellbeing results from a change in
the composition of output – i.e., occurs through the switch to cleaner technology in
production.
It is also worth highlighting the positive impacts of a reduction in poverty.
Poverty reduction positively influences wellbeing through several, mutually-rein-
forcing, channels. It is good for the environment because poor people do not have
the choices to consume environmentally-friendly products.6 It is also good for social
cohesion because poverty is a source of all kinds of social tensions. Through these
impacts, it then creates a series of second-round positive influences on wellbeing
(including social and environmental resilience), which we have suppressed in our
numerical simulations.
Poverty-reducing measures provide examples of policy interventions that have
multiple benefits. These are the ideal types of interventions when we are dealing
with multi-dimensional and mutually dependent environmental, social, and economic
systems.
In the next Chap. 4, we construct a more “pragmatic” version of our stylised model
and use it to provide more examples of how these types of models work. We do this
through an analysis of the effects of different types of policy interventions. The com-
mon thread that runs through all these examples is the interaction of environmental,
social, and economic factors in influencing overall wellbeing. Policy advisers can
use these models in designing policy packages that exploit the trade-offs and com-
plementarities between these various influences, towards increasing wellbeing on a
sustainable basis.
Non-technical
1. Explain what you understand from the term, “steady state properties of a model”
– and how this is different from the “equilibrium properties of a model”.
2. Explain why the “steady state properties” of the three models presented in this
chapter are different. What are the fundamental differences?
6 This is referring to the composition, as distinct from the scale, effects of poverty reduction on
pollution.
3. Explain why we would construct different types of models. How would we choose
which model to use?
4. Critically assess what role evidence plays in the choice of models. What types of
evidence are used?
5. Assess and explain the policy implications of the three different models used in
this chapter.
6. Assess and explain the implications of the three models for the choice of policy
instruments.
7. Explain, using the full stylised model, how we would prioritise the choice of
policies and policy instruments.
Technical
1. Use the models to discuss the complementarities between environmental, social,
and economic factors on wellbeing, and identify policy interventions and choice
of instruments that would be able to leverage these complementarities towards
increasing wellbeing on a sustained basis.
2. Suggest a couple of policy interventions that would have significant and sustained
positive impacts on wellbeing – how did you go about coming up with your
selection?
3. What instruments would you use in implementing your policy choices, in the
answers you gave to the previous question? Explain.
Readers who wish to learn more about building and using integrated models towards
informing policy advice should study carefully Acemoglu et al. (2012). Also very
useful in this context is Arrow et al. (2012).
See also Galor and Weil (1999, 2000) on the desirability of building unified
models.
Chapter 4
Operational Policy-Simulation Model
Abstract We concluded Chap. 3 with a few numerical simulations, to demonstrate

how environmental, social, and economic factors interact in influencing wellbeing.
We deliberately used a very condensed version of the stylised model of Chap. 2 for
those numerical simulations. Our purpose was simply to highlight the operation of
some of these key interactions. If we wish to use such a model for policy purposes,
we have to work with a complete model. Is that feasible? Can the stylised model
of Chap. 2 be operationalised as a tool for the design and prioritisation of public
policies aimed at improving wellbeing? Building such an operational policy model
is currently work in progress at the New Zealand Treasury. The purpose of this chapter
is to summarise the modeling strategy that is guiding this effort. We do so with the
assistance of diagrams; the underlying mathematical model can be found in King
(2018), which is available online. The model represents the interactions among the
eleven key influences on wellbeing, which are the components of the OECD Better
Life Index (BLI) as presented in Fig. 1.1, as well as between these influences and the
capital stocks that comprise “comprehensive wealth” (as we discussed in Chaps. 1
and 2). Our discussion also covers the purpose of, and sensible uses for, the model, as
well as ways of interpreting and communicating the results of the policy simulations
to be performed with such a model. At the end of the chapter, we trace the results of a
few policy simulations with the help of diagrams, but in a qualitative way that simply
highlights the directions of responses to policy adjustments. The purpose of these
exercises is again, as in the simulations conducted at the end of Chap. 3, to highlight
the interactions between environmental, social, and economic factors in influencing
wellbeing, as a basis for guiding policy choices and prioritisation. The quantification
of these simulations, so that they can inform actual policies, is work in progress at the
New Zealand Treasury. Nevertheless, even the qualitative model is sufficient to guide
the policymakers in their choices of sets of complementary policies that are required
to guide the system towards an improvement in wellbeing – taking into account the
complementarities and tradeoffs between the responses to policy changes.
This chapter explains:
• How the components of the OECD Better Life Index (BLI) are used in constructing
an operational policy-simulation model.
https://doi.org/10.1007/978-981-13-6104-3_4
112 4 Operational Policy-Simulation Model
• How this model is broadly consistent with the full stylised model of Chap. 2.
• What the operational policy-simulation model would be used for – how it could
assist in policy choices.
• What guidance the operational policy-simulation model could provide in the choice
of policy priorities and/or policy instruments.
4.1 Introduction
Let us go back to Chap. 1 and start with Fig. 1.1. This figure represents the components
of the OECD’s Better Life Index (BLI), which in turn are the key influences on
wellbeing. We explained in Chap. 1 that the choice of these particular influences is
based on empirical analysis and research.
In the generic model presented in Sect. 2.2.1 (in Chap. 2), which is a model based
on Arrow et al. (2012), we conceptualised the components of “comprehensive con-
sumption” as corresponding to the components of the OECD’s BLI. Overall well-
being is thus a function of “comprehensive consumption”. The object of interest is
intergenerational wellbeing and its sustainability. Wellbeing is positively influenced
by comprehensive consumption, and “comprehensive wealth” (which is comprised
of various capital assets) is the source of comprehensive consumption.
Then in Sect. 2.2.2, we developed a particular benchmark stylised model, based
substantially on Acemoglu et al. (2012), to demonstrate the linkages between inter-
generational wellbeing, comprehensive consumption, and comprehensive wealth in
a dynamic optimisation context. We then used this stylised model as a platform on
which we built an enhanced model that integrates environmental, social, and eco-
nomic factors, and associated externalities, as essential and complementary influ-
ences on wellbeing – and therefore of primary interest for public policy.
Our emphasis in Chaps. 2 and 3 was on developing and then comparing the steady-
state properties of these benchmark and expanded stylised models – the latter includ-
ing public goods associated with resilience to systemic risks, social cohesion, sus-
tainability, and so on.
Here we present the modeling strategy for an “operational version” of the stylised
model of Chap. 2. The ultimate purpose of this work, which is in progress at the New
Zealand Treasury, is to construct a tool for the design and prioritisation of public
policies aimed at improving wellbeing. The model of this chapter and the stylised
model of Chap. 2 are not identical, but they share the following key features:
• Improving intergenerational wellbeing as the objective of public policy.
• The choice of the components of the OECD BLI as the main influences on well-
being.
• An integrated model of environmental, social, and economic factors affecting
wellbeing.
• The modeling of the interaction between the capital stocks representing “compre-
hensive wealth” and the key influences on wellbeing.
4.1 Introduction 113
• The recognition that various influences (such as good health and a clean environ-
ment) influence wellbeing both directly, and through their impact on productivity
and real income.
• Production uses both “dirty” and “clean” technology.
• A distinctive role for government as the collective agent for households, commu-
nities, and businesses for improving wellbeing.
• A small open economy model, with linkages to the rest of the world through flows
of goods and services, immigration, and capital.
• The market-facing influences on wellbeing are pulled together through a set of
market-clearing conditions.
The main differences between the two models relates to the assumptions made
about the actors (such as households, businesses, and so on) whose choices and
behaviours drive the model. All actors in the stylised model of Chap. 2 make their
choices on the basis of optimisation principles. The consequences of this choice, for
example in terms of the non-linearities it leads to in the aggregated (“macro”) version
of the model, are of little consequence or concern. In this chapter, on the other hand,
we adopt a more agnostic approach to the choice of behavioural assumptions, with
a view to keeping the resulting operational policy-simulation model as simple and
tractable as possible.
4.2 Model Overview and Structure
4.2.1 Links between Different Influences on Wellbeing
The purpose of this model is to support policy design, identify policy choices, and
help prioritise policy options, when the objective of public policy is to improve
intergenerational wellbeing. To that end, it is intended to serve as a tool for integrated
thinking about the key influences on wellbeing, including trade-offs and synergies
for policies. It is an operational version of the stylised model of Chap. 2. This means
it must be able to answer practical policy questions in a suitable timeframe. As a
tool for thinking, it should raise new questions, make implicit assumptions explicit,
allow the anticipation of otherwise unanticipated consequences, and break down
silos between different policy areas. As the model is focused on the trade-offs and
synergies between the different influences on wellbeing, it covers a significant breadth
of subjects. To remain tractable and simple enough to be useful, this breadth comes
at the sacrifice of depth and detail.
The quantification of the model is developed in King (2018). It is a basic top-
down stock-and-flow model including a Computable General Equilibrium (CGE)
model of an open economy. It is rich in detail about the behaviours of multiple
household types, businesses, production processes, international linkages, and the
role of government. Households can produce goods and services at home, have them
supplied by the government, or purchase them from firms. Firms buy labour and
capital from households for use in production, and can import and export goods. The
government taxes income and consumption, and uses those funds to provide goods
and services to households, pay interest on debt, and invest in physical infrastructure.
The model essentially consists of sets of direct and indirect influences on wellbe-
ing and their interactions. The direct influences on wellbeing are simply the eleven
components of the OECD BLI, as captured in Fig. 1.1. These do not form a com-
plete model by themselves. There are a number of supporting elements required to
complete the model (and represent the indirect influences on wellbeing), as well
as accommodating a variety of policy and other experiments within the model. This
includes a production sector (which influences wellbeing indirectly through its effect
on, among other things, the quality of the natural environment) and a government
sector, as well as interactions with the rest of the world (such as migration).
Many of the interactions between the different influences on wellbeing occur in
the “investment” equations of the model. These investment equations describe how
the stocks (also called capitals) in the model change from one time period to the
next, typically in response to changes in the stocks that relate to other influences on
wellbeing.
4.2.2 Modeling Strategy: Direct Influences
The evolution of the components of the OECD BLI (i.e., the direct influences on
wellbeing), obeys the standard equation of motion:
K t+1 = (1 − δ) K t + It (4.1)
What “drives” the model are the investment terms in each of these equations. These
also provide the vehicles through which public policy influences wellbeing.
These vehicles are described, in this chapter, verbally and with the help of dia-
grams.1
4.2.2.1 Social Connections (Community)
Humans are a social species, and as such good social connections are extremely
important for our wellbeing. The quality of family and other relationships are one of
the most important factors affecting subjective wellbeing. Good social connections
can improve our health, increase our resilience, and increase the likelihood of finding
a good quality job. Poor social connections lead to isolation and loneliness, as well
as making it much harder to get things done.
Like the other main influences on wellbeing in this model, social connections is
modelled as a stock (K ) that can be invested in for improvement, with the subscript
1 As already stated, a method for the model quantification can be found in King (2018).
4.2 Model Overview and Structure 115
SC referring to social connections. In that sense, K SC,t can be interpreted as part of

K s – the social capital in Eq. (2.30) (it is, however, unrelated to Sc – the science-
trained labour in Eq. (2.15)). Thus, the stock of social capital changes over time
according to:
K SC,t+1 = (1 − δ SC ) K SC,t + I SC,t ∀t (4.2)
where δ SC is the natural rate of decline in social connections.

The influence of social connections on wellbeing is then given by the level of social
connections (a procedure that is applied consistently for all of the direct influences
on wellbeing):
W SC,t = K SC,t ∀t (4.3)
As we will explain below, K SC,t and I SC,t , like all other variables in the model,
are initially measured in their natural units, but then converted to indices in the
operational version of the model.
While levels of social connection are significantly affected by factors outside this
model, such as the “friendliness of people”, there are a number of factors included in
the model which can also improve or worsen the chances of good social connections.
A change in any of these can cause a proportional change in social connections
(which assumes linearity that can, strictly speaking, be true only close to a steady
state). These factors include health-adjusted leisure time, volunteering, employment,
formal education, personal safety, health, and civic engagement. A large population,
high immigration levels, or high levels of wealth inequality can decrease social
connections:
I SC,t = ξ SC,i (Nt − Nt−1 ) (4.4)
i
where N refers to each of the individual influences on wellbeing referred to above

and ξ SC,i are their relative weights, identified by sociologists.
The evolution of all of the main influences on wellbeing, as well as their impact
on wellbeing, follow the same structural form – see Eq. (4.1); also (4.4), where the
subscript SC can be substituted for a subscript representing a different capital – and
therefore will not be repeated. What we will do below however, very briefly, is to
identify the main influences on each of the investment equations associated with each
of the direct influences on wellbeing.
4.2.2.2 Health
Good health is a crucial part of wellbeing. Without it, we feel uncomfortable at best,
and it also prevents us from improving other aspects of our wellbeing. In the worst
case scenario, poor health leads to (early) death, beyond which earthly wellbeing is
no longer relevant.
It also has large impacts on how we can spend our time. Poor health effectively
reduces our ability to use our time well, reducing the intensity with which we can
work, play, and study, and increasing the time needed for personal care, includ-
ing resting, and receiving medical treatments and monitoring. This applies both to
physical and to mental health.
Our health level is affected by many factors, some of which also relate to other
aspects of our wellbeing. There is a strong degree of continuity, with people who
currently have good health likely to continue to have good health in the future;
and those who do not, continuing to have poor health. When we do get sick, the
level of healthcare we have access to (relative to the level we need) determines how
quickly (or whether) we get better. Wealth, education, leisure, social connections,
and housing can all improve health for the better, as can life satisfaction and personal
safety. Unemployment, pollution, and a degraded natural environment can make it
worse.
The average level of health is modelled here as a stock, just like social connections,
and again evolves over time according to the standard stock-flow equation as in (4.2)
above.
As with the investment term associated with social connections (4.4), here too the
health investment term – change in health level – is determined by adding the change
in each of the factors that affects health, multiplied by an effectiveness parameter for
that factor. This assumes that each of these factors acts independently and linearly
on health. The factors that affect health are the excess of demand for health services
above supply; the change in biodiversity; the change in the wealth, formal education,
quantity of housing, and unemployment of the household type; life satisfaction;
social connections; and personal safety of the household. Housing quality and health-
adjusted leisure are included relative to an “ideal” value. The level of pollution is
used here, rather than the change in pollution, as it is the level of pollution, and not
the change in pollution, that is important for health changes (unlike other variables
that affect health).
Most of these values are determined in other parts of the model, demand for and
supply of healthcare being the exception. Demand for healthcare is modelled as a
function of current health, with healthier people requiring less healthcare than those in
poorer health. This way of modelling healthcare assumes that demand for healthcare
increases rapidly as health levels decline. Note that it does not take account of the
increase in demand for healthcare services due to greater wealth or income which
can create higher expectations.
Healthcare supply, on the other hand, is given by the total of home production (e.g.,
carers), market purchases of healthcare, and government supply. Thus, supply and
demand of healthcare services do not necessarily match, and if demand for healthcare
services exceeds supply, this will have a negative impact on health through the health
investment equation.
4.2.2.3 Jobs
While many people bemoan the need to have a job, it is generally recognised that
the benefits of having a good-quality job extend beyond the income it provides.
A good-quality job can provide people with meaning and purpose, independence,
stability, status, and skills development; as well as the means to provide for their
economic wellbeing. Being unemployed (i.e., willing to work but not having a job),
is generally considered bad for one’s wellbeing, creating financial stress, boredom,
and negatively affecting self-esteem.
The evolution of jobs is one of the influences on wellbeing that is not modelled as
a stock-flow equation. The employment rate and the long-term unemployment rate
comprise two of the indicators for the influence of jobs on wellbeing. The other two
indicators are wage rates and a measure of job security. The impact of the influence
of jobs on wellbeing is measured as the arithmetic mean of these four job-related
factors.
4.2.2.4 Income and Wealth
The OECD BLI (see Fig. 1.1) includes income and wealth as the main components
of what we refer to as the direct “economic” influences on wellbeing.
In the model presented in King (2018), the economic influences on wellbeing are
jointly represented by the simple average of the indexes for three economic variables
– net income, total consumption, and financial wealth. This follows the method used
by the OECD (see Durand 2015), where the indicators, after conversion to indexes,
are combined using simple arithmetic averages to derive indexes for each aspect of
wellbeing.
In the model, income may be generated from the return on the ownership of finan-
cial wealth, and wages earned from work. Wealth, under the “economic” category,
refers to financial wealth. Consumption refers to the consumption of market goods
and services.
4.2.2.5 Environment
The quality of the environment we live in influences our wellbeing in several differ-
ent ways. High levels of pollution negatively impact physical health. Lack of green
spaces negatively affects mental health. Production depends on accessing natural
resources and being able to emit pollution to natural sinks. Biodiversity contributes
to technological progress (for example, biomining for new drugs; biomimicry in
engineering). A natural, unpolluted environment also provides a variety of recre-
ational opportunities, and its existence is valued for its own sake by many people
who may never visit an untouched habitat.
In our model, three aspects of the environment are considered: resources, pollu-
tion, and biodiversity. Both the level of pollution, as well as a measure of biodiversity,
are used as indicators for the environmental influences on wellbeing. On the other
hand, while the level of resources is important for production (as discussed later in
this chapter), it does not have a direct impact on wellbeing.
Pollution is emitted into the natural environment by production. The environment

can absorb and neutralise some pollution, but pollution flows in excess of this level
buildup in the environment, damaging biodiversity and creating health risks. Most
types of pollution form geographic concentrations, often in the area they are emitted,
but sometimes elsewhere such as polluted streams emptying into a lake. A few forms
of pollution do not become geographically concentrated, but have global impacts,
such as carbon dioxide.
The total level of pollution experienced by society is again governed by a standard
equation of motion, of type (4.1). The investment term in that equation has two key
contributors, namely the level of pollution emitted by production in each time period,
and the spending (and effectiveness of that spending) by government primarily on
clean-up.
Biodiversity is represented as an environmental stock. It is valuable as an enabler
for production and for health, as well as in its own right. As a stock, its evolution is
again governed by an equation of type (4.1). The investment term in that equation
has two key contributors, namely spending by government on increasing biodiversity,
and the negative effects of pollution on biodiversity.
4.2.2.6 Personal Safety
Personal safety is primarily an indication of the crime (particularly violent crime)

level in the society – the lower the level of crime, the greater the level of personal
safety. People who have a lack of personal safety will usually be living their lives in
fear. This has effects on their physical and mental health. It may restrict their activities,
indirectly affecting their ability to earn an income, get educated, and generally enjoy
life. Trust is reduced, impacting negatively on communities and business.
The evolution of the average level of personal safety is modelled as a standard
equation of motion. The main contributors to the investment term in that equation,
multiplied by an effectiveness parameter for that contributors, are the excess need
for policing and other crime prevention above the supply of policing and crime
prevention services, social connections, net income of households, and the level of
formal education. Poor mental health and long-term unemployment increase the risk
of crime and therefore decrease personal safety.
Most of these contributors are determined elsewhere in the model, demand and
supply for crime prevention being the exception. Demand for crime prevention is
modelled as a function of current personal safety, with safer societies requiring less
crime prevention than those that are more dangerous.
Supply of policing and security, on the other hand, is given by the total of home pro-
duction (e.g. good parenting, escorting family members at night), market purchases
of security and prevention, and government supply of crime prevention services
including policing, rehabilitation, and prevention.
The level of personal safety is then used as impact-measure for the influence of
personal safety on wellbeing.
4.2.2.7 Housing
The need for shelter is one of the basic necessities of life, and the quality of that shelter
can have a large impact on wellbeing. The location of the housing is important – if it is
not close to friends and families, schools, workplaces, and amenities, too much time
and money must be spent on transport or alternatively social connections, education,
and income will suffer. This is also dependent on the type of transport available.
Good quality housing also has enough room for its occupants. Overcrowding
increases the risk of infectious diseases and can also be stressful and unpleasant
for the occupants. The quantity of housing is also closely linked to the price of
housing – the cheaper housing is, the less likely it will be overcrowded to save
money. The price of housing is also important in its own right. Because housing
is a necessity, it is usually a high priority in a household’s budget, and expensive
housing reduces the amount of money available for other consumption including
food, healthcare, and education.
The arithmetic mean of the indexes of these four indicators, namely quality, quan-
tity, price, and location of housing is used in the model to represent the impact of
housing on wellbeing.
4.2.2.8 Work-Life Balance
We all only have twenty-four hours in a day, but we can choose to use that time in many
different ways. Some of that time is spent in activities that are directly pleasurable,
and some is spent doing things that are necessary for (indirectly) improving our
wellbeing such as working. Work-life balance describes the trade-off between two
aspects of modern life – the time desired for leisure and personal care on the one
hand, and the time that must be spent working (both paid work and home production),
on transportation, and on education, in order to live and improve other aspects of
wellbeing on the other hand.
In this model, the time available for leisure and personal care is defined as the time
left over after the other activities have taken place, as a proportion of time. Health
and social connections, in particular, are then identified as the main influences on
the conversion of that leisure time to “quality time” – the individual’s ability to
enjoy and make the most of that time, and in the amount of time they are able to
dedicate to personal care and rest. Thus, “quality adjusted” leisure time becomes the
impact-measure of work-life balance on wellbeing in the model.
4.2.2.9 Education
Many people value education, knowledge, and skills for their own sake. Learning
new skills and knowledge often creates a sense of achievement and satisfies our
curiosity, as well as increasing employability, and making labour more productive. It
also assists with civic engagement, and improves our ability to look after our health,
as well as the environment.
However, it takes a significant investment of both time and income to obtain a
good education and acquire a reasonable level of skills. People spend an average
of 17.5 years in formal education in OECD countries (see OECD-Education 2017)
mostly before entering the workforce. They then continue to develop skills on the job
(and elsewhere) throughout their working career and sometimes beyond. Typically,
the government will pay for most of the cost of primary and secondary education,
but a much smaller proportion of tertiary education.
In this model, we allow for two types of education-related human capital – formal
education and skills. Formal education occurs outside the workplace, typically (but
not exclusively) at a school, university, or polytechnic. It takes time away from other
activities a household may perform (including earning an income), and must be paid
for either by the government or the household. It often leads to a qualification.
Skills are defined as the result of on-the-job training and experience, and are
harder to measure. They are effectively paid for by producers, who must sacrifice
productive time from their employees to invest in their skills. In return, producers
benefit from the increase in skills through improved productivity. As much of this
training is provided on-the-job or through other forms of experience, there are no
direct costs to be paid for.
In the model, both formal education and skills evolve over time according to a
standard equation of motion. The investment terms in those separate equations are
influenced by the quantity of effort put into increasing formal education and skills
respectively.
The arithmetic mean of these two stocks (i.e., the levels of formal education and
skills) is then used as a measure of the impact of education on wellbeing.2
4.2.2.10 Subjective Wellbeing
Subjective wellbeing is measured by life satisfaction. It is broader than happiness,

and includes how people feel about their lives as a whole rather than their current
emotional state. It is commonly measured using the Cantril Ladder (see e.g., Adler
and Fleurbaey 2016), with people asked to give their position on a ladder, where zero
represents the worst possible life and ten represents the best possible life for them.
There are many aspects of life that affect life satisfaction and are included in this
wellbeing model. Health, net income, social connections, formal education level,
job security, personal safety, and altruism (measured as time spent volunteering)
all improve life satisfaction. Unemployment, time spent commuting, pollution, and
whether you are less financially wealthy than your neighbours, can decrease life
satisfaction.
The level of subjective wellbeing represents its impact on wellbeing in the model.
Subjective wellbeing is determined by the history of each individual, with more
2 King (2018) provides the sub-models for formal education and skilling.
recent events having a larger effect, and negative events also having a larger effect
than positive events. Events that affect subjective wellbeing include changes in health,
income, employment, and wealth, among others.
4.2.2.11 Civic Engagement
Civic engagement is the term used to describe how engaged people in the society are
with the way they are governed.
In this model, civic engagement is treated as a stock that changes as other aspects
of society change. As such, its evolution is again represented by a standard equation
of motion.
The main contributors to the investment term in the equation of motion are the
change in education, net income, social connections, immigration, and employment.
It is also affected by the change in leisure time available, without which people may
not have the time to engage with government even if they have both the desire and
the ability.
The influence of civic engagement on wellbeing is then represented by a measure
of the stock of civic engagement.
4.2.3 Modeling Strategy: Production Sector
4.2.3.1 Overview
As we already observed above, the eleven key influences on wellbeing, which have
been described in the previous section, do not form a complete model by themselves.
There are a number of supporting elements required to complete the model, and to
allow for a variety of policy and other experiments within the model. This includes
a production sector and a government sector, as well as interactions with the rest of
the world.
Production is the process that converts a certain set of inputs into a set of outputs,
which provide some of the sources of wellbeing. At the highly abstract level required
in this model, the inputs include capital and labour, and production is enabled by
technology, personal safety, infrastructure, social connections, and the natural envi-
ronment. The outputs include the desirable consumption goods and services, as well
as undesired pollution.
The production sector is an important supporting element of wellbeing for house-
holds and general society. The production sector provides the jobs that enable job-
related wellbeing. Those jobs, and the use of capital, provide the income for the
economic aspect of wellbeing. They also provide most of the goods and services
for consumption (again contributing to the economic aspect of wellbeing). Housing,
healthcare, transportation, policing and security, and education services are provided
by the production sector. Production also affects the environment, albeit negatively.
In this model, production is treated similarly to how it would be treated in a static

CGE model, as is household consumption choices, and provides some of the key
components for our numerical simulations. The closure of the model is provided by
a combination of variables exogenous to the whole model, as well as the variables cal-
culated using the other parts of the model, particularly the stock and flow equations.
Here we outline the key components of the model with the help of diagrams.
There are several different divisions in the way production is modelled here. First,
goods and services are divided into extracted goods and produced (final) goods. The
former do not have any intermediate inputs and are not available as consumption
goods. The latter use extracted goods as intermediate inputs and are available for
consumption. Second, goods and services can be produced in different ways, based
on the environmental friendliness of the method of production. Finally, many of the
produced goods are important for aspects of wellbeing beyond the economic aspect,
such as for housing or healthcare, and these are produced by specific production
sectors.
4.2.3.2 Extraction
Extraction goods are used for intermediate consumption by final production sectors,
but are not available for final consumption by households or government. Conceptu-
ally, they include primary production goods like coal, fish, water, and sunlight that
are available “for free” from the environment, but must be extracted using labour
and capital in the form of machinery. For modelling purposes, we assume that no
intermediate inputs from other industries are required to produce extraction goods.
Production of extraction goods is enabled by levels of social connections, personal
safety, biodiversity, and government-provided infrastructure such as roading. It is
also enabled (unlike final goods) by the level of resources available, as the more
resource there is, the easier it is to extract. These enablers of production are things
that are necessary for production, but are neither used up nor directly paid for by
extraction producers. This is illustrated in Fig. 4.1.
Two generic extraction goods are modelled – a flow resource and a stock resource.
4.2.3.3 Final Production
Final production goods are all other (i.e., excluding extraction) goods and services.
They are sold directly to the final consumer, which may be the government or house-
holds. They require extraction goods as well as factors of production such as capital
and labour for their production. They also require most of the same enablers as
extraction production, namely social connections, personal safety, biodiversity, and
government-provided infrastructure. This is illustrated in Fig. 4.2.
The final goods production has a number of sectors, several of which are used
for specific purposes for other parts of the model. Education, healthcare services,
transport, policing and security, and good and bad quality housing, are all included
Fig. 4.1 Production process for extracted goods
as final production sectors and described further in relation to the appropriate aspects
of wellbeing. In addition, there is an “other” sector for all other types of goods and
services produced for consumption in the economy.
4.2.3.4 Clean and Dirty Processes
The environmental friendliness of the production process is modelled using “clean”

and “dirty” production processes, very much in accordance with the benchmark
Fig. 4.2 Production process for final goods
stylised model of Chap. 2. Any given production process for any particular good sits
somewhere on a continuum between “clean”, which is low-polluting and generally
environmentally friendly, and “dirty”, which is not. In this model, we do not use the
function ξ E,Y (Ytd /Yt ) (see Eq. (2.84)). A dividing line is placed on the dirty-clean
continuum, halfway between the two extremes, and all processes on the clean side are
grouped as “clean”, and all others (on the dirty side) are grouped as “dirty’. Clean
and dirty processes are modelled using the same equations, but the parameters in
those equations can take different values, which will be averages of all the different
processes in the group. Note that although the dividing line is placed halfway between
“clean” and “dirty”, this does not necessarily mean that half of production is clean
and half is dirty.
We assume that pollution is produced by each process in proportion to the level of
production using that process, for each extraction or final good. Clean processes will
produce less pollution per unit of production than dirty processes, but the modelling
method is the same for both. Note that a final goods producer could reduce their
level of pollution by reducing this parameter (e.g., switching to a cleaner process),
thus modifying the position of the producer on the dirty-clean continuum, and also
by reducing the proportion of extraction goods required in production.
4.2.4 Modeling Strategy: Market Clearing
This model assumes that all markets – both for factors of production, and for goods
and services – clear. Thus, demand equals supply for all factors of production, and
goods and services, and this is used to determine the prices of the factors of pro-
duction, and the quantities of each good or service. The full set of equations can be
found in King (2018).
4.2.5 Modeling Strategy: Rest of the World
This model is for a small, open economy, and thus interactions with the rest of the
world are important. These interactions may take the form of imports and exports of
goods and services (both extraction and final), of capital, and of people (migration).
The mathematical models for these interactions are proposed in King (2018). There
are also other interactions with the rest of the world, most notably the interchange
of ideas, that are not included in this model.
As this model is for a small economy, it is assumed that this economy does not have
an effect on international prices (in international currency) or international levels of
personal safety and so on, and so they are exogenous. For simplicity, they are also
chosen to be constant over time.
The exchange rate is chosen for simplicity to be also exogenous and constant.
The exchange rate can then be used to convert the international prices for imports
and exports into local currency.
4.2.6 Modelling Strategy: Government
The government makes up a large component of the economy, and it has the ability to
influence much of the rest of the economy and society more broadly. It is also tasked
with running the country. In a democracy, governments that do a better job of run-
ning the country well – presumably by at least maintaining the wellbeing of its
citizens – are more likely to be re-elected. Thus, governments are motivated to
improve the wellbeing of their citizens, and have a number of tools to achieve
this. This means that including a government sector in this wellbeing model is
highly important, and even more so as the model is intended for use as a numer-
ical government-policy simulation tool.
4.2.6.1 Fiscal Policy
The main role of the government in the model is to raise income through taxes for
spending and redistribution. Three types of taxes are collected in the model: income
tax, consumption tax, and pollution tax. Note that this does not include any form of
company or production tax. The government uses this income for spending, investing,
and providing transfers back to households. It must also pay interest on any debt it
owes. However, it does not need to maintain a balanced budget, and in any given year
it can lend or borrow to make up the difference between income and expenditure.
For the purposes of this model, we assume that the government requires govern-
ment debt to be maintained at a fixed ratio of GDP.
4.2.6.2 Spending
The government spends its revenue for a variety of different purposes, and part
of the government budget process is deciding how to divide the available funds
among these options. In this version of the model, the government can spend its
funds on decreasing pollution, increasing biodiversity, increasing personal safety,
on providing healthcare and formal education services, and on other consumption
goods and services. These are decided exogenously, and are kept constant. Most of
these are decided at the societal level.
4.2.6.3 Government Wealth
The government carries a significant level of wealth on its balance sheet, and this is
assumed to be physical wealth in the form of infrastructure, which is an enabler of
production. As with other stocks in the model, infrastructure levels change according
to the standard evolution equation. There are many different choices governments
could make for investment in infrastructure. Here, we assume that infrastructure
investment occurs at a rate that counteracts depreciation and allows for population
growth. This essentially keeps the level of infrastructure constant relative to popula-
tion. This provides for a baseline level of maintenance, but does not take into account
common variations around this that are typically due to more ad-hoc decisions by
the government.
4.2.6.4 Transfers
Transfers are the cash and in-kind payments made to households, usually to redis-
tribute wealth within a society or to alleviate poverty. These are modelled in two
groups. First, are the transfers that could be defined as social welfare payments that
vary according to the conditions in the economy and society. They include unem-
ployment benefits (paid to all who are unemployed), and other payments such as
government-provided support to those who need it. The remaining transfers are cal-
culated as a residual so that the government can achieve its fiscal policy.
4.2.7 Modelling Strategy: Calculation of Indexes
In this model, variables are initially presented in their “natural units”. These are the
units that they are usually measured in, for example, New Zealand dollars for income,
time-proportion for time (convertible to hours per day by multiplying by 24; or days
per year by multiplying by 365), and proportion for level of employment. This means
that each variable in the model has its own units, which may or may not be shared
with other variables.
Using natural units makes it easier to interpret the results, and it avoids problems
such as finding shadow prices to express all variables in dollar terms. However, it
leaves us with the challenge of how we compare different variables in a meaningful
way. Which is larger, if income has increased by $1,000 per year and leisure time by
1 h per day?
To get around this problem, all variables that are used as indicators of wellbeing
are calculated both in their natural units and as an index. The indexes can then be
compared to one another and used in calculations that combine them. This then
creates the next problem: how should the indexes be calculated?
There are many different ways indexes can be calculated, some of which are
discussed in an OECD handbook Nardo et al. (2005). For the purposes of this model,
there are some particular features that it would be desirable for the indexes to exhibit.
These are as follows:
• Shows changes consistently over time: the variables in this model evolve over
time, and it is important to be able to tell how things are changing in time.
• Shows differences between household types consistently: different types of house-
hold may respond to changes in different ways, and this needs to be visible to
compare the effects of wellbeing on the different household types.
• Both good and bad effects move in the same direction: indexes are easier to combine
and interpret if up always means better. However, for some variables such as
pollution, an increase is a bad thing. For such variables, the index should increase
if the value of the variable decreases.
• Relative changes should only appear if there are also absolute changes: this means
the relative values should be calculated by comparing to fixed values.
• Uses only data available in the model: data from external sources can be hard to
come by for this model.
Indexes in this model are calculated as the percentage improvement from a par-
ticular base value. The base VB is given by the societal value of that variable at time
t = 0. In the usual situation where the results of the model are given as the difference
between a business-as-usual case and a scenario, the base value will always be the
business-as-usual value at time t = 0. Using this method, an index value of 1 means
the variable is the same as the base. An index value of 1.1 means the variable is 10%
better (not necessarily larger!) than the base; and an index value of 0.9 means it is
10% worse than the base.
As a single base is used for all time periods, this method shows changes consis-
tently over time. By using the same base for different household types, the differences
between them can be distinguished and are treated consistently. Because the inverse
calculation is used for variables where an increase is bad for wellbeing, an increase
in the index always represents an increase in wellbeing. The choice of a fixed base
means changes in the index will only appear if the value of the underlying variable
changes. Finally, the choice of base described here is calculated as a fundamental
part of the model. Thus this method satisfies the important features required for an
index in this model.
4.2.8 Modelling Strategy: Calculation of Wellbeing
Wellbeing is a heterogeneous concept, made up of many different aspects. Any person

or household will put a unique weighting on each aspect that may change over time.
Thus, many different indexes for wellbeing have been proposed, but there is no “best”
index for all people, or even a large group of people (see Adler and Fleurbaey 2016).
All such indexes require value judgements unique to any individual. For this reason,
the emphasis on the model results should be on the indexes for each of the main
influences on wellbeing and the values that are used to calculate them.
Nevertheless, there is some value in calculating an overall value for wellbeing.
In particular, it can provide a consistent way for assessing different policies which
contain different trade-offs, provided the results are taken as approximate and the
reasons for the results understood.
The choice of mathematical form for combining the key influences on wellbeing
is in many respects arbitrary. However, it is desirable that it shows diminishing
improvements in wellbeing from increases in any particular influence on wellbeing,
and that the consequence of any particular influence on wellbeing reaching zero is
catastrophic. While a number of equational forms might meet these criteria, we have
chosen a weighted geometric mean (a standard Cobb-Douglas-style equation) for
period wellbeing.
In addition to the overall wellbeing calculated for each time period, an intergen-
erational wellbeing value (i.e., wellbeing aggregated over time) can be calculated to
aid comparisons. People may be willing to sacrifice some wellbeing now for greater
wellbeing in the future, and consider their overall wellbeing throughout their life to
be greater for that sacrifice. Note however that intergenerational wellbeing is even
less meaningful than the overall period wellbeing, as it retains all the problems of
that value, and then adds in the problems of time discounting, which are also driven
by value judgements unique to any individual.
Typically, good things in the future are rated less highly than good things now
for a number of reasons including the fact that the future is uncertain. This implies
that the future is discounted, and we assume it is discounted at a constant rate. We

assume that we can write intergenerational wellbeing as the discounted sum of the
wellbeing in each period. However, this does imply that the wellbeing of people in
the distant future is discounted to almost zero and therefore unimportant relative to
the present – sometimes known as the dictatorship of the present.3
4.3 Policy Simulations
Before we turn to some examples of qualitative or directional simulations aimed at

assisting with policy evaluations, some summary comments are appropriate.
4.3.1 Summary
In summary, this chapter presented a verbal and diagrammatic summary of an opera-

tional policy model of wellbeing, which can be used for numerical simulations. The
full set of equations for the model is proposed in King (2018). The model is designed
to support policy developments and decisions, by providing a tool for thinking about
changes in the main influences on wellbeing, including trade-offs and synergies for
policies. It does this by modelling the linkages between the eleven categories of influ-
ences on wellbeing used by the OECD in their Better Life Index, in combination with
a small CGE model and stock-and-flow equations.
Data is always a challenge for a model of this sort, as it requires a large number
of parameters and initial values. Many of these (especially the standard economic
values) are available from national statistics and other appropriate sources; or can be
calibrated from such data. Others may need to be proxied from international data, and
some will be unavailable from reliable sources. All need to be tested with sensitivity
analyses, and for those that are completely unavailable this will be crucial. Note that
as results will often be given qualitatively and relative to a baseline, the results might
not be as sensitive to the parameters as might be expected at first.
There is also much to learn about policy by applying the model. Without a model
of this type, the second, third and higher order effects (particularly those in different
policy areas from the original) are very difficult to anticipate or attribute to their
original causes. This type of model allows all those effects to be recognised, as well
as helping us obtain some understanding of how significant they are. For example,
if an increase in health spending by the government is modelled, the model results
indicate that a small negative effect on the natural environment could occur. This is
not an intuitively obvious result, but tracing through the model logic and results shows
it is not a spurious result: more health spending results in better health, better health
3 Chichilnisky (1997) proposes a modification to the standard discount equation that corrects for this
effect, by adding a non-zero term in the limit t → ∞. (See Eqs. (2.101) and (2.102) in Sect. 2.8.4.)
results in higher productivity, higher productivity results in more consumption and

production, more production results (all else equal) in more pollution and therefore
a poorer natural environment. This is a chain of impacts that is too long for most
people to perform mentally without the aid of a model, but the consequences could
be important. Any of the parameters and exogenous variables in the model can be
shocked as a policy change. Shocks can also be added to many other variables,
especially the investment variables (for example, the destruction of government-
owned physical capital by the 2016 Kaikoura earthquake in New Zealand).
The model can be used for both single policies and for policy packages. In the case
of policy packages, it can be used to see how the effects of different policies balance
each other out. It can also be used to help determine which policies should be included
in a package. It helps this in two ways. First, it provides a consistent framework for
policies from different fields (e.g., education and health) to be compared with each
other and in a broader way than tools such as cost-benefit analysis allow. Second,
policies can be chosen on the basis of their ability to mitigate some of the undesirable
trade-offs (such as the negative environmental effects of a health policy) inherent in
the choice of policies that are the cornerstones of a package.
4.3.2 Using the Model
Our particular interest in this book is in the interaction between environmental, social,
and economic factors in influencing wellbeing, and the challenges policy makers
face in exploiting the trade-offs and complementarities between these spheres of
wellbeing in designing policy packages that improve wellbeing on a sustainable
basis. Since we are interested in using the model to assist with policy design, we also
need to be very aware of the critical need to present and communicate the results of
the numerical policy simulations in a way that maximises their impact. To this end,
it is important to appreciate that policy advisers want to know: what policy levers
to use, the quantum of change required, quantification of the full set of effects over
time, highlighting the complementarities and trade-offs in outcomes, and all of this
presented in a simple-to-understand way, usually with the assistance of graphs and
diagrams.
In what follows, building on the health-policy example we referred to above, we
first set the scene with a simple example that shows how an investment towards
improving health works through the system to influence wellbeing. In doing so, we
also trace through the positive and negative effects of this policy intervention. Once
we are aware of the trade-offs involved (e.g., between an increase in productivity and
a deterioration of environmental quality), we can start thinking about complementary
policies that are designed to reduce the negative effects of policy interventions, with
a view to increasing the aggregate positive effects of policy packages on wellbeing,
which is ultimately our primary interest.
4.3 Policy Simulations 131
Fig. 4.3 Main effects of a health spend increase
The common thread running through all four policy-simulations we present below
(although in qualitative and diagrammatic form) is the interactions between environ-
mental, social, and economic spheres of wellbeing in response to policy interventions.
4.3.2.1 Investing in Health
In this first simulation, we trace the effect of an increase in government spending on

healthcare. Figure 4.3 shows the channels through which this affects wellbeing.
As shown in Fig. 4.3, the increase in health level increases firm productivity, the
ability to benefit from consumption of formal education, and adjusted leisure time
(which then gives improved work-life balance). The improved firm productivity
means there is more production and higher wages, and so more consumption of all
types of market goods, including of education and housing goods. The increase in
education creates a virtuous cycle, further improving firm productivity. However, it
also uses time, which reduces leisure. As this version of the model does not have a
feedback mechanism to balance this, the time budget for the household may eventu-
ally be violated. While this is unlikely to happen in reality, it does highlight that this
virtuous cycle cannot continue indefinitely. Moreover, the additional production cre-
ates additional pollution, which has a negative effect on the environment (although
very small with the parameters used here).
There is a second effect of increasing health spending; namely, the effect caused
by the increase in government spending. This reduces the amount of government
transfers to households, which reduces household income. However, the increase in
wages due to increased productivity more than cancels this effects out, and household
income increases. As household income increases, savings and therefore wealth also
increase. As household wealth is used as capital for production, this further increases
the amount of production that can occur.
We would naturally expect the increase in health spending to have the largest
wellbeing effect on the health component of the OECD BLI, as well as a generally
positive effect on other influences on wellbeing. Note however, that the effect on
health would be somewhat smaller than the size of the increase in spending—this
is because some private spending on health is crowded out, and there are diminish-
ing returns from healthcare spending. Overall, one would expect increasing health
spending to be beneficial for almost all aspects of wellbeing.
4.3.2.2 Investing in Skills
In this second scenario, we consider the impacts of an increase in the level of time
invested by firms in skill development. This would be operationalised by reducing the
percentage of time spent by an employee “at work”, in favour of an increase in time
allocated to skilling. Higher skills should increase productivity, which would then
have a positive effect on wellbeing through an increase in production and therefore
consumption. Nevertheless this could potentially be outweighed by the decrease in
production due to the time spent on skills development instead of production. As
shown in Fig. 4.4, this decrease in production would then flow on to a decrease in
consumption, including housing, ehealthcare, and formal education, giving basically
opposite results to the previous scenario where firm productivity increased.
This suggests that skill development, if taking time away from production, should
be carefully balanced with the productivity enhancements it can bring. It should only
be ongoing if it does bring sufficient ongoing productivity improvements. Otherwise,
it can result in lower wellbeing through reduced production and consumption, flowing
through to other aspects of wellbeing.
4.3.2.3 Subsidising Clean Technology
In this scenario, we increase the weight of the use of clean technology in production.
This should improve the quality of the natural environment, but might at the same
time increase production costs. The effects on each aspect of wellbeing are traced in
Fig. 4.5.
Fig. 4.4 Main effects of a skills increase
The primary effect of increasing the proportion of clean over dirty technology is
a decrease in pollution, which improves environmental wellbeing both by slowing
the growth in the level of pollution and in reducing the effect of pollution on bio-
diversity. The improvement in pollution level and biodiversity has a positive effect
on health; and also improves firm productivity. This then improves the other aspects
of wellbeing. The negative effects of increased production from greater health and
greater productivity need not cause enough of an increase in pollution to noticeably
reduce the good environmental effects of switching to cleaner technologies. This is
partly because the increased production is also from cleaner technology (Fig. 4.5).
Note that the results of this scenario suggest that switching to cleaner technology
and therefore improving the environment need not come at a cost to production.
4.3.2.4 Investing in Social Connections
Humans are a social species, and as such good social connections are extremely
important for our wellbeing. The quality of family and other relationships are one of
the most important factors affecting subjective wellbeing. Good social connections
can improve our health, increase our resilience, and increase the likelihood of finding
a good quality job. Poor social connections lead to isolation and loneliness, as well
as making it much harder to get things done.
In this final scenario, we increase the initial stock of social connection, and trace
the effects on each aspect of wellbeing through Fig. 4.6.
This has qualitatively similar results to the first scenario where we increased health
spending. In this case, the higher social connections stock increases the benefits of
Fig. 4.5 Main effects of a clean tech increase
leisure time as well as increasing firm productivity (but not education). Note that this
does not increase health directly, as health level changes when the level of healthcare
consumption (as part of overall consumption) changes. As shown in Fig. 4.6, the
improvement in firm productivity then flows through to the other aspects of wellbeing
in the same way as for the first scenario, increasing production and consumption,
education, housing, and pollution. Work-life balance is higher from the start, as the
value of leisure time is improved by better social connections.
Fig. 4.6 Main effects of a social connections increase

Non-technical
1. Explore and discuss the various types of “policy-simulation models”.
2. Discuss how the components of the OECD Better Life Index (BLI) are used in
constructing the “operational policy-simulation model” of this chapter.
3. Analyse how the “operational policy-simulation model” relates to the full stylised
model of Chap. 2.
4. Discuss what “numerical policy-simulation models” are used for – how do they
assist in policy choices.
5. Critically assess what guidance “numerical policy-simulation models” provide
towards the choice of policy priorities and/or policy instruments – and how they
are used for this purpose.
Technical – and Extensions
1. Outline the similarities and differences between the “operational policy-simulation

model” of this chapter and the “stylised model” of Chap. 2.
2. Outline the similarities and differences between the “operational policy-simulation
model” of this chapter and the viability-theory based model of Chap. 6.

Readers who want to learn about CGE models can read Bergman (2003) and Wing
(2004).
Part II
Part II of the book comprises the last three chapters.

These chapters explore the implications of fundamental uncertainty and adaptive
complexity for the design and implementation of public policy.
If we do not know how the future will evolve, and we cannot predict the reactions
of individuals and communities to various types of policy interventions, how do we
design and implement a wellbeing-focused public policy?
We argue that in this type of environment, the primary focus of public policy
needs to shift to building resilience. Resilience-building needs to have two aims if it
is to be effective in enhancing the lives people choose to live. First, it needs to aim at
protecting society from major catastrophes. Second, it has to create an environment
that encourages social cohesion and creativity.
These extensions need not be limited to conceptual discussions. They can be for-
mally incorporated into quantitative models. In this context, we offer viability theory
as a promising modelling device, as a substitute for optimisation models when the
latter are inapplicable.
We can use models based on viability theory to simulate the effects of policy
interventions on the “viability kernel” (or the “wellbeing frontier”). Investments that
expand the “viability kernel” are in effect expanding the playpen within which valued
and diverse lives can be lived reasonably safely. That is the best public policy can
aspire to, in a complex and fundamentally uncertain world.
There is one more critical implication of radical (or fundamental) uncertainty
and adaptive complexity, and that relates to both the design and implementation of
public policy. Unless communities are fundamentally involved in the design and
implementation, as well as the evaluation, of public policy, policy will not have an
enduring effect on enhancing wellbeing.
Chapter 5
Wellbeing-Enhancing Public
Policy - Extensions
Abstract Recognising the prevalence of radical uncertainty and complexity, in a

world populated by billions of individuals pursuing improvements in their very
diverse valued lives, has a fundamental impact on the policy setting. There needs
to be no change in the objective of improving individual and community wellbeing
on a sustained basis. The main impact is on the focus of public policy, as well as
the choice of policy instruments and implementation mechanisms. The search for
solutions to optimisation problems is replaced by a focus on investing in resilience –
characterised by adaptability of the environment, society, and economy – to emerg-
ing situations that cannot be predicted or controlled. At the system level, resilience
rests on two foundations: protecting comprehensive wealth from catastrophes, while
at the same time nourishing human inventiveness – natural inquiry through experi-
mentation (see Mokyr 2016). At the centre of the enquiry are the ecosystems within
which individuals and communities live and operate, including the institutions that
shape their incentives. This revised policy framework embraces localism, experimen-
tation, small steps in implementation, reversibility and fast failures, and planning on
surprises. It is built around the benefits of decentralisation and emergence. Collec-
tive wellbeing is pursued not by searching for investment levers that will enhance
well-defined social outcomes, but rather by making it easier for people to pursue
the varied lives they value, with no pre-defined desired aggregate social outcomes.
“What works” is defined by communities, not by public servants, although the latter
can play a constructive and critical role in helping communities achieve the outcomes
they value.
This chapter helps readers:
• Explain what is meant by “complexity” and “radical uncertainty”. Readers will
also be able to distinguish between “radically uncertain” and “risky” situations.
• Explain why and how the introduction of “complexity” and “radical uncertainty”
to the stylised model of Chap. 2 makes any difference, especially for the design
and focus of public policy.

https://doi.org/10.1007/978-981-13-6104-3_5
140 5 Wellbeing-Enhancing Public Policy Policy - Extensions
• Critically assess how the choice of policy priorities and policy instruments is
affected in a model that allows for “complexity” and “radical uncertainty”.
• Apply “complexity economics” and “radical uncertainty” to the design of public
policy and policy implementation in different circumstances.
5.1 Introduction
This chapter and the next should be viewed as companions. They are both concerned
with the implications of introducing “radical uncertainty” and “complexity”1 to the
stylised model of Chap. 2. This chapter provides a qualitative discussion of the sub-
ject. It highlights, and justifies, the emergence of resilience as the main focus of
public policy when such extensions are introduced. The next Chap. 6 introduces and
applies viability theory as one possible formal method of dealing with these exten-
sions. Viability theory is particularly suited to formally modelling dimensions of
resilience that relate to our capability to monitor the system and steer it away from
catastrophic boundaries or thresholds, as well as building shock-absorbing capacity
through appropriate investments.
We continue to assume that most individuals and communities constantly strive
to pursue the lives they value, subject to their capabilities and opportunities (i.e.,
substantive freedoms), and conditioned by the incentives they are subject to. They
do so in a social setting. We do not know, nor do we wish to pass judgement on, what
lives individuals and communities value. What a wellbeing-focused public policy
strives to do, is to first create the environment where the pursuit of valued lives is
possible, and then support the acquisition of the capabilities and opportunities (i.e.,
substantive freedoms) that enable the pursuit of these valued lives.
We extend the stylised model of Chap. 2 to recognize that there are two other
crucial factors that cannot be ignored: radical uncertainty (see King 2016, Chap. 4)
and complexity (see Arthur 2014).
Under radical uncertainty, with the set of possible states of the world and/or
their probabilities unknown, optimisation at an individual or policy level becomes
impossible. We simply do not know what the future will look like, although we may
have a broad idea of the types of catastrophes our environment, society, and economy
are most likely to be exposed to.
Under complexity, particularly in a social context, actions and outcomes cannot be
separated – there is “reflexivity”: there are no external (exogenous) social or economic
facts that are independent of our actions. Individuals’ actions and interactions with
others influence the social and economic outcomes that emerge, which in turn lead
to reactions, and on it goes (see Soros 2003).
The stylised model of Chap. 2 did highlight the interdependence of environmental,
social, and economic influences in the generation of sustained wellbeing. It was also
1 Thetype of complexity we are referring to here is often distinguished by the term “adaptive
complexity” or similar terms (see e.g., Haldane 2013; Von Hayek 1989).
able to identify the key linkages between these sets of influences. It was even able to
isolate the key domains of a public policy that pursues intergenerational wellbeing. It
was not of much help however in telling us how and when we should implement these
policy interventions. This is precisely where the literature that emphasises complex-
ity, radical uncertainty, and satisficing (vs. optimising) individuals and businesses,
adds particular value. This literature does not question or challenge the appropri-
ateness of sustained wellbeing as the objective of public policy. It enhances policy
design and implementation by contributing to the choice of appropriate instruments
and implementation mechanisms, as well as evaluation methods.
This shift of policy setting increases relevance, but at the expense of precision.
There are no specific policy buttons to push, to achieve well defined outcomes with
any degree of certainty. Instead, the framework is one that emphasises the creation of
environments (or ecosystems) within which individuals and communities can con-
tinue to pursue the kinds of lives they value, and possibly even improve them. There
is an equal emphasis on protecting valued ways of living by preventing environmen-
tal, social, and economic catastrophes. Finally, there is a strong insistence on the
absolute necessity of involving the communities for which policies are designed, in
the design, implementation, and evaluation of public policies.
Before we go any further, we should note that we will be using the term “ecosys-
tem” quite a bit in what follows. We use this term in the sense of a community of
interacting organisms, and their physical and natural environments. These organisms
can be human beings, non-humans, as well as institutions (such as the government,
courts of law, and banks). These often take the form of complex networks or inter-
connected systems. Their effective and efficient functioning is critical for, among
others, social cohesion, environmental health, and financial stability.
A diagrammatic representation of all the component forces, including public pol-
icy and institutions, at play in this complex system of interactions is captured in
Fig. 5.1, which we have borrowed from Costanza et al. (2017). The place and role of
government is embedded in the right-hand box, as part of built, human, and social
capital.
Among other matters, the figure highlights the interaction of various types of
services, provided by different bodies, to make the whole system work, while also
keeping it at a safe place. As Costanza et al. (2017) explain, we have:
• “Provisioning services” that refer to the interaction between natural capital and
other forms of capital to produce food, timber, fibre, and so on. They give the
example of fish delivered to people as food, requiring fishing boats (built capital),
fisher-folk (human capital), and fishing communities (social capital).
• “Regulating services” which, through similar interactions between various forms
of capital, provide for example, flood control, storm protection, water regulation,
human disease regulation, water purification, air quality maintenance, pollination,
pest control, and climate control. They give the example of storm protection by
coastal wetlands, which requires built infrastructure, people, and communities to
be protected.
Fig. 5.1 Policy model
• “Cultural services” which produce recreation, aesthetic, scientific, cultural identity,

sense of place, or other ‘cultural’ benefits. The example provided here is of a
recreational benefit that requires a beautiful natural asset (a lake), in combination
with built infrastructure (a road, trail, dock, etc.), human capital (people able to
appreciate the lake experience), and social capital (family, friends, and institutions
that make the lake accessible and safe).
• “Supporting services”, which refer to processes such as soil formation, primary
productivity, biogeochemistry, nutrient cycling, and provisioning of habitat.
The question is what possible role(s) can public policy play in this type of envi-
ronment and setting, towards enhancing human wellbeing on a sustained basis.
5.2 A Radically Different Policy Framework
“Traditional economics views the economy in a fairly mechanistic way. If people

are rational and we want to change their behaviour, then we just need to change
their incentives. Thus, a lot of policy is conducted through tinkering with the tax
code or subsidies. For example if one wants more innovation, give an R&D tax
credit; if one wants less smoking, tax it heavily. Of course people aren’t immune
to such incentives, but often the response is far less than policymakers would like”
(see Beinhocker 2017). And often there are unintended consequences because people
adjust their behaviours and adapt to the changed circumstances (including incentives)
that the new environment provides (see, by way of example, Thaler 2016).
5.2 A Radically Different Policy Framework 143
How then could policy respond, purposefully and productively, to serving inter-
generational wellbeing when the future is fundamentally uncertain and the world we
are living in (including the environment, society, and economy that surround us) is
complex?
Once we introduce radical uncertainty and complexity to our framework, the
importance of resilience is naturally elevated. At a system level, protecting compre-
hensive wealth from catastrophes becomes paramount.2 There is also a complemen-
tary shift of emphasis to building resilient ecosystems that fertilise human inven-
tiveness – to deliberately investing in harvesting diverse and contingent community
responses to emerging environmental, social, and economic challenges. “Optimising
behaviour is a special case of a more general theory of behaviour under uncertainty.
And in situations of radical uncertainty, where it is impossible to optimise, a new
approach is required. I have suggested a possible starting point with the idea of cop-
ing strategies” (King 2016, p. 366). This also corresponds with what evolutionary
economists refer to as “learned routines” and “built in adaptability” (see Nelson
2017, 2018).
Before we go any further, we need to remind ourselves what we mean by
“resilience”, especially in a systems and dynamic context. In Chap. 2, we defined
resilience as the capability of the economy and society to respond to, and continue to
operate reasonably effectively and efficiently in the aftermath of, significant systemic
shocks to comprehensive wealth – measured, in principle, in terms of the speed and
cost of adaptation.
A system may show resilience to major systemic shocks not necessarily by return-
ing exactly to its previous state following a shock, but instead by finding different
ways to carry out essential functions; that is, by adapting. We therefore think of
resilient environmental, social, and economic systems as ones that have the capabil-
ity to: withstand sudden shocks, adapt to changing contexts, and recover to a desired
state (either the previous one or a new one), while preserving the continuity of its
operations. Thus resilience encompasses both recoverability (the capacity for speedy
recovery after a crisis) and adaptability (timely adaptation in response to a changing
environment).
To paraphrase (North 1994), in a dynamic context, it is adaptive rather than alloca-
tive efficiency that is the key to long run wellbeing. Successful economic, political,
and social systems have evolved flexible institutional structures that can survive the
shocks and changes that are part of successful evolution.3
In a world of complexity and radical uncertainty, individuals and communities
do not know the best course of action open to them. In fact, there is no objective,
independent of our actions, best course of action to be discovered and implemented.
Individuals constantly strive and experiment towards improving their lives, or pre-
serving the ways of living they value. The focus of policy shifts to encouraging
2 See Broadberry and Wallis (2017) and Haldane (2018) on the critical influence of resilience-
building economic and social institutions on the progress of human prosperity since the Industrial
Revolution.
3 For an overview of modern evolutionary economics, see Nelson (2017).
and incentivising individuals to continue doing so, while expanding the boundaries
within which this ongoing experimentation takes place, and monitoring and ensuring
that these boundaries are not crossed. Formalising and operationalising this policy
framework is the subject matter of the next Chap. 6.
5.3 Key Policy Focus is on Building Resilience
So what would a public policy that is focused on building resilience to unknown

systemic shocks look like? How would it be framed? How would it be implemented
so that adaptive complexity is taken into account as well?
The foundations of such a policy would include (see Scott 1998):
• a recognition of the limits of what we are likely to know about a complex and
functioning order;
• a recognition that individuals and communities will respond and adapt to whatever
changes are put in place;
• the contributions to resilience of social, natural, regional diversity;
• the indispensable role of practical knowledge, informal processes, and improvisa-
tion in the face of unpredictability;
• an appreciation of the complementary roles of “the centre” and the local/regional
communities; and
• in that very context, the advantage that “the centre” has in seeing the big picture,
identifying trends early, and intervening in a timely fashion in epidemics.
One of the main teachings of complexity economics is that creative ideas are
primarily generated bottom up, not top down – through interactions and cross-
fertilisations of ideas promulgated by human contacts via various means. A critical
role for the policy maker is that of supporting and connecting. Creativity remains the
source of innovation, technological progress, and prosperity under all frameworks –
how we encourage and support it is the point of difference. Think of a landscaper
instead of an architect – the focus is on creating the right ecosystems for people, com-
munities, and ideas to flourish. It is critical to ensure that individuals are incentivised
to do the best they can for themselves, while taking account of (i.e., internalising)
the consequences of their actions on others. Hence the critical role of conditionality,
where and when appropriate, in the effective and efficient operation of a wellbeing
state. The other, complementary, role of public policy is to provide insurance against
system-level catastrophes – i.e., systemic resilience.
In a complex and evolving world dominated by radical uncertainty, the scope and
instruments of public policy available to a wellbeing state need to be expanded to
include the “comprehensive ecosystem that surrounds all decision makers”. Instead
of using direct interventions to achieve certain goals, the government could rather
try to influence the rules of the game. The rules of the game are the institutions
that define the individual agents’ action sets. Instead of solving problems for the
agents, the government could create institutions that make it easy for agents to solve
5.3 Key Policy Focus is on Building Resilience 145
problems for themselves. These rules of the game may influence the market dynamics
of the complex system in a desired way. They may even include ‘supernudges’ that
have an effect on agents’ preferences and norms (See Colander and Kupers 2014).
Thus the proposed response to complexity and radical uncertainty is to shift the
focus of policy from direction to nourishment — to building resilience to potential
systemic risks that threaten valued ways of living, as well as encouraging creative
responses, through cooperative community mechanisms (see also Ostrom 2009).
This type of policy framework embraces localism, encouragement of experimen-
tation at local level, small steps in implementation, reversibility and fast failures, and
planning on surprises. It is built around the benefits of decentralisation and emer-
gence. Collective wellbeing is pursued not by searching for investment levers that
will enhance well-defined social outcomes, but rather by making it easier for people
to pursue the varied lives they value, with no pre-defined desired aggregate social
outcomes.
As Beinhocker (2017) argues, we must take a more deliberately evolutionary
view of policy development. Rather than thinking of policy as a fixed set of rules
or institutions engineered to address a particular set of issues, we should think of
policy as an adapting portfolio of experiments that helps shape the evolution of the
economy and society over time.4
There are three principles underpinning this approach:
1. Rather than predict we should encourage experimentation in communities and at
local levels – and then supporting local experiments, aimed at addressing local
problems, when they are having locally desired outcomes. Policymaking often
starts with an engineering perspective – there is a problem and government should
fix it. So we look for an “optimal solution” or maybe a “second best solution”. An
alternative approach is to create locally driven portfolios of small-scale experi-
ments trying a variety of solutions, see which ones work, scale-up the ones that
are working, and eliminate the ones that are not. Such an evolutionary approach
recognises the complexity of social-economic systems, the difficulty of predicting
what solutions will work in advance, and difficulties in real-world implementa-
tion. Failures then happen on a small scale and become opportunities to learn
rather than hard-to-reverse policy disasters.
2. Policies and institutions should be made as adaptable as possible. A better
approach is to create rules that provide general frameworks, but then adapt to
specific circumstances. One example is how California’s building codes have
succeeded in reducing energy consumption. Rather than try to predict the state
of energy efficiency technologies in future years, the regulators created a set of
general performance standards that automatically ratchet-up as the state of tech-
nology improved.
3. And finally, policymakers need to think of themselves less as social engineers and
more as “system stewards”. As Michael Hallsworth from the Institute for Gov-
4 This
broadly also corresponds to Markovian policies, derived in feedback form via dynamic pro-
gramming, as opposed to open-loop optimal policies derived from the maximum principle (see e.g.,
Haurie et al. 2012).
ernment (IFG) explains, rather than engineering specific outcomes, government’s

role as system steward is to create the conditions in which interacting agents in the
system will adapt towards socially desirable outcomes. Policy design and imple-
mentation are thought of as integral rather than separately, and mechanisms for
feedback and continuous learning and improvement are built in from the begin-
ning. (See Beinhocker 2017, from where this material is sourced, for a more
detailed discussion of all these points.)
5.4 Critical Role of Political Economy and Institutions
5.4.1 Political Economy Considerations
“A major challenge for these more adaptive approaches to policy is the political
difficulty of failure. Learning from a portfolio of experiments necessitates that some
experiments will fail. Evolution is a highly innovative but inherently wasteful process
– many options are often tried before the right one is discovered. Yet politicians are
held to an impossibly high standard, where any failure, large or small, can be used
to call into question their entire record.
“Likewise, politicians are always expected to have clear plans, and simple, easy
to understand answers in which they have unshakeable confidence. You would never
hear a politician give a speech where she or he says ‘It is a complex problem, we’re
not sure what to do. But we have several good ideas that we’ll try on a small scale.
We’ll then ramp up the ones that work and close down the ones that don’t, and then
have a good shot at solving it.’ For some reason we don’t mind such an approach
when it is used by doctors looking for new drugs, energy companies looking for oil,
or venture capitalists looking for the next big idea. But we seem to prefer politicians
who tell us the world is simple and predictable, even though we know it to be complex
and unpredictable.
“So an explicit, widespread use of new economic approaches to policymaking
may require some education of citizens, the media and politicians themselves on the
risks of overconfident top-down solutions, and the importance of small-scale failures
as a way to learn and prevent large-scale disasters.”
“This new economic thinking may also provide the foundation for new political
narratives. Eric Liu and Nick Hanauer, in their 2011 book The Gardens of Democ-
racy, explore the possible shape of such a narrative. They liken the narratives of tradi-
tional economics to ‘machine-thinking’ and advocate a shift to ‘garden-thinking’ that
emphasises the dynamic, constantly evolving nature of the economy, and the inter-
connectedness of society. The state then plays the role of a gardener helping create
the conditions in which the garden of society can flourish.” (Beinhocker 2017)
5.4 Critical Role of Political Economy and Institutions 147
5.4.2 Institutions
In almost all policy discussions, there is an ongoing reference to “the right institu-
tions” as part of the solution. This is especially so when we are dealing with complex
problems and operating in an environment of radical uncertainty, when ensuring time
consistency of policies and policy frameworks is particularly challenging.
The literature on development economics has converged on the view that “institu-
tions play a vital role in shaping social, economic, and political incentives, reducing
transaction costs and uncertainty, and promoting long-run economic growth” (see
Bennett et al. 2016, p. 1).
In this context, (North 1994, p. 2) describes institutions as, “the humanly devised
constraints that structure human interaction. They are made up of formal constraints
(rules, laws, constitutions), informal constraints (norms of behaviour, conventions,
and self imposed codes of conduct), and their enforcement characteristics. Together
they define the incentive structure of societies and specifically economies”. Hodg-
son Geoffrey (2006) offers a wider definition of institutions as systems of established
and embedded social rules that structure social interactions. Institutions do not only
constrain and influence the choices and actions of individual consumers and busi-
nesses, but they also perform a critical enabling function – and their evolution is
endogenous (see Arvanitidis 2004; Acemoglu and Robinson 2013).
According to North (1990, 1991), institutions comprise both “informal con-
straints” (such as sanctions, taboos, customs, traditions, codes of conduct) and “for-
mal rules” (such as constitutions, laws, property rights). The former are part of the
heritage we refer to as “culture”, while the latter are created by the state. In both
cases, their fundamental role has been to create order and reduce uncertainty (see
Bennett et al. 2016, p. 7).
5.4.2.1 Institutions and Wellbeing
Material resources are necessary but not sufficient conditions of broader wellbeing
(see Stiglitz et al. 2010; Porter et al. 2014). When wider wellbeing is the object of
public policy, what additional institutions, if any, should we be thinking of as sources
of wellbeing? In an uncertain and complex world, how would these institutions
enhance the capabilities and opportunities (i.e., substantive freedoms) of individuals
and communities to live the kinds of lives they have reason to value? And equally to
the point, what else should public policy strive to provide and achieve towards this
end, in addition to devising institutions?
Bennett et al. (2016) traces the origins of this exploration to the classical
economists Adam Smith, John Stuart Mill, and David Ricardo, stating that they
“stressed the importance of institutions, public policy, and the rule of law in shap-
ing economic performance and human wellbeing” (pp. 5–6). The modern version of
this line of exploration, under the themes “fundamental (as distinct from proximate)
causes of economic growth and development” and “deep roots of economic devel-
opment”, is well represented by the works of Acemoglu and Robinson (2013) and
Spolaore and Wacziarg (2013), among many others. See also Hodgson (2014) for a
broader discussion of institutions in the context of Evolutionary Economics.
Before we address this question with the help of a few examples, we present
some empirical evidence to highlight the critical role of institutions in the pursuit of
intergenerational wellbeing.
For a review of the literature, including empirical evidence, on the interface between
culture, institutions, and economic outcomes, see Alesina and Giuliano (2015) and
Gorodnichenko and Roland (2011). One of the main findings of this strand of lit-
erature is that, “individualistic cultures that place value on personal freedom, self-
reliance, creative expression, affective autonomy, and reward individuals for their
accomplishments with higher social status, tend to experience faster rates of inno-
vation and economic growth compared to more collectivist societies” (see Bennett
et al. 2016, p. 9).
For an early contribution regarding the positive impact of wellbeing for the devel-
opment of good institutions, see Lipset (1959). Bennett et al. (2016) provides a useful
and recent survey of related literature. Faria et al. (2016) provide further empirical
evidence supporting the potential virtuous cycle between institutions and sources of
wider wellbeing. On the endogeneity of institutions in the long run, also see Rodrik
(2004) and Acemoglu et al. (2005).
This does not mean that we are faced with a circularity that leads no room for
deliberate policy. On this point, see the argument and evidence provided, among
others, by: Rodrik (2004), Acemoglu et al. (2014), Gründler and Krieger (2016), Sen
(2009), and Sen (2017).
Also see Aidt (2009) and (2011) for examples of empirical evidence on the neg-
ative effects of bad institutions on economic and social outcomes.
More to the point, given our particular focus on wider wellbeing, rather than
exclusively on economic growth or material prosperity, is the relatively recent and
emerging literature on the interface between formal and informal institutions, and
wellbeing. “More recent studies have started examining the relationship between
formal and cultural institutions, and a variety of other objective and subjective social
outcomes such as trust, tolerance, crime, poverty, or subjective wellbeing. Berggren
and Nilsson (2016), for example, find suggestive evidence that the institutions of eco-
nomic freedom increase tolerance, measured by the willingness to (e.g.,) let atheists
and homosexuals speak, keep books in libraries, and be a college level teacher.
These authors suggest that this relationship may work through reducing feelings of
tension and conflict. In this sense, institutions may play a vital role in creating a non-
discriminatory and inclusive environment that fulfils the basic psychological need
for relatedness.
An important insight from this literature, for instance, is that institutions may have
a positive impact on psychological wellbeing beyond their direct impact on a variety
of socio-economic outcomes: a concept known as procedural utility (see Frey et al.

2004). The idea here is that people care not merely about outcomes, but also about
the processes that lead to these outcomes. Individuals, for instance, may experience
a higher level of subjective wellbeing if they believe that they are treated in a way
they consider just or fair regardless of the material outcomes. For instance, the right
to participate in the political process, measured by the extent of direct democratic
rights across regions, is strongly correlated with subjective wellbeing (see Frey and
Stutzer 2002).
Nikolaev and Bennett (2016) show that people who live in countries with insti-
tutions consistent with the principles of economic freedom experience a greater
perception of freedom of choice and control over their life. They furthermore find
suggestive evidence that a potential mechanism that explains this relationship is the
perception of procedural fairness and social mobility. An emerging literature on the
relationship between formal institutions and subjective wellbeing also finds a positive
correlation between the institutions of economic freedom and subjective wellbeing,
most often proxied by survey measures of life satisfaction (for a review of this liter-
ature (see Kešeljević et al. 2016). (This general discussion is sourced from Bennett
et al. 2016, pp. 11–12.)
Although these results are interesting and noteworthy, they should be received
with a degree of caution. One should be particularly careful about attributing causal-
ity from institutions to all the outcomes cited above. By way of example, while
Singaporeans enjoy a great degree of economic freedom, one cannot say the same
about the broader freedom of choice they enjoy. Furthermore, it is doubtful that the
positive impact of an institution on subjective wellbeing can last if people cannot
observe or experience the outcomes the institution was supposed to deliver, even if
they like the principles on which the institution had been created.
5.4.2.3 Examples - Building Resilience to Systemic Risks
We now turn to exploring, with the help of examples, how we can build resilience to
systemic risks by investing in the right types of institutions. We start with building
resilience to financial-system risks.
Resilience to Systemic Risks to the Financial System
A proposal for public policy formulation and implementation, in the midst of radi-
cal uncertainty and complexity, is extremely well developed by King (2016) in the
context of monetary and financial policy. It is a great example of how all the key
dimensions of a comprehensive policy framework, of the sort identified above by
Beinhocker (2017), can be delivered in practice.
An efficient and stable financial system is a critical ingredient of economic infras-
tructure as a platform for sustained overall wellbeing. How can public policy support
an efficient and effective financial system, while at the same time ensuring sustainable
financial stability by protecting it from potential catastrophes?
We are dealing with a complex system of intricately interrelated networks of

financial institutions (see Haldane and May 2011). These institutions play some
critical roles — including: money creation (banks create deposits to finance risky
lending), payments services, financial intermediation between borrowers and lenders,
and maturity transformation (linking the present to the future).
The key steps, and associated principles, are:
• Identify the key catastrophe, or source of catastrophe, that this critical infrastruc-
ture or ecosystem is potentially exposed to.
• Determine the institutions that need to be established to prevent it (if it is under
human control or influence) or adapt to it (if it is outside of human control or
influence).
• Make sure that the presence of such institutions does not blunt the incentives
of private individuals, businesses, or communities to do the best they can for
themselves, while internalising the consequences of their actions on others.
• Make sure that the institutions established to deal with the fundamental systemic
catastrophe are adaptable – they have the capacity to evolve in response to the
evolution of the environment (including the impact of emerging technologies) in
which they operate.
As King (2016) convincingly argues, the main catastrophe the financial system,
and through it the wider economic system, is exposed to is the drying up of liquidity
for the whole financial system in a financial crisis (as it did in 2007/8). Avoidance of
this systemic catastrophe through liquidity provision to the whole system is a major
public good. Only central banks can perform this function. How they perform this
function, and what incentives and constraints they wrap around it, is critical for the
overall ongoing functioning of the system in an effective and efficient manner.
How the central bank can insure against financial catastrophe, by providing liq-
uidity in an effective and efficient manner, has evolved over time. King argues that the
traditional lender-of-last-resort (LOLR) mechanism for providing system liquidity
has reached its used-by date, and it needs to be replaced by a pawn-broker-for-all-
seasons (PFAS) strategy, which he argues is a coping strategy in the face of radical
uncertainty.
“The aim of the PFAS is threefold. First, to ensure that all deposits are backed
by either actual cash or a guaranteed contingent claim on reserves in the central
bank. Second, to ensure that the provision of liquidity insurance is mandatory and
paid for upfront. Third, to design a system which in effect imposes a tax on the
degree of negative externalities generated in our financial system – private financial
intermediaries should bear the social costs of their actions” (ibid, p. 271).
In implementing this set of measures, we need to “ensure that the risks involved in
money and banking are correctly identified and borne by those who enjoy the benefits
from our financial system” (ibid, p. 251). The measures we put in place should also
ensure that, first, private institutions do internalise the wider (social) consequences of
the risks they incur in the pursuit of profits, mainly through imposing higher equity
ratios on their total assets; and second, that their creativity in using their balance
sheets is not blunted through inappropriate regulations.
Since systemic risks cannot be privatised, we need to put measures in place to

minimise their occurrence. These measures include: significantly higher capital-
adequacy ratios; leverage ratios; liquidity ratios; pre-insurance; and collateralisation.
The technicalities associated with these terms need not distract us here. Suffice it to
say, they are all measures targetted at making sure that the shock absorbing capacity
of the financial system as a whole, as well as individual financial institutions, are
strengthened.
Equally important is ensuring that these generic restrictions do not interfere with
how financial institutions manage their balance sheets [assets (loans) and liabilities
(deposits)]: “We could leave the banks free – to lend money as they please, provided
we no longer allowed them to manufacture the money they lend. In short: nationalise
money, but do not nationalise banking” (see Fisher 1936).
Resilience to Environmental Systemic Risks
It is worth exploring whether, and if so how, King’s suggested approach to developing
resilience (coping strategies) to systemic financial risks, in the broader context of
complexity and radical uncertainty, can be extended to other systemic domains –
e.g., environmental and social.
As a reminder, let us go back and start with the key steps, and associated principles,
highlighted in the previous section, but suitably adapted to environmental risks.
In the context of environmental infrastructure and ecosystem services, the equiv-
alents of liquidity (for the financial system) are the climate system and biodiversity
(see De Perthuis and Jouvet 2015). Climate warming and a gradual extinction of
species-diversity are the two sources of environmental disaster.5 These are protected
by natural regulatory systems needed for the reproduction (sustainability, viability)
of natural capital stocks.
The critical difference between financial and environmental ecosystems is that,
in the case of the latter, once we breach the critical boundary(ies), there may be no
turning back – so there needs to be a comparatively much greater focus on prevention
through appropriate policy interventions.6
These policies come in two categories, just as they do with financial system risks.
First, we need to make sure that we do not come anywhere near an environmental
catastrophe.7 Second, and as a supporting precautionary measure, we need to ensure
that all the complementary mechanisms are working to guide both consumption and
production, towards products and services using cleaner technologies.
What we are trying to achieve is to avoid an ecosystem collapse, in the form of
an irreversible environmental disaster; do so in an effective and efficient manner; do
so in an equitable way, thinking of cross-society and inter-generational equity; while
5 Needless to say, this would be contested on the grounds that it is too narrow. There appears to be a
broader consensus around the nine interlinked planetary boundaries identified by Rockström et al.
(2009). This is not critical for our purposes.
6 This is not to deny that, for example, an overfished fishery can possibly return to its pristine state,
or at least become productive again.

7 The way we will express this in Chap. 6 is that none of the economic state variables should leave
the viability kernel.

at the same time, using a suite of policy instruments, switch both consumption and
production towards clean products and services.
The set of policies that are aimed at keeping the ecosystem away from catastrophic
boundaries are generically referred as “prudential policies” (see Carlin and Bowles
2017, Chap. 20), and typically have caps on various activities (such as emissions) as a
critical ingredient. As with policies that are aimed at an efficient and stable financial
system, these caps are complemented by market mechanisms such as “trading” –
thus “cap and trade” policies are common (as with emissions trading). Prudential
policies are policies that place a very high value on reducing the likelihood of a
disastrous outcome, even if this is costly in terms of other objectives foregone. Such
an approach is often advocated where there is great uncertainty about the conditions
under which a disastrous outcome would occur.
Complementary incentive measures include large-scale deployment of environ-
mental pricing (e.g., carbon price to deal with greenhouse gas emissions); R&D
subsidies favouring the switch to “clean” technology; and taxes and other penalties
against the use of “dirty” technology.
A critical ingredient is obtaining a political mandate, for the external costs of
development and production to be absorbed by those who reap most benefit. The
obstacle is the interface between politics and vested interests. That is why education
is such a critical ingredient of the solution; when the people understand and insist on
(say) carbon tax, politicians will introduce it.
Finally, we need to look for integrated and mutually reinforcing environmental,
social, and economic solutions towards achieving sustainability (such as education,
and poverty reduction).
Resilience to Social Systemic Risks
What would a catastrophe in, or collapse of, a social ecosystem look like? In recent
times, Yugoslavia and Syria provide a flavour of the answer. Plenty of other examples
exist throughout history. It results in the total collapse of social cohesion and possibly
social order.
The obvious systemic answer to immunising the social system to such a catastro-
phe is sustainable social cohesion.
Going back to the Wellbeing Triangle in Fig. 1.5 of Chap. 1, specific policies
towards that end have to be strongly grounded in the history, culture, and values of
the society one is living in.
In the context of New Zealand, there are (historical) bicultural and (emergent)
multicultural dimensions to ensuring social cohesion.
In the Maori-Pakeha (European) bi-cultural context, social cohesion has been sus-
tained through the recognition of, and partial compensation for, historical injustices,
education, and a deliberate set of attempts to promote and imbed bi-culturalism.
What also helps a lot in New Zealand’s specific context is the role that sport and
education play as catalysts towards building “bridging capital”.
What about the equivalents of market mechanisms, and associated incentives, that
are so integral to solutions for economic, financial, and environmental systems – do
they exist in the broader social sphere? The answer is yes, and it takes the form of
strong and diverse communities. Coupled with economic prosperity and a sense of
equity (or fairness), these provide strong incentives for people to live in harmony.
This provides a nice introduction to “localism” as a key ingredient of, and a policy
instrument for building, environmental, social, and economic resilience to potentially
catastrophic shocks.
5.5 Case for Localism
While most of the policy discussions above are framed in a national context, in fact
communities, localities, and regions have a critical role to play in the pursuit of
sustained wellbeing.
The role of a citizen-centred public policy is to support people and communities in
their efforts to improve their lives – by enhancing their opportunities and capabilities
(i.e., substantive freedoms) to pursue the lives they have reason to value, while
respecting and celebrating the diversity of valued lives.
In a country where different communities, cities, and regions are prospering,
potentially in very different ways, people have far wider opportunities to live the
kinds of lives they want to live, without leaving the country.
Central government agencies, assisted by technology that enables the collection
and sharing of information on what various communities are trialing to improve
their lives, could provide a live database for everyone to contribute to, and share
information about their experiments in improving their communities’ lives. Central
agencies also have access to the analytical capabilities to assess what works and what
does not, towards achieving community-defined objectives, to aid with investment
decisions.
Several people, including (Hayek 2013), note that the fundamental case for local-
ism (i.e., deliberately promoting and supporting the flourishing of local communities
and economies), is that local government is closer to the people it serves, it is better
able to reflect the needs of communities, it is critical for the health and vitality of
democracy, and it provides a greater variety of choice that can accommodate different
valued lives.
This increased devolved responsibility needs to be accompanied by increased
sources of funding – with associated strong controls and incentives. As Crampton
and Acharya (2015) argue, Regional and Local Councils should be given a share
of additional tax revenues generated when economic growth-enhancing policies are
successfully implemented in the regions.
This all fits in nicely with a very rich literature on complexity economics, insti-
tutional economics, and evolutionary economics, which are totally in tune with the
Darwinian theory of evolution – arguing that a deliberate policy of encouraging
different local experiments, and then selecting and amplifying the successful ones,
can help complex economic and social systems evolve and flourish. The distinctive
role of policy is to encourage, nudge, and support, while putting controls in place to
prevent the building of excesses (see Arthur 2014; Ostrom 2009).
In general, communities and regions want to own and solve their own problems;
they are keen to pursue a partnership model with the central government. They want
central government to facilitate, and participate in, exploratory conversations around
the respective roles and responsibilities of central and local government in promoting
regional economic development and broader wellbeing. One of the skills they do not
have is setting up controlled trials and then evaluating their performance. Nor do they
have the knowhow, technology, and resources, to collect and share the data gathered
from these trials so that they contribute to broader learning.
5.6 Setting Policy Priorities in a Complex and Uncertain

World
Let us return to the critical question posed by Martin and Pindyck (2015): faced
with numerous potential catastrophes, with uncertainties surrounding occurrences
and timings, which should society attempt to avert? Of course, we can also ask
the question in a positive way: faced with numerous investment opportunities at a
systems level, which should society prioritise?
This is also the type of question that the broader policy-oriented literature on
the United Nations Sustainable Development Goals is increasingly focused on (see
Griggs et al. 2013): taking into account the strong dependencies between these goals,
which should we prioritise so that we achieve the highest sustainable social return
on our investment?
The answer suggested by the above analysis, as well as the equations of motion
describing systemic environmental, social, and economic resilience in Chap. 2, is
that these priorities should be set by identifying policy interventions that have mul-
tiple, mutually reinforcing, benefits. From this perspective, and by way of just one
example, the reduction of poverty, broadly defined (see Alkire 2016), is a potentially
prime policy target. As we tried to demonstrate and highlight with the help of a
few numerical simulations in Chap. 3: reduction of poverty helps reduce inequity
and increase social cohesion; through that channel, it increases potential economic
growth, as well as increasing social and economic resilience; and it also directly
reduces environmental degradation and increases environmental resilience because
poverty is one of the biggest enemies of the environment.
As to how we do this, the answer provided in this chapter is clear: working with,
and through, communities, finding out what works in and for those communities,
and then investing in the solutions that work towards making them available across
the country.
5.7 A Simple but Complex Example 155
5.7 A Simple but Complex Example
We end this chapter with an example drawn from Eppel and Karacaoglu (2017). The
purpose of this example is to demonstrate how, even a seemingly very straightforward
policy intervention, implemented in a social setting, does not result in the anticipated
and desired outcomes, but rather requires ongoing substantive involvement of the
communities to make it work. There are other, and more complex, examples cited in
the same paper.
In setting the scene for this example, and very much in accordance with the
material of this chapter, Eppel and Karacaoglu (2017) note: “Public policy and man-
agement take place in a multi-centred and networked world in which what govern-
ments do is only one influence among many, and is arguably less influential than the
multiplicity of ongoing interactions among government and non-government actors
in producing the outcomes that result. This multi-centredness creates institutional
collective action dilemmas and fragmentation, which makes governance inherently
complex and the achievement of particular outcomes uncertain. It is the interaction of
a multiplicity of individual actors at the micro level that creates a desired macro-level
pattern, but it is also impossible for any actor to fully understand and control all the
actions that will create that outcome. In short, you cannot draw up a plan, follow the
steps, and end up where you planned” (p. 380).
The particular example involves early childhood vaccination.
“Vaccination against a raft of diseases in early childhood, such as polio, whooping
cough, mumps, measles and rubella, is relatively easy to argue for, on the basis
of the evidence of the individual and collective longer-term benefits to individual
and societal health, and to future costs. Investment in vaccination has been well
justified, in New Zealand and elsewhere, and targets have been set for the District
Health Boards (DHBs). However, what all the DHBs have amply demonstrated, since
vaccination of all children was made a health target, is that achieving the desired result
is not straightforward and requires many actors to be involved to achieve a positive
result.”
“In this example, there is both a macro and a micro problem to be solved, and the
knowledge and tools required at these different levels of complexity are different. At
the macro or national policy level, clinicians, epidemiologists, policymakers and so
on can tell us what the level of vaccination in the community is, and at what level
it is likely to be effective for lowering the risk and cost of disease to society as a
whole. The macro problem analysis might give us a detailed map of the areas and
populations where vaccination levels are too low for the avoidance of future risk and
cost. It will not help at all with understanding the particularities and peculiarities of
those areas and populations where the vaccination level is deficient.”
“For this, a more micro view at the level of the individual and their context is
needed. What is influencing the individual parents and their decisions about whether
or not to vaccinate their children and how does this vary across the areas of concern?
This is likely to be a multi-causal and changing story over time, as past educative
efforts take effect and new groups move into an area. Thus even when the micro
analysis leads to an understanding of the values and behaviours in a particular com-

munity so that these might be influenced in favour of higher levels of vaccination,
the community will continue to evolve and change in non-linear and unpredictable
ways (e.g., new waves of immigrants or changes in the economic circumstances of
an area).”
“As a result, ongoing achievement of higher vaccination levels cannot be assured.
There is a need for co-evolution between the strategies of the providers of vacci-
nations, the various social and health professionals working on this problem, and
the communities in question. Or to put it differently, [...] there is a need to span the
implementation gap through for example more decentralised decision making that
allows for co-production in the achievement of vaccination targets involving health
professionals, communities, and individual families.”
“[...] The macro level learning process can identify the populations and locations
for particular investment focus based on available historic data. However, this level
of information is useless for informing who and where the people most in need of the
intervention are now and how they might be induced to participate in a vaccination
programme. This needs a process of micro level prospective learning about a particu-
lar community through engagement with and in that community by the professionals
involved; understanding how it operates, what its values and fears are and so on will
be necessary for vaccination rates to change.”
“Few public policy areas likely to be the target of social investment are as ‘simple’
as the vaccination example. In most cases the end point/outcome of the investment
is less well defined and even more contested than the benefits of vaccination are for
a minority who resist attempts to encourage vaccination” (pp. 383–4).
1. Explain what is meant by “complexity” and “radical uncertainty”.

2. What is the difference between between “radically uncertain” and “risky” situa-
tions?
3. Explain why and how the introduction of “complexity” and “radical uncertainty”
to the stylised model of Chap. 2 makes any difference.
4. Critically assess how the choice of policy priorities and policy instruments is
affected in a model that allows for “complexity” and “radical uncertainty”.
5. Explain the impact of “complexity” and “radical uncertainty” on the focus and
design of public policy.
6. A town recently badly suffered from the consequences of unexpectedly severe
flooding. The Council is intending to respond by building tall flood-walls. You
have been employed as an adviser. What would your advice be? Explain.
7. A particular country is exposed to a major influx of migrants. You have been
asked to advise on how they should be responding. What would be your advice
and why?
5.7 A Simple but Complex Example 157

Readers who wish to pursue the subject of radical uncertainty, especially in a policy
context, would find reading (King 2016) extremely useful. The book also provides a
comprehensive reading list on the topic.
Readers who wish to pursue the subject matter of complexity, again in a policy
setting, would find it very useful to start by studying (Arthur 2014) and selectively
reading the references cited in that book.
Chapter 6
Viability Theory for Policy Formulation
Abstract In this chapter we build on the concepts of radical uncertainty and com-
plexity. We introduce these concepts into our policy discussion in a quantitative
manner. We earlier suggested that one of the key outcomes of allowing for radical
uncertainty and complexity in policy formulation, is a shift in the focus of policy
towards building resilience to systemic risks, and a much greater reliance on com-
munities in the implementation of such resilience-focused policies. In this chapter,
we introduce and propose viability theory as a mathematical tool for representing,
designing, and implementing a resilience-focused public policy. The chapter has
four purposes. First, to introduce viability theory into the policy tool kit. Second,
to explain when this tool is a useful alternative or complement to other policy tools
that are based on an optimisation framework. Third, to explain viability theory, and
differentiate it from optimisation theory, with the help of a few examples. Finally, to
apply viability theory to the problem at hand, which is to design policies focused on
building resilience to systemic risks, when the policy environment is characterised
by radical uncertainty and complexity. In this context, and within the framework
provided by viability theory, public policy has two (complementary) purposes. First,
to expand the “viability kernel”. Second, to ensure that the system remains within the
kernel when it is subjected to external shocks. To provide continuity with the rest of
the book, our examples are again centred on the interaction between environmental
(emissions), social (inequality), and economic influences on wellbeing.

• Explain the difference between optimisation and viability models.

• Critically assess the different uses and advantages/disadvantages of these models.
• Understand the concept of “calibration” and be able to explain how models are
calibrated.
• Explain and assess when viability models are best suited to use for policy-
informing advice.

https://doi.org/10.1007/978-981-13-6104-3_6
160 6 Viability Theory for Policy Formulation
• Explain and comparatively assess the similarities/differences and advantages/

disadvantages of the policy-simulation model of Chap. 4 and the model of this
chapter.
• Judge how viability models can be used to prioritise policy options.
• Critically assess the use and potential role of viability models when the world
we are developing policy advice for is characterised by “radical uncertainty” and
complexity.
6.1 Introduction
According to Simon (1955), some agents are only boundedly rational and do not
optimise a utility function to establish optimal – usually unique – economic evolu-
tions. Instead, these agents seek evolutions – possibly multiple – that remain within a
predetermined set of constraints. By way of example, such constraints may concern
health and/or environmental, social, and economic requirements, reflecting the will
of individuals and communities. Of course, fulfilling this aim may become a goal of
the government as well, acting as an agent for the community. If so, then maintain-
ing these evolutions inside the constraint set will be satisfactory to the government,
and the policies that fulfil this aim will be satisficing.1 Looking for strategies that
maintain evolutions of a dynamic system in a constraint set is a problem of viability
theory.
A diagrammatic representation of what policy is trying to achieve is captured in
Fig. 6.1, which we have borrowed from Raworth (2017). In our example, and in the
context of the integrated environmental, social, and economic framework we have
been working with throughout the book, what we are trying to do is to maintain the
system within the green band labelled, “the safe and just space for humanity”.
The government, as our representative, is looking for a feasible strategy (possibly
a vector strategy) that could maintain the aforementioned variables in a closed set
of constraints. Such a strategy, called viable, can be obtained as a solution to a
viability problem.
Now compare and contrast this with a situation where the government is looking
for an optimal (instead of a viable) policy solution.
The formulation of an optimisation problem for a dynamic system requires a
utility function, which calls for – in the case of multiple stakeholders (e.g., workers
and businesses), or influences on wellbeing (e.g., a clean environment and housing)
– weights to be assigned to the stakeholders’ (frequently) conflicting objectives. In
a democracy, assigning these weights is a politically sensitive process.
We have seen many examples of utility functions in Chap. 2. Consider e.g., W0 in
(2.9) – the summed or (integrated) discounted utility flows of current and future gen-
erations. There are no (explicit) weights in this case because there are no (explicit)
multiple stakeholders, or influences on wellbeing. However, this dynamic optimisa-
1 This neologism was introduced by Simon (1955).

Fig. 6.1 Safe place
tion procedure requires a value for ρ – the discount rate. In fact, the discount rate
weighs the generational utilities in the sum (2.9).
Similar to the assumption on values of the weights, assuming a particular discount
rate is politically sensitive. In particular, too high a discount rate neglects the stream
of utilities of the future generations. As a consequence, a strategy which maximises a
sum of thus discounted utilities will disadvantage the future generations. A possible
way to equilibrate the needs of the current and future generations is to maximise
W0 that depends on ω in (2.101), which is the Chichilnisky wellbeing function (see
Chichilnisky 1997). However, here too, we need to assume a value for ω – another
politically motivated parameter.
None of these problems is present when strategies are derived through viability
theory. Comparatively speaking, the constraints on clean air, water pollution levels,
inequality, economic growth, consumption, etc. should be agreed considerably easier
than the above weights. This is so because constraints refer to measures and quantities,
and knowledge on what the unacceptable levels of pollution, inequality, etc. are, is
usually available. A satisficing policy can maintain the economy in the constraint set
forever. This means that, among other things, policy advice based on viability theory
does not discriminate against any generation.
Furthermore, policy advice based on viability theory is more robust to parameter

uncertainty than “optimal” advice. An obvious reason for this is that a viable policy
requires less parameters than an optimal policy (e.g., no discount rate).
Another feature motivating the use of viability theory in policy advice is that opti-
misation is routinely concerned with strategies that steer an economy toward a single
steady state. If there is uncertainty regarding the model, then the claim that a single
state is what the government wants is difficult to justify. This is not the case for policy
advice based on viability theory; such advice can comprise several strategic options
out of which the government might select one, through some form of prioritisation.
In summary, if the government is able and willing to specify their preferences in
terms of constraints imposed on socio-economic indices of interest, then a viability-
theory approach to policy appears to provide the appropriate modelling framework
for generating supporting policy advice.
An important aspect of policy design is to devise resilient policies to allow for,
and hedge against, adverse effects of complexity and uncertainty. Some of these may
be accommodated in our models. Resilience can be naturally defined as a problem
of keeping some key economic indices in a closed set – so, it can be modelled as a
problem of viability theory.
As stated in the Preface to the book, under radical uncertainty, with the set of
possible states of the world and/or their probabilities unknown, optimisation at an
individual or government policy level becomes impossible. Here viability theory
might be of help. If, at least, a range of possible shocks is known, we can add
this information to the system’s dynamics. This will mean that there will be more
directions along which the system can evolve, and that some evolutions may be faster
than under certainty. As a result, the “viability kernel” (see below for a definition)
will probably shrink. This will alert the government that there are fewer viable states
than under certainty. Providing this information comes as a benefit of computing our
models.
As long as the viability kernel is non-empty, keeping the system in the constraint
set and away from non-viable states is possible. This represents one critical dimension
of a resilient system.
The next Sect. 6.2 provides an introduction to viability theory, followed by an
illustration of the differences between an optimal and a viable policy using a simple
economic example. We will then use viability theory and another simple model to
obtain policies that can keep environmental emissions under control. This will be
followed by replacing emission control with the degree of income inequality as a
constraint in the model. Then, in the final section of the chapter, we introduce both
emission and inequality controls as joint constraints into the model. Thus we retain
the common thread that runs through the book, focusing on the interaction between
environmental, social, and economic influences on wellbeing, as a key consideration
in designing policies aimed at increasing wellbeing on a sustained basis.
6.2 The Basics of Viability Theory 163
6.2 The Basics of Viability Theory
6.2.1 An Introduction to Viability Theory
Viability theory is a relatively new branch of mathematics (see e.g., Aubin et al.
2011). Overall, viability theory is concerned with identifying viable evolutions –
trajectories that do not violate some set of the state constraints over a given (possibly
infinite) time frame.
Definition 6.1 A viability domain is a set of initial states from which viable trajec-
tories originate.
It is, however, the viability kernel, which is the largest viability domain that provides
the basic tool for analysing constrained evolutions. We define formally the viability
kernel in Eq. (6.4) in the next section.
Viability theory is the study of dynamic systems that explores the set of initial
conditions which will generate evolutions that obey the laws of motion of a system,
while remaining in a certain closed set of state constraints for the duration of the
evolution. The state constraints can be imposed on some key economic indices like
inflation, consumption, etc. .
If the dynamic system is a control system, then the strength of controls (i.e.,
magnitude of instruments) will determine which initial states the system can originate
from, while remaining in the closed set. These are referred to as “safe” states. We call
the largest set of such initial conditions, the viability kernel. If the viability kernel is
non-empty, then the controller (in our case the government) knows which states are
“safe”. The government also knows the complement to the safe states, from which
no evolution can be confined to the constraint set, for the given strength of controls.
Constraining the size of control (instrument) adjustments is an important realistic
feature of solutions based on viability theory. For example, the speed with which
taxation can change may be constrained by an election promise. Varying this speed
and/or the bounds for the key economic indices will produce different solutions –
i.e., different viability kernels, representing different levels of resilience.
Most viability theory applications have focused on environmental policy – see
for example Béné et al. (2001), Martinet and Doyen (2007), De Lara et al. (2006),
and Martinet et al. (2007). Viability theory has also been applied to macroeconomics
(see e.g., Clément-Pitiot and Saint-Pierre 2006; Clément-Pitiot and Doyen 1999;
Krawczyk and Kim 2009; Krawczyk and Judd 2016; Bonneuil and Saint-Pierre 2008;
Bonneuil and Boucekkine 2014; Krawczyk and Sethi 2007).2 Further down in this
2 Forrigorous introductions to viability theory see Aubin (1991), Quincampoix and Veliov (1998),
Veliov (1993) and Aubin et al. (2011). Viability theory has also been applied to finance (see Pujal and
Saint-Pierre 2006), managerial economics (see Krawczyk et al. 2012), as well as microeconomics
(see Krawczyk and Serea 2013).
chapter, we will present two economic applications of viability theory. They will
illustrate how a government interested in policy design may profit from solving a
viability theory problem.
In the next section we present the concepts of viability theory which are essential
to our analysis.
6.2.2 The Viability Kernel – The Largest Viability Domain in

the Constraint Set
The core ingredients of a viability problem3 are:
1. A continuum of time4 values, Θ ≡ [0, T ] ⊆ R+ , where T > 0 is problem hori-

zon; can be finite or infinite.
2. A vector of n real-valued state variables, x(t) ≡ [x1 (t), x2 (t), . . . , xn (t)] ∈ Rn ,
t ∈ Θ that together represent the dynamic system in which we are interested.
3. A constraint set, K ⊂ Rn , which is a closed set representing some modal and
normative constraints imposed, or to be imposed, on these state variables. Viola-
tion of these constraints means that the system has become non-viable. Thus in
seeking viable trajectories, we always want to ensure that ∀t (t ∈ Θ) x(t) ∈ K .
4. A vector of real-valued controls, u(t) ≡ [u 1 (t), u 2 (t), . . . , u m (t)] ∈ Rm , t ∈ Θ.
5. Some normative constraints on the controls. In this chapter, we assume that u ∈ U
where U is the set of control vectors available at each state. (In general, the set
U can depend on x(t).)
6. A set of n real-valued first-order differential equations,
ẋ(t) = f (x(t), u(t)) (6.1)

u(t) ∈ U (x(t)) (6.2)
where
⎡ ⎤ ⎡ ⎤
ẋ1 (t) f 1 (x(t), u(t))
⎢ ẋ2 (t) ⎥ ⎢ f 2 (x(t), u(t)) ⎥
⎢ ⎥ ⎢ ⎥
ẋ(t) ≡ ⎢ . ⎥ and f (x(t), u(t)) ≡ ⎢ .. ⎥. (6.3)
⎣ .. ⎦ ⎣ . ⎦
ẋn (t) f n (x(t), u(t))
Each function f i : Rn × Rm → R, i = 1, 2 . . . n specifies, jointly with (6.2), the

range of velocities of the corresponding variable xi , at the state position x(t) ∈ Rn .
3 Thissection draws extensively from Krawczyk and Pharo (2013, 2014a) and Krawczyk and Judd
(2016).
4 A similar formulation could be made for a viability problem in discrete time.
Formulation (6.1)–(6.2) states that, at x(t) ∈ Rn , the change in the system’s state
– i.e., the system’s velocity ẋ(t) – can only be as much as the control from U (x(t))
allows. Thus, (6.1)–(6.2) jointly define the system’s dynamics F(·).
Definition 6.2 If K ⊂ Rm represents the closed set of constraints that state x(t)
must satisfy for all t, then we say that the system’s dynamics F(·) are compatible
with K if there exist states in K from which evolutions can originate that obey the
constraints indefinitely.
Such states form a viability domain.

Definition 6.3 Let K be a closed set in Rn . The viability kernel for dynamics F,
denoted: V F (K ) ⊂ K , is the largest possible viability domain under F. Mathemati-
cally,

V F (K ) ≡ x(0) : ∃x(t) originating at x(0), satisfying (6.1) − (6.2) and K ∀t
(6.4)
We can also say that the viability kernel V F (K ) ⊂ K is the set of all points that are
viable in K under F.5
Take, for example, an economy that is controlled by limited tax-rate adjustments.
Set K may be the Cartesian product of constraint intervals on debt and consumption.
The kernel V F (K ) will then consist of a combination debt and consumption levels
which can be maintained in V F (K ) ⊂ K forever by some limited tax adjustments.
Establishing the viability kernel V F (K ) = ∅ solves the viability problem. That is,
“good” – viable – states x ∈ V F (K ) are separated from “bad” – non-viable – states
x∈ / V F (K ). Where F represents a control model, this has important implications
for policy making, in that it can be used to formulate control rules that maintain the
system’s sustainability, as follows.
Consider a viable point x ∈ V F (K ). Let W (x) be a set of control policies available
at this x. Because x ∈ V F (K ) is a member of the viability kernel, then there must
exist at least one policy in W (x) that keeps the system’s evolution inside V F (K ).
Let WV (x) ⊂ W (x) denote the collection of these policies.6 With this notation, the
following sustainable policy rule can be formulated:
if x ∈ V F (K ) then apply an instrument u ∈ WV (x) . (6.5)
The above “generic” rule can be decomposed into two normative directives for a
given viability problem, where fr denotes the set frontier (or boundary):
i. in the interior of the viability kernel V F (K ) \ fr V F (K ) use any control from
WV (x);
5 Where K itself is a viability domain for F, K will be the viability kernel of itself.
6 They depend on x and can be called Markovian; Aubin (1997) calls them regulation maps.
Fig. 6.2 The viable and

non-viable trajectories
ii. when one gets “near” to the boundary and in particular when one is on the
boundary of the kernel fr V F (K ), an extreme instrument, or a specific path,
must be followed that takes us away from the boundary.7
We can say that the above controls are compatible with the inertia principle (see
Aubin 1991; Aubin et al. 2011), which states that self-sustaining dynamic systems
will not alter their controls except when the viability of the system is at stake.8 We
can similarly say that these controls are realisations of a satisficing policy postulated
in Simon (1955) – so long as viability is not threatened, any control is “good enough”.
Figure 6.2 provides a geometric interpretation of the concepts presented above for
a control problem on T = ∞, with otherwise deterministic dynamics. The state con-
straint set K is represented by the yellow (or light shadowed) round shape contained
in the state space (where X denotes the state space; here, X ≡ R2 ). The viability
kernel for the constraint set K , given controls from set U (x) and the system dynam-
ics F, is the brownish (darker) shadowed contour. The solid and dash-dotted lines
symbolise system evolutions. If an evolution begins inside the viability kernel V F (K )
then we have sufficient controls to keep it in the constraint set K for t ∈ Θ. The
system evolution represented by the trajectories that start inside the kernel (solid
lines) are viable in K i.e., they remain in K . Actually, since Θ = [0, ∞], the evo-
lutions remain also in V F (K ) ⊂ K . This is not the property of the other trajectories
(dash-dotted lines) that start outside the kernel. They leave K in finite time.
7 Unlessa steady state has been reached.

8 Anexample of this may be the actual “real world” behaviour of inflation-targetting central banks,
who will often avoid changing interest rates for as long as they can.
6.2.3 Viability Versus Optimality
6.2.3.1 Generic Differences
It cannot be over-emphasised that solving a viability theory problem is a different

proposition than computing optimal control.9
A crucial difference between these two approaches is that the formulation of a
viability-theory problem explicitly includes the set of acceptable states in K , whereas
in the optimisation approach, the constraints that define K are usually implicit in the
loss function. An immediate consequence of this is that problems modelled using a
viability approach do not need to specify utility or loss functions in order to formulate
policy rules. Therefore, there is no need to calibrate such functions, which would
amount to a subjective appraisal of which constraints are more important. Instead, it is
only necessary to determine the bounds of the set K – a potentially much simpler task,
given that such bounds (be they normative or modal) are often trivially observable.10
Optimisation determines the best strategy in the sense of an adopted utility func-
tion. However, e.g., in fisheries’ management, maximisation of profit may frequently
lead to non-smooth (also called bang-bang) strategies. Shutting the fishery down
every second year or so may be optimal for the fish biomass and also for the dis-
counted profit on [0, ∞]. Yet socially, such a policy would be undesirable. A solution
of a viability problem will tell the manager when and where – given the tolerable
smoothness (“strength”) of controls – viable strategies can be applied. Also, viability
theory helps decentralise the management process. Given the one and unique opti-
mal strategy, the manager “must” follow this strategy. Knowing the viability kernel
gives liberty to the manager as to what to do inside the kernel. This is because the
manager is certain that the system will remain within the viability constraints, as
using a predetermined strategy on the kernel boundary will keep the system in the
kernel.
It follows that knowing the viability kernel, which is a solution of a viability prob-
lem, makes the government aware of the locus of states in which the dynamic system
can continue to exist, for a given “strength” of implementable controls. Viability is
hence a generalisation of stability (and therefore closely related to resilience, which
can be defined as the system’s property to return to the desired state, after the system
was shaken by a shock), rather than optimality. This is so, because the kernel is a
closed set and it can be characterised by some measure that the distance between
two states in the kernel will never exceed. More importantly for the regulator, the
information about the kernel is sufficient for a realisation of a “satisficing” policy,
as opposed to an optimal one. In addition, the rather relaxed approach advocated by
directive (i.), Sect. 6.2.2, offers the government a possibility to strive to achieve other
goals (e.g., political), when more than one control u ∈ WV (x) is available. These
9 An optimal control problem can however be a special case of a viability problem if constraints are
so tight that only one path is considered viable.

10 E.g.,the inflation band in New Zealand, as in several other countries, is specified by a Policy
Targets Agreement signed between the Minister of Finance and the Governor of the Central Bank.
goals might not have been used for the specification of the constraint set K – perhaps
they were difficult to specify mathematically or they arose after the viability kernel
had been established; or they are considered merely “nice to have” – i.e., “wants”
rather than “needs”.
6.2.3.2 To Optimise or Not...
An analytical description of a viability kernel is rare but possible. We will describe

analytically a viability kernel for a stylised scarce commodity consumption problem
to see how viability theory might be useful in deriving rules that otherwise could be
thought of as behavioural. We will also suggest a possible optimisation problem as a
counterpart to the viability problem, to highlight some particular features of the latter.
The System’s Dynamics and a Viability Problem
Our commodity is of limited supply M > 0. The limit could be the total electricity
supply from a large trader, a city bus carrier’s total passenger capacity, a monopolistic
oil provider’s supply, etc. In each of these cases, even if the commodity supplier
is a private firm, it will be held responsible by the government if the delivery of the
commodity fails. The fines applied by the government are usually very high and can
threaten the firm’s survival. On the other hand, a zero demand for the commodity
would also be disastrous to the supplier. Hence, ideally from the supplier’s (and gov-
ernment’s) point of view, the demand for the commodity should be contained between
x and M, where x is some minimum demand level, satisfactory from the firm’s point
of view in that the firm will be able to satisfy it, while making positive profits.
The commodity that we are interested in is such that one can develop an “appetite”
for it.11 – i.e., one usually wants to consume more of the commodity in the next period
than now. This will be the case for electricity, public transport usage, chocolate, drugs,
heating oil, and many more. For example, if we find oil economical this winter, we
might want to use more of it next year.
Let x(t) denote demand for the commodity at time t ∈ Θ. Suppose that the
changes in the demand are proportional12 to the amount consumed at present. So, the
demand variation can be written as ẋ = ax(t) where a > 0. Clearly, if our “appetite”
is not modified, we will soon want to consume more than M, thus exhausting the
supply.
Think of compensation p(t)x(t) (where p(t) ≥ 0 is price) paid to the commodity
provider. Paying the compensation should slow down the demand. We conjecture
that the following equation captures the consumption habit of x:
ẋ(t) = ax(t) − p(t)x(t). (6.6)
11 Aubin (1997) discusses several simple viability problems. In particular, an affine and bilinear
system dynamics are considered (see ibid, pp. 46–51). Our model is inspired by the former but has
a similar structure to the latter, with the economic interpretation expanded.
12 Linearity of the changes is not crucial for the subsequent analysis.
If price p(t) changes with x(t) as a result of an appropriate pricing strategy p(x(t)),
then there is a chance of keeping x(t) in [x, M]. E.g., if x(t) is small and close to x
and the price is high, then dropping the price should prevent the demand for x(t) to
diminishing below x. If, on the other hand, x(t) is close to M and the price is low,
increasing the price should prevent the demand from growing above M.
A way of managing the commodity through strategy p(x(t)) can be inferred from
the viability kernel established for the dynamics (6.6) and a constraint x ≤ x(t) ≤ M.
The prices will also be constrained: by a price cap, p, related to inflation, or above
which consumers switch to a different product or technology; and by a lower bound
p, which may come from some profitability condition p · x ≥ P > 0.
Mathematically, the supplier may be seeking a satisficing strategy p̃(x(t)) for
t ∈ Θ such that
• given the state equation (6.6),
(x(t), p(t)) ∈ [x, M] × [ p, p], ∀t ∈ Θ . (6.7)
In this case, evidently, the constraint set is represented by:
K ≡ [x, M] × [ p, p] . (6.8)
The above constraints may need be supplemented if the commodity price is

sticky,13 or should be sticky (i.e., smooth). If there is a limitation on how fast the
price may change, then – if e.g., demand is accelerating – a price path that assures
x ≤ x(t) ≤ M ∀t will have to be special. Therefore the commodity price will have
to be controlled by the supplier with an additional constraint14 on an admissible level
of stickiness c,
ṗ ∈ [−c, c] . (6.9)
An Optimisation Problem
The supplier may, however, be ambitious and strive to obtain an optimal solution to
their problem, which could be formulated as follows:
13 E.g., high demand could cause the “clearing” price to rise faster than inflation. However, the
supplier may not want to feel some kind of a social condemnation caused by the prices behaving
too differently from inflation. Also, the supplier may not want to change the menu too frequently.
All that means is that the price cannot jump.
14 We want to mention that relationship like (6.9) (also see (6.23), (6.38) and (6.53)–(6.54)) is
frequently called – in viability theory – a differential inclusion, see e.g., Aubin et al. (2011). A
differential inclusion is a generalisation of the concept of an ordinary differential equation where
the right hand side of the equation is a multi-valued map. In other words, for a given state-space
point, a differential inclusion tells us about the many possible evolutions of the state variables from
that point, rather than a unique evolution determined by an ordinary differential equation. The “non-
ordinary” feature of (6.9) is that its right-hand side is an interval [−c, c] and not a single value.
The “bunch” of possible state evolutions for the system’s dynamics can be obtained by solving the
system’s equations for the entire contents of interval [−c, c].
• given the state equation (6.6), compute p̂(t) ∈ [ p, p] such that
T
e−ρt p(t)x(t) − W1 [max(0, x(t) − M)]2 − W2 [max(0, x − x(t))]2 dt .
0
(6.10)
is maximised where ρ is the discount rate, and W1 , W2 > 0 are penalty coefficients
on the square functions [. . . ]2 (which are justifiable but non-unique). Notice that we
have assumed away the supply cost (or absorbed it in p(t)).
Maximisation of (6.10) in p(t) is a difficult problem to solve. It is an optimal
control problem15 with a non-linear, non-differentiable objective function. It also
contains three arbitrary parameters: ρ, W1 , W2 .
However, if the fines to pay for disrupting supply are very large, then W1 , W2 are
very big. If, furthermore, ρ is small, then the problem boils down to maintaining
the demand and price in rectangle [x, M] × [ p, p], which is identical with problem
(6.7).
We observe that the arbitrary constants ρ, W1 , W2 do not enter the specification
of problem (6.7). On the other hand, maximisation of (6.10) cannot be performed
without knowing their values.
The Viability Kernel
We will solve the problem of selecting (x(t), p(t)) ∈ K for all t ∈ Θ, subject to the
system’s dynamics (6.6) through a viability analysis. This means we will establish a
Markovian price policy p̃(x(t)) ∈ [ p, p] that keeps (x(t), p(t)) in K . The problem
of how one variable – here p(t) – should behave so that all variables are constrained
– here (x(t), p(t)) ∈ K ∀t ∈ Θ – is a typical viability-theory problem.
Additionaly we assume that the commodity is non-stockable (e.g., because of
some fire regulations or perishability) and we cannot buy more of it than we are able
to consume at once.
Based on what we have described above, we propose that the described process
has the following dynamics F (compare (6.1)–(6.2)).
ẋ = ax(t) − p(t)x(t)
(6.11)
ṗ = u ∈ [−c, c].
Let us calibrate this model as follows. Let the maximum supply M be 1 and
x = 0.1; price p(t) can vary between 0.02 and 0.2 . Let the coefficient of appetite
growth a = 0.05 and price stickiness c = 0.001.
We can see the constraint set K in Fig. 6.3 as the yellow (light) shaded rectangle
whose vertices are (0.1, 0.02), (1, 0), (1, 0.2), (0.1, 0.2). If price p(t) = a, the line
p(t) = a defines a collection of steady states – see the solid line at p(t) = 0.05 and
also refer to system (6.11). However, each state on this line is unstable. Indeed, below
this line, p < a ⇒ ẋ > 0, and if the price drops below a and remains constant, the
demand will drift to the right; see the dash-dotted evolution marked C.
15 Maximisation of (6.10) in a Markovian (i.e., feedback) strategy would be even more difficult.
Fig. 6.3 The viability kernel
Above the line, p > a ⇒ ẋ < 0 and, if the price remains unchanged, the demand
will drift to the left – see the evolution marked C’. It is clear that if we are off steady
state and do nothing, we will hit one of the physical constraints on x in finite time.
To prevent exhaustion (x = M) or extinction (x = x), the price must change.
The viability kernel V F (K ) will be established through an explicit calculation
of the critical evolutions that bring the system, under the full speed c (or −c), to a
steady state that intersects with the constraints. In Fig. 6.3, these critical evolutions
are represented by the thick red dash-dotted lines that terminate at (1, 0.05) – the
upper frontier of V F (K ), and (0.1, 0.05) – the lower frontier. States (1, 0.05) and
(0.1, 0.05) are the “last” acceptable steady states. The kernel V F (K ) is represented
by the curvilinear pentagon delimited by the red dash-dotted lines and fragments of
the constraint set K , as shown in Fig. 6.3 – see the green-grey shape.
The analytical forms of the frontier evolutions are obtained as solutions of (6.11)
which are:
x
p(x) = a + 2(−u) ln , (6.12)
C1
for the top frontier of the kernel, and

x
p(x) = a − 2(−u) ln . (6.13)
C2
for the bottom frontier of the kernel. Calibrated for u = −0.001 (the top frontier of
the kernel, C1 = 0.1), they are:

p(x) = 0.05 + 0.002 ln (10 x) , (6.14)
and for u = 0.001 (the bottom frontier of the kernel, C2 = 1)16 :

p(x) = 0.05 − −0.002 ln (x) . (6.15)
To steer away from M, when p < a, we should increase p. Similarly, for p > a,
we should decrease p to avoid x = x. With ṗ > 0 and, respectively, ṗ < 0, the price
will go to a steady state on the line p = a. This is sufficient for x to remain constant,
which might be desired by the regulator. The dotted lines in Fig. 6.3 show the system’s
trajectories inside the kernel. When prices change at the “full velocity”, c = 0.001 on
evolution A and c = −0.001 on evolution A’, both evolutions terminate at a steady
state inside V F (K ) ⊂ K . So, for evolutions like A and A’ the strategy of changing p
at full speed is sufficient to prevent x from crossing the boundary of the constraint
set K . The evolutions C and C’ are for u = 0. As shown in Fig. 6.3 they are viable
because they can intersect with evolutions which terminate at a steady state.
However, the evolutions like those originating from B, B’ or B1 ’, which change
at the full speed, are bound to violate one of the constraints in finite time. This is
because the price cannot change fast enough. (For |c| > 0.001 the viability kernel
will be larger. See Fig. 6.24 in the section on resilience – Sect. 6.4.) We may say
that if the system has somewhat drifted to B, B’ or B1 ’, it is too late to stabilise it.
Evidently, the supplier should have acted earlier.
Now, we can see how a viability analysis could help the supplier to establish a
viable price strategy. Should there be no other considerations regarding price, the
policy can be17 :
⎧
⎪
⎪ if (x, p) ∈ V F (K )\fr V F (K ), apply any price adjustment u ∈ [−c, c];
⎪
⎪
⎪
⎪
⎨
if (x, p) ∈ fr V F (K ) (x, p) : p < a , increase price u = c ;
⎪
⎪
⎪
⎪
⎪
⎩if (x, p) ∈ fr V F (K ) (x, p) : p > a ,
⎪
decrease price u = −c ;
(6.16)
where denotes intersection. In words, use an admissible price change inside the
kernel and change the price at full speed at the frontiers.
Should a steady state ( p = a) be achieved, it might be maintained, until a shock
strikes, by the following policy rule:

if (x, p) ∈ V F (K ) (x, p) : p = a , keep price steady u = 0 . (6.17)
If the shock does not push the system outside the kernel, viability will not be lost and
using (6.17) can continue. If the shock pushes the system outside the kernel, special
16 Remember that ln(x) is a negative number for x ≤ 1; its product with −0.002 is positive.
17 Denote the kernel’s boundary by fr V F (K ); hence V F (K )\fr V F (K ) is the kernel’s interior.
crisis control needs be implemented, see Doyen and Saint-Pierre (1997) and Béné
et al. (2001).
As observed earlier (after rule (6.5)), the policy advice can be passive: if the
current point is within the viability kernel, then apply any u ∈ [−c, c] (see the first
“if” in (6.16)). The policy will be active (i.e., p = c or p = −c), like in the second
and third “if” in (6.16), if the current point is at the frontier of the viability kernel.
When the policy is passive, other goals not included in K (e.g., the integral (6.10)
could be minimised; a steady state could be selected with demand above, or below,
a certain value) may be realised.
We notice that the curvilinear boundaries of the kernel have an attractive economic
interpretation. For example, consider the lower part of V F (K ). For a given price p
and consumption x such that point ( p, x) is below the steady state line, consumption
of a rational consumer can grow until the limit of the viability kernel is reached (just
imagine a continuation of evolution C until the red dash-dotted line is attained). Then,
on the frontier, the active policy – second “if” in (6.16) – tells the consumer, how the
price must, and will, evolve so that x ≤ M. This is important for the consumer who
might be solving their own optimisation (or viability) problem where a demand law
will be part of the problem specification. Notice that the equation of the frontier is a
dynamic inverse demand law, which we have obtained endogenously.
6.3 Applications
This section is intended to demonstrate how viability theory can be used to design
wellbeing-enhancing policies, when the main focus of policy is on creating and
protecting a set of safe environmental, social, and economic ecosystems within which
individuals, communities, and businesses can pursue their objectives.
We are exploring how a viable public policy can be devised for the system pre-
sented in Fig. 1.2. The regulator first quantifies some constraints on the four capital
assets (i.e., comprehensive wealth) that are fenced by the five policy domains. The
constraints may reflect efficiency levels for each capital asset18 or some desirable
bounds. Then, using the equations of motion from Chap. 2, the existence of admissible
strategies (for taxation and/or other instruments, if available) that maintain the four
capital assets inside their respective constraint intervals could be established. The
existence of such strategies will imply non-emptiness of the viability kernel for this
system. Then, constraints on wellbeing, represented by the five socio-economic pol-
icy domains (social cohesion, equity, economic growth potential etc. – see Fig. 1.2),
could be imposed. The resulting strategies could then be verified to establish whether
they can maintain the system inside the viability kernel and keep the indices within
their respective constraints. If such strategies exist, then they are the candidates to
be applied by government as sustainable public policies. Below, in Sect. 6.3.1 we
18 For example, we know that if economic capital is larger than its golden rule steady state value,
then that capital level is inefficient.

first examine public policies for sustainable emissions, and then, in Sect. 6.3.2, for
sustainable inequality. In Sect. 6.3.3, we look for policies guaranteeing joint sustain-
ability of emissions and inequality, and in Sect. 6.4, we study the resilience of an
economic system to various types of shocks.
6.3.1 Sustainable Emissions19
In this subsection, by way of demonstration through examples, we use viability the-

ory, in the context of a calibrated neoclassical economic model, to identify sustain-
able economic states. We identify economic states that, under smooth adjustments
to abatement policy, remain in the efficiency region and, additionally, the level of
emission in each such state satisfies an acceptability constraint.
6.3.1.1 Capital, Consumption, and Emissions Control – a Stylised

Model
In the model below, as in Fig. 1.2, multiple capital assets are necessary to assure
sustainability. In our model, economic capital carries the main responsibility. More-
over, we introduce a separate state variable for consumption, which acts as a proxy
for social cohesion, as well as serves other purposes such as maintaining acceptable
health levels.20 Sustainability consists of keeping emissions in an acceptable range
while assuring that capital and consumption remain within their limits of acceptabil-
ity as well. We suggest that the model here is a reduced version of the wellbeing
frontier introduced in Fig. 1.2. Our aim is to establish the safe space as depicted in
Fig. 6.1.
Output y is produced using a neoclassical production function. In per capita terms,
the production function is A1−α k α where α ∈ (0, 1), k is the per capita capital stock,
and A is the exogenously given labour-augmenting technology, which is constant in
our model.21 Obviously, k, as well as the other state and decision variables (here, c, z
and v), are time-dependent. However, to unburden the notation, the time argument t
will be dropped unless necessary.
When private economic actors make a market-facing decision, they treat the total
pollution level – denoted X – as given, thereby causing too much emissions. In a
situation like this, the government will impose emission controls.
19 This subsection draws from Kim and Krawczyk (2017).

20 Consumption is here defined as a state variable. We derive a consumption equation from a trans-
formation of the Euler equation, obtained as a solution to private agents’ utility maximisation. See
footnote 22, in Sect. 6.3.1.1, for a brief explanation and consult the publications listed there.
21 Here we depart slightly from the stylised model presented in Chap. 2 and use e.g., Lee (2012)
or Krawczyk and Judd (2016) who assume that the level of technology is given and fixed. In “real
life” technology improves but the improvement process is indeterministic. Therefore, our results
correspond to the worst case scenario of technological progress.
6.3 Applications 175
Emission control is represented in our model by the variable z = 1 − θ , where θ ∈

[0, 1], represents the proportion of output y devoted to abatement z (see Eq. (6.19)
below). More abatement (i.e., higher θ , therefore lower z) reduces pollution X . This
relationship is captured by the following equation:
X = A1−α k α z β , where β > 1 . (6.18)
The parameter β > 1 assures abatement has a positive but diminishing marginal
impact on emission reduction.
However, the effort expended towards emission reduction takes away resources
for abatement. The output, net of abatement cost, which can be used for consumption
or capital accumulation, is then given by:
y = A1−α k α (1 − θ ) = A1−α k α z (6.19)
where y is the per capita output.

The following economic capital state equation represents the economy-wide
resource constraint:
k̇ = A1−α k α z − δk − c (6.20)
where c is per capita consumption and δ > 0 is depreciation. We see that net invest-
ment is positive when output A1−α k α z (net of abatement cost) exceeds the sum of
capital depreciation and consumption δk + c.
Because of the technology of abatement, a trade-off occurs between consumption
and the environment: the greater the effort expended towards emission control θ
(the smaller z), the less output is produced (or is available for consumption, and
other forms of investment). In a situation like that, the government needs to select
z to maintain a compromise between cleanliness of the environment and sufficient
output, capable of maintaining acceptable levels of consumption (hence the link of
consumption to social cohesion).
The other state variable in this model is consumption. We obtain the equation of
motion for consumption from the Euler equation in the representative agent’s utility
function maximisation22
c
ċ = α A1−α k α−1 z − δ − ρ . (6.21)
σ
∞ c1−σ − 1
22 Briefly, apply the maximum principle to maximise, in c, e−ρt dt where ρ > 0 is
0 1−σ
the rate of time preference and σ > 0 is the relative risk aversion parameter, subject to (6.20). Take
the shadow price of capital λ > 0, which is equal to the marginal utility of consumption, and obtain
1
c = 1 . Then, (6.21) follows from the equation for shadow price λ̇. For more details see e.g.,
λσ
Krawczyk and Judd (2016) or Driscoll-O’Keefe and Krawczyk (2016).
So, consumption increases as long as the real interest rate, α A1−α k α−1 z − δ, exceeds
the time preference rate ρ.23 We can see that consumption may slide toward the trivial
steady state of zero if depreciation or consumer’s time preferences are large relative
to marginal productivity of capital. The latter is controlled by emission control z: too
low z implies c → 0. Therefore the government has to be careful in selecting z.
We also note here the interpretation of σ , as the coefficient of relative risk aversion
(1/σ is the elasticity of inter-temporal substitution). The same interpretation concerns
consumption σ in Eqs. (6.34) and (6.51).). High values of σ , which characterise risk
averse agents, make the speed of consumption changes slower than under low values
of σ . This may be partly a reflection of lower consumption rates of highly risk averse
individuals who “prefer” to save, to build shock absorbers, rather than consume.24
We will see in Fig. 6.25 in the section on resilience – i.e., in Sect. 6.4 – that, indeed,
a larger σ corresponds to the larger viability kernel that better absorbs shocks than
the smaller kernel.
The government needs to control z to maintain k, c within some agreed bounds,
considering the dynamics implied by the private agent’s intertemporal choices
described by Eqs. (6.20) and (6.21).
However, one more concern for the government is to limit the level of emission
X to below X̄ ; so,
X = A1−α k α z β ≤ X̄ (6.22)
where X̄ may be selected as a limit above which the society’s health will suffer.
Because it is by and large impossible to change the levels of z instantaneously,
the government acts incrementally by selecting the speed u ∈ U with which z can
change. This is captured by the following differential equation coupled with a speed
constraint:
ż = v where v ∈ V . (6.23)
In this one-dimensional case, U is just the interval [v, v̄].

Overall, the economic dynamics at hand is represented by (6.20), (6.21) and (6.23)
and jointly referred to as the map F (compare F defined by (6.1)–(6.2) in Sect. 6.2.2).
We recognise that this system is nonlinear and has multiple steady states. We can
see that the dynamics (6.20), (6.21) and (6.23) will be difficult to control to satisfy
(6.22). From (6.20) and (6.23), we know that increasing v to increase z (so decreasing
abatement) will accelerate capital growth. But, as shown in equation (6.21), this will
not necessarily increase consumption because the marginal product of capital is low
for large capitals. As a result, the parenthesised term in (6.20) may become negative
and consumption will decrease. Decreasing z to diminish emission X slows down
23 We claim in Sect. 6.1 that an economic viability problem does not require specification of the
discount rate, or time-preference. Therefore, we say, viable policy does not suffer from a gener-
ation bias. This is true in the government’s problem. The emission-control strategy, computed in
Sect. 6.3.1.3, is generation independent. The time-preference rate ρ in (6.21) is entirely due to the
standard formulation of the representative agent problem.
24 It can also reflect the strong possibility that risk averse individuals would like to ascertain whether
a shock to income is temporary or permanent, before they adjust their consumption rates in response.
capital growth and, unless capital is very small, can also have a negative impact on
consumption growth.
6.3.1.2 The Viability Problem and Calibration
The government will want to know if there exist abatement strategies that conserve
the efficiency of the system – i.e., the system remains within the agreed constraints
of the state variables (see footnote 18 in Sect. 6.3.1.1), and is also sustainable – i.e.,
the strategies do not result in violation of the environmental constraint (6.22).
The constraint set for the state variables is defined as the set K :

K ≡ (k, c, z) : k ∈ [k, k̄], c ∈ [c, c̄], z ∈ [0, 1] , (6.24)
compare item 3. in Sect. 6.2.2 and the definition of the constraint set K (6.8). The
limits k, k̄, c, c̄ are normative, determined e.g., by the golden rule, and positivity of
consumption for subsistence. The constraint for z is modal – i.e., the model would
be uninterpretable for z ∈ / [0, 1]. Furthermore, keeping (k, c, z) ∈ K is a result of
u ∈ [v, v̄], where the decision about the size of v and v̄ rests on the government’s
desire and abilities concerning the speed of emission-control adjustments.
In this case, the viability problem for dynamics F – (6.20), (6.21), (6.23) – consists
of establishing the viability kernel composed of initial positions (k(0), c(0), z(0))
from which evolutions emanate, that remain in K and, additionally, (k(t), z(t)) must
be such that X (t) ≤ X̄ ∀t. In other words, so that it does not cross the boundary of
K, the economy should remain inside the viability kernel, i.e., in the “safe space”, in
the sense of Fig. 6.1.
We will assume that ρ = 0.04, α = 0.4, σ = 2, δ = 0.1 characterise, in broad
terms, a reasonably industrialised economy, composed of rational agents interested
in the near future.25
The other parameters that need to be calibrated are A, β, the limits k, k̄, c, c̄, v, v̄,
and the emission bound X̄ .
Following Stokey (1998) (Sect. 4), we propose β = 3. This means that the abate-
ment technology is fairly efficient: for z = 0.5 which reduces output by 50%, emis-
sions will be reduced by 87.5%.
We now normalise the unit of capital measurement. Let the steady state of capital
ks equal one when there is no abatement (z = 1). Then, for ċ = 0 and when ks = 1
and z = 1, (6.21) yields:
1−α
1
δ+ρ 5
A= = 0.35 3 ≈ 0.1738 . (6.25)
α
25 For example, for the representative agent’s time preference ρ = 0.04, what can happen in 50 years
e−0.04·10 0.67
is about 5 times less important than what will happen in 10 years. Indeed, −0.04·50 = ≈ 5.15.
e 0.13
However, as commented on σ in footnote 23, the government’s viability problem is generation-
neutral.
We propose that the lower bound on capital should be 10% of ks , so k = 0.1. We

will set the capital upper bound k̄ in relation to the golden rule. The golden rule capital
maximises steady state consumption. From (6.20), in steady state c = A1−α k α − δk.
1−α 1−α 1
αA
It is maximised for k = = 1.2237 and reaches 0.2571. We will set
δ
k̄ = 1.25, which is slightly above the golden rule capital stock.
To propose the limit c, it will be convenient to know the steady state consumption
cs that corresponds to ks = 1 and z = 1. From (6.21), cs = A1−α − δ = 0.25. We
set a lower limit c = 0.01, which is 1/25 of the consumption steady state without
abatement. We set c̄ = 0.26 as the upper bound, which is slightly above the golden
rule level of consumption.
From (6.22), the no-abatement steady state emission X s = A1−α = 0.35. We will
assume that X̄ = 0.275 is the allowable emission level, which constitutes about 80%
of X s .
Finally, the emission-control adjustment speed v will be selected from [−0.25,
0.25].
In the end, we can write the set K as the following Cartesian product:
K = [0.1, 1.25] × [0.01, 0.26] × [0, 1] . (6.26)
Our aim is to establish a “safe space” inside K . Remaining in that space will guarantee
that the economy will not move out of K – compare Figs. 6.1 and 6.2.
6.3.1.3 The Viability Kernel and Viable Policies
Capital-Consumption Dynamics
Consider Fig. 6.4. It illustrates the behaviour of the calibrated system in the k, c
space for two different emission-control levels: z = 1 (left panel) and z = 0.5 (right
panel). It is apparent that active emission control (see the right panel steady state), is
associated with lower capital and lower consumption than when emissions remain
unabated, as in the left panel.
Furthermore, Fig. 6.4 confirms the known economic property of the capital-
consumption relationship – that a steady state can only be achieved through a saddle
path (see the black dotted lines in each panel). When an economy is above the sad-
dle path, consumption grows but capital collapses. When an economy is below the
saddle path, capital grows but consumption declines. We also notice that the further
from a saddle path it is, the faster the economy moves – this is symbolised by the
length of the arrows.
The question which the government would like us to answer is whether there are
paths of z (actually, v) which can steer the economy toward a saddle path that will
lead the economy to a steady state inside K .26 If an economy is to be steered towards a
26 Orbiting in K would also be acceptable.

Fig. 6.4 k, c quivers for z = 1 and z = 0.5
saddle path, some diverging arrows in Fig. 6.4 will have to react to the changing z and
turn toward a saddle path. However, given the limited size of z, the turning may not be
sufficient to change the economic evolution, if the arrow corresponding to the current
(k, c) is “long” (i.e., if the economy is moving fast away from the saddle path). The
set of k, c from which the turning is sufficient to change the economic evolution is
the viability kernel. Looking at Fig. 6.4 where the “short” arrows are limited to close
neighbourhoods of the saddle paths, we can anticipate that this calibrated model may
have a rather small viability kernel.
To repeat, in this model, k, c, and z are state variables. Economic agents only know
the current value z that is controlled by v.27 The government treats the first order
conditions of private economic agents as a given and selects v to smoothly adjust
z over time to retain k, c, and z in the state constraint set K . However, a smooth
adjustment of z will exist only if k, c, z belong to the viability kernel.
The process of determining the viability kernel (i.e., solving a viability problem), is
computationally intensive. However, thanks to some specialised software, computing
viability kernels for simple models (at present up to five state variables and two
controls) is possible. The software we use is VIKAASA, briefly introduced in an
appendix to this chapter.
The application of VIKAASA enabled us to obtain the viability kernel V F (K ),
presented in Fig. 6.5 left panel whose dimensions are the state variables k, c, z. The
box represents the constraint set K . Here, the grey “boulder” represents the viability
kernel.28 It comprises the “viable” states from which the economy can generate
27 If the economic agents (perhaps mistakenly) treat z as a constant, then they have a standard
optimal control problem. Its Markovian solution is a consumption function c(k; z), obtained from
the Euler equation in the agent’s optimisation problem. The agents constantly revise their plan, in
light of new information about z; however, the consumption function remains the same.
28 A few explanatory remarks on the method of interpreting the boulders may be necessary. The
dimensionality of our problem is 3 (i.e., there are three state variables k, c, and z). Emission X is
Fig. 6.5 3D slices of the viability kernel
evolutions terminating at a steady state inside K .29 In the complement of the kernel,
the states are “non-viable”.
The purple (smaller) boulder comprises the states from which the economy can
generate viable evolutions, such that X ≤ X̄ ∀t.
In the right panel, the same colour convention is applied. Here the slices are drawn
in k, c, X for all admissible values of z and y. Emission X is on the vertical axis, for
which we do not draw an upper limit because X is unbounded for the unconstrained
emission problem.
The following observations can be made regarding the left panel of Fig. 6.5:
i. the boulders in the left panel are “thin” along the z-axis, which indicates that there
are numerous combinations of (k, c, z), for which there is no path of z, resulting
from admissible adjustments u ∈ U , which could keep the economy in K ;
ii. however, the boulders become moderately “thicker” for low z (i.e., for stricter
emission controls) – hence, there are more viable (k, c) combinations in
economies that engage in strong emission control than in those that do not;
iii. viable consumption is closely aligned with capital; in fact, the viable (k, c) com-
binations form a narrow viability “corridor” which we attribute to the necessary
proximity of k, c to the saddle-paths, shown in Fig. 6.4.
The right panel confirms the alignment of the variables (k, c). Furthermore, we see
that unabated emission will not exceed 0.35 and many economic states have emission
levels below 0.275 – i.e., lower than the imposed upper limit on emission.
Now, consider Fig. 6.6. Here, we see 2D slices of the viability kernel, in coordi-
nates (k, c) – left panel and in (k, z) – right panel. The left panel shows the 2D-slice
through (z, X, y); the right panel is the 2D slice through (c, X, y).
We can see in the left slice that, when consumption is large, the viability kernel
(grey boulder) does not overlap with (k, c) that guarantee X ≤ X̄ (purple boulder).
also a variable of interest in this problem and so is, to a limited extent, economic output y. These
values (i.e., k, c, z, X , and output) are computed by VIKAASA and stored in a 5D array. Hence, the
3D boulders in Fig. 6.5 are 3D “slices” through the 5D array. In the following figures, the system’s
slice through z : emissions control=all, x: emissions=all, output=all slice through consumption=all, x: emissions=all, output=all
Fig. 6.6 2D slices of the viability kernel
This is so, because, when consumption is large, output will not suffice to maintain
capital and keep abatement high (small z) – see Eq. (6.20). We can also see more
clearly than in Fig. 6.5 that, for viability, consumption has to be somewhat propor-
tional to capital – to belong to the “corridor”, referred to in item iii above. In the right
panel, the states characterised by large capital and lax emission controls, cannot be
viable.
Evidently, for viability, a point must lie in both slices – i.e., in the kernel.
Examples of Viable and Non-viable Evolutions
We will now illustrate our claim that the viability kernel is a collection of all initial
conditions of the system, from which viable evolutions originate. To repeat, the
viable evolutions satisfy all imposed constraints, on the state trajectory, controls,
and emissions, for ∀t ∈ Θ. We will illustrate this by showing that viable evolutions
originate within the kernel, and by indicating that if an evolution originates outside
the kernel, at least one of the imposed constraints, normative or modal, will not be
satisfied.
There are 10 points of initial conditions, marked by the numbers 1–10, displayed
in each panel of Fig. 6.6. These initial conditions represent 10 possible economic
states – e.g., (k1 , c1 , z 1 ) = low capital, medium-small consumption, weak emis-
sion control , (k2 , c2 , z 2 ) = very low capital, very small consumption, very weak
emission control , etc. While evolutions originating at each point are not necessarily
unique, their general directions are indicative of where an economy can evolve to,
from a given point.
The lines which emanate from the points (1, 3, 5, 6), terminate at points inside
the kernel and represent viable evolutions. The lines from the points 9 and 10, which
crash through a boundary of K , represent non-viable evolution. (Yet, their initial
dynamics will also be discussed with the help of 2D slices. The variables, through which the 5D
array is sliced, are listed in each picture’s title.
29 Actually, in V (K ) ⊂ K , because the horizon is infinite.
F
Fig. 6.7 Time profiles of viable evolutions
conditions may look “promising” because they start from within the (c, z) slice of the
viability kernel). At the points (2, 4, 7, 8), the initial combinations of k(0), c(0), z(0)
are already outside the (c, z) slice of the viability kernel. This makes these points
evidently nonviable.
Some of these evolutions are also represented as time profiles in Figs. 6.7 and 6.8.30
Here are the descriptions of the evolutions originating at each position.
1. Low capital, medium-small consumption, weak emission control, modelled as
(k, c, z) = [0.2769, 0.1062, 0.8846].31
This point is viable. This point may be typical of a less industrialised (but not
starving) country, which would like to strengthen emission control. The corre-
sponding evolution from this point is represented by the black dashed lines in
30 The time profile figures display the time paths not only of k, c, z, and X , but also of output y,
√
emission control adjustments v, and “velocity”. The velocity is the Euclidean norm k 2 + c2 + z 2 ,
which informs us about the steadiness of the economy. Evidently, the closer the velocity is to zero,
the closer the economy is to a steady state. Unsurprisingly, all velocities converge to near-zero in
Fig. 6.7 where viable evolutions are plotted, whereas they fail to converge to zero in Fig. 6.8 where
non-viable evolutions are plotted.
31 These numbers result from a particular discretisation of the set K implemented by VIKAASA,
necessary to obtain the viability kernel and the economic evolutions.

Fig. 6.8 Time profiles of non-viable evolutions
each panel of Fig. 6.6.

Increasing emission control from this point is very successful: capital and con-
sumption grow and emission diminishes. Such is this case because decreasing
z in an economy with low k and c can generate growth of each of these vari-
ables. This is mainly because capital productivity is high for low capitals (see
Eqs. (6.20), (6.21)).
The time profiles corresponding to this evolution are shown in Fig. 6.7 also, as
the black dashed lines.
2. Very low capital, very small consumption, very weak emission control, modelled
as (k, c, z) = [0.1442, 0.0485, 0.9615].
This is a non-viable state shown as a red dot in the left part of each panel of
Fig. 6.6. Notice that the dot in the right panel is outside the kernel.
Capital productivity is high for low capital, which will grow rapidly. However,
increasing emission control (i.e., lowering z) and growing capital diminishes
the parenthesised term in (6.21) to below (δ + ρ) – depreciation augmented by
time preference. As a result, consumption crashes through its lower bound in
period 54 (see the red dash-dotted line in Fig. 6.8). Keeping z large is not viable
either – capital grows but so do emissions. This is a “sad” scenario of a poor
country that wants to grow and control emissions.
3. Medium-low capital, very low consumption, ambitious emission control, mod-

elled as (k, c, z) = [0.3654, 0.03885, 0.1154].
This point is viable. The successful evolution to a steady state, which is at the
kernel’s boundary, is shown as the black dotted lines in Figs. 6.6 and 6.7.
The emission control program can be maintained, albeit at a lower level (higher
z). However, maintaining this emission-control level is costly, and the (already
low levels of) capital and consumption will decrease.
4. Very low capital, very small consumption, very ambitious emission control,
modelled as (k, c, z) = [0.1, 0.02923, 0.0385].
This is clearly a non-viable state. It is shown as a red diamond in the left part
of each panel of Fig. 6.6, outside the kernel.
Even if capital productivity might be elevated, the emission control level is so
high that both capital and consumption crash through their respective lower
bounds (in period 4, not shown in the time profile figures) – compare case 2
above.
5. High capital, low consumption, ambitious emission control, modelled as
(k, c, z) = [0.8077, 0.0956, 0.1154].
This point is viable. The successful evolution to a steady state, which is in the
middle of the kernel, is shown as the black dash-dotted lines in Fig. 6.6. This
is the case of a well-capitalised economy which ends up poorer if an ambitious
abatement program is implemented. Emission control is relaxed (z increases)
because emission is very small, given the original small z. This scenario does
not stop capital from decreasing because z grows gradually and consumption
is still relatively high. However, consumption also decreases but the drop is
small because capital productivity matches δ + ρ fast. As a result, this economy
stabilises with low emissions and medium-small capital and consumption. The
time profiles corresponding to this evolution are shown in Fig. 6.7, also as the
black dash-dotted lines.
6. High capital, very high consumption, very lax emission control, modelled as
(k, c, z) = [0.9404, 0.2600, 0.8462]. The corresponding emission level is
0.2069.
This economic position is viable; i.e., a strategy for v exists (hence z), which can
keep the entire economic evolution inside the viability kernel V F (K ) ⊂ K , and
such that X ≤ 0.275. Figure 6.6 presents this evolution as the solid black (rather
short) lines in each panel. Figure 6.7 shows the corresponding time profiles also,
as the black solid lines. The evolution is indeed short; it takes 9 periods to reach
a steady state.
We can see that the steady state is here achieved by relaxing the emission control
even further, that is, up to the value of z = 0.9156. This enables the economy
to preserve consumption and only marginally decreases capital. Failure to do so
could lead the emission control to drain the capital (after some time), and the
economy would have K left before a steady state could have been achieved.
7. Very high capital, very high consumption, very lax emission control, modelled
as (k, c, z) = [1.1173, 0.2517, 0.8462]. The corresponding emission level is
0.2217.
Figure 6.6 shows this position as located to the right from the previous point
(case 6.) in each panel, and represented as a red star. Because of higher capital,
the initial emission is higher than before (X = 0.2217 vs. X = 0.2069) but is
still under the upper bound X̄ = 0.275. However, in each panel display this state
is shown as non-viable even if it has more capital, and consumption is higher,
than in the previous case 6.
This is because the emission path violates the emission constraint in t = 2 – see
the emission time profile (7) in Fig. 6.8. Nevertheless, the evolution from this
position stabilises in t = 10 which makes it viable in the unconstrained, viability
problem – see the grey boulder in Fig. 6.7.
8. Very high capital, high consumption, minimum emission control, modelled as
(k, c, z) = [1.1173, 0.2215, 0.9615]. The corresponding emission level is X =
(0.35) · (1.1173)0.4 · (0.9615)3 = 0.3252 > X̄ .
This economic position marked as the red “o” in Fig. 6.6 (right panel) is non-
viable because it fails the emission constraint at time 0 and, also, is outside the
(unconstrained) viability kernel. (Last but not least, we have found no strategy
for v that could stabilise the economy in K in finite time.)
Figure 6.6 right panel suggests that nearly any level of capital may be viable. However,
in the left panel of this figure we realise that only those capital values which are
“aligned” with consumption will be viable. We now examine what happens to an
economy rich in capital that is not aligned with consumption.
9. High capital, very small consumption, active emission control, modelled as

(k, c, z) = [0.8077, 0.0485, 0.1154]. This economic position is below the ker-
nel slice in the left panel of Fig. 6.6: capital dominates consumption.
Figures 6.6 and 6.8 depict the evolution from this state as red dashed lines. Con-
sumption slides to below the lower limit within 40 periods, so this economic
position is non-viable. The system’s dynamics are such that the relaxation of
the emission control (hence increasing z) is sufficient to maintain almost steady
capital, which eventually grows. However, the growth of z is insufficient to pre-
vent consumption from vanishing – see Eqs. (6.20), (6.21) (and consider the
explanation concerning case 2 above).
10. Low capital, medium consumption, weak emission control, modelled as
(k, c, z) = [0.2769, 0.1542, 0.8846]. This economic position is to the left of
Fig. 6.9 Transition evolutions from unconstrained to constrained emissions
the slice in Fig. 6.6 left panel – consumption dominates capital.
Figures 6.6 and 6.8 show the evolution from this state as the red-black dashed lines.
Capital diminishes to below its lower limit within 7 periods, so this economic position
is non-viable. The system’s dynamics (6.20), (6.21) are such that the relaxation of
the emission control facilitates the growth of consumption. Consumption increases
and violates its upper bound. Capital cannot sustain this growth and dissipates.
Transition of an Economy from Unconstrained to Constrained Emissions
Our results enable us to analyse strategies that can lead an economy with uncon-
strained emissions to an economy with constrained emissions. See the solid (blue)
lines in Fig. 6.9 transiting from the grey boulder to the purple boulder.
If (k, c, z) ∈ V F (K ), such that X ≥ X̄ , the government will seek strategies to
lower emissions and keep the economy in V F (K ). In other words, the government
will seek strategies to transit (k, c, z) to the purple boulder, if (k, c, z) is in the grey
boulder but remains outside the purple boulder. Strategies like that should exist in
economies whose capital and consumption are medium-high to high, and emission
control is practically non-existent. Lowering z in these economies will diminish
emission and can lead the economy to an emission-constrained steady state without
Fig. 6.10 Transition profiles
leaving V F (K ). Figure 6.9 illustrates such transitions, represented by the solid blue
lines, which go from the grey to the purple boulder. It is evident from the figure that
(again), for viability, capital-consumption combinations need be reasonably aligned.
The time profiles resulting from the emission-control adjustments v are shown in
Fig. 6.10. See the bottom-left panel for the profiles of v. It is evident from this panel
that the adjustments need to be performed “carefully”. Each initial economic position
requires its own emission-control profile. This is because z has to be lowered in a
manner that maintains the right-hand sides of Eqs. (6.20), (6.21) non-negative during
the adjustment process, when these right-hand sides approach zero. Expectantly,
output diminishes; however, capital and consumption increase. Although increasing
capital diminishes capital productivity, the productivity still dominates the term δ +
ρ in (6.21).
6.3.1.4 Conclusions
Knowledge of the viability kernel can tell the government, from which states the con-
straint set K cannot be attained, for a given size of instrument adjustments. This is an
important conclusion of a study based on viability theory: unlike traditional optimal
solutions that uniquely determine a possibly non-smooth strategy, knowledge of the

viability kernel informs the government about attainability and non-attainability of
goals, by smooth strategies.
Knowledge of the viability kernel can also tell the government up to which level
of shocks the economy is resilient.
Believable calibration of our model allows us to formulate policy advice. For
example, governments should be cautious in selecting initial levels of emission-
control adjustments v when they embark on an emission control program. Among
many observations, we note that countries which embark on an ambitious abatement
program may fail to keep their economies within the state constraints if their present
levels of capital and consumption are low.
6.3.2 Sustainable Inequality Policy32
Similar to Sect. 6.3.1, in this subsection we will use viability theory in the context
of a calibrated neoclassical economic model. This time, we will identify sustainable
economic states. whereby sustainability concerns inequality. This means that we will
identify economic states that – under smooth adjustments to a taxation policy – will
remain in the system’s efficiency region and that, additionally, an inequality measure
will be satisfied in each such state.
Given the formal similarity of the models used here and in Sect. 6.3.1, the structure
of this subsection follows the previous section’s.
6.3.2.1 Capital, Consumption, Debt, Taxation and Inequality – a

Stylised Model
As depicted in Fig. 1.2, multiple capital assets are necessary to sustain equity which,
in this section, is represented by constrained inequality. As in Sect. 6.3.1, in the current
model, economic capital is responsible for sustaining equity and an acceptable level
of consumption As in Sect. 6.3.1.1, we postulate here that consumption is aligned
with social cohesion, and can capture other things such as a health level. In the
model, we will also have debt B, and taxation τ . The latter will control the process
of economic growth.
So, in this subsection, sustainability means keeping inequality in a tolerable range
while assuring that capital, consumption, debt, and taxation remain within their limits
of acceptability. Again we suggest that the model here is another reduced version of
the wellbeing frontier introduced in Fig. 1.2.
Output is produced using a neoclassical production function. In per capita terms,
the production function is:
y = Ak α 1−α (6.27)
32 This subsection draws from Krawczyk and Townsend (2015a, b, c).

where α ∈ (0, 1), k is the per capita capital stock, c is consumption, is labour and
A is total factor productivity, which is constant in our model.33
The government taxes income to finance debt (B), as well as exogenous gov-
ernment spending (G).34 Capital depreciates at rate δ. A representative household
discounts35 future consumption at rate ρ. Labour supply and savings are both
endogenous, allowing each economic state to be characterised by the state vector
x(t) = (k(t), c(t), B(t), τ (t)) where t is time. To unburden the notation, the time
argument t will be dropped unless necessary.
When private actors make a market-facing decision, they treat inequality as given,
thereby – most likely – causing too much inequality by their selfish choices. In a
situation like this, government will impose inequality controls.
Inequality and the Relative Factor-Share36
Our measure of income inequality in this chapter is the share of income taken by the
top earners. We will look at the top 10%, the top 1%, and the top 0.1% of earners.
The study of income inequality is the study of a distribution, which no single
statistic can fully describe. As demonstrated by Atkinson (1970) this is not a trivial
concern. In his comparison of twelve countries, India is the third most equal by one
measure and the eleventh by another. Shifts of the Pareto index (which parameterised
a Pareto distribution) and the ever-present Gini coefficient in government statistics
can correspond to shifts anywhere along the distribution. Percentile ratios – say,
the cut-off of the top quintile divided by the cut-off of the bottom quintile – are
also popular with government statistical agencies. However, they ignore the tails of
the distribution and so ignore the extremes which correspond to the most visible
inequality. We believe that it is the ratio of the incomes of the very wealthy compared
to the rest which has inflamed popular rhetoric.37 This justifies our interest in the
shares of income taken by the top earners.
Consider the following economic assumptions:
1. all income is factor income;
2. workers own no capital and capitalists do not work;
3. the top group’s income is capital income;
4. capital income is not more pro-cyclical than labour income.
We contend that assumptions 1.–3. are satisfied in the top income group. If so, and if
assumption 4. is also satisfied, then the relative factor-share will carry information
about inequality.
33 See footnote 21.

34 Judd (1987) allows labour and capital income to be taxed at different rates. For simplicity, we
assume that each factor income is taxed at the same rate. For a viability model with two taxation
rates see Driscoll-O’Keefe and Krawczyk (2016).
35 See footnote 23.
36 In the following paragraph we rely on Krawczyk and Townsend (2015b) for some details.
37 Remember the Occupy Movement’s slogan? We are the 99%!
Table 6.1 The relationships between inequality and the relative factor-share
10% share 1% share 0.1% share
Effect of relative 6.534 *** 6.660 *** 1.624 ***
factor-share (2.341) (1.307) (0.302)
R2 0.136 0.389 0.578
Number of years 68 68 62
ADF test on residuals −3.406 *** −3.709 *** −3.481 ***
ADF lags 1 1 1
Robust standard errors in brackets. P-values follow ∗ ∗ ∗ < 0.01 < ∗∗ < 0.05 < ∗ < 0.1. The num-
ber of lags in the ADF test were selected with the Akaike information criterion; the ADF tests have
neither drift terms nor time trends
Fig. 6.11 10, 1 and 0.1% income shares, versus the relative factor-share
Statistical evidence of a strong relationship between the share of income taken

by the top earners and the relative factor-share has been found by Krawczyk and
Townsend (2015b), in the New Zealand context.38
With the shares of total income calculated from the World Top Incomes Database
(see Alvaredo et al. 2014) and the factor ratio from the income measure of national
accounts,39 Krawczyk and Townsend (2015b) have concluded that more than 50% of
variability of the 0.1% income share, and nearly 40% of the 1% share, are explained
by the relative factor-share.
pagination
Table 6.1 includes Krawczyk and Townsend (2015b)’s results of both the regres-
sion analysis and a co-integration test. Figure 6.11 shows plots of income shares
against the relative factor-share. All three slope estimates are positive, significant,
and able to explain a substantial proportion of the variation in inequality. This is
strong evidence that some relationship exists between the relative factor-share and
inequality, as expressed by the top income shares. Furthermore, moving up the income
distribution the estimates are more significant and able to explain an increasing pro-
38 We have not tested that relationship for other countries.

39 Krawczyk and Townsend (2015b) obtained data from 1972 with Statistics New Zealand’s Infos-
hare tool; older data were obtained from the latest revision available in the New Zealand Official
Yearbooks.
portion of variation. This is what one would expect: while inequality towards the
middle of the distribution may still be caused somewhat by wage inequality, at the
extreme end much income is from capital, and so increases in wealth will be reliant
on greater returns to capital.
Based on these results Krawczyk and Townsend (2015a) propose that the relative
factor-share is an adequate proxy for inequality. We will now find a closed-form
expression for the relative factor-share.
Thanks to Turnovsky and García-Peñalosa (2008) we know that the
aggregate behaviour of models with only wealth heterogeneity will be identical to the
aggregate behaviour of a corresponding representative agent model. Thus, models
of aggregate income received by labour and capital can be robust to heterogeneity in
the agents generating that income. We will then use a representative agent model to
derive the relative factor-share.
A Mathematical Model for the Relative Factor-Share
The following economic capital state equation represents the economy-wide resource
constraint:
k̇ = Ak α 1−α − δk − c − G (6.28)
where c is per capita consumption, is labour, A is total factor productivity, and α

is the Cobb-Douglas output elasticity. Furthermore, δ > 0 is depreciation and G is
government expenditure. We see that investment is positive when output y (6.27)
exceeds the sum of capital depreciation, consumption, and government spending.
We note an obvious trade-off between consumption and economic growth: the
more consumption, the slower the growth rate. In a situation like that, the govern-
ment needs to select taxes τ – present in Eq. (6.34) below – to maintain a compromise
between current consumption and the capital growth capable of maintaining accept-
able levels of future consumption.
If λ (λ > 0) is the marginal value of capital at time t (i.e., the shadow price in the
representative agent’s optimisation problem), then it follows from the maximisation
of the representative household’s utility function u:
c1−σ 1+η
u = u(c) − v( ) where u(c) ≡ and v( ) ≡ V , (6.29)
1−σ 1+η
that, on an infinite horizon with discount rate ρ, with σ > 0, η > 0, V > 0,
λ̇ = λ(ρ − r̄ ) . (6.30)

∂y
Here, r̄ = (1 − τ ) − δ is the after tax marginal product of capital, or interest
∂k
rate (or rent). Expanding r̄ in (6.30) yields:
1−α
λ̇ = λ ρ − (1 − τ ) α A −δ . (6.31)
k
We can see that the growth of shadow prices can be slowed down, or made negative,
with a low tax rate and high capital productivity. This can help the household to
increase consumption.
du 1
The marginal utility of consumption can be expressed as both: = σ and as
dc c
du
= λ.40 Hence,
dc
1
c = 1/σ , (6.32)
λ
which yields, after differentiation41
−1 1 −1 1+σ
ċ = · · λ̇ = c λ̇ . (6.33)
σ λ1+1/σ σ
Using (6.31) and (6.33), after some simplifications, we get:
ρ + δ − α A k α−1 1−α
(1 − τ )
ċ = − c · . (6.34)
σ
Consumption has a trivial steady state but can also grow to reach a positive steady
state if the discount rate or depreciation are small relative to capital productivity.
Low taxation can also help increase consumption.
Government debt will grow proportionately to the interest rate, while also reflect-
ing the government surplus G − T , where T is the tax collected:
Ḃ = r̄ B + G − T. (6.35)
If we allow capital depreciation to be tax-deductible, the expression for tax becomes:
∂ ∂y
T = τk y − δk + τ = τ Ak α 1−α
− τ δk (6.36)
∂k ∂
Combining the last two expressions results in the following debt dynamics:
Ḃ = r̄ B − τ Ak α 1−α
+ G + τ δk . (6.37)
That is, debt can diminish if output is large and/or if the tax rates are high (the tax
deduction for capital depreciation is small relative to the other terms in (6.37) if δ is
small).
To fully describe the evolutions of the states, the differential equations for capital,
consumption, and government debt (6.28), (6.34), (6.37) need to be completed by
a differential equation for the tax rate τ , coupled with a speed constraint (compare
40 The latter is the agent’s first order optimality condition. Actually, there are two such first-order
conditions. The other is η = λ(1 − τ )w.

41 The adjoint state λ(t) will not ‘jump’ because x(t) will not be on the boundary of K .
(6.23)):
τ̇ = u ∈ [u, ū], u ≤ 0, ū ≥ 0 (6.38)
This condition determines how smoothly the government can adjust the tax rates.
Labour in (6.28), (6.34), (6.37) can be expressed as a function of capital and
consumption (hence, is endogenous).
• Wages w equal the marginal product of labour:
dy A(1 − α)k α
w= = α
. (6.39)
d
If capital increases when labour remains constant, the wages will grow.
• In equilibrium, the marginal utility of consumption weighted by the net wages must
be equal to the marginal disutility from labour. If w increases, c will also increase if
the value of remains constant:
(1 − τ )w η
=V . (6.40)
cσ
• Using (6.39) to substitute wages in (6.40) and solving for labour yields:
α+η
1
(1 − τ )(1 − α)Ak α
= , (6.41)
V cσ
which shows that labour depends on capital, consumption, and taxation.

Overall, the economic dynamics at hand are represented by (6.28), (6.34), (6.37)
and (6.38), and will be jointly referred to as the map F.42 Compare F defined by
(6.1)–(6.2) in Sect. 6.2.2 and also by (6.20), (6.21), (6.23) in Sect. 6.3.1.
We recognise that F is non-linear and has multiple steady states. We can see
that such dynamics will be difficult to control, to keep the underlying economy in a
constrained set.
The relative factor-share is the ratio of net capital income to net labour income,
χ ≡ r̄ k/w̄l where r̄ and have already been derived; w̄ is wages multiplied by 1 − τ .
Expressing χ in terms of k and c, we have:
1−α α+η
1
r̄ k α V cγ
χ≡ = −δ k η(1−α) (A(1 − α))−(η+1) (6.42)
w̄ 1−α 1−τ
As we have shown in Fig. 6.11 and Table 6.1, this ratio correlates with income
inequality.
42 We could use (6.41) to substitute labour into (6.28), (6.34), (6.37). However, the resulting expres-
sions are long and complicated, and have no intuition.

When δ = 0, (6.42) simplifies to a constant, suggesting that in a one-tax model

without depreciation, χ would depend only on α, the Cobb-Douglas output elasticity.
When δ > 0, the relative factor-share will depend on both the direct effects of taxation
– see ‘1 − τ ’ in the denominator of the first term in the bracketed expression – and
also on the indirect effects of taxation on capital and consumption, captured by
(6.28) and (6.34). If, through an indirect effect, higher taxes decrease capital and
consumption, the long run effects of higher taxes on the relative factor-share can be
ambiguous.
Given the results on the correlation between inequality and the relative factor-share
as shown in Table 6.1, we will assume that constraining χ in (6.42) also constrains
inequality.
6.3.2.2 Viable Inequality Problem and Calibration
The government will want to know if there exist taxation strategies that conserve
the efficiency of the system – i.e., the system remains within the agreed constraints
on the state variables (we again direct the reader to footnote 18 in Sect. 6.3.1.1),
while ensuring that inequality remains sustainable – i.e., the taxation strategies do
not result in the violation of an inequality constraint imposed on χ .

K ≡ (k, c, B, τ ) : k ∈ [k, k̄], c ∈ [c, c̄], B ∈ [B, B], τ ∈ [τ , τ ] . (6.43)
The limits43 k, k̄, c, c̄, B, B, τ , τ are normative, determined e.g., by the golden
rule, positivity of consumption for subsistence, credit history, and election promises.
Furthermore, the time-dependent (k, c, B, τ ) are a result of the application of u ∈
[u, ū], where the decision about the size of u and ū rests on the government’s political
will concerning the speed of the taxation-rate adjustments.
With K defined in (6.43), the viability problem for dynamics F – here, (6.28),
(6.34), (6.37) and (6.38) – consists of establishing the viability kernel composed of
initial positions (k(0), c(0), B(0), τ (0)) from which evolutions emanate that remain
in K and – additionally:
0 < χ ≤ χ̄ . (6.44)
In other words, these evolutions – so that they do not cross the boundary of K – must
originate inside the viability kernel (i.e., from within the “safe space” in the sense of
Fig. 6.1).
Using a stylised steady state k s = s = 1 with no taxes and no government expen-
diture, we calibrate A and V and obtain A = 0.2093, V = 0.2989.44 We then assume
that government expenditure G is constant and set at 10% of no-tax steady-state out-
put; G = 0.1 · A = 0.0209.
43 Note the formal similarity between (6.43) (6.24), (6.8) and item 3 in Sect. 6.2.2.
44 We assume α = 0.43, ρ = 0.04, η = 1, σ = 0.5, δ = 0.05.
In our model, capital will be within 10 and 200% of the no-tax steady-state capital
stock, k ∈ [0.1, 2] and consumption should not deviate too far from a long-run
equilibrium (see Krawczyk and Judd 2016), c ∈ [0.015, 0.225]. Debt may grow to
350% of the maximum steady-state capital stock and also drop somewhat below zero,
so B ∈ [−1, 3.5]. The tax rate cannot be less than zero, and can at most be equal to
80%, τ ∈ [0. 0.8].
We will first require χ to be ≤0.4, and then require it to be ≤0.25. The relative
factor-share χ = 0.4 corresponds, per our OLS estimates in Table 6.1, to the top 1%
taking 5.7% of income and to the top 0.1% taking 1.1% of income. The value of
χ = 0.25 corresponds to the top 1% taking 4.7% and to the top 0.1% taking 0.9%.
(In New Zealand the top 1% currently take about 8% of national income. In the
mid-1980s they were taking about 5.5%.)
Thus the constraint set, for which we will find the viability kernel, is:
K × [0, χ ] , (6.45)
where χ is either 0.4 or 0.25 and the calibrated
K ≡ {[0.1, 2] × [0.015, 0.225] × [−1, 3.5] × [0, 0.8]} .
We finally need to specify the maximum tax-rate adjustment speed – the quantum
by which the government can increase or decrease the tax rate within a year. We set
the maximum speed as 20% points per year, u ∈ [−0.2, 0.2].
6.3.2.3 Sustainable Inequality45

Here, we analyse the viability of relative factor-share constraints.
We first ask how a relative factor-share constraint shrinks the viability kernel.
If this constraint shrinks the viability kernel, the economic states, which without
this constraint are viable, become non-viable. If so, we will conclude that there is
a sharp trade-off between low inequality (equity) and the state variables. This will
imply that an economy that adheres to the inequality constraint needs to avoid certain
combinations of the state variables, which are otherwise viable. If the the viability
kernels before and after the constraint imposition remain similar, then there is no
significant trade-off between inequality and the other variables. Second, we will ask
how government debt affects our conclusions.
These questions are a sample from the many questions viability theory could
answer – we could have just as easily asked whether requiring smoother tax adjust-
ments shrinks the kernel significantly, whether higher consumption than the current
c is viable, whether economies with high elasticities of substitution shrink more
45 This section draws from Krawczyk and Townsend (2015c).

quickly when χ constraints are imposed, and whether the ability to tax capital and
labour income at different rates expands the kernel.
The explanatory remarks on the kernel interpretation, provided in footnote 28, are
also relevant for the understanding of the following Figs. 6.12, 6.13 and 6.15.
However, the state dimensionality of our problem is now 4 (i.e., k, c, B, τ ). Adding
adjustments u, inequality χ , output, and labour requires a 7D array to store the kernel
coordinates. Therefore, the 3D “boulders” (spanned in 3D) displayed in the figures
below are convex hulls of slices through the 7D array.
Figure 6.12 shows 3D kernel slices for the kernels produced by three different sets
of constraints.46 All three slices require χ ≥ 0. The first has no further constraint.
The second requires χ ≤ 0.4. The third requires χ ≤ 0.25. The projections of these
slices onto the capital-consumption plane are shown in black.
The kernels and their projections do shrink as the inequality constraint is tightened.
The lowest levels of capital become non-viable when the relative factor-share is
required to be less than or equal to 0.4. There is a much larger reduction in the kernel
when the constraint is lowered to χ ≤ 0.25. At this point, only high levels of capital
are viable. This suggests that economies with little capital may find themselves unable
to constrain inequality, and that in general the trade-off between equality and capital
can be significant. We also note a somewhat less sharp trade-off between equality
and consumption.
Fig. 6.12 Viability kernels for different relative factor-share constraints
Fig. 6.13 Viability kernels for different relative factor-share constraints; starting debt B=3.5
46 The slices in Fig. 6.12 include points regardless of their initial debt level, provided that government
debt could be controlled to remain in [−1, 3.5], whereas those in Figs. 6.13 and 6.15 require debt
to start at some level.
From the black shapes in these figures (and also in those below) we infer that, in
this model – as well as in Sect. 6.3.1 on sustainable emissions – to ensure viability,
consumption has to be somewhat proportional to capital i.e., to belong to a “corridor”,
we referred to in item iii, in Sect. 6.3.1.3.
We now ask whether low inequality targets remain viable when a government has
significant debt. Figure 6.13 requires debt to start equal to 3.5.
The high-debt kernels are noticeably different to those in Fig. 6.12. In particular,
a high relative factor-share is impossible with high levels of government debt. The
left-most panel has no constraint on the relative factor-share, but nonetheless the
factor-share is always <0.4.
Why High Debt Assists Equity
To see why high debt assists equity we need to examine why a high relative factor-
share is incompatible with low taxes. We note that the relative factor-share is not
directly a function of debt. However, the relative factor-share is decreasing in tax
rates – see (6.42) – if high tax rates reduce capital income more than they reduce
labour income. Thus if high debt requires high tax rates, then the corresponding
relative factor-share will be low. To demonstrate this, consider the converse case
of low tax rates. Low tax rates will lead to a high χ , but are only viable with low
initial debt. Set τ = 0.2, k = 0.2 and c = 0.02. This corresponds to χ = 0.3549. We
impose only a bottom constraint χ ≥ 0, and compare the trajectories which begin
with debt = 0 to those with debt = 3. We use VIKAASA to see which taxation
strategy can keep the economy within the constraints on B, c, k and τ .
Figure 6.14 shows the time profiles of consumption, debt, tax, and the relative
factor-share. If debt is high (dashed lines) we must increase taxes, otherwise debt
will crash through the upper bound. But increasing taxes (at full speed, to keep debt
<3.5) pushes consumption below its minimum constraint, showing us that the initial
point is not viable. When we do not increase taxes (solid line) we go above the debt
constraint. When we start from zero debt (dotted lines) it can be stabilised without
increasing taxes, allowing us to simultaneously stabilise consumption. We can see
that high inequality is viable with low debt, but not with high debt.
Our viability kernels suggest that highly indebted economies will have neither
high inequality nor low tax rates. We provide some evidence for a negative relation-
ship between taxation rates and inequality observed in some selected countries, in
Appendix 6.1.
Low Debt Economies
The above analysis considered high-debt economies. In contrast, Fig. 6.15 includes
only viable points which start with debt equal to 0. The slice which required χ ≤ 0.25
had too few viable points to generate a 3D figure, and is thus excluded.
The low debt kernels, both unconstrained and with χ ≤ 0.4, have far fewer viable
points than their high debt counterparts in Fig. 6.13. This is surprising; intuitively,
lower debt would give governments more flexibility. The low-debt kernels are so
small because of the bottom constraint on debt, B ≥ −1. This constraint can be
justified – perhaps we are concerned about the macroeconomic impact of a savings
Fig. 6.14 Time profiles of high debt (solid lines without tax changes, dashed with tax changes)
and low debt (dots)
Fig. 6.15 Viability kernels with debt = 0
glut. In any case, the bottom constraint is needed by VIKAASA which requires a
compact constraint set. Given that constraint, a low debt economy with high taxation
rates and high capital would quickly accumulate excess savings.
6.3.2.4 Conclusions
Our viability kernels have provided us with two conclusions. First, we find a signif-
icant trade-off between equality and efficiency. Both the viability kernels and their
projections shrunk as we lowered the relative factor-share constraint. This has left
large areas of low capital and consumption non-viable, albeit deemed efficient. Sec-
ond, the trade-off between equality and efficiency does not become more stark in
high debt countries – in fact, the high tax rates those countries require will provide
them with low inequality anyway.
As in the previous model, believable calibration of our model allows us to for-
mulate policy advice. For example, governments should be cautious about the initial
levels of capital and consumption, when a new taxation strategy is contemplated. If
those levels are low, the economy may be outside the viability kernel. This means that
no admissible taxation policy can keep the economy inside the set K . In such a situ-
ation, restoring economic viability (e.g., through international aid) should be given
first priority, rather than embarking on an ambitious inequality-reduction program.
6.3.3 Viability with Inequality and Emissions Constraints
We will now show how viability theory can help solve a sustainable policy problem,
in which the government seeks to constrain inequality and emissions jointly. This
will be a step toward a comprehensive solution of the policy problem depicted in
Fig. 1.2. Instead of proposing policies concerning separate wellbeing items, one at a
time, as in Sects. 6.3.1 and 6.3.2, we will compute the viability kernel for a problem
which combines the constraints on both emissions (6.22) and inequality (6.44).
6.3.3.1 Limitations of Our Model and of Numerical Simulations
This subsection has a further, and more critical, purpose and message as well. A
resilience-focused public policy, that is necessarily based on limited information on
key ingredients such as individuals’ preferences, will want to expand the viability
kernel, surrounded by the wellbeing frontier, as much as possible. This would help
increase the resilience of the system to exogenous shocks. This requires comple-
mentary investments in all components of comprehensive wealth. Thus, by way of
example, concurrently investing in “clean technology” and in labour skills that are
able to use that technology, by using the tax revenue collected from production that
uses the “dirty technology”, would eventually increase the productive capacity of the
economy,47 reduce the level of emissions, and improve income inequality. While the
47 Building clean capital requires time. Investing in clean technology and in labour skills takes
resources away from dirty capital and thus dirty capital will diminish. In the transition, we may
experience a decline big enough to push the economy outside K .
model of this chapter is not rich enough to simulate such an example, the simulations
presented below suggest this outcome.
A further, closely related, point is also strongly supported by the numerical simula-
tions of this section. If we keep the structure of production, technology, preferences,
and so on, unchanged and keep adding more constraints to the viability problem,
the viability kernel will keep shrinking. In fact, in quite a few cases, these additional
constraints will convert viable systems to non-viable systems. This is intuitively very
obvious. By way of example, if one starts by only focusing on economic growth,
and then adds constraints on inequality and emissions to the viability problem, a
seemingly viable system could become non-viable if economic growth was being
generated by destroying the environment and creating huge inequalities. Thus, if one
wishes to maintain positive economic growth, while also ensuring a clean environ-
ment and a reasonable level of equality, one needs to deliberately change the structure
of production and consumption, through deliberate investments in “clean technol-
ogy” and associated labour skills. This could accommodate the additional constraints,
reflecting dimensions of life we care about, without shrinking the viability kernel.
6.3.3.2 Capital, Consumption, Debt, Taxation, Inequality and

Emissions – A Stylised Model
The inclusion of emission control z into the model that generates the relative factor-
share requires reformatting the system’s dynamics (6.28), (6.34), (6.37) and (6.38).
The main feature of the new dynamics is the production function:
y = Ak α 1−α
z (6.46)
which depends on emission control z ∈ [0, 1]. For comparison, see Eqs. (6.19) and
(6.27), in which the above symbols have been used. A consequence of this production
function is the emission output function:
X = Ak α 1−α β
z , (6.47)
where the symbols are as in (6.18).

Allowing for (6.46), the economy-wide resource constraint becomes:
k̇ = z Ak α 1−α
− δk − c − G . (6.48)
We see that investment is positive when output y – now dependent on emission

control z – exceeds the sum of capital depreciation, consumption and government
spending. Evidently, low z – i.e., high abatement – diminishes investment.48
48 This is a reflection of the limitations of the specific model we are using. E.g., low z could be
achieved for a larger A – total factor productivity. Then low z would not necessarily come at the
expense of aggregate productive investment.
Maximisation of the household utility function (6.29), with capital given by (6.48),
yields the necessary optimality condition for the marginal value of capital λ. The
condition looks identical with (6.30); however, here, the marginal product of capital
1−α
∂y
is = αz A , hence rent differs from that in (6.31) and equals:
∂k k
1−α
r̄ = (1 − τ ) αz A −δ . (6.49)
k
For small z, we recognise the possibility of low capital productivity, and even of
negative rent, if depreciation δ is not insignificant. The latter will be exacerbated by
large capital. With (6.49), the equation for λ̇ becomes:
1−α
λ̇ = λ ρ − (1 − τ ) αz A −δ . (6.50)
k
As in Sect. 6.3.2, the first order optimality condition for consumption is c = λ−1/σ .
Combining it with (6.50) yields:
1−α

ρ − α zA k
− δ (1 − τ )
ċ = − c · . (6.51)
σ
As before, consumption has a trivial steady state but can grow to reach a positive
steady state if the discount rate or depreciation are small relative to capital produc-
tivity. Low taxation and little abatement (i.e., large z) can also help increase con-
sumption. Perhaps counter-intuitively, large capital combined with high abatement
(low z) can cause significant consumption decay.49
Government debt changes according to (6.35). Allowing for z in the marginal
product of capital yields:
Ḃ = r̄ B − τ z Ak α 1−α
+ G + τ δk . (6.52)
That is, debt can diminish – or grow more slowly – if output is large and/or if the tax
rates are high (the tax deduction for capital depreciation is small relative to the other
terms in (6.52) if δ is small). Evidently, high z (i.e., low abatement) – because it is
associated with higher level of production and therefore income – speeds up debt
repayment and so B will grow more slowly, or diminish, when abatement is low.
However, z is also present in r̄ (see (6.49)). There, high z (low abatement) means
high rent, which accelerates debt. Therefore, the net outcome of high z on future
debt will depend on the current levels of capital and debt.
49 See the explanation provided in footnote 48.

To fully describe the evolutions of the states, the differential equations for capital,
consumption, and government debt (6.48), (6.51), (6.52) need to be complemented
by two differential equations for the tax rate τ and emission control z, each coupled
with the rate’s speed constraint:
τ̇ = u ∈ [u, ū], u, ū ≥ 0 (6.53)

ż = v ∈ [v, v̄], v, v̄ ≥ 0 (6.54)
where v, as in Sect. 6.3.1, is the speed of adjustment of emission control. The above
equations determine how smoothly the government can adjust the tax and emission-
control rates.
Labour (as in Sect. 6.3.2) is endogenous in (6.48), (6.51), (6.52). I.e., it can be
expressed as a function of capital and consumption. The wage rate (w) equals the
marginal product of labour:
dy z A(1 − α)k α
w= = α
(6.55)
d
and is negatively affected by emission control. In equilibrium, the marginal utility of
consumption, weighted by net wages, must be equal to the marginal disutility from
labour (compare (6.40))
(1 − τ )w
= V η. (6.56)
cσ
Substituting the wage rate and solving for labour yields:
α+η
1
(1 − τ )(1 − α)z Ak α
= , (6.57)
V cσ
which depends on capital, consumption, taxation, and emission-control rates.

Overall, the economic dynamics at hand is represented by (6.48), (6.51), (6.52)
and (6.53)–(6.54). As in Sects. 6.3.1 and 6.3.2, we will refer to these dynamics as
map F.
The relative factor-share χ (which we treat as a proxy for inequality) now
becomes:
1−α α+η
1
r̄ k α V cσ
χ≡ = −δ k η(1−α) (z A(1 − α))−(η+1) (6.58)
w̄ 1−α 1−τ
When δ > 0, the relative factor-share will depend on both the direct effects of taxation
and emission control, and also on the indirect effects of taxation and emission control
on capital and consumption, captured by (6.48) and (6.51). Interestingly, mainly
because −(η + 1) < 0 (so, z is in the “denominator”), the direct effect of hight
abatement – i.e., low z – results in decreasing inequality. However, if, as a result of the
indirect effect, higher taxes and emission control decrease capital and consumption,
then the long run effects on the relative factor share needs be confirmed numerically.
6.3.3.3 The Model Calibration
The government will want to know if there exist taxation and emission control strate-
gies that maintain efficiency,50 while ensuring that both inequality and emissions
remain at sustainable levels – i.e., the strategies do not cause violation of the con-
straints imposed on χ and X for all t ∈ Θ.

K ≡ (k, c, B, τ, z) : k ∈ [k, k̄], c ∈ [c, c̄], B ∈ [B, B], τ ∈ [τ , τ ], z ∈ [z, z̄] . (6.59)
The limits k, k̄, c, c̄, B, B, τ , τ are normative, determined e.g., by the golden
rule, the positivity of consumption for subsistence, credit history, and election
promises. The constraints on z, [z, z̄], are modal. Furthermore, the time-dependent
(k, c, B, τ, z) are a result of the application of u ∈ [u, ū] and v ∈ [v, v̄], where the
decision about the sizes of u, ū, v, v̄ rests on the government’s political will concern-
ing the speed of their policy adjustments (perhaps, promised at the recent election).
With K defined in (6.59), the viability problem for dynamics F – here, (6.48),
(6.51), (6.52) and (6.53)–(6.54) – consists of establishing the viability kernel
composed of initial positions (k(0), c(0), B(0), τ (0), z(0)), from which evolutions
emanate that remain in K (6.59) and, additionally:
0 < χ ≤ χ̄ and X ≤ X̄ . (6.60)
In other words, evolutions that do not cross any boundary of K must originate inside
the viability kernel (i.e., in the “safe space” in the sense of Fig. 6.1).
For the model calibration we will add the stylised steady state z s = 1 to the stylised
states defined in Sect. 6.3.2.2. Then, for the same parameters as in Sect. 6.3.2.2,
we obtain the same A = 0.2093, V = 0.2989, the corresponding steady state con-
sumption cs = 0.1593 and G = 0.0209. Then, we use the same arguments as in
Sect. 6.3.2.2 to determine the current set of constraints K , which we complete by the
modal constraint on z ∈ [0, 1].
Therefore the set K is now:
K ≡ {[0.1, 2] × [0.015, 0.225] × [−1, 3.5] × [0, 0.8], × [0, 1]} . (6.61)
We complete the viability constraint formulation by specifying:
0 ≤ χ ≤ 0.4 and X ≤ 0.105 . (6.62)
50 For
the meaning of efficiency, please refer back to the footnote 18 in Sect. 6.3.1.1. Also, note the
formal similarity between (6.59), (6.43) (6.24), (6.8) and the item 3 in Sect. 6.2.2.
Fig. 6.16 Viability kernel slices for χ ≤ 0.4 in the model without z and in the model with z = 1
The latter limit -0.105 - is approximately 50% of the unabated steady state X s when
ks = s = 1; i.e., X s = 0.2093 · 1α · 11−α · 1β = 0.2093. Therefore the constraint
set, for which we will find the viability kernel, is K in (6.61) completed by (6.62).
We will keep the tax-rate adjustment speed the same as in Sect. 6.3.2.2 (i.e., u ∈
[−0.2, 0.2]). Finally, we need to specify the maximum emission-control adjustment
speed – the quantum by which the government can increase or decrease the emission
control within a year. We set this maximum speed as 25 percentage points per year,
v ∈ [−0.25, 0.25].
6.3.3.4 The Viability Kernels When χ and X are Constrained Jointly
Are Current Results Compatible with Those When Only χ is Constrained

We will compare the two 3D slices shown in Fig. 6.16. Both are drawn in (k, c, χ ).
The left one is computed for the four state-variable model proposed in Sect. 6.3.2;
the right one is computed for the five state-variable model proposed in this section.
Both slices include points regardless of their initial debt, taxation rate and emissions
(the right one).
The left slice is a copy of the middle panel in Fig. 6.12. The right slice is the
kernel’s slice through z = 1, without the emission constraint.
The slices look by-and-large identical.51 This implies that the model of the current
section, with z, generalises the results from the model without z. Furthermore, the
models coincide for z = 1.
Slices When χ and X are Constrained Jointly
Figure 6.17 shows two 3D slices of the viability kernel computed for inequality and
emissions constrained jointly, 0 < χ ≤ 0.4 and X ≤ 0.105. The left slice is plotted
51 Thetiny differences come from that “slicing” through B - left panel - does not have to generate
the same points that “slicing” through B and z = 1 - right panel - does.
Fig. 6.17 The viability kernel slices for inequality and emissions
in the same coordinates k, c, χ , and scale, as are the slices in Fig. 6.16, drawn for
no constraint on emission. In the right slice, the vertical coordinate are emissions X
(rather than inequality χ , as in the left panel). We can see from the left slice that
adding emission control z, and a related constraint, to a model that – in Sect. 6.3.2
– was controlled by the tax-rate adjustment only, has produced a different viability
kernel.
Overall, the left slice is smaller than the slices in Fig. 6.16. This was expected
because more constraints are to be satisfied now than before, so there are fewer points
which satisfy the two constraints (6.60) than the one (6.44). Furthermore, both slices
indicate a shift of the kernel toward lower values of capital and consumption. While
this may be not so intuitive at first glance, it could have been expected because it
is conceivable that adding a constraint on emissions will reduce the level of output,
relative to the model in Sect. 6.3.2 where emissions are unconstrained. If so, adding
this constraint will reduce ceteris paribus both capital and consumption. A more
detailed explanation follows.
• First, the “boulders” in Fig. 6.17 contain states of low capital and consumption that
were non-viable in Fig. 6.16. This means that now there are taxation and abatement
policies which can generate the desired measures of inequality and emissions with
low capital and consumption. Figure 6.18 helps understand why this is possible.
Consider the state [k(0), c(0), B(0), τ (0), z(0)] = [0.575, 0.05, 0.5, 0.6,
0.6667], from which all but the emission-control time profiles begin. Let us look
at the green solid lines. The selection of z(0) = 0.6667 – see the fifth panel – not
only keeps emissions low – see the seventh panel – but also slows down (or decreases)
capital growth – see Eq. (6.48). This prevents capital from rising fast, which would
have meant a rapid fall of capital productivity. As a result, rent r̄ will not diminish
much – see Eq. (6.49). Diminishing taxation – see the fourth panel – will improve
rent, and the consumption decline will slow down. All these changes will decelerate
the economy, as measured by velocity – see the eighth panel in Fig. 6.18 – to below
the Stopping Tolerance,52 which confirms viability of the considered state.
52 See Appendix 6.2.

Fig. 6.18 Time profiles of two economies with medium capital and low consumption. The green
solid lines correspond to an economy regulated jointly by the tax and emission-control changes; the
purple dashed lines correspond to an economy regulated by the sole tax changes; the black dotted
lines correspond to lowering taxes
In the k-c plane in Fig. 6.19, this evolution is shown as the solid green line orig-
inating at A whose coordinates – on this plane – are [k(0), c(0)] = [0.575, 0.05].
The grey shape is the “shadow” from the boulder in Fig. 6.17. The purple shape is
the shadow from the boulder in Fig. 6.16. The rectangle represents the projection of
the constraint set K onto the k-c plane.
Now, let us look at the purple dashed lines in Fig. 6.18, which originate from
[k(0), c(0), B(0), τ (0), z(0)] = [0.575, 0.05, 0.5, 0.6, 1]. The lines are computed
by VIKAASA as best in the sense of trying to keep the economy in K with χ < 0.4
and velocity below a stabilisation threshold. We notice that satisfying all these require-
ments appears impossible and the initial point is deemed non-viable.53
This is because the system’s dynamics changes substantially if z is kept at 1 forever
– see the fifth panel. In particular, capital growth is not slowed down by emission
control. The possibility of increasing consumption, and decelerating capital growth, is
hampered by the diminishing capital productivity precipitated by the capital growth.
As a result, consumption crashes through the lower boundary in period 18. This
53 The purple slice was computed for z = 1 and without the constraint on emissions. So, X (0) > X̄
in the seventh panel of Fig. 6.18 is of no concern.
Fig. 6.19 The viability kernel slices for constrained inequality and emissions. The purple shape:
only inequality is constrained; the grey shape: both are constrained
renders the starting point non-viable. We also notice that the fast growing capital
crashes through the upper boundary in period 20.
In the k-c plane in Fig. 6.19, this evolution is shown as the dashed purple line
originating at A. The point A is outside the viability kernel for the non-abating
economy i.e., it is outside the purple shape. Expectantly, the purple dashed line,
which represents the evolution from A with z = 1, crashes through the consumption
lower boundary (compare the dashed purple line in the second panel in Fig. 6.18).
Interestingly, an experiment of lowering taxes to zero (in two periods, see the
black dotted lines) increases consumption but also raises inequality to above 0.4.
Furthermore, low taxation would release a lot of labour, and capital would grow even
faster than under the previous taxation policy, shown by the blue dashed line. As a
consequence, capital productivity would decrease and consumption would decline.
We conclude that abatement (i.e., z < 1) can have a stabilising effect on an econ-
omy with low consumption.
• Second, the other important difference between the slices in Fig. 6.17, on the one
hand, and those in Fig. 6.16 on the other hand, is that states with high capital and
consumption are non-viable now (i.e., when the government controls z).
We see in Eq. (6.47) that capital k cannot be large, especially if z is close to 1, if
emissions X are to be below X̄ = 0.105.
Consider the point B in Fig. 6.19 of coordinates [k(0), c(0), B(0), τ (0), z(0)]=
[1.2083, 0.12, 3.125, 0, 1]. Capital and consumption are quite high here. This point
Fig. 6.20 The viability kernel slice in k, c, z
is viable when emissions are not constrained but non-viable when they are. The
reason for the latter situation is that the emissions X (0) = 0.2482 > 0.105. This is
sufficient for non-viability of this point, if emissions are constrained. Intuitively, high
capital generates too much emissions when z = 1.
Ceteris paribus, any point with capital larger than at B will cause X > 0.105 and be
non-viable. A solution to produce less emissions and make such points viable should
consist of diminishing z. However, the grey shape in Fig. 6.19 is a slice obtained as
an amalgamation of the slices of the viability kernel for all z. So, according to VI-
KAASA, there are no strategies for z, which could make k ≥ 1.2083 viable. This is
non-intuitive and requires an explanation.
Look at Fig. 6.20 where z is on the vertical axis. The shadow on the k-c plane is the
same as the shadows in Fig. 6.17. We see no viable points for z < 0.5 in the figure.
Neither can we notice any firm relationship between capital k and emission control z
that would create more viable points for large capital through strong abatement i.e.,
low z. This is unexpected because large capital causes emissions to rise (see (6.47)),
so z should be rather small to enable abatement.
The four points in Fig. 6.20 – marked A, C, D are E – are k-c-z projections
of the starting points of the four evolutions (see Fig. 6.21), which we will use to
explain why combinations of medium to large capital k and intensive abatement
(low z) are non-viable. The point A is of coordinates: [k(0), c(0), B(0), τ (0), z(0)] =
Fig. 6.21 Time profiles of viable and non-viable evolutions for diminishing z
[0.575, 0.05, 0.5, 0.6, 0.6667] and thus corresponds to the evolution shown as the
green solid lines in Figs. 6.19 and 6.18. This is the same point A as in Fig. 6.19 where
its projection on the plane k-c is shown; here, this point is in the 3D space k-c-z.
The points C and D differ from A on the last coordinates which are, respectively,
z(0) = 0.5 and z(0) = 0.3333. They are marked, respectively, by the black and red
solid lines in Fig. 6.21. The point denoted by the cross is their projection onto k-c.
The point E corresponds to high capital and consumption and intensive abatement.
Its coordinates are [k(0), c(0), B(0), τ (0), z(0)]= [1.2083, 0.12, 0.5, 0.6, 0.3333];
its projection onto k-c is the four-arm star. The thin dotted lines are added to help
interpretation of the locations of one point relative to the other points. An evolution
from E is marked by the red dashed lines in Fig. 6.21.
Let us now look at Fig. 6.21 to examine what impact decreasing z has on viability,
if capital is medium to large.
As already stated, the solid green line, the same as in Fig. 6.18 and commented
there, is shown as a reference time-profile. This evolution originates at the viable
point A. The economy at this point is with medium capital, low consumption and
moderate abatement; debt is low and the taxation rate is rather high. What happens
to this economy when z is dropped to z(0) = 0.5 (in Fig. 6.20, see point C that is
below the green boulder) – visualise the black lines. Expectantly, emissions drop –
see the seventh panel – and all state variables remain in K . The only reason for this
point to become non-viable is a negative value of χ (0) = −1.033.
The same result – a negative value of χ (0) – is obtained when z(0) drops further,
to z(0) = 0.3333 (point D) and also when this level of emission control is applied to
high capital and consumption (point E). The evolutions from these points are marked
by the red solid and, respectively, dashed lines.
A negative value of the relative factor share χ – (6.58) – indicates that rent r̄ –
(6.49) – is negative. Notwithstanding the difficulty of interpreting negative inequality,
∂y
a negative relative factor share would require < δ – capital productivity should
∂k
be less than depreciation. Such a situation would not be sustainable: capitalists do not
invest if they are receiving a negative return. Thus requiring χ > 0 is an important
constraint on how the economy works. This constraint is therefore modal.
We will now look at r̄ in (6.49) to see when it can be negative and how a low z and a
high k contribute to this condition. To help with intuition, we will assume that c, B, τ
change slowly in comparison to z and k.54 This will allow us to rewrite (6.49) as:
Fig. 6.22 When rent is negative or positive
54 Indeed, c and B appear smoother in the time profile figures, than k. Emission control z can move
fast if u changes fast while taxation τ can be kept constant.
Fig. 6.23 When emissions are below or above X̄ = 0.105
z
r̄ (k, z) = (1 − τ ) 1−α
Φ −δ (6.63)
k
where Φ is a positive number that depends on c, B and τ .
In Fig. 6.22 we can see how r̄ (k, z) depends on k and z around [c, B, τ ] =
[0.12, 0.5, 0.6]. The “tiled” plane is at the value of zero so, any combinations of
k, z which give a point on the “sail” below the 0-plane corresponds to a negative
rent. This is precisely the image predicted by the expression (6.63): if z is small
and/or k is large, then rent can become negative. Economically, capital productivity
is smaller than the depreciation rate (see (6.49)). In particular, the rent is negative
when k = 0.575 or k = 1.2083 for many “small” values of z. This explains why the
viability slices in Figures 6.17, 6.19, and 6.20 do not contain values of high capital
or intensive abatement.
However, not all large capital values can be viable when z is large. This is because
when both k and z are large, emissions are high as per Eq. (6.47) and can exceed X .
In Fig. 6.23 we can see how X (k, z) depends on k and z around [c, τ ]=[0.12, 0.6]
(B does not impact emissions directly). The tiled plane is at the value of X = 0.105 so,
any combinations of k, z which give a point on the sail above this plane correspond to
exceeding emissions. This is precisely the image predicted by the expression (6.47):
if both z and k are large, then the economy produces too much emissions. This poses a
challenge to the government of how to select z: too large z endangers the satisfaction
of the emission constraint; too small z will diminish capital productivity below a
profitability threshold.
6.3.3.5 Conclusions
Our viability kernel analysis suggests that governments that aim at an economy
which is socially and environmentally sustainable should not always follow the old
economic “wisdom” that says that more is better. In particular, high capital may
be difficult to maintain because, on the one hand, it requires small z to satisfy the
emission constraint but, on the other hand, too low z may lower capital productivity
to below a profitability threshold.
As in the previous models, believable calibration of our model can help to formu-
late achievable goals and policy advice that leads to their fulfilment. As before, the
crucial information for the government will be whether the current economic state
is within the viability kernel.
While the model of this section is good enough to highlight some of the key trade-
offs involved in managing the viability kernel, it is not rich enough to identify the
types of complementarities that need to be exploited towards expending the viability
kernel. In fact some of our key policy messages could be very much strengthened
with a richer model that would accommodate, for example, both the production and
consumption of “dirty” and “clean” products - e.g., cars that use fossil fuels versus
electric cars. In this sense, the stylised model of Chap. 2 is richer than the simulation
models used in this chapter – simply reflecting current computational limitations.
Thus, in interpreting the results presented in this section, and drawing policy advice
from them, one should be aware of the limitations of the special model used in this
section, especially its very aggregated nature.
6.4 Resilience in a Viability Theory Model
Resilience is one of the main themes of this book. As we say in Sect. 2.8.1, “resilience
refers to the capability of the economy and society to respond to, and continue to
operate reasonably effectively and efficiently in the aftermath of, significant systemic
shocks to comprehensive wealth...”.
6.4.1 How to Accomplish Resilience?
We believe that strategies used by governments should not only satisfy the indi-
viduals’ requirements in “normal” times; they should also invest in resilience to
major systemic shocks (i.e., be able to assist individuals through “difficult” times).
6.4 Resilience in a Viability Theory Model 213
In particular, economic capital could have been severely damaged by an earthquake,

and consumption may have dropped drastically because of a financial crisis. For such
times, the government should have invested in infrastructures that are able to respond
effectively and efficiently. However, as the comparison between an economy without
emission control (see the kernel slices in Fig. 6.16), and the one with emission control
(see the kernel slices in Fig. 6.17) suggests, the levels of such investments should be
chosen carefully. Pushing too much capital might not be a solution because it might
move the economy outside the viability kernel.
Resilience can be realised in various ways.
1. In Sect. 2.8.2 state variables Γt E and Γt S are proposed to capture the building of
resilience-enhancing infrastructures (see Éqs. (2.86) and (2.87), and also footnote
34).
2. Similar to what we say about resilience in Sect. 2.8.1, De Lara (2017) says, in a
more mathematical context than ours, that resilience is a system feature that, after
a shock, enables the system to return to an acceptable “regime”, that is, the state
and control paths must return to a set of acceptable paths.
3. Resilience cannot be absolute – a sufficiently strong earthquake will destroy any
house. In the Preface we propose that if certain wellbeing indices remain in a
closed set for the life time (perhaps infinite) of the system, which the indices
describe, then the system is resilient. If shocks hit and move the system inside
the viability kernel, then there is always a strategy to regain the system’s stability.
Clearly, the larger the kernel, the larger the shock which the system can withstand.
4. In case a shock pushes the system outside the viability kernel (because the anti-
shock measures have been insufficient), a special strategy needs to be applied to
return the system to the viability kernel.
The four models dealt with in this chapter (scarce commodity consumption –
Sect. 6.2.3.2, emission control – Sect. 6.3.1, inequality – Sect. 6.3.2, and emission
control and inequality jointly – Sect. 6.3.3) are algebraically simple and not (easily)
amenable to the addition of resilience-enhancing infrastructures à la item 1 above.
Below we discuss how analysing viability kernels, obtained in the above sections,
can help governments assess and improve resilience.
6.4.2 Kernel Width
Knowledge of the viability kernel can tell the government up to which level of shocks
the system is resilient.
If shocks concern one state variable at a time then, rather than speaking about the
kernel “size” as in item 3. above, we can use the width of the kernel’s 2D slice as a
crude but rather robust measure of the system’s resilience to shocks.
Take the model of emission control discussed in Sect. 6.3.1. It can be seen in
the k, c-slice in Fig. 6.6 left panel55 that if the capital-consumption choices keep the
system in the middle of the slice, then capital shocks of up to 0.25 of the steady state
capital (when no abatement takes place, see Sect. 6.3.2.2) should not destabilise the
system. Respectively, tolerable consumption shocks can be assessed with the help of
the same slice.
We can say that keeping this system in the middle of the viability kernel makes
it resilient to shocks of up to the magnitudes read from the figure. This means that,
after a shock, the system will still be inside the viability kernel where a viable policy
exists to stabilise this system inside the kernel. Evidently, the system’s resilience to
shocks diminishes as it approaches the kernel’s boundary. In particular, the system
will lose resilience if it is at the kernel’s boundary.
6.4.3 Faster Instrument Adjustments
Consider the model of scarce commodity consumption in Sect. 6.2.3.2. Figure 6.3
shows the viability kernel V F (K ) obtained for that problem when the maximum
speed of the price adjustment (i.e., price stickiness) is |c| = 0.001. The same kernel
is now shown in Fig. 6.24, see the central green-grey shape surrounded by the dashed
brown line.
Suppose a shock brought the dynamic system to point B’, which is outside this
viability kernel. The price drop on the blue dashed line emanating from B’ is −0.001,
the maximum allowed in Sect. 6.2.3.2. The evolution represented by this line misses
V F (K ) and crashes through the minimum demand boundary of K .
However, the black dotted line, also emanating from B’, terminates at the steady
state line p = 0.05. The reason is that the price-adjustment speed on the black line is
twice as fast (i.e., c = −0.002) as the blue line. Furthermore, we notice that the black
line is contained in the light-green shape – the new viability kernel V F (K ) ⊃ V F (K ),
computed for the maximum speed of price adjustment |c | = 0.002.56
We can also see that the black-dotted evolution from B, see the bottom of the
figure, also remains in V F (K ), while the blue-dashed evolution crashes through the
maximum demand boundary. This is consistent with what happens around B’: the
faster evolution stays in the viability kernel V F (K ) but not the slow one. Finally, the
point B1 ’ lies outside V F (K ) and, given the adjustment-speed restrictions, no line
that emanates from this point remains in the constraint set.
As a result of the modification of the maximum instrument-adjustment speed, the
viability kernel has increased from V F (K ) to V F (K ). This has improved resilience of
55 Inthe k, z-slice (right panel) almost all points in [0.1, 1.25] × [0, 1] appear viable. It is the
elongated shape of the k, c-slice in the left panel which tells us about viable combinations of
k and c.
56 This maximum speed defines the new dynamics F , in which c is replaced by c , see
Eqs. (6.6), (6.9).

6.4 Resilience in a Viability Theory Model 215
Fig. 6.24 The augmented viability kernel V F (K )
the system because “larger” shocks, which could not have been contained in V F (K ),
will now be contained in V F (K ), built for faster adjustments.
6.4.4 Risk Aversion
As observed in Sect. 6.3.1 – consumption, expressed through (6.21) (also through

(6.34) and (6.51)), grows slower for large σ s than for small σ s. In utility modelling,
large σ s characterise risk averse individuals. We want to shed light on whether risk
aversion can have an impact on resilience.
Figure 6.25 shows two slices cut from two different viability kernels obtained for
the emission control problem, discussed in Sect. 6.3.1. The central (grey) slice was
obtained for σ = 2; the surrounding (cream-white) slice – for σ = 3. Both slices are
cut through the viability kernels obtained for unconstrained pollution. The central
slice (for σ = 2) is identical with the (light) grey slice shown in the left panel of
Fig. 6.6. We notice that the larger slice in Fig. 6.25 (obtained for σ = 3) extends to
the right-bottom corner and contains the smaller slice. Evidently, the larger slice
characterises a more resilient economy.
The extension of the slice with σ = 3 toward lower consumption (and capital) sug-
gests that socio-economic systems composed of risk averse agents are more resilient
to consumption (and capital) drops than economies composed of risk lovers. We
suggest the following explanation.
Fig. 6.25 Augmented viability kernel for risk averse individuals
In dynamic socio-economic systems, it is natural for higher levels of risk aversion

to translate into higher saving rates (to build shock absorbers) and lower consump-
tion rates. Thus, a highly risk averse individual would react much more slowly to,
for example, positive income shocks by not simply consuming the extra income
immediately – they would tend to save a greater portion of it. The system image of
this, assuming that higher saving rates are associated with higher investment rates
in all types of capital, including human and social capital as well as financial and
economic capital, would be more resilient economies to environmental, social, and
economic shocks.
In summary... knowing the boundaries of the viability kernel of a system enables
us to say up to which level of shocks the system is resilient. Enhancing a particular
system’s resilience could be realised by rebuilding the system so that its parameters
(e.g., A, δ, α) “improve”.57 However, this rebuilding cannot be realised by means of
changing z or τ , and is beyond the scope of the models considered in this chapter.
57 Obviously,
a larger value of A means technological improvement. Whether increasing the other
model parameters means improvement may be a value judgment.
6.5 Conclusions 217
6.5 Conclusions
The purpose of this chapter was to show, with the help of simple models, how
viability theory can be used to design public policy, in a world characterised by
radical uncertainty and complexity, when the objective of public policy is to support
sustainable wellbeing. While, for the moment, computational limitations prevent us
from working with a model that is as rich as the stylised model of Chap. 2, we have
nevertheless achieved our purpose even within the limitations of the simple models
of this chapter.
Public policy pursues its objective by trying to expand the opportunities and
capabilities of individuals and communities to live the kinds of lives they value. This
is what we have called the “wellbeing frontier”, which we have formalised as the
viability frontier – the solution of the viability problem is the viability kernel which
corresponds to the area ring-fenced with the wellbeing frontier of Fig. 1.2.
We do not have all the information required (say on individuals’ preferences) to
solve a dynamic optimisation problem. However, even within the constraints of the
simple models of this chapter, we can show how, by specifying broadly acceptable
levels of minimum per capita consumption (as a proxy for social cohesion) and
efficiency of capital (as a proxy for the potential growth rate of the economy), and
by placing reasonable constraints on the degree of income inequality and level of
emissions, we can represent the key policy dimensions of the wellbeing frontier.
We have shown how to use simple but effective levers to keep the system within
the viability kernel. This is a formal way of deriving and presenting, what we have
referred to as a “sustainable solution” to the wellbeing problem. In doing so, and
with the help of numerical simulations, we were also able to study the trade-offs and
complementarities involved among the various outcomes that have an influence on
wellbeing.
1. Explain the difference between optimisation and viability models.
2. Critically assess the different uses and advantages/disadvantages of these two
types of models.
3. What do we mean by “calibration” - explain how models are calibrated.
4. Explain and assess when viability models are best suited to use for policy-
informing advice.
5. Explain and comparatively assess the similarities/differences and advantages/
disadvantages of the policy-simulation model of Chap. 4 and the model of this
chapter.
6. Assess how viability models can be used to prioritise policy options.
7. Critically assess the use and potential role of viability models when the world we
are making policy in is characterised by “radical uncertainty” and complexity.
For rigorous introductions to viability theory see Aubin (1991), Quincampoix and
Veliov (1998), Veliov (1993) and Aubin et al. (2011).
Readers who are interested in learning more about viability theory and its appli-
cations to policy problems will find Krawczyk and Pharo (2013) a very useful intro-
duction to the subject.
Readers who are advanced mathematically will profit from studying Cardaliaguet
et al. (1999).
Appendix 6.1: Empirical Evidence
The Emission-Control Model

We have suggested in Sect. 6.3.1.3 that the current levels of capital and consumption
may determine a country’s willingness to actively pursue an emission-control pol-
icy. An example of a high-capital high-consumption country that invests heavily in
renewable energy sources (mainly wind farms) is Germany. Its neighbour, Poland,
is less industrialised, consumes less and constantly requests permission (from the
European Commission) to postpone the application of the tight European emission
rules.
The Inequality Model
One of the conclusions of Sect. 6.3.2.3 is that high values of the relative factor share
are viable with low debt, but not with high debt. By extension, this means that highly
indebted economies will have neither high inequality nor low tax rates.
This does indeed seem to be the case. Japan has the highest public debt in the
world, with public debt in 2010 equalling 206% of GDP (The World Bank 2015).
Our model predicts that Japan will have neither high inequality nor low tax rates: this
is correct, Japan’s 1% share in 2010 was 9.51% (Alvaredo et al. 2014) and its top
marginal tax rate was 40% (National Tax Agency 2010). In contrast the country in
the Alvaredo et al. (2014) database with the highest 1% share in 2010 was Columbia,
where the wealthiest 1% take 20.45% of national income. Columbia had debt equal
to only 38% of GDP (The World Bank 2015); the top tax rate was 33%.
Appendix 6.2: A Method for Finding Viability Kernels
VIKAASA58 is a suite of MATLAB® programmes that approximate viability kernels.

The algorithms applied by VIKAASA use truncated stabilisation and an approach
suggested in Gaitsgory and Quincampoix (2009).
VIKAASA can be used either as a set of MATLAB® functions, or via a GUI.59
The GUI can specify the viability problem, run the kernel approximation algorithms
58 See Krawczyk and Pharo (2011) and Krawczyk and Pharo (2014b); also Krawczyk et al. (2013).
59 VIKAASA is also compatible with GNU Octave, though its GUI is not.
6.5 Conclusions 219
Fig. 6.26 VIKAASA main window
and display the results. A detailed (though somewhat outdated) manual for VIKA-
ASA can be found in Krawczyk and Pharo (2011). The latest version of VIKAASA
is available for download at Krawczyk and Pharo (2014b). In Fig. 6.26, we show the
main window of VIKAASA.
In this paper, our algorithm solves a truncated optimal stabilisation problem for
each element of K h ⊂ K , a discretisation of K . For each x h ∈ K h , VIKAASA
assesses whether a dynamic evolution originating at x h can be controlled to a (nearly)
steady state without leaving the constraint set in finite time. Those points that can be
brought close enough to such a state are included in the kernel while those that are
not are excluded. This algorithm we use is called the inclusion algorithm, as opposed
to the exclusion algorithm, both explained in Krawczyk et al. (2013) and Krawczyk
and Pharo (2011). We note that this computational method will miss some viable
points if they are viable only because the evolutions starting at them are large cycles.
However, we did not encounter similar points in our experiments.
Chapter 7
Themes
Abstract This chapter summarises the key themes of this book and highlights the
challenges that lie ahead. Once we specify the generation of shared (across society
and generations) and sustainable wellbeing as the main outcome that public policy
aims to achieve, and we appreciate the multiplicity and complementarity of spheres
of wellbeing, it becomes self-evident that we need to think of environmental, social
and economic policies in an integrated way.
Keywords Intergenerational wellbeing · Complexity · Fundamental uncertainty ·

Resilience · Viability
This book presents and integrates eight big themes, all centred on a wellbeing-focused
public policy:
• The ultimate purpose of public policy is to improve the lives of individuals and
communities on a sustained basis, as they choose to live them – i.e., their wellbeing.
• Wellbeing is a function of multiple influences (collectively referred to as “com-
prehensive consumption”), which are in turn sourced from multiple capital stocks
(collectively referred to as “comprehensive wealth”).
• Public policy attempts to enhance wellbeing by increasing the opportunities and
capabilities (i.e., substantive freedoms) of individuals and communities to pursue
the lives they value; it does so by protecting, shaping, and ensuring widespread
access to “comprehensive wealth”.
• Adopting an integrated approach to environmental, social, and economic policy is
critical; implementation is through appropriately prioritised investments in infras-
tructures, including institutions (collectively referred to as “ecosystems”).
• The prioritisation of these investments is assisted by numerical policy-simulation
models.
• When radical (or substantive) uncertainty and complexity are allowed for, the
objective and domains of public policy remain unchanged, but the focus of imple-
mentation shifts to building resilience to systemic risks as well as nourishing
creativity. Again we rely on simulation models to prioritise investments, in con-
sultation with communities.

https://doi.org/10.1007/978-981-13-6104-3_7
222 7 Themes
• Whether a particular public investment is worth undertaking or not, as well the

prioritisation across investment options, is based on cost-benefit analyses, where
the “currency” of comparison is intergenerational wellbeing.
• Whether public policy in general is succeeding or not is evaluated on the basis of
three complementary assessment mechanisms:
i. objective measures of the main influences on, or dimensions of, wellbeing;
ii. survey-based measures of subjective wellbeing outcomes; and
iii. the evolution of the “wellbeing frontier” over time.
The question has been asked, “what have we learned as a result of this effort that
we did not know before?”
We derive three lessons from the material presented in this book. First, a public
policy that is focused on improving individual and collective wellbeing on a sustained
basis must use a framework that recognises the interdependencies between environ-
mental, social, and economic influences on wellbeing. This framework needs to be
then used as a platform for building models that identify and exploit the trade-offs and
complementarities involved as a basis for formulating policy advice. Second, when
radical uncertainty and complexity are key features of the world for which policy is
designed, and in which it is implemented and evaluated, we need to be prepared to
complement optimisation theory and associated models with others; we have sug-
gested viability theory as one such complement. Finally, in a world characterised by
complexity and radical uncertainty, both the design, as well as the implementation
and evaluation, of public policy need to involve communities in substantive ways.
Once we specify the generation of shared and sustainable wellbeing as the main
objective of public policy, and we appreciate the multiplicity and complementarity
of influences on wellbeing, it becomes self-evident that we need to think of environ-
mental, social, and economic policies in an integrated way.
In this book, we provided one framework, and a small sample of theories and
models, to facilitate such integrated thinking in designing policies. In doing so,
we demonstrated that while this endeavour is difficult, it is possible. The models
we have suggested highlighted the key complementarities and trade-offs between
policies. Our models also helped us emphasise and demonstrate that ignoring these
linkages will potentially lead to policies that harm intergenerational individual and
community wellbeing.
One of the main contributions of this book is the attempt to integrate different
frameworks and models, spread across a wide-ranging literature, towards enhancing
public policies aimed at enhancing intergenerational wellbeing. While the academic
literature tends to emphasise points of difference and distinction, given our policy
focus, we have instead tried to pull together common threads from this wide litera-
ture towards formulating an integrated, but still rigorous, approach to policy design,
implementation, and evaluation.
Policy making is particularly exciting in the pursuit of a shared vision for the
future of any country (in our case, New Zealand). It is in this context that the focus
shifts strongly to a search for mutually reinforcing actions that expand our collective
wellbeing-frontier. Trade-offs are replaced by complementarities as the main focus of
7 Themes 223
policy advice. And once that vision is understood and shared, the democratic process,
if it is functioning properly, endogenises and builds support for wellbeing-enhancing
and time-consistent public policies.
The models we have used also help us identify the distinctive role of a government
that is focused on enhancing the capabilities and opportunities (i.e., substantive free-
doms) of individuals and communities to pursue the kinds of lives they have reason
to value. As it happens, an open-minded reading of the wider literature on related
subjects clearly suggests that the policy recommendations of any integrated approach
(i.e., the what dimensions of policy), irrespective of the types of models used, are
likely to be very similar.
On the other hand, approaches to policy implementation (i.e., the how dimensions
of public policy) vary significantly, especially when radical (or fundamental) uncer-
tainty and complexity are allowed for. Then the primary focus of policy shifts to
enhancing resilience – characterised by adaptability of the environment, society, and
economy – to catastrophes that cannot be anticipated in advance. At the system level,
resilience rests on two foundations: protecting comprehensive wealth from catastro-
phes, while at the same time nourishing human inventiveness. There is thus a shift of
emphasis away from a policy setting that is focused on finding optimal solutions to
systems of equations, centred on well-defined objective functions, towards building
resilient environmental, social, and economic ecosystems (including institutions).
We offer viability theory as one possible substitute for intertemporal optimisation
models, to aid policy makers in devising policies that can be used to ensure that the
whole integrated system remains within sustainable bounds, while individuals and
communities pursue the lives they value in diverse ways.
The revised policy framework also embraces localism, experimentation at commu-
nity levels, small steps in implementation, reversibility and fast failures, and planning
on surprises. Collective wellbeing is pursued not by searching for investment levers
that will enhance well-defined social outcomes, but rather by making it easier for
people to pursue the varied lives they value, with no pre-defined desirable aggregate
social outcomes. “What works” is defined by communities, not by public servants,
although the latter can play a constructive and critical role in helping communities
achieve the outcomes they value.
Some readers of our work, especially academic colleagues, have suggested that
we have been “trying to swallow a whole elephant”, instead of “cutting it into chunks
and then digesting it in smaller pieces”. This would be a typical academic reaction, by
those (appropriately and justifiably) focused on formulating and solving well-defined
problems, using tractable models. This is a luxury policy advisers cannot afford. The
policy world is a “messy” but nevertheless exciting, worthwhile, and promising one.
Inevitably, the contents of this book are suggestive and exploratory in nature. We
encourage others to pursue this line of enquiry, bringing to the problems at hand any
set of tools that may help progress this exploration.
Glossary
A
A – Technological change; exogenously given labour-augmenting technology
Aic , Aid – Productivity (or “quality”) of machines using c – “clean” and d “dirty”
technologies, respectively
j
Āt – The world technology frontier
α – Cobb-Douglas output elasticity
B
B – All government activities (subsidies and penalties) directed at influencing the
behaviour of private people, institutions and communities (a control variable); debt
level
B – Debt
BtR F A – Bank holdings of foreign risk-free assets
B l (= Bcl + Bdl ) – Bank loans to domestic entrepreneurs (for investing in “clean”
and “dirty” technology)
B d – Bank deposits held by domestic households
B, B – Lower and upper debt limits
l
Bss – Steady-state level of bank loans
d
Bss – Steady-state level of bank deposits
BssR F A – Steady-state level of bank holdings of foreign risk-free assets
C
c – Maximum speed of price changes, price stickiness
C(·) – A vector of consumption flows that includes not only marketed consumption
goods, but also various others including leisure, arts, health services, and consump-
tion services supplied by nature – i.e., it represents comprehensive consumption
y
Ct – Aggregate domestic consumer demand for the final product
https://doi.org/10.1007/978-981-13-6104-3
226 Glossary
C y – The consumption of the single (or composite) marketed product Y

C x – Consumption of leisure
C h – Consumption of good health
y x
Css , Css , Css
h
– Steady-state levels of the consumption of products, time, and good
health, respectively
c B – Cost of all government activities (subsidies and penalties) directed at influencing
the behaviour of private people, institutions and communities
cΩ W (ΩtW ) – Cost functions for the delivery of positive externalities, or reduction of
negative externalities, to individuals
y
cΩ y (Ωt ) – Cost functions for the delivery of positive externalities, or reduction of
negative externalities, to businesses
c(Rt ) – The per unit extraction cost for the exhaustible resource, expressed in units
of the final output (Yt )
δ
δ – Depreciation rate; rate of capital depreciation
δ E – Rate of environmental regeneration
δ Γ E – Rates of regeneration or enhancement in environment.
δ Γ S – Rates of regeneration or enhancement in social resilience
δ K s – The rate of enhancement of social capital
δ P O V – Rate of change of poverty
E
E D – Time allocated to education
Sc
E D y,t – Income (part of output) invested in education
E – Environmental quality
E ss – Steady-state quality of the environment
E, S, Γ E , Γ S – Indices ∈ (0, 1), representing the degree of environmental quality,
social cohesion, and the quality and effectiveness of measures that enhance environ-
mental and social resilience to potential major shocks to each, respectively
η j ∈ (0, 1) – The probability function η j : R+ → [0, 1], which represents the
likelihood of successful innovation
F
F(·) – System’s dynamics.
Ft – Immigration – is measuring the number (stock) of migrants at the beginning of
period t
F i , i ∈ {A, L , L s , Sc, M c } – Stocks of foreign unskilled and skilled labour, scien-
tists, and “clean” machines respectively
M
Fss j – Steady-state levels of stocks of foreign machines of type (j) imported
FssL , FssL s , FssSc – Steady-state stocks of foreign skilled and unskilled labour and sci-
entists
Glossary 227
G
G – Level of government spending
Γt E – Degree of resilience – environment
Γt S – Degree of resilience – social infrastructure
γ V,P O V – The positive effect of an increase in overall wealth on poverty reduction
y
Γss , ΓssS , ΓssE – Steady-state degrees of resilience to potential systemic risks to eco-
nomic and social infrastructure, and the environment, respectively
H
H̃t – Time invested in health
j
H̃y,t for j ∈ {L , L s , Sc} – Income (part of output) invested in health services
H ∈ (0, 1) – “Health index” (representing the stock of health as part of our human
capital)
χ – Relative factor-share; the ratio of net capital income to net labour income; the
potential distortionary effects of government activities
I
i l – The nominal interest rate at which banks lend to businesses (entrepreneurs)
i d – The nominal interest rate that banks pay to depositors
ı̄ – The international wholesale nominal interest rate
l l d d
i ss , rss , i ss , rss – Steady-state nominal as well as “real” loan and deposit interest rates
I j,t – Generic investment in asset, or capital stock (j)
I SC,t – Investment in social connections (SC)
K
k – Per capita capital stock
K – Capital stock
K ⊂ Rn – Constraint set - a closed set representing some modal and normative
constraints imposed, or to be imposed, on the state variables (x)
K(s) – The stocks of a set of capital assets at time s (i.e., comprehensive wealth)
K b – Bank capital (or net worth) held by domestic households or overseas investors
K s – Social capital
K p – Physical (human-made) capital (or machines) used in production
K h – Human capital
K j,t – Generic capital stock (of j)
K SC,t – Stock (a capital asset) of social connections (SC)
s
K ss – Steady-state level of social capital
p
K ss – Steady-state level of physical capital
b
K ss – Steady-state level of bank capital
228 Glossary
L
– Labour
L – Unskilled labour
L s – Skilled labour
L dj,ss – Steady-state demand for use with the two technologies j ∈ {c, d}
L ds,ss – Steady-state skilled-labour demand
L dss – Steady-state unskilled-labour demand
L ss,ss – Steady-state supply of skilled labour
L H̃ – The (health-adjusted) labour required to operate the machines using the “dirty
technology”
L s H̃ – The (health-adjusted) labour required to operate the “clean technology”
λ – Marginal value of capital; marginal utility from consumption
λ K i (t) – The (spot) shadow price of the ith capital asset at time t
λT (t) ≡ ∂ W/∂t – Shadow price of time
M
M > 0 – Maximum amount of generic commodity
Mt – The total quantity of machines used in production in period t
m id , m ic – The continuum (indexed by i) of machines (physical capital) using d – the
“dirty” and c the “clean” technologies, respectively
j
m i,ss – Steady-state level of demand for machine using technology type (j)
j
m ss – Steady-state machine productions using technologies j ∈ {c, d}
(1 + μ) – (where μ > 0) is the factor by which innovation increases the quality (or
productivity) of a domestically-used machine when the scientists are successful in
innovating
monc – Monitoring costs incurred by banks
N
N X – Time allocated to non-leisure activities
P
p(·) – Price
y
Pt – Price of the final output (Y)
y
Pss – Steady-state level of the price of the final output (Y)
y
Πt – The profit level from the production of the final output in period t
ψ – The unit cost, inclusive of transaction costs (measured in units of the final output)
m
pt j – The price of the machine type j – “clean” or “dirty”
j
pi,ss – Steady-state level of prices for machine type (j)
P O Vt – Poverty – measured by the percentage of people whose ownership of com-
prehensive wealth is below a certain threshold
P O Vss – Steady-state level of degree of poverty
Glossary 229
Q
Q t – Inequality – measured by the variance of wealth ownership across different
clusters of individuals with different skills
Q ss – Steady-state degree of inequity
R
R̃ – Use of the non-renewable (exhaustible) resource (R)
R̃ss – Steady-state level use of the non-renewable natural resource
r̄ – Rent
r K p – The (“real”) return to physical capital – K p
r K b – The (“real”) return to bank equity (capital) – K b
r d – The real interest rate paid by banks on bank deposits – (B d )
rtl – The interest rate producers need to pay on the loans they obtain from banks to
pre-pay for labour services and machinery at the beginning of each period, before
they can produce and sell their produce at the end of the period
r̄ , r d , r l – “Real” (inflation-adjusted) interest rates – generic, deposit, loans
ρ – Discount rate
S
s, t (0 ≤ t ≤ s) – Time indices
S – Social cohesion
Sat·,· – Saving out of market income, with the first superscript referring to the type
of household making the saving and the second to the form in which saving is taking
place
Sct – The total number of scientists available at the beginning of period t
Sc j – The group of scientists (or engineers) working with technology j – those who
target working with the “clean” (Scct ) or “dirty” (Scdt ) technologies for the period
ahead
Scd,ss , Scc,ss – Steady-state allocations of scientists (or research) to “dirty” and
“clean” technologies respectively
s
Scss – Steady-state supply of scientists
Sss – Steady-state degree of social cohesion
σ – Coefficient of relative risk aversion
T
T > 0 – Problem horizon (can be finite or infinite); level of tax collected
T R – Time allocated to training
Ls
T R y,t – Income (part of output) invested in training
τ – The uniform tax rate on gross (before tax) income
Θ – A continuum of time values, Θ ≡ [0, T ]
230 Glossary
θ y ∈ (0, +∞) – The elasticity of substitution (in production) between the “dirty”
and “clean” technologies
U
U, U (x), W, W (x) – Control set; controls available at x
U (·) – A mapping from Rm C to R1+ where m C is the number of consumption goods.
Hence U (C(s)), t ≤ s ≤ ∞ represents the collective utility flow at time s
u, v – Rate of instrument changes; tax adjustment rate, emisson-control adjustment
rate; control
[u, ū], [v, v̄] – Intervals for rates of instrument changes
u(· · · ) – Utility function (household’s)
u(·) – Utility from consumption
v(·) – Disutility from labour
Υtc – Transaction costs (in real output terms) per unit of consumption expenditure
y
Υt – An average measure of monetary transactions costs associated with the pur-
chase of labour services and machinery in the production of the final output
V
Vt – Society’s comprehensive wealth
v j ≡ V j /V, j ∈ {L , L s , Sc} – The share of total private wealth owned by each of
the three clusters of individuals (j)
V F (K ) – Viability kernel for system’s dynamics F and constraint set K
W
W A (t) – Intergenerational wellbeing at t
W1 , W2 > 0 – Penalty coefficients on the square functions [. . . ]2
ws ≥ w – The market wage of skilled labour
w – The market wage of unskilled labour; wage rate
w Sc ≥ w – The market wage (or salary) for scientists
W SC,t – Level of wellbeing attributed to social connections
wss – Steady-state level of unskilled wages
wss
s
– Steady-state level of skilled wages
wss
Sc
– Steady-state level of scientist’s wages or salaries
ω ∈ [0, 1] – The weight we put on the welfare of future generations versus current
generations
X
x(t) – Generic state variable (can be a vector)
ẋ(t) – The system’s velocity
X – Time allocated to leisure activities; state space; level of emission
y
X t – The export of the final product
Glossary 231
y
X ss – Steady-state level of the export of the final-good
ξ E,Y – Environmental degradation resulting from production of output Y
ξ E,P O V – Measures the impact of poverty on environmental quality
ξ S,i > 0, i ∈ F, Q, P O V – Measure the rates of degradation of social cohesion
ξ i ∈ (0, 1) – The rate of depreciation of i, for i ∈ {L , L s , Sc, M c }
Y
y – Output produced using a neoclassical production function
Y – Units of final output
Y d – The quantity of output obtained through “dirty” technology
Ytd /Yt – Measure of the extent to which overall production uses “dirty” technology
(imbedded in “dirty” machines)
Yss – Steady-state level of final-good output
Z
z – Emission control; abatement level indicator
Z̃ – Use of the renewable resource (Z ) in production
Z t χ – The wedge between output and consumption arising from the potential dis-
tortionary effects of government activities
References
Abdallah, S., Thompson, S., Michaelson, J., Marks, N., & Steuer, N. (2009). The happy planet index
2.0: why good lives don’t have to cost the earth. New Economics Foundation.
Acemoglu, D., Aghion, P., Bursztyn, L., & Hemous, D. (2012). The environment and directed
technical change. The American Economic Review, 102, 131–166.
Acemoglu, D., Aghion, P., & Zilibotti, F. (2006). Distance to frontier, selection, and economic
growth. Journal of the European Economic Association, 4, 37–74.
Acemoglu, D., Akcigit, U., Hanley, D., & Kerr, W. (2016). Transition to clean technology. Journal
of Political Economy, 124, 52–104.
Acemoglu, D., Johnson, S., & Robinson, J. A. (2005). Institutions as a fundamental cause of long-run
growth. Handbook of Economic Growth, 1, 385–472.
Acemoglu, D., Naidu, S., Restrepo, P., & Robinson, J. A. (2014). Democracy does cause growth.
Technical Report. National Bureau of Economic Research.
Acemoglu, D., & Robinson, J. A. (2000). Why did the west extend the franchise? Democracy,
inequality, and growth in historical perspective. The Quarterly Journal of Economics, 115, 1167–
1199.
Acemoglu, D., & Robinson, J. A. (2013). Why nations fail: The origins of power, prosperity, and
poverty. New York: Crown Business.
Acemoglu, D., & Robinson, J. A. (2015). The rise and decline of general laws of capitalism. The
Journal of Economic Perspectives, 29, 3–28.
Adler, M. D., & Fleurbaey, M. (2016). The Oxford handbook of well-being and public policy.
Oxford: Oxford University Press.
Agénor, P. R., & Neanidis, K. C. (2015). Innovation, public capital, and growth. Journal of Macroe-
conomics, 44, 252–275.
Aghion, P., Akcigit, U., & Howitt, P. (2018). What do we learn from Schumpeterian growth theory?
Handbook of economic growth. Amsterdam: Elsevier.
Aghion, P., Howitt, P., & Mayer-Foulkes, D. (2005). The effect of financial development on con-
vergence: Theory and evidence. The Quarterly Journal of Economics, 120, 173–222.
Aidt, T. S. (2009). Corruption, institutions, and economic development. Oxford Review of Economic
Policy, 25, 271–291.
Aidt, T. S. (2011). Corruption and sustainable development. International Handbook on the Eco-
nomics of Corruption, 2, 3–51.
Alesina, A., & Giuliano, P. (2015). Culture and institutions. Journal of Economic Literature, 53,
898–944.

https://doi.org/10.1007/978-981-13-6104-3
234 References
Algan, Y., & Cahuc, P. (2013). Trust, growth and well-being: New evidence and policy implications.
Handbook of economic growth. Elsevier.
Alkire, S. (2005). Valuing freedoms: Sen’s capability approach and poverty reduction. Oxford:
Oxford University Press on Demand.
Alkire, S. (2016). The capability approach and well-being measurement for public policy. The
Oxford handbook of well-being and public policy.
Alvaredo, F., Atkinson, A.B., Piketty, T., & Saez, E. (2014). The world top incomes database.
Retrieved December 8, 2014 from http://topincomes.g-mond.parisschoolofeconomics.eu/.
Angelopoulos, K., Benecchi, A., & Malley, J. (2017). Can subsidising job-related training reduce
inequality? CESifo Working Paper No. 6605.
Antonelli, C. (2016). Endogenous innovation: The creative response. Economics of Innovation and
New Technology, 6, 1–30.
Arrow, K. J., Dasgupta, P., Goulder, L. H., Mumford, K. J., & Oleson, K. (2012). Sustainability and
the measurement of wealth. Environment and Development Economics, 17, 317–353.
Arrow, K. J., Dasgupta, P., Goulder, L. H., Mumford, K. J., & Oleson, K. (2013). Sustainability and
the measurement of wealth: Further reflections. Environment and Development Economics, 18,
504–516.
Arthur, W. B. (2014). Complexity and the economy. Oxford: Oxford University Press.
Arvanitidis, P. (2004). The diverse traditions of institutional economics: Conceptual and method-
ological underpinnings. University of Thessaly, Greece, Discussion Paper Series, 10, 317–50.
Asheim, G. B., & Weitzman, M. L. (2001). Does NNP growth indicate welfare improvement?
Economics Letters, 73, 233–239.
Atkinson, A. B. (1970). On the measurement of inequality. Journal of Economic Theory, 2, 243–263.
Atkinson, A. B. (2015). Inequality. Cambridge: Harvard University Press.
Au, J., Coleman, A., Sullivan, T., et al. (2015). A practical approach to wellbeing based policy
development: What do new Zealanders want from their retirement income policies. Working
Paper 15/14. New Zealand Treasury Working Papers.
Aubin, J. P. (1991). Viability theory. Systems & control: Foundations & applications. Boston:
Birkhäuser. https://doi.org/10.1007/978-0-8176-4910-4.
Aubin, J. P. (1997). Dynamic economic theory: A viability approach. Berlin: Springer.
Aubin, J. P., Bayen, A. M., & Saint-Pierre, P. (2011). Viability theory: New directions (2nd ed.).
Berlin: Springer. https://doi.org/10.1007/978-3-642-16684-6.
Baujard, A., & Gilardone, M. (2017). Sen is not a capability theorist. Journal of Economic Method-
ology, 24, 1–19.
Beinhocker, E. (2017). How profound changes in economics make left versus right debates irrele-
vant. Evonomics.
van den Belt, M. (2018). Nyc’s billion dollar tip for New Zealand. The New Zealand Herald.
Benczur, P., Karagiannis, S., & Kvedaras, V. (2017). Finance and economic growth: Financing
structure and non-linear impact. Technical Report. Joint Research Centre, European Commission
(Ispra site).
Béné, C., Doyen, L., & Gabay, D. (2001). A viability analysis for a bio-economic model. Ecological
Economics, 36, 385–396.
Benjamin, D. J., Heffetz, O., Kimball, M. S., & Szembrot, N. (2014). Beyond happiness and satis-
faction: Toward well-being indices based on stated preference. The American Economic Review,
104, 2698–2735.
Bennett, D. L., Nikolaev, B., & Aidt, T. S. (2016). Institutions and well-being. European Journal
of Political Economy, 45(supplement), 1–10.
Berggren, N., & Nilsson, T. (2016). Tolerance in the United States: Does economic freedom trans-
form racial, religious, political and sexual attitudes? European Journal of Political Economy, 45,
50–70.
van den Bergh, J. (2009). The GDP paradox. Journal of Economic Psychology, 30, 117–135.
van den Bergh, J., et al. (2015). Green agrowth as a third option: Removing the GDP-growth
constraint on human progress. Technical Report. WWWforEurope.
References 235
Bergman, L. (2003). An introduction to computable general equilibrium modeling. Stockholm

School of Economics.
Berthe, A., & Elie, L. (2015). Mechanisms explaining the impact of economic inequality on envi-
ronmental deterioration. Ecological Economics, 116, 191–200.
Bertram, G., & Terry, S. (2013). Green border control: Issues at the environment/economy border.
Bitzer, J., Gören, E., & Hiller, S. (2015). International knowledge spillovers: The benefits from
employing immigrants.
Blanchard, O. (2015). Blanchard: Looking forward, looking back. IMF survey magazine.
Boarini, R., Comola, M., Smith, C., Manchin, R., & De Keulenaer, F. (2012). What makes for a
better life? The determinants subjective well-being in OECD countries-evidence from the Gallup
world poll. OECD Statistics Working Papers 2012, 0_1.
Boarini, R., & d’Ercole, M. M. (2013). Going beyond GDP: An OECD perspective. Fiscal Studies,
34, 289–314.
Boarini, R., Kolev, A., & McGregor, A. (2014). Measuring well-being and progress in countries
at different stages of development: Towards a more universal conceptual framework. OECD
Development Centre Working Papers, 1.
Bonneuil, N., & Boucekkine, R. (2014). Viable Ramsey economies. Canadian Journal of Eco-
nomics/Revue canadienne d’économique, 47, 422–441.
Bonneuil, N., & Saint-Pierre, P. (2008). Beyond optimality: Managing children, assets, and con-
sumption over the life cycle. Journal of Mathematical Economics, 44, 227–241.
Bosetti, V., Cattaneo, C., & Verdolini, E. (2015). Migration of skilled workers and innovation: A
European perspective. Journal of International Economics, 96, 311–322.
Boston, J. (2016). Governing for the future: Designing democratic institutions for a better tomorrow.
Emerald Group Publishing.
Boston, J., & Chapple, S. (2014). Child poverty in New Zealand. New Zealand: Bridget Williams
Books.
Bourdieu, P. (1986). The forms of capital. In J. Richardson (Ed.) Handbook of theory and research
for the sociology of education. New York: Greenwood.
Braunerhjelm, P., & Henrekson, M. (2016). An innovation policy framework: Bridging the gap
between industrial dynamics and growth. In Essays in public sector entrepreneurship (pp. 95–
130). Cham: Springer.
Breunig, R. V., & Majeed, O. (2016). Inequality or poverty: Which is bad for growth?
Broadberry, S., & Wallis, J. J. (2017). Growing, shrinking, and long run economic performance: His-
torical perspectives on economic development. Technical Report. National Bureau of Economic
Research.
Brown, D., Woolf, J., & Smith, C. (2012). An empirical investigation into the determinants of life
satisfaction in New Zealand. New Zealand Economic Papers, 46, 239–251.
Bruni, L., Comim, F., & Pugno, M. (2008). Capabilities and happiness. Oxford: Oxford University
Press.
Bruton, H. J. (1998). A reconsideration of import substitution. Journal of Economic Literature, 36,
903–936.
Buchanan, C., & Hartley, P. R. (2000). Equity as a social goal. New Zealand Business Roundtable.
Buera, F. J., Kaboski, J. P., & Rogerson, R. (2015). Skill biased structural change. Technical Report.
National Bureau of Economic Research.
Cadogan, B. (2013). Welfare policy: Governance history and political philosophy. New Zealand
Treasury: background paper for the 2013 statement on the long-term fiscal position.
Campiglio, E. (2016). Beyond carbon pricing: The role of banking and monetary policy in financing
the transition to a low-carbon economy. Ecological Economics, 121, 220–230.
Cardaliaguet, P., Quincampoix, M., & Saint-Pierre, P. (1999). Set-valued numerical analysis for
optimal control and differential games. Annals of the International Society of Dynamic Games,
4, 177–247. https://doi.org/10.1007/978-1-4612-1592-9_4.
Carlin, W., & Bowles, S. (2017). CORE eBook. CORE.
236 References
Casey, G., & Galor, O. (2014). Population dynamics and long-run economic growth. MPRA Paper
62598.
Chakravarty, S., & Lugo, M. A. (2016). Multidimensional indicators of inequality and poverty. In
The Oxford handbook of well-being and public policy.
Chichilnisky, G. (1997). What is sustainable development? Land Economics, 73, 467–491.
Clark, A. E. (2016). SWB as a measure of individual well-being. The Oxford handbook of well-being
and public policy.
Clément-Pitiot, H., & Doyen, L. (1999). Exchange rate dynamics, target zone and viability. Uni-
versity of Paris Discussion Paper.
Clément-Pitiot, H., & Saint-Pierre, P. (2006). Goodwin’s models through viability analysis: Some
lights for contemporary political economics regulations. In 12th international conference on
computing in economics and finance, Cyprus. Conference maker.
Colander, D., & Kupers, R. (2014). Complexity and the art of public policy: Solving society’s
problems from the bottom up. Princeton: Princeton University Press.
Collier, P. (2013). Exodus: How migration is changing our world. Oxford: Oxford University Press.
Conti, J., Holtberg, P., Diefenderfer, J., LaRose, A., Turnure, J. T., & Westfall, L. (2016). Interna-
tional energy outlook 2016 with projections to 2040. Technical Report. USDOE Energy Infor-
mation Administration (EIA), Washington, DC (United States). Office of Energy Analysis.
Costanza, R. (2015). Ecosystem services in theory and practice. Valuing Nature: Protected Areas
and Ecosystem Services, 6.
Costanza, R., de Groot, R., Braat, L., Kubiszewski, I., Fioramonti, L., Sutton, P., et al. (2017).
Twenty years of ecosystem services: How far have we come and how far do we still need to go?
Ecosystem Services, 28, 1–16.
Cowen, T. (2007). Caring about the distant future: Why it matters and what it means. The University
of Chicago Law Review, 74, 5.
Coyle, D. (2014). Measuring economic progress. VOX CEPR Policy Portal, 17 February.
Coyle, D. (2015). GDP: A brief but affectionate history. Princeton: Princeton University Press.
Crampton, E., & Acharya, K. (2015). In the zone: Creating a toolbox for regional prosperity.
Wellington: The New Zealand Initiative.
Czaika, M., Parsons, C., et al. (2015). The gravity of high-skilled migration policies. IMI policy
paper.
d’Albis, H., Boubtane, E., & Coulibaly, D. (2016). Immigration policy and macroeconomic per-
formance in France. Annals of Economics and Statistics/Annales d’Économie et de Statistique,
279–308.
Dalziel, P., & Saunders, C. (2014). Wellbeing economics: Future directions for New Zealand. Bridget
Williams Books.
Dalziel, P., & Saunders, C. (2016). Treasury’s refreshed views on New Zealand’s economic strategy:
A review article. New Zealand Economic Papers, 1–17.
Dalziel, P., Saunders, C., Fyfe, R., & Newton, B. (2009). Sustainable development and cultural
capital. Statistics New Zealand.
Dalziel, P., Saunders, C., Tait, P., Saunders, J., Miller, S., Guenther, M., et al. (2017). Rewarding
responsible innovation when consumers are distant from producers: Evidence from New Zealand.
International Food and Agribusiness Management Review, 1–18.
De La Croix, D., & Michel, P. (2002). A theory of economic growth: Dynamics and policy in
overlapping generations. Cambridge: Cambridge University Press.
De Lara, M. (2017). A mathematical framework for resilience: Dynamics, strategies, shocks and
acceptable paths. Technical Report. https://hal.archives-ouvertes.fr/hal-01578049.
De Lara, M., Doyen, L., Guilbaud, T., & Rochet, M. J. (2006). Is a management framework based
on spawning stock biomass indicators sustainable? A viability approach. ICES Journal of Marine
Science, 64, 761–767.
De Perthuis, C., & Jouvet, P. A. (2015). Green capital: A new perspective on growth. New York:
Columbia University Press.
References 237
Di Tella, R., & MacCulloch, R. (2008). Gross national happiness as an answer to the Easterlin
Paradox? Journal of Development Economics, 86, 22–42.
Di Tella, R., MacCulloch, R. J., & Oswald, A. J. (2001). Preferences over inflation and unemploy-
ment: Evidence from surveys of happiness. The American Economic Review, 91, 335–341.
Dolan, P., & Fujiwara, D. (2016). Happiness-based policy analysis. In The Oxford handbook of
well-being and public policy.
Doyen, L., & Saint-Pierre, P. (1997). Scale of viability and minimal time of crisis. Set-Valued
Analysis, 5, 227–246.
Driscoll-O’Keefe, M., & Krawczyk, J. B. (2016). Viability of an economy with constrained inequality
in a two tax system. Technical Report 7. School of Economics and Finance Victoria University
of Wellington.
Drupp, M. A. (2016). Limits to substitution between ecosystem services and manufactured goods
and implications for social discounting. Environmental and Resource Economics, 1–24.
Duraiappah, A. K., & Muñoz, P. (2012). Inclusive wealth: A tool for the United Nations. Environment
and Development Economics, 17, 362–367.
Durand, M. (2015). The OECD better life initiative: How’s life? and the measurement of well-being.
Review of Income and Wealth, 61, 4–17.
Eiffe, F. F. (2010). Amartya Sen reading Adam Smith. History of Economics Review, 51, 1–23.
Engelbrecht, H. J. (2014). Hamlet without the prince: The capital approach to development, the New
Zealand Treasury’s living standards framework and policy making. Prometheus, 32, 245–263.
Eppel, E., & Karacaoglu, G. (2017). Social investment in a complex and radically uncertain world.
In J. Boston and D. Gill (Eds.) Social investment – a New Zealand policy experiment. Bridget
Williams Books. (pp. 378–404).
Etilé, F., Frijters, P., Johson, D. W., & Shields, M. A. (2017). Modelling heterogeneity in the
resilience to major socioeconomic life events. PSE Working Papers No. 2017-09.
Evans, L., Quigley, N., et al. (2013). Intergenerational contracts and time consistency: Implications
for policy settings and governance in the social welfare system. Technical Report. New Zealand
Treasury.
Faria, H. J., Montesinos-Yufa, H. M., Morales, D. R., & Navarro, C. E. (2016). Unbundling the
roles of human capital and institutions in economic development. European Journal of Political
Economy, 45, 108–128.
Farina, F., et al. (2015). Development theory and poverty. A review. Technical Report. AISSEC DP
2015-05.
Feldman, M., Hadjimichael, T., Lanahan, L., & Kemeny, T. (2016). The logic of economic devel-
opment: A definition and model for investment. Environment and Planning C: Government and
Policy, 34, 5–21.
Fernández, A., & Tamayo, C. E. (2015). From institutions to financial development and growth:
What are the links? Journal of Economic Surveys, 31, 17–57.
Ferreira, F. H., & Peragine, V. (2015). Equality of opportunity: Theory and evidence. Policy Research
Working Paper no. WPS 7217. World Bank Group.
Ferreira, S., Hamilton, K., & Vincent, J. R. (2008). Comprehensive wealth and future consumption:
Accounting for population growth. The World Bank Economic Review, 22, 233–248.
Fioramonti, L. (2017). Wellbeing economy: Success in a world without growth. South Africa: Pan
Macmillan.
Fisher, I. (1936). Money. Adelphi Company.
Fleche, S., Smith, C., & Sorsa, P. (2012). Exploring determinants of subjective wellbeing in OECD
countries. OECD Statistics Working Papers.
Fleurbaey, M., Blanchet, D. (2013). Beyond GDP: Measuring welfare and assessing sustainability.
Oxford: Oxford University Press.
Fontana, G., & Sawyer, M. (2015). The macroeconomics and financial system requirements for a
sustainable future. In Finance and the macroeconomics of environmental policies (pp. 74–110).
Berlin: Springer.
238 References
Frey, B. S., Benz, M., & Stutzer, A. (2004). Introducing procedural utility: Not only what, but also
how matters. Journal of Institutional and Theoretical Economics JITE, 160, 377–401.
Frey, B. S., & Stutzer, A. (2002). What can economists learn from happiness research? Journal of
Economic Literature, 40, 402–435.
Friedman, B. M. (2006). The moral consequences of economic growth. Society, 43, 15–22.
Gaitsgory, V., & Quincampoix, M. (2009). Linear programming approach to deterministic infinite
horizon optimal control problems with discounting. SIAM Journal of Control and Optimization,
48, 2480–2512. https://doi.org/10.1137/070696209.
Galama, T. J., Morgan, R., & Saavedra, J. E. (2017). Wealthier, happier and more self-sufficient:
When anti-poverty programs improve economic and subjective wellbeing at a reduced cost to
taxpayers. Technical Report. National Bureau of Economic Research.
Galor, O., & Weil, D. N. (1999). From Malthusian stagnation to modern growth. The American
Economic Review, 89, 150–154.
Galor, O., & Weil, D. N. (2000). Population, technology, and growth: From malthusian stagnation
to the demographic transition and beyond. American Economic Review, 806–828.
García-Montalvo, J., Reynal-Querol, M., et al. (2017). Ethnic diversity and growth: Revisiting the
evidence. CEPR Discussion Paper No. DP 12400.
Gershman, B. (2016). Long-run development and the new cultural economics. Working Papers
2016-06, American University, Department of Economics.
Gintis, H. (2014). Review of David Colander and Roland Kupers - complexity and the art of public
policy: Solving society’s problems from the bottom up. Amazon.com Customer Review, May 24.
Gleeson-White, J. (2015). Six capitals, or can accountants save the planet?: Rethinking capitalism
for the twenty-first century. New York: W. W. Norton & Company.
Gleisner, B., McAlister, F., Galt, M., & Beaglehole, J. (2012). A living standards approach to public
policy making. New Zealand Economic Papers, 46, 211–238.
Gorodnichenko, Y., & Roland, G. (2011). Which dimensions of culture matter for long-run growth?
The American Economic Review, 101, 492–498.
Gough, I. (2015). Climate change and sustainable welfare: The centrality of human needs. Cam-
bridge Journal of Economics, 39 (5), 1191–1214.
Greasley, D., Hanley, N., Kunnas, J., McLaughlin, E., Oxley, L., & Warde, P. (2013). Comprehensive
investment and future well-being in the USA (pp. 1869–2000). Stirling Economics Discussion
Papers 2013-06, University of Stirling, Division of Economics.
Greasley, D., Hanley, N., McLaughlin, E., & Oxley, L. (2014). The emperor has new clothes:
Empirical tests of mainstream theories of economic growth. Stirling Economics Discussion
Papers 2014-08. University of Stirling, Division of Economics.
Greasley, D., McLaughlin, E., Hanley, N., & Oxley, L. (2017). Australia: a land of missed oppor-
tunities? Environment and Development Economics, 1–25.
Green, A., Janmaat, G., & Cheng, H. (2011). Social cohesion: Converging and diverging trends.
National Institute Economic Review, 215, R6–R22.
Griggs, D., Stafford-Smith, M., Gaffney, O., Rockström, J., Öhman, M. C., Shyamsundar, P., et al.
(2013). Policy: Sustainable development goals for people and planet. Nature, 495, 305–307.
Growiec, K., Growiec, J., Kaminski, B., et al. (2017). Mapping the dimensions of social capital.
SGH KAE Working Paper Series No. 2017/025.
Gründler, K., Krieger, T. (2016). Democracy and growth: Evidence from a machine learning indi-
cator. European Journal of Political Economy, 45, 85–107.
Guild, R., Matsumoto, I., et al. (2013). Moving from risk to resilience: Sustainable urban develop-
ment in the pacific. Pacific Studies Series. Asian Development Bank.
Haldane, A. G. (2013). Rethinking the financial network. In Fragile stabilität—Stabile fragilität
(pp. 243–278). Berlin: Springer.
Haldane, A. G. (2015). Growing, fast and slow. University of East Anglia 17.
Haldane, A. G. (2018). Ideas and institutions–a growth story. Bank of England Speech. The Guild
Society. University of Oxford.
Haldane, A. G., & May, R. M. (2011). Systemic risk in banking ecosystems. Nature, 469, 351.
References 239
Hamilton, K. (2000). Genuine saving as a sustainability indicator. OECD proceedings: Frameworks

to measure sustainable development (pp. 65–78).
Hamilton, K., & Hepburn, C. (2014). Wealth. Oxford Review of Economic Policy, 30, 1–20.
Hanley, N., Oxley, L., Greasley, D., McLaughlin, E., & Blum, M. (2016). Empirical testing of
genuine savings as an indicator of weak sustainability: A three-country analysis of long-run
trends. Environmental and Resource Economics, 63, 313–338.
Hansen, L. P. (2014). Nobel lecture: Uncertainty outside and inside economic models. Journal of
Political Economy, 122, 945–987.
Hardt, L., & O’Neill, D. W. (2017). Ecological macroeconomic models: Assessing current devel-
opments. Ecological Economics, 134, 198–211.
Haurie, A., Krawczyk, J. B., & Zaccour, G. (2012). Games and dynamic games. Business series
(Vol. 1). Singapore: World Scientific – Now Publishers.
Hausman, D., McPherson, M., & Satz, D. (2016). Economic analysis, moral philosophy, and public
policy. Cambridge: Cambridge University Press.
Hayek, F. A. (2013). The constitution of liberty: The definitive edition (Vol. 17). Abingdon: Rout-
ledge.
Heckman, J. J., & Corbin, C. O. (2016). Capabilities and skills. Journal of Human Development
and Capabilities, 17, 342–359.
Heckman, J. J., & Mosso, S. (2014). The economics of human development and social mobility.
Annual Review of Economics, 6, 689–733.
Helliwell, J. F., Layard, R., & Sachs, J. (2015). World Happiness Report 2013.
Helm, D. (2010). Rethinking the economic borders of the state. The Social Market Foundation
(November 2010).
Hermele, K., et al. (2015). How economists understand (or not) the relationship between the real
and financial economy. FESSUD Papers 82.
Hodgson, G. M. (2014). On fuzzy frontiers and fragmented foundations: Some reflections on the
original and new institutional economics. Journal of Institutional Economics, 10, 591–611.
Hodgson Geoffrey, M. (2006). What are institutions? Journal of Economic Issues, 40, 1–24.
Human (2013). The human capital report. World Economic Forum.
Hyman, M. (2008). Stabilizing an unstable economy. Yale University Press.
Imperative, S. P. (2017). Social progress index. Social progress imperative.
Institute, L. (2016). Legatum prosperity index. Legatum Institute.
Jakab, Z., Kumhof, M. (2015). Banks are not intermediaries of loanable funds–and why this matters.
Jones, C., & Vollrath, D. (2015). Introduction to economic growth (3rd ed.). New York: W. W.
Norton and Company.
Judd, K. L. (1987). The welfare cost of factor taxation in a perfect-foresight model. Journal of
Political Economy, 95(4), 675–709.
Karacaoglu, G. (2015). The New Zealand Treasury’s living standards framework-exploring a
stylised model. Working Paper 15/12. New Zealand Treasury.
Karacaoglu, G., Au, J. (2018). It is not just the economy, stupid. In S. Levine (Ed.) Stardust &
substance. Victoria University Press (pp. 229–250).
Kešeljević, A., et al. (2016). Institutional origins of subjective well-being: Estimating the effects of
economic freedom on national happiness. Journal of Happiness Studies, 17, 659–712.
Kim, K., & Krawczyk, J. B. (2017). Sustainable emission control policies: Viability theory approach.
Seoul Journal of Economics, 30, 291–317.
King, A. (2018). Living standards analysis model: The first prototype. Working Paper 18/05. New
Zealand Treasury. ISBN 978-1-98-855677-2.
King, M. (2016). The end of alchemy: Money, banking, and the future of the global economy. New
York: W. W. Norton & Company.
Kratena, K. (2015). Thematic report: Macroeconomic models including specifically social and
environmental aspects. Technical Report. WWWforEurope deliverable.
Krause, A. (2014). Happiness and work. IZA Discussion Paper, No 8435, August.
240 References
Krawczyk, J. B., & Judd, K. L. (2016). A note on determining viable economic states in a dynamic
model of taxation. Macroeconomic Dynamics, 20, 1395–1412.
Krawczyk, J. B., & Kim, K. (2009). “Satisficing” solutions to a monetary policy problem:
A viability theory approach. Macroeconomic Dynamics, 13, 46–80. https://doi.org/10.1017/
s1365100508070466.
Krawczyk, J. B., & Kim, K. (2014). Viable stabilising non-Taylor monetary policies for an open
economy. Computational Economics, 43, 233–268.
Krawczyk, J. B., & Pharo, A. (2011). Viability kernel approximation, analysis and simulation
application – VIKAASA manual. SEF Working Paper 13/2011. Victoria University of Wellington.
http://hdl.handle.net/10063/1878.
Krawczyk, J. B., & Pharo, A. (2013). Viability theory: An applied mathematics tool for achieving
dynamic systems’ sustainability. Mathematica Applicanda, 41, 97–126. http://wydawnictwa.ptm.
org.pl/index.php/matematyka-stosowana/article/view/409.
Krawczyk, J. B., & Pharo, A. (2014a). Manual of VIKAASA 2.0: An application for computing and
graphing viability kernels for simple viability problems. SEF Working Paper 08/2014. Victoria
University of Wellington. http://researcharchive.vuw.ac.nz/handle/10063/3432.
Krawczyk, J. B. & Pharo, A. (2014b). Viability kernel approximation, analysis and simulation
application – VIKAASA code. http://socsol.github.io/vikaasa/.
Krawczyk, J. B., Pharo, A., Serea, O. S., & Sinclair, S. (2013). Computation of viability kernels: A
case study of by-catch fisheries. Computational Management Science, 10, 365–396. https://doi.
org/10.1007/s10287-013-0189-z.
Krawczyk, J. B., & Serea, O. S. (2013). When can it be not optimal to adopt a new technology?
A viability theory solution to a two-stage optimal control problem of new technology adoption.
Optimal Control Applications and Methods, 34, 127–144. https://doi.org/10.1002/oca.1030.
Krawczyk, J. B., & Sethi, R. (2007). Satisficing Solutions for New Zealand Monetary Policy. Tech-
nical Report No. DP2007/03. Reserve Bank of New Zealand. http://www.rbnz.govt.nz/research_
and_publications/articles/details.aspx?id=3968.
Krawczyk, J. B., Sissons, C., & Vincent, D. (2012). Optimal versus satisfactory decision making:
A case study of sales with a target. Computational Management Science, 9, 233–254. Published
online: February 2012. https://doi.org/10.1007/s10287-012-0141-7.
Krawczyk, J. B., & Townsend, W. (2015a). Basic understanding of social inequality dynamics.
Working Paper Series 4154. Victoria University of Wellington, School of Economics and Finance.
Krawczyk, J. B., & Townsend, W. (2015b). New Zealand inequality and the struggle between capital
and labour. Working Paper Series 4156. Victoria University of Wellington, School of Economics
and Finance.
Krawczyk, J. B., & Townsend, W. (2015c). Viability of an economy with constrained inequality.
Working Paper Series 4689. Victoria University of Wellington, School of Economics and Finance.
Kroll, C., & Delhey, J. (2013). A happy nation? opportunities and challenges of using subjective
indicators in policymaking. Social Indicators Research, 114, 13–28.
Krugman, P. (1979). A model of innovation, technology transfer, and the world distribution of
income. Journal of Political Economy, 87, 253–266.
Kubiszewski, I., Costanza, R., Franco, C., Lawn, P., Talberth, J., Jackson, T., et al. (2013). Beyond
GDP: Measuring and achieving global genuine progress. Ecological Economics, 93, 57–68.
Laeven, L., Levine, R., & Michalopoulos, S. (2015). Financial innovation and endogenous growth.
Journal of Financial Intermediation, 24, 1–24.
Larsen, C. A. (2014). Social cohesion: Definition, measurement and developments. Working Paper,
Institut for Statskundskab, Aalborg Universitet.
Layard, R., & O’Donnell, G. (2015). How to make policy when happiness is the goal. World
Happiness Report 2015, 76–87.
Lee, J. (2012). Examining a green growth model for policy implications. Seoul Journal of Eco-
nomics, 25, 57–87.
Levine, R. (2005). Finance and growth: Theory and evidence. In P Aghion and S Durlauf (Eds.),
Handbook of Economic Growth.
References 241
Lipset, S. M. (1959). Some social requisites of democracy: Economic development and political
legitimacy. American Political Science Review, 53, 69–105.
Liu, M. Y. (2016). Political economy approach of institutional research. Technical Report. Univer-
sity Library of Munich, Germany.
Markey-Towler, B. (2016). Economics cannot isolate itself from political theory: A mathematical
demonstration. SSRN Electronic Journal. October.
Martin, I. W., & Pindyck, R. S. (2015). Averting catastrophes: The strange economics of scylla and
charybdis. The American Economic Review, 105, 2947–2985.
Martinet, V., & Doyen, L. (2007). Sustainability of an economy with an exhaustible resource: A
viable control approach. Resource and Energy Economics, 29, 17–39.
Martinet, V., Thébaud, O., & Doyen, L. (2007). Defining viable recovery paths toward sustainable
fisheries. Ecological Economics, 64, 411–422. https://doi.org/10.1016/j.ecolecon.2007.02.036.
Mazzucato, M. (2013a). The entrepreneurial state: Debunking the public versus private myth in
risk and innovation. London: Anthem.
Mazzucato, M. (2013b). Financing innovation: Creative destruction versus destructive creation.
Industrial and Corporate Change, 22, 851–867.
Mazzucato, M. (2015). The green entrepreneurial state. SPRU Working Paper Series.
Mazzucato, M. (2016). From market fixing to market-creating: A new framework for innovation
policy. Industry and Innovation, 23, 140–156.
Mazzucato, M., Perez, C. (2014). Innovation as growth policy: The challenge for Europe. SPRU
Working paper Series.
Mazzucato, M., & Semieniuk, G. (2017). Financing renewable energy: Who is financing what and
why it matters. Technological Forecasting and Social Change.
Mazzucato, M., & Wray, L. R. (2015). Financing the capital development of the economy: A
Keynes-Schumpeter-Minsky synthesis.
McCloskey, D. N. (2014). Exordium. University of Illinois at Chicago.
McCloskey, D. N. (2016). Bourgeois equality: How ideas, not capital or institutions, enriched the
World. Chicago: University of Chicago Press.
Merton, R. C., & Bodie, Z. (1995). A conceptual framework for analyzing the financial system. The
global financial system: A functional perspective (pp. 3–31).
Michalopoulos, S., Laeven, L., & Levine, R. (2009). Financial innovation and endogenous growth.
Technical Report. National Bureau of Economic Research.
Miller, N. (2006). Externalities and public goods. Notes on Microeconomic Theory. Chapter 8.
Mokyr, J. (2016). A culture of growth: The origins of the modern economy. Princeton: Princeton
University Press.
Naqvi, S. A. A., & Engelbert, S. (2017). Directed technological change in a post-Keynesian eco-
logical macromodel. Working Papers PKWP1714, Post Keynesian Economics Society (PKES).
Nardo, M., Saisana, M., Saltelli, A., Tarantola, S., Hoffman, A., & Giovannini, E. (2005). Handbook
on constructing composite indicators.
National tax agency (2010). Income tax guide for foreigners. http://www.nta.go.jp/tetsuzuki/
shinkoku/shotoku/tebiki2010/pdf/43.pdf.
Nations, U. (2017). The Sustainable development goals report. United Nations.
Nelson, R. R. (2017). Economics from an evolutionary perspective. LEM Papers Series 2017/18.
Nelson, R. R. (ed.) (2017). An overview of modern evolutionary economics. New York: Columbia
University Press.
Nikolaev, B., & Bennett, D. L. (2016). Give me liberty and give me control: Economic freedom,
control perceptions and the paradox of choice. European Journal of Political Economy, 45, 39–52.
Nordhaus, W. D., & Tobin, J. (1972). Is growth obsolete? In Economic research: Retrospect and
prospect, volume 5, economic growth. NBER (pp. 1–80).
North, D. C. (1990). Institutions, institutional change and economic performance. Cambridge:
Cambridge University Press.
North, D. C. (1991). Institutions. Journal of Economic Perspectives, 5, 97–112. https://doi.org/10.
1257/jep.5.1.97.
242 References
North, D. C. (1994). Economic performance through time. The American Economic Review, 84,
359–368.
O’Donnell, G., Deaton, A., Durand, M., Halpern, D., & Layard, R. (2014). Wellbeing and policy.
Legatum Institute.
OECD. (2015). How’s life.
OECD. (2016). Better life index.
OECD. (2017a). Better life index.
OECD. (2017b). How’s life? 2017: Measuring well-being.
OECD-Education. (2017). Education at a glance: OECD indicators 2017. OECD.
Office, U.N.D.P.H.D.R. (1999). Human Development Report: Background papers. Human Devel-
opment Report Office, United Nations Development Programme.
Ostrom, E. (2009). Nobel prize lecture.
Panizza, U., et al. (2017). Non-linearities in the relationship between finance and growth. Economics
Section, The Graduate Institute of International Studies. Technical Report.
Pelenc, J., Ballet, J., & Dedeurwaerdere, T. (2015). Weak sustainability versus strong sustainability.
Brief for GSDR United Nations.
Pervaiz, Z., & Chaudhary, A. R. (2015). Social cohesion and economic growth: An empirical
investigation. Australian Economic Review, 48, 369–381.
Phelps, E. S. (2013). Mass flourishing: How grassroots innovation created jobs, challenge, and
change. Princeton: Princeton University Press.
Popov, A. (2017). Evidence on finance and economic growth. ECB Working Paper. Technical
Report.
Popp, D., Newell, R. G., & Jaffe, A. B. (2010). Energy, the environment, and technological change.
Handbook of the Economics of Innovation, 2, 873–937.
Porter, M. E., & Heppelmann, J. E. (2014). How smart, connected products are transforming com-
petition. Harvard Business Review, 92, 64–88.
Porter, M. E., Stern, S., & Green, M. (2014). Social Progress Index 2014. DC: Social Progress
Imperative Washington.
Pujal, D., & Saint-Pierre, P. (2006). Capture basin algorithm for evaluating and managing complex
financial instruments. In 12th International Conference on Computing in Economics and Finance,
Cyprus. Conference maker.
Putnam, R. D. (2001). Bowling alone: The collapse and revival of American community. New York:
Simon and Schuster.
Qasim, M., Oxley, L., McLaughlin, E., et al. (2018). Genuine savings as a test of New Zealand weak
sustainability. Technical Report. University of St. Andrews, School of Geography and Sustainable
Development.
Quincampoix, M., & Veliov, V. M. (1998). Viability with a target: Theory and applications. Appli-
cations of Mathematical Engineering, 47–58.
Ravlik, M. (2014). Determinants of international migration: A global analysis. Higher School of
Economics Research paper No WP BRP 52/SOC/2014.
Raworth, K. (2017). Doughnut economics: Seven ways to think like a 21st-century economist.
Hartford: Chelsea Green Publishing.
Robeyns, I. (2005). The capability approach: A theoretical survey. Journal of Human Development,
6, 93–117.
Rockström, J., Steffen, W., Noone, K., Persson, Å., Chapin III, F. S., Lambin, E., et al. (2009).
Planetary boundaries: Exploring the safe operating space for humanity. Ecology and Society, 14.
Rodríguez-Pose, A., & von Berlepsch, V. (2017). Does population diversity matter for economic
development in the very long-term? historic migration, diversity and county wealth in the US.
CEPR Discussion Paper No. DP12347.
Rodrik, D. (2004). Getting institutions right. CESifo DICE Report 2, 10–15.
Roos, M. W. (2015). The macroeconomics of radical uncertainty. Ruhr Economic Paper No 592.
References 243
Saam, M., Papageorgiou, C., & Schulte, P. (2014). Elasticity of substitution between clean and dirty
energy inputs-a macroeconomic perspective. ZEW - Centre for European Economic Research
Discussion Paper No 13-087.
Sachs, J. D. (2015). Investing in social capital. World Happiness Report (pp. 152–166).
Sadetskaya, K. (2014). Higher living standards in New Zealand. New Zealand Treasury Internal
Note.
Sager, L., et al. (2017). Income inequality and carbon consumption: Evidence from environmen-
tal Engel curves. Technical Report: Grantham Research Institute on Climate Change and the
Environment.
Saunders, C., & Dalziel, P. (2017). Twenty-five years of counting for nothing: Waring’s critique of
national accounts. Feminist Economics, 23, 200–218.
Savioli, M., & Patuelli, R. (2016). Social capital, institutions and policymaking. Quaderni - Working
Paper DSE No 1070.
Schmidt, P. (2015). Market failure versus system failure as a rationale for economic policy? A
critique from an evolutionary perspective. Technical Report. Papers on Economics and Evolution.
Schreyer, P. (2016). GDP. In The Oxford Handbook of Well-Being and Public Policy.
Schumpeter, J. A. (1947). The creative response in economic history. The Journal of Economic
History, 7, 149–159.
Schumpeter, J. A. (1982). The theory of economic development: An inquiry into profits, capital,
credit, interest, and the business cycle (1912/1934), January 1, 1982, 244. New Jercy: Transaction
Publishers.
Scott, J. C. (1998). Seeing like a state: How certain schemes to improve the human condition have
failed. New Haven: Yale University Press.
Sen, A. (2004). Rationality and freedom. Cambridge: Harvard University Press.
Sen, A. K. (2009). The idea of justice. Cambridge: Belknap.
Sen, A. K. (2017). Collective choice and social welfare. Penguin.
Simon, H. A. (1955). A behavioral model of rational choice. Quarterly Journal of Economics, 69,
99–118. https://doi.org/10.2307/1884852.
Smith, A. (1790). The theory of moral sentiments. Library of economics and liberty.
Smith, A. (1937). The wealth of nations [1776].
Smith, C. (2015). Measuring well-being: Progress to date and remaining challenges. OECD Pre-
sentation Slides. KDI School, Sejong; 24 November.
Smith, C. (2017). Doing something about poverty in New Zealand. McGuinness Institute Think
Piece 26.
Soros, G. (2003). The alchemy of finance. New Jercy: Wiley.
Spence, M. (2014). Five reasons for slow growth. Project Syndicate, 29.
Spolaore, E., & Wacziarg, R. (2013). How deep are the roots of economic development? Journal
of Economic Literature, 51, 325–369.
Stiglitz, J. E. (2015). New theoretical perspectives on the distribution of income and wealth among
individuals: 4 parts. Technical Report. National Bureau of Economic Research.
Stiglitz, J. E., Sen, A., Fitoussi, J. P., et al. (2010). Report by the commission on the measurement of
economic performance and social progress. Paris: Commission on the Measurement of Economic
Performance and Social Progress.
Stokey, N. L. (1998). Are there limits to growth? International Economic Review, 1–31.
Sunaga, M., et al. (2015). A Schumpeterian growth model with financial intermediaries. Discussion
Papers in Economics and Business 15-19, Osaka University, Graduate School of Economics and
Osaka School of International Public Policy (OSIPP).
Tanzi, V. (2011). Government versus markets: The changing economic role of the state. Cambridge:
Cambridge University Press.
Thaler, R. H. (2016). Behavioral economics: Past, present and future. The American Economic
Review. 106(7), 1577–1600.
The World Bank. (2015). Quarterly public sector debt. http://data.worldbank.org/data-catalog/
quarterly-public-sector-debt.
244 References
Thompson, M. J. R. (2015). Social capital, innovation and economic growth. NIPE Working Paper
Series, 3, 1–24.
Thomson, A. (2013). Indicators for measurement of the living standards framework in New Zealand.
New Zealand Treasury Internal Note.
Tirole, J. (2017). Economics for the common good. Princeton: Princeton University Press.
Toledo, A., et al. (2016). Eradicating poverty, reducing inequality, and promoting sustainable
development. Papers presented at global meetings of the emerging markets forum, Emerging
Markets Forum.
Turnovsky, S. J. (2013). The relationship between economic growth and inequality. New Zealand
Economic Papers, 47, 113–139.
Turnovsky, S. J., & García-Peñalosa, C. (2008). Distributional dynamics in a neoclassical growth
model: The role of elastic labor supply. Journal of Economic Dynamics and Control, 32, 1399–
1431.
Turnovsky, S. J., & Mitra, A. (2013). The interaction between human and physical capital accumu-
lation and the growth-inequality trade-off. Journal of Human Capital, 7, 26–75.
Van Zanden, J. L., Baten, J., Mira d’Ercole, M., Rijpma, A., Smith, C., & Timmer, M. (2014).
Global well-being since 1820. OECD Publishing.
Varvarigos, D., Xin, G., et al. (2015). Social Interactions, the Evolution of Trust, and
Economic Growth. Discussion Papers in Economics 15/05, Department of Economics, University
of Leicester.
Veliov, V. M. (1993). Sufficient conditions for viability under imperfect measurement. Set-Valued
Analysis, 1, 305–317. https://doi.org/10.1007/bf01027640.
Von Hayek, F. A. (1989). The pretence of knowledge. The American Economic Review, 79, 3–7.
Voosholz, F. (2014). The influence of different production functions on modeling resource extraction
and economic growth. Technical Report. CAWM Discussion Paper, Centrum für Angewandte
Wirtschaftsforschung Münster.
Walzer, M. (1983). Spheres of justice: A defense of pluralism and equality. New York: Basic Book.
Ward, G. (2015). Is happiness a predictor of election results? CEP (Centre for Economic Perfor-
mance) Discussion Paper No 1343.
Waring, M., & Steinem, G. (1988). If women counted: A new feminist economics. San Francisco:
Harper & Row.
Watkins, A., & Ferrara, E. L. (2005). Ethnic diversity and economic performance. Journal of
Economic Literature, 43, 762–800.
Watson, D., Maître, B., Whelan, C. T., & Russell, H. (2016). Technical Paper on Measurement of
Multidimensional Quality of Life in Ireland. ESRI.
Weitzman, M. L. (1976). On the welfare significance of national product in a dynamic economy.
The Quarterly Journal of Economics, 90, 156–162.
Weitzman, M. L. (2009). Income, wealth, and the maximum principle. Cambridge: Harvard Uni-
versity Press.
Welsch, H., & Biermann, P. (2016). Poverty is a public bad: Panel evidence from subjective well-
being data. Technical Report. Oldenburg Discussion Papers in Economics.
Wing, I. S. (2004). Computable general equilibrium models and their use in economy-wide policy
analysis. Technical Note. Joint Program on the Science and Policy of Global Change, MIT.
Yamamura, E. (2016). Social conflict and redistributive preferences among rich and poor: Testing
the hypothesis of Acemoglu and Robinson. Journal of Applied Economics, 19, 41–63.
Yusifzada, L., & Mammadova, A. (2015). Financial intermediation and economic growth. Working
Paper 1091. William Davidson Institute.
Index
A D
Adaptive complexity, 25, 140 Debt, 188
Differential inclusion, 169
Discount rate, 129, 161, 176
Dynamic programming, 18
B
Biodiversity, 117, 122, 126, 133
Bonding capital, 73 E
Bonding social capital, 38, 83 Economic capital, 37
Boulder, 179, 180, 185–187, 196, 206 Ecosystem, 141
Bridging capital, 74, 152 Education, 119, 121, 126, 131, 132, 134
Bridging social capital, 38, 83 Emission control, 200
Environment, 121, 129, 132
Equation of motion, 114, 115
Equity, 74, 188
C Euler equation, 175
Cantril ladder, 120 Exhaustible resource, 92
Capital equation, 114, 115 Extraction goods, 122
Civic engagement, 121
Community, 114
Compatibility of dynamics and constraints, F
165 Final goods, 122
Complements, 49 Fiscal policy, 126
Comprehensive consumption, 30
Comprehensive wealth, 28, 31, 37, 112
G
Computable General Equilibrium (CGE),
Gini coefficient, 189
113, 122, 129
Golden rule, 178
Conditionality, 144 Government, 125
Constrained control adjustments, 163 Government transfers, 126, 132
Constraint, modal, 203, 210
Constraint, normative, 203
Consumption, comprehensive, 8 H
Coping strategy, 143, 150 Happiness, 120
Crime, 118 Health, 115, 129, 131, 133
Critical policy tradeoffs, 101 Healthcare, 116, 121, 126, 131, 132

https://doi.org/10.1007/978-981-13-6104-3
246 Index
Housing, 119, 121, 131, 132, 134 Pareto index, 189

Human capital, 37 Policy advice, 188, 199, 212
Policy analysis - benchmark model, 97
Policy analysis - stylised model, 97
I Policy - comprehensive wealth, 102
Income, 117, 132 Policy - finance, 102
Inequality, 82, 83, 188 Policy - numerical simulations, 103, 105
Infrastructure, 122, 126 Policy - poverty reduction, 103
Intergenerational wellbeing, 31 Policy, resilient, 162
Investment, 34 Policy - wellbeing state, 102
Investment approach, 10, 17 Pollution, 117, 126, 130, 132
Investment equation, 115 Procedural utility, 149
Production, 121, 131, 132, 134
Production, clean, 123, 132
J Production, dirty, 123
Jobs, 116, 133 Production processes, 123
Prudential policies, 152
K
Knowledge, 119 R
Radical uncertainty, 25
Relative factor-share, 189–191, 193, 195–
L 197, 199, 202
Laissez faire benchmark model, 91 Rent, 191, 201, 205, 210, 211
Laissez faire stylised model, 93 Representative household (agent), 176, 177,
Leisure, 119, 131, 134 189, 191
Localism, 153 Resilience, xiii, xv, 40, 41, 72, 76, 78, 79,
143, 144, 167, 212
Resources, 117, 122
M Rest of the world, 125
Maori-Pakeha, 152
Market-clearing condition, 125
Market failure, 5 S
Markovian policy/strategy, 165, 170 Safety, 118, 121, 122, 126
Maximum principle, 18, 175 Satisficing policy, 25, 161
Measure of income inequality, 189 Shadow prices, 33
Micro-foundations, 35 Skills, 119, 132
Migration, 125 Small, open economy, 125
Modal constraints (bounds), 167 Smooth strategy, 188
Model dynamics, 100 Social and cultural capital, 38
Social cohesion, 28, 31, 36, 38, 40, 41, 48,
65, 66, 71, 73, 75–77, 82, 174, 188
N Social connection, 114, 122, 133
Natural capital, 39 State state outcomes, 100
Neoclassical model, 188 Steady state models, 91
Normative constraints (bounds), 167 Sticky price, 169
Strategy, non-smooth, 167
Strategy, satisficing, 169
O Strategy, viable, 160
OECD BLI, 8, 112, 114, 129, 132 Stylised model, 95
Subjective wellbeing, 120
Substitutes, 49
P Sustainability, 75, 188
Parameter uncertainty, 162 Sustainability, strong, 13, 52
Index 247
Sustainability, weak, 9, 13, 52, 75 Viability kernel, 25, 151, 159, 162, 163, 165,
Sustainable economic states, 188 166, 169, 171, 173, 179, 180, 184,
Sustainable policy rule, 165 187, 194, 195, 197, 203, 205, 212–
214, 216–219
Viability theory, 140, 159, 160, 163, 222
T Volunteering, 120
Taxation, 188 Viability theory problem, xv
Taxes, 126
Time consistency, 17, 18
Time constraint, 119 W
Wealth, 117, 132
Wealth, comprehensive, xi, 3, 9
Welfare state, 10
U Wellbeing, 3
Utility function, 160 Wellbeing frontier, 3, 28, 222
Wellbeing function, xi, 32, 128
Wellbeing, intergenerational, xi
V Wellbeing state, 10
Viability domain, 163, 165 Work-life balance, 119

Intergenerational Wellbeing and Public Policy: Girol Karacaoglu Jacek B. Krawczyk Anita King

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Intergenerational Wellbeing and Public Policy: Girol Karacaoglu Jacek B. Krawczyk Anita King

Uploaded by

Copyright:

Available Formats

Girol Karacaoglu · Jacek B.

ISBN 978-981-13-6103-6 ISBN 978-981-13-6104-3 (eBook)

Jacek B. Krawczyk dedicates his contribution

Anita King dedicates her contribution to

Philosophical and Institutional Foundations

be designed and implemented to achieve desired outcomes (especially social and

Where appropriate and relevant, the substantive chapters include references to

Organisation of the Book

of the main motivations behind this particular example is to demonstrate the

Wellington, New Zealand Girol Karacaoglu

2.5 Finance - Money, Credit, and Banking . . . . . . . . . . . . . . . . . . . . 58

4.3 Policy Simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

© The Author(s), under exclusive license to Springer Nature 3

• Appreciate the nature, and significance, of empirical evidence in informing the

1.2 Justification for Public Policy

1.3 Purpose of Good Public Policy

economies is influenced by viable policies.

Fig. 1.1 OECD BLI

and strong sustainability.

to as “genuine saving” (or “comprehensive investment”) (see Hamilton and Hepburn

1.4 Domain of Good Public Policy

investment approach to public policy”.

13 An equivalent classification, using different terminology, distinguishes between the mid-20th

Fig. 1.2 Wellbeing frontier

Fig. 1.3 Ecosystems

1.5 Policy Instruments

In a complex and evolving world dominated by “radical uncertainty”, and satisfic-

1.6 Delivery Mechanisms

circumstances, the primary focus of policy shifts to enhancing resilience – which, at

Fig. 1.4 Resilience

desirable aggregate social outcomes. Solutions are developed by communities, not

1.7 Time Consistency

How do we ensure time-consistency (i.e., ongoing alignment of public policy with

Fig. 1.5 Wellbeing triangle

1.8 Measuring Progress on Wellbeing

are available for all OECD countries.

Fig. 1.6 New Zealand’s wellbeing performance

Fig. 1.7 New Zealand’s wellbeing progress

1.9 Measuring the Quality of Public Policy

Fig. 1.8 Wellbeing triangles

1.10 Empirical Evidence

This subsection is substantially based on the material provided by Smith (2015).

1.11 Methodology and Structure of Presentation

1.11.1 Why Mathematical Models

options. Often, especially in a world of uncertainty and complexity, it is wise to

1.11.2 What Types of Models

22 See Roos (2015) for an example of this literature.

1. Assess the following statement: “Economics is a value-free social science.”

Suggestions for Further Reading

Abstract This chapter presents a formal “stylised model” in an attempt to pull

Fig. 2.1 Stylised model

2.2 Consumption, Wellbeing, and Wealth

2.2.1 Generic Model

2.2.1.1 Motivation and Microfoundations

The generic model is grounded in a framework for thinking about wellbeing-

Definition 2.2 Comprehensive wealth is defined by the stocks of various types of

The source of comprehensive consumption is comprehensive wealth. Some capital

2.2.1.2 Mathematical Model

We now present the skeletons of a mathematical version of the narrative provided in

Time is continuous and is denoted by s and t (0 ≤ t ≤ s ≤ ∞). The horizon is

where WA (t) denotes intergenerational wellbeing at t, which depends on C(·) – a

In expression (2.2), W (t) depends directly on t, to reflect the potential impact of

λT (t) ≡ ∂ W/∂t (2.5)