N O .

3 0

State-Society Synergy for Accountability
Lessons for the World Bank






N O .

3 0

State-Society Synergy for Accountability
Lessons for the World Bank


Washington, D.C.

Copyright © 2004 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing: April 2004
printed on recycled paper

1 2 3 4 06 05 04
World Bank Working Papers are published to communicate the results of the Bank's work to the development community with the least possible delay. The manuscript of this paper therefore has not been prepared in accordance with the procedures appropriate to formally-edited texts, and the World Bank accepts no responsibility for errors. Some sources cited in this paper may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank cannot guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply on the part of the World Bank any judgment of the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. The World Bank encourages dissemination of its work and normally will grant permission for use. Permission to photocopy items for internal or personal use, for the internal or personal use of specific clients, or for educational classroom use, is granted by the World Bank, provided that the appropriate fee is paid. Please contact the Copyright Clearance Center before photocopying items. Copyright Clearance Center, Inc. 222 Rosewood Drive Danvers, MA 01923, U.S.A. Tel: 978-750-8400 • Fax: 978-750-4470. For permission to reprint individual articles or chapters, please fax your request with complete information to the Republication Department, Copyright Clearance Center, fax 978-750-4470. All other queries on rights and licenses should be addressed to the World Bank at the address above, or faxed to 202-522-2422. ISBN: 0-8213-5831-6 eISBN: 0-8213-5832-4 ISSN: 1726-5878 This paper was written by John Ackerman, PhD candidate in Sociology at the University of California at Santa Cruz. It was commissioned, developed, and produced by the Civil Society Team and the Public Sector Group for the Latin American and Caribbean Region at the World Bank. Cover photo by Curt Carnemark

Library of Congress Cataloging-in-Publication Data
State-society synergy for accountability: lessons for the World Bank. p. cm.—(World Bank working paper; no. 30) Includes bibliographical references. ISBN 0-8213-5831-6 1. Developing countries—Economic policy—Citizen participation. 2. Economic development—Social aspects—Developing countries. 3. Political ethics—Developing countries. 4. Responsibility—Developing countries. 5. Developoing countries—Politics and government—21st century. 6. Civil society—Developing countries. 7. World Bank. I. World Bank. II. Series. HC59.7.S7587 2004 352.3'5—dc22 2004042227



v 1 3 7 13 14 17 23 25 28 31 33 37 43 45 49 51

Acknowledgments Executive Summary 1. Economic Development and Good Government 2. Accountability and Civil Society 3. Case Studies Case #1: Participatory Budgeting in Porto Alegre, Brazil Case #2: Mexico’s Federal Electoral Institute Case #3: Police and School Reform in Chicago Case #4: Decentralization and Rural Development in Mexico Case #5: Toxics Release Inventory in the United States Case #6: Women’s Police Stations in Brazil Case #7: Grass-Roots Anti-Corruption Initiatives in India 4. Lessons for the World Bank 5. Conclusion Annexes A. Summary of Case Studies B. Summary of Lessons for World Bank Staff Bibliography

Table 1: Four Dimensions of State Capacity 4

Figure 1: Accountability and Civil Society 11




his report benefited from comments by Jonathan Fox, David de Ferranti, Katherine Bain, Linn Hammergren, Patti Petesch, Susan Rose-Ackerman, Catalina Smulovitz, Michael Walton, Ernesto May, and the members of the LCR Public Sector Group.




he contemporary era of globalization and market liberalization by no means implies the end of the state. The institutional, technical, administrative and political capacities of the state are now as important as they ever have been. Good government continues to be an absolutely necessary prerequisite for successful economic growth in the developing world. One of the most effective ways to strengthen the basic capacities of the state is through the construction of an honest, efficient, and effective (or, in other words, accountable) bureaucratic apparatus. This task should be at the top of the agenda for almost all states. How can this be done? There are two general strategies for the achievement of this task. One option is to strengthen the “command-and-control” elements of government. The alternative is for governments to open up their bureaucracies to pressure “from below.” This involves constructing a healthy relationship between state and society so that social actors and individual citizens are empowered to oblige the government to uphold the rule of law and fulfill its promises. Although the top-down approach is clearly important and necessary in most countries of the developing world, especially given the chronic weakness of their bureaucratic apparatuses, the focus of this paper is on the latter approach. Specifically, how can the relationship between state and society be transformed from a process of particularistic demands, concessions and manipulations to a healthy engagement that produces a solid bureaucratic apparatus and policy outcomes that are in the interest of the public as a whole? This is the central question that guides the study below. The paper first surveys the literature on accountability and establishes a categorization of the different ways by which civil society can interact with the state in order to improve accountability. It then explores in detail seven case studies of successful experiences of “state-society synergy for accountability.” The studies draw from a wide range of different contexts (Brazil, India, Mexico, the United States) and from a variety of different areas of government activity (corruption control, environmental regulation, poverty reduction, election monitoring, infrastructure provision, school reform, police reform).




The paper concludes with a series of conceptual and practical lessons for World Bank staff on how best to initiate, design, and implement successful pro-accountability mechanisms grounded in state-society synergy. Some of the most important lessons include the need to fully institutionalize participative mechanisms, to involve societal actors from the very beginning of the design stage of the process, to open up participation to a wide diversity of social and political actors, and to complement decentralization with centralized supervision. In the end, this paper argues that marketization is not the only way to tap into the energy of society for the improvement of governance. Instead of sending sections of the state off to society, it is often even more fruitful to invite society into the inner chambers of the state.



The agonies of collapsed states such as Liberia and Somalia demonstrate all too clearly the consequences of statelessness. Good government is not a luxury but a vital necessity, without which there can be no development, economic or social. —Anthony Chhibber, as cited in Wallis and Dollery (2001)

capable, efficient and effective state apparatus is necessary to maintain the political stability, competitive markets, social policies, rule of law, and infrastructure investments required for vibrant economic growth in the developing world. Contrary to popular opinion, this is particularly the case under conditions of market liberalization and globalization. As many authors have argued in recent years (for example, Haggard and Kaufman, 1995; Evans, 1995; Grindle, 1996; World Bank, 1997; Vellinga, 1998), it is a mistake to equate economic success in the globalized world with a simple reduction in the size or the range of the state’s activities. To the contrary, global production chains and rapid communications demand a qualitative transformation and restructuring of the role of the state. Indeed, in recent years, most states in the developing world have even been under pressure to increase their capacity. Witness, for instance, the intense international pressure on otherwise market-friendly presidents like Mexico’s Vicente Fox to increase government revenues through tax hikes. Loans need to be paid back; basic services must be provided to the poorest citizens; workers need to be helped in their transition to a more urban, industrialized, and market-based economy; strategic investments in education and infrastructure have to be made; markets must be regulated; and peace and stability need to be maintained. As Merilee Grindle (1996) has argued, in order to carry out these and other tasks the state must strengthen its capacity in at least four different areas (see Table 1). First, its institutional capacity must be assured through the imposition of “authoritative and effective ‘rules of the game’ to regulate economic and political interactions.”





Dimension Institutional Capacity Definition Authoritative and effective “rules of the game” to regulate economic and political interactions. Ability to assert the primacy of national policies, legal conventions, and norms of social and political behavior over those of other groupings. The ability to set and manage effective macro-economic policies. A cadre of well-trained economic analysts and managers. Well-staffed and appropriately placed units for policy analysis. Important role for technical input and information in decision making. Effective administration of basic physical and social infrastructure. Ability to perform basic administrative functions essential to economic development and social welfare. Effective and legitimate channels for societal demand making, representation, and conflict resolution. Responsive political leaders and administrators. Societal participation in decision-making.

Technical Capacity

Administrative Capacity

Political Capacity

Source: Adapted from Grindle 1996.

This means that the state ought to have the “ability to assert the primacy of national politics, legal conventions, and norms of social and political behavior over those of other groupings.” Second, the state’s technical capacity has to be consolidated. This implies “the ability to set and manage effective macroeconomic policies,” which in turn requires “a cadre of well-trained economic analysts and managers [and] well-staffed and appropriately placed units for policy analysis.” Third, states need to strengthen their administrative capacity, which involves the “ability to perform basic administrative functions essential to economic development and social welfare.” Finally, states in the developing world have to develop a full political capacity, which requires the construction of “effective and legitimate channels for societal demand making, representation and conflict resolution.” The contemporary era of globalization and market liberalization by no means implies the end of the state. The institutional, technical, administrative and political capacities of the state are now as important as they have ever been. Good government continues to be an absolutely necessary prerequisite for successful economic growth in the developing world. Looking at the Latin American and Caribbean Region (LCR), there are three fundamental threats to the construction of good societal participation in the region, namely corruption, clientelism, and capture. All three of these phenomena refer to the use of public office for private gain, and their impact goes far beyond the simple diversion of funds. Corruption, in addition to directly enriching individual bureaucrats, distorts markets and hampers service delivery (RoseAckerman, 1999). Clientelism, in addition to unfairly channeling public resources to specific client groups, alters the dynamics of political competition and leads to the ineffective provision of public services (Fox, 1994). Capture, in addition to providing rents to specific economic actors, also greatly alters markets and worsens the position of consumers, workers, and the environment vis á vis corporations (Stigler, 1971). In the end, only by assuring the public interest character of the state will good governance and the economic growth that accompanies it be possible in LCR. The poor are often hard hit by such “particularistic” uses of the state. This is because the poor usually have few resources available to play such particularistic games. Societal participation




is therefore necessary not only for economic growth in general, but also for combating poverty in particular. As the case studies will further illustrate, the relationship between societal participation and development effectiveness stems from numerous causal links, including: 1. Participatory planning processes promote decisionmaking based on needs identified by potential beneficiaries, thus yielding a better targeting of resources; 2. Social accountability provides greater oversight of public resource management to discourage corruption, clientelism, and capture; and, 3. Participatory monitoring and evaluation gauges client opinion and satisfaction to advise in the improvement of the quality of services. One of the most effective ways to strengthen the four basic capacities of the state and to combat corruption, clientelism and capture is through the construction of an honest, efficient and effective (or, in other words, accountable) bureaucratic apparatus. This task should be at the top of the agenda for almost all states in the developing world. How can this be done? There are two general strategies for the achievement of this task. On the one hand, governments can strengthen the “command-and-control” elements of public administration. This involves forcing bureaucrats into line “from above” by more rigorously enforcing performance targets and by making them more vulnerable to punishment by superiors. On the other hand, governments can open up their bureaucracies to pressure “from below.” This involves constructing a healthy relationship between state and society so that social actors and individual citizens are empowered to oblige the government to uphold the rule of law and fulfill its promises. This second strategy constitutes one of the central thrusts of what has been called the “New Public Management”1 and in general involves what McCubbins and Schwartz (1984) have called the “fire alarm” approach as opposed to the “police patrol” approach to assuring the accountability of public servants. Although the top-down approach is clearly important and necessary in many countries of the developing world, especially given the chronic weakness of their bureaucratic apparatuses, the focus of this paper is on the latter approach. Specifically, how can the relationship between state and society be transformed from a process of particularistic demands, concessions, and manipulations to a healthy engagement that produces a solid bureaucratic apparatus and policy outcomes that are in the interest of the public as a whole? This is the central question that guides the discussion below. Chapter 2 provides a general overview of the concept of accountability. It begins with a review of some of the most important literature on this topic. Then it offers a categorization of the different ways in which civil society can interact with the state in order to improve government accountability. Chapter 3 includes various case studies of successful examples of what is referred to as “statesociety synergy for accountability.” The emphasis is on successful cases because, as Judith Tendler (1997) has pointed out, “the mainstream donor community’s advice about public-sector reform arises from a literature that looked mainly at poor performance… This means that countries and the experts that advise them have few models of good government that are grounded in these countries’ own experiences.” Development professionals are acutely aware of the ways that governments fail. There is a desperate need for sustained study of successful government innovations in order to inspire and direct positive action. This study offers possible strategies for reflection and discussion by World Bank staff. These case studies present just a sampling of the innovative work currently being conducted in this sector. In addition, it should be noted that the evaluation
1. There is, of course, also a “top down” version of the New Public Management which involves the use of strict performance contracts with each segment of government, as in New Zealand (Wallis and Dollery, 2001). Here I refer to the “bottom up” version of the New Public Management which is intimately linked to the empowerment of civil society (Behn, 2001).

of the impact of the relationship between state and civil society on poverty reduction and the public policy process remains a work in progress. Each of the cases presented offers mechanisms that demonstrably increase civic engagement. Each anticipates an outcome of improved public sector performance, but evaluations identifying precise determinants and relationships of causality on these types of initiative are still few and far between.2 As the forms of social accountability are varied, it is also possible that their impact is not uniform. Further review may find that “accountability” as active participation in the policy process has different consequences for policy quality than does accountability as social monitoring and control. Participation as voice may also differ from participation in the decision-making process itself. Nonetheless, several of the cases presented suggest positive results. Further investigation is required to provide additional evidence on if and how increased civic involvement can improve public policy performance and poverty reduction. The case studies draw from a wide range of different contexts (Brazil, India, Mexico, U.S.) and from a variety of different areas of government activity (corruption control, environmental regulation, poverty reduction, election monitoring, infrastructure provision, school reform, police reform). The diversity of cases is important for two reasons. First, it allows the reader to gain a full perspective on the many different ways in which states and societies can work together to solidify accountability. Second, it lends scientific weight to the conclusions. When the same fundamental issues come up repeatedly throughout a wide diversity of cases, it is much more likely that the issues are fundamental to the phenomenon itself, and not due to specific contextual factors. There are, of course, some inherent limitations to research based on case studies. For instance, small data sets prohibit the use of the most sophisticated statistical methods. In addition, there is always the possibility that the cases selected are unique, thereby limiting the application of the lessons learned from them. The conclusions that arise from case studies should subsequently be put to the test using quantitative methods. Nevertheless, the case study method also has many unique benefits. The close exploration of particular cases is the only way to discover the specific mechanisms that lead to success or failure. In addition, only by understanding the details of implementation can the researcher see whether and how the discourse of participation actually leads to the implementation of participatory schema on the ground. Studies based on large data sets necessarily sacrifice much of the useful detail embedded in case studies. Following the case studies, in Chapter 4 some overall lessons for World Bank staff that arise out of this paper are outlined. This chapter offers both conceptual and practical lessons on how best to initiate, design, and implement successful pro-accountability structures grounded in statesociety synergy. Finally, the paper concludes with a brief overview of the central points of the paper and some suggestions with regard to directions for future research.

2. Both the Latin America and Caribbean Region (LCR) and the Poverty Reduction and Economic Management (PREM) anchor of the World Bank are at present engaged in a quantitative evaluation exercise precisely to provide this type of additional insight on the impact of empowerment type initiatives on development effectiveness and other outcomes.



In both the North and South, shortcomings in conventional accountability systems—secrecy in auditing, ineffective policy reviews in legislatures, the electorate’s difficulty in sending strong signals to decision makers between elections, excessive delays in courts and inadequate sanctions for failure to apply administrative rules or respect standards—have created pressures for better channels for vertical information flows and stronger accountability relationships between state agents and citizens —Goetz and Jenkins (2001)


ood government does not emerge spontaneously or naturally out of the good hearts of individual bureaucrats and politicians. It is the result of a tough, and often conflict-ridden, process of institutional design. The principle element that assures good government is the accountability of public officials. This involves both answerability, or “the obligation of public officials to inform about and to explain what they are doing” and enforcement, or “the capacity of accounting agencies to impose sanctions on powerholders who have violated their public duties”(Schedler, 1999a). Although some individual officials may never need institutional structures to assure their commitment to the public good, most do need it at least some of the time. The only way to guarantee good government is by institutionalizing a powerful accountability structure that holds every public official responsible for his/her actions as a public servant. Although the term accountability is often used as if it referred only to the prevention of illegal or corrupt activity by public officials, it should also include the design and implementation of government programs within the general understanding of basic “public duties” for which public servants are responsible. For instance, according to Samuel Paul (1992) “accountability means holding individuals and organizations responsible for performance measured as objectively as possible.” This means that in addition to the honest handling of public funds, bureaucrats should also be responsible both for the fulfillment of predetermined policy goals and for the responsiveness of policies to the specific needs of the public, and particularly to the needs of those who most depend on government services: the poor. Therefore, in addition to answerability and




enforcement, should perhaps add receptiveness, or the capacity of officials to take into account the knowledge and opinions of citizens, as a third central element of accountability.

Categories of Accountability
Robert Behn (2001) follows in this line of thought by defining three broad categories of accountability. First, there is accountability for finances that “focuses on financial accounting—on how the books are kept and how the money is spent.” Second, there is accountability for fairness where “to ensure that government and its officials pay careful attention to ethical standards…we create rules to codify exactly what we mean, operationally, by fairness and equity. These rules create processes and procedures that, if followed, ensure that government has been equitable-that it has treated its citizens fairly.” Third, like Paul, there is accountability for performance that goes beyond prudent spending and fair treatment to involve the successful accomplishment of public purposes. Behn explains that while “accountability for finances and accountability for fairness reflect concerns for how government does what it does…we also care what government does— what it actually accomplishes.” For the author, the central question in this third category of accountability is “are the policies, programs, and activities of government producing the results that they were designed to produce?” For the sake of clarity and simplicity, in this paper a more simple two-sided dividing line will be applied between legal accountability and performance accountability. “Legal accountability” involves keeping public officials in check by making sure that they respect the legal order both in their administrative tasks and in their relationship to society at large. “Performance accountability” involves the successful implementation of policies designed to benefit the public in general, with a particular emphasis on policies that attend to the needs of the poor. The first type of accountability has to do with respect for the rule of law and preventing the abuse of public office. The second type of accountability involves questions of efficiency, effectiveness, and fairness.

Mechanisms of Accountability
What then are the most powerful and effective accountability mechanisms that help discipline and direct the actions of public officials? Free and fair elections are one of the most common and effective of such mechanisms. Through periodic elections, top government officials are forced to take into account the mandate of the electorate (receptiveness), inform the public about what they have done while in office (answerability), and then be judged by the people (enforcement). Under this arrangement, political leaders who work for the common good are supposed to be reelected, and leaders who use public office for particularistic ends are supposed to be removed from office. Proof of the effectiveness of this mechanism is the fact that the recent arrival of electoral democracy in most of LCR has indeed improved the accountability of public officials, leading to important advances in good government. Nevertheless, there are also unfortunately many examples of worsening accountability situations in Latin America’s fledgling democracies. Indeed, as Guillermo O’Donnell (1994) has written, many of the new democracies in LCR are best understood as “delegative democracies” whose actions are in essence unaccountable to anyone except for the president himself. There are both structural and contextual explanations for this failure of formal electoral democracy to bring about good governance in LCR. There are at least three different structural problems with elections as accountability mechanisms in general. First, and most obviously, elections only hold elected officials accountable. The vast majority of public officials are appointed bureaucrats who are not directly accountable to the public through the electoral process. Although elected politicians have strong incentives to keep bureaucrats in line in order to protect their own reputations, the ease with which blame can be shifted to the bureaucrat himself/herself or to other elected officials greatly dilutes the indirect accountability effect of elections on bureaucrats. Second, because elections only occur once every few years and force an incredible diversity of opinions and evaluations together into a single ballot, it is virtually impossible for elections to give clear accountability signals to individual office holders (Manin, Przeworski and




Stokes, 1999). Indeed, as authors like James Fearon (1999) have argued, citizens themselves often consider their vote to be more about “picking the best person for the job” than about sanctioning or holding accountable past office holders. In this case, the accountability signal is diluted even further. Third, even if the accountability signal were somehow clearly discernible, the fact that most politicians are elected by only a small portion of the population often forces politicians to favor patronage, or corruption, over initiatives that would bring long-term benefit to the public as a whole (Varshney, 1999). If the weaknesses of existing democratic institutions in the LCR are added to these more general structural problems with elections, the situation gets even worse. The gap between political and civil society, the clientelistic nature of many political parties, the excess private funding for candidates, and the lack of public information about the general workings of government and even less information about the specific behavior of individual office holders in the LCR weakens even further the effectiveness of elections as mechanisms of sanction and control in the region. In short, for both structural and contextual reasons, free and fair elections are simply not enough to guarantee good government in the LCR. Consequently, vertical accountability mechanisms, like elections, that require government officials to appeal “downwards” to the people at large need to be complemented by horizontal accountability mechanisms that require public officials and agencies to report “sideways” to other officials and agencies within the state itself. As O’Donnell (1999) has written, “horizontal accountability” is “the existence of state agencies that are legally enabled and empowered, and factually willing and able, to take actions that span from routine oversight to criminal sanctions or impeachment in relation to actions or omissions by other agents or agencies of the state that may be qualified as unlawful.” Examples of horizontal accountability mechanisms include institutions like human rights ombudsman, independent electoral institutes, corruption control agencies, legislative investigative commissions and administrative courts. All of these are public institutions that are specifically designed to evaluate, control and direct the behavior of other government officials. Nevertheless, as with elections, the record of such agents of horizontal accountability has been quite irregular throughout the LCR. Although some institutions like Brazil’s Public Prosecutor, Peru’s Ombudsman and Mexico’s Federal Electoral Institute (IFE) stand out as examples of innovative successes in horizontal accountability mechanisms (Sadek and Batista Cavalcanti, 2000; Santistevan, 2000; Woldenberg, 2002), most other institutions have been much less successful in carrying out their tasks. Once again we are faced with the puzzle of uneven development of political institutions in the region. What accounts for the wide variations in both vertical and horizontal accountability enforcement between different countries in the LCR and between different institutions within the same country?

The Participation of Civil Society
This paper highlights the importance of one crucial explanatory factor: the direct involvement of civil society.3 The participation of community groups, NGOs and citizens in general is one of the most important factors that explains successful accountability arrangements. The participation of civil society can reinforce structures of accountability in three different ways. First, direct monitoring and pressure from civil society actors as well as popular votes on particular issues or policies can complement elections as additional elements of vertical accountability. This increases the frequency and the clarity of the accountability signals that citizens send to public officials and thereby improves their vertical accountability towards the public. Examples of such participation include everything from surveys that monitor the effectiveness of public services, to media exposés of bureaucratic wrongdoing, to the organization of plebiscites and referendums.
3. The term “civil society” refers to both organized and unorganized citizens acting independently from government, political parties and the profit motive in order to transform society and governance. This includes religious and professional organizations, labor unions, grassroots organizations, and NGOs, but also reaches beyond these groups to include the participation of citizens outside of formal organizations.



For the purposes of this paper, this form of civil society participation is called direct vertical accountability in order to distinguish it from electoral vertical accountability, which is grounded in the formal, periodic election of members of the three branches of government. Second, in addition to the direct effects of external pressure “from below,” the action of civil society can strengthen the effectiveness of existing horizontal accountability mechanisms within the state itself. As Jonathan Fox (2000) has written,
civil society demands for state accountability matter most when they empower the state’s own checks and balances. By exposing abuses of power, raising standards and public expectations of state performance, and bringing political pressure to bear, they can encourage oversight institutions to act, as well as to target and weaken entrenched opponents to accountability.

Here the idea is not the direct action against public wrongdoing by society as with vertical accountability, but the more indirect mechanism of pressuring existing agencies to do their jobs effectively. For example, in their analysis of the social response to two extra-judicial killings in Argentina, Catalina Smulovitz and Enrique Peruzzotti (2000a, 2000b) documented how the combination of mobilization, legal action and media exposure can effectively guarantee that the judicial system operates impartially, even when the perpetrators are well connected or even part of the government apparatus itself. Another example of this type of civil society participation would be the human rights ombudsmen, a clearly “horizontal” institution whose actions are consistently strengthened and often directed by social demands. This form of participation is referred to as society-driven horizontal accountability. Third, in addition to pressuring from the outside and reinforcing existing control mechanisms within the state, civil society actors can participate directly in the government’s own institutions of horizontal accountability. Such participation breaks down the division between vertical and horizontal accountability mechanisms, since it involves the participation of “vertical” actors in “horizontal” mechanisms. It is therefore a “hybrid form of accountability” that might best be called “diagonal accountability” (Goetz and Jenkins, 2001). For Anne Marie Goetz and Rob Jenkins, this form of civil society participation is special because it “represents a shift towards augmenting the limited effectiveness of civil society’s watchdog function by breaking the state’s monopoly over the responsibility for official executive oversight.” Examples of this form of participation would include citizen advisory boards that fulfill public functions like auditing government expenditures, supervising procurement, or monitoring elections. Following Goetz and Jenkins, this form of participation is referred to as diagonal accountability. Both state and society actors usually play crucial roles in bringing about these three forms of accountability. “Direct vertical” mechanisms like plebiscites and referendums require both the support of the legislature and a vibrant civil society that is willing to get out the vote. “Societydriven horizontal” mechanisms like human rights ombudsmen require both social pressure and enlightened bureaucratic action. “Diagonal” mechanisms also depend on the willingness of a government willing to open up its accounts to independent actors as well as the will of civil society participants to take the time to look over and evaluate these accounts. This is not to say that “state-society synergy” (Evans, 1996a, 1996b) always takes place. States often unilaterally organize the participation of civil society, and societies frequently unilaterally press their demands on government. Yet, the most productive results arise when both sides actively participate in opening spaces for civil society participation. Unilateral state action often ends in manipulation or “reverse vertical accountability” (Fox, 2000) in which citizens end up being accountable to government officials for their actions. At the same time unilateral social action often results in repression and violence by the state. The effectiveness of “state-society synergy” is by no means dependent on the existence of consensus, value sharing or even trust between state and civil society actors. Indeed, conflict and suspicion generate greater oversight, which often yield greater success in producing state-society synergies. As Catalina Smulovitz (2000) has pointed out, it is often the case that “the social trust









that results from value-sharing weakens citizens’ oversight and control capacities of what rulers do, and increases, in turn, the chances of opportunistic actions by one of them.” It is important not to fall prey to depoliticized or neutral ideas of civil society that see “cooperative” or “moderate” forms of social organization as the only ones that can positively influence the construction of accountability arrangements. Smulovitz has made this point very clear. “I am not arguing, as some have [for example, Almond and Verba, 1963; Weingast, 1997], that an autonomous civil society is important because citizens share values that sustain the benefits of self-restraint. I am arguing that an autonomous civil society is important because it implies the existence of multiple external eyes with interests in the enforcement of law and denunciation of non-obedience.” From the above discussion, the following diagram can be derived, (Figure 1) summarizing the theoretical framework:




his chapter examines several relatively successful attempts to construct “state-society synergy for accountability.” Each case study identifies the different forms that civil society participation takes (direct vertical, society-driven horizontal, and diagonal) as well as the types of accountability (legal and performance accountability) that are strengthened. Particular emphasis is placed on issues of institutional design to derive concrete lessons for reformers on what does and does not work. In addition, each case study explores the origins of the particular accountability arrangement/institutional design. Effective accountability mechanisms never simply arise out of enlightened thinking by individual politicians, but always also emerge out of complex social processes of conflict and negotiation. The cases are organized according to the level at which state actors have explicitly encouraged the participation of civil society in the structuring of accountability arrangements. The first case presents the example of participatory budgeting in Porto Alegre, where the local government has made an effort to include social actors fully in the process of policy design and monitoring. Then the case of the Federal Electoral Institute (IFE) in Mexico is discussed, where the federal government has successfully balanced the construction of a highly professional, powerful and autonomous government agency with the full involvement of societal actors in program direction, operation and monitoring. Following this, the case of police and school reform in Chicago is explained, where the government also actively encouraged direct citizen participation in government, but refrained from directly stimulating civil society organization and from giving legal status to some important elements of the participatory scheme. The fourth case study discusses decentralization and rural development in Mexico. Here the opening up of the state is less wholehearted than the previous three cases. Nevertheless, an alliance between reformist public officials and strong civil-society organizations managed to construct participatory accountability relationships despite the resistance of other sectors. Fifth, the case of the Toxics Release Inventory in the United States is presented, where the government almost accidentally ended up playing the role of external prompter, as opposed to direct organizer




and stimulator, of civil society participation for accountability. Sixth, the less successful case of women’s police stations in Brazil is discussed, where, after an initial stage of state-society synergy, the state ended up overly bureaucratizing and mainstreaming the original proposal. Finally, a pair of cases of “social auditing” from India demonstrate that state-society synergy is possible even under conditions of a recalcitrant state in a particularly difficult and “secret” area of government activity. These Indian cases show that although a progressive government or a significant group of “reformists” within government definitely helps the accountability process, constructive accountability relationships between state and society also can arise out of social mobilization alone.

Case #1: Participatory Budgeting in Porto Alegre, Brazil
The Porto Alegre city government represents one of the most effective strategies for state-society collaboration for accountability in the developing world. Since 1989, when the Worker’s Party (PT) first won the city government, Porto Alegre has placed spending decisions for over ten percent of its annual budget in the hands of the people. Every year, more than 14,000 citizens in this city of 1.3 million participate in neighborhood meetings as well as 16 regional and 5 thematic assemblies to establish priorities for government investment in infrastructure and basic social services. Each assembly elects two council members to serve on a citywide Council of Participatory Budgeting (COP), the organ responsible for putting together the final citywide budget plan. At each level of the process (neighborhood, district, citywide) decisions are made through intense negotiation and the use of sophisticated weighted voting systems designed to assure a fair distribution of resources, with an emphasis on helping the most needy areas of the city. At the end of the process, the proposed budget is then submitted to the local legislature for final approval and promulgation. During the following year, the regional and thematic assemblies, council members and neighborhood groups evaluate the previous year’s negotiation process and monitor the implementation process of the previous year’s budget. In Porto Alegre (and many other cities governed by the PT in Brazil since 1989, like Belo Horizonte and Betim [Nylen, 2002]) the PT has created an alternative mechanism of accountability through direct and constant citizen participation in the fundamental tasks of governing. The participatory process is responsible for establishing agreements between citizens and representing the interests of the people at large with respect to specific spending and implementation decisions.

Diagonal Accountability in Action
The participatory budgeting process provides an excellent example of diagonal accountability. Ordinary citizens are involved directly in the planning and supervision of public spending, activities normally under the exclusive purview of public officials. This arrangement is clearly a step beyond both the traditional watchdog or society-driven horizontal role of civil society as well as protest or referendum based direct vertical roles for social actors. Instead of trying to influence policy from the outside, the citizens of Porto Alegre are invited inside the governmental apparatus itself, thus confusing the neat horizontal-vertical framework for understanding accountability mechanisms. This arrangement has had a clear impact both on questions of legal accountability and performance accountability in Porto Alegre. First, because it institutionalizes the existence of “multiple external eyes” (Smulovitz, 2000), it has drastically reduced the possibilities and incentives for corrupt behavior on behalf of bureaucrats, thereby strengthening legal accountability. Each neighborhood and region is clearly informed as to the exact amount of funds that will be invested in which products and services in its area. Even more importantly, because the citizens themselves participate in designing the budget, they feel that they have a personal stake in making sure the government complies with it. As Zander Navarro (1998) has written,
by introducing an unprecedented transparency in the formation, allocation and implementation of the municipal budget, “opening” it to the general scrutiny by the citizenry, the [PT government in




Porto Alegre has] dramatically reduced the room for petty and backstage arrangements linking civil servants to private interests. Not to mention major illegal and/or ethically illicit proposals, rendered impossible when all acts and intentions are so loudly publicized.

Second, the budgeting process improves performance accountability by reducing clientelism by opening alternative channels for the participation of civil society. The crucial element is the entirely open and public nature of the budget assemblies. Any adult can attend, speak and vote in the assemblies.
The success of the participatory budgeting (PB) as a participatory policy is connected to the fact that it offers an alternative to the so-called tradition of political mediators [or “clientelism”], a tradition in which politicians distribute material goods as a favor. The relevant phenomenon, in the case of the PB, is the capacity it has of transferring from the political mediators to the population the decision of the distribution of material goods through the creation of a set of public elements: assemblies, lists of the previous access to goods, necessity criteria. (Avritzer, 2000)

Under a clientelist scheme, if and when citizens become dissatisfied with the way particular organizations are representing them in the assemblies, it is extremely easy to form a new group and thereby gain access to special organizational representation in the popular votes. This leads to easy “exit” options for members of clientelistic groups where “voice” is not an effective form of protest. Third, performance accountability also is improved by limiting the capture of state institutions by wealthy interests. Popular participation helps replace the power of money with the power of voice. The special design of Porto Alegre’s system reinforces this tendency even further. The algorithm used for determining budget priorities intentionally tilts investments towards poorer neighborhoods. Due to this built in pro-poor bias, the same need presented by two neighborhoods is much more likely to be implemented in the poorer one than the wealthier one. As a result,
of the hundreds of projects approved, investment in the poorer residential districts of the city has exceeded investment in wealthier areas … Each year, the majority of the 20 to 25 kilometers of new pavement has gone to the city’s poorer peripheries. Today, 98 percent of all residencies in the city have running water, up from 75 percent in 1988; sewage coverage has risen to 98 percent from 46 percent; in the years between 1992-95, the housing department (DEMHAB) offered housing assistance to 28,862 families, against 1,714 for the comparable period of 1986-1988; and the number of functioning public municipal schools today is 86, against 29 in 1988. (Baiocchi, 2001)

Origins of the Participatory Budgeting Process
The origins of this successful pro-accountability arrangement can be found in both society and the state. There are three different means by which social actors and practices shaped the PB process. First, the idea of instituting a participatory budget had its origins within civil society. It was the Union of Residents’ Associations of Porto Alegre (UAMPA) that first advocated the introduction of such a mechanism in 1986 (Avritzer, 2000). Second, an independently dynamic civil society has been willing to occupy the spaces that have been opened “from above” by the city government. Avritzer establishes that the expression “participatory budget” did not exist in the PT’s electoral platform for city government in 1988. It only spoke of developing “popular councils.” The design of today’s PB arrangement only arose after a period of intense negotiation and participation both within the new government and between the new government, civil society groups, and involved citizens. Third, the particular institutional form developed by the Porto Alegre government was largely modeled on already existing practices of deliberation and negotiation. “It is also due to the adaptation of the institutional form of participation to the pre-existing praxes between community



players that the participation of these players is so big. In other words, from the institutional point of view it seems to be of fundamental importance that the proposals for public participation operate according to pre-existing praxes” (Avritzer, 2000). State actors have also played an important role in originating and maintaining the PB process. The simultaneity of PB in Porto Alegre with the rise of the PT to power is not a coincidence. The PB has been a central element in the PT’s broader project to use state power to bring about a long-term transformation in the social bases of power and authority. In addition, the particular institutional design chosen by the state reformers went far beyond incorporating previously existing organized civil society to actively promote the development of new social actors. As Abers (1998) has written, “Porto Alegre provides us with an unusual example in which state reformers took on a more proactive role, not only providing an enabling environment, in which the formation of civic groups was explicitly promoted, but also working directly with local communities to help them organize.”

Lessons Learned
This case provides many lessons for future pro-accountability state reformers. First, poor, uneducated people can and do effectively participate in the activities of governance. Baiocchi (2001) documents that 60 percent of the participants in the PB process in Porto Alegre have no high school education and 30 percent earn less than two times the minimum wage (US$62 per month, Nov. 1999), an income that leaves them below the poverty line. Avritzer’s (2000) survey shows that 46 percent of the participants have not completed elementary school. Abers (1998) reveals that while in 1991 29 percent of Porto Alegre’s residents earned three times the minimum wage or less, 45 percent of the budget participants fit this profile. This last statistic is particularly interesting because it shows that the underprivileged not only actively participate but that they even participate more, relative to their size in the population, than better off groups. In short, there is no reason to hesitate to design participative accountability mechanisms in poor, or even destitute, regions. This brings us to the second lesson. It appears that governments can only get back as much as they put in to efforts to activate civil society participation for accountability. Authentic opening and active involvement are crucial for success. In Porto Alegre, citizens are taken out of their usual role as only “advisors” or information providers to government projects and thrust directly into the decision making process itself. Public opinion surveys and open forums for the expression of opinions are clearly useful, but they do not create the active engagement and sense of responsibility as do truly participatory processes like the PB process in Porto Alegre. In addition, there is an important difference between simply opening spaces for the participation of previously organized groups and communities and getting actively involved in stimulating the participation of citizens in general. A central element of the Porto Alegre experience is that it makes participation open to the entire population and, even more importantly, actively encourages the participation of unorganized citizens through the use of government employed community organizers. Without such involvement, “participation” schemes can easily end up only strengthening previously existing clientelistic networks and unequal intra-community power relations. Third, governments need to take civil society into account in the design of the participative mechanisms themselves. This is an absolutely crucial point that is all too frequently ignored. Porto Alegre is special not only because citizens have an unprecedented level of involvement in the PB process itself, but also because they were directly involved in the design of the process itself. The PB did not simply arise out of the minds of enlightened bureaucrats. It originated in civil society, was pushed forward by social actors and was ultimately modeled on previously existing practices in civil society by a new government that itself consisted mostly of individuals who had made their careers as community and social activists. Participatory mechanisms usually sustain the mark of their birth. The best way to achieve full participation as an end point is to start off with it at the beginning.




Fourth, it is important to increase the benefits and to reduce the costs of participation. On the one hand, one of the crucial elements of success of PB in Porto Alegre is that the expansion of participation coincided with a substantial increase in tax revenues4 that enhanced the government’s capacity to fulfill community demands. Participation became obviously worthwhile because it brought real benefits to the communities involved. On the other hand, the government also made great efforts to reduce the costs of participation. The government sent officials to the neighborhood and district meetings instead of requiring residents to travel to city hall. In addition, community organizers were sent out into the communities and District Administrative Centers were set up. This brings us to our fifth and final point that has to do with the issue of decentralization. According to Fung and Olin Wright (2001), the Porto Alegre experience is an excellent example of how a healthy balance can be struck between “devolution” and “centralized supervision and coordination”. Although devolution and decentralization are important because they bring government closer to the people, if carried out blindly, they tend to reinforce inequalities both within the newly “autonomous” local units as well as between them. Local power holders are allowed to run free, and underprivileged localities are abandoned to their own devices. Decentralization is only productive if the center remains responsible for the supervision and coordination of activities in the local units. This tension is creatively and effectively resolved in the case of Porto Alegre by the involvement of the central city government at all levels of participation and the careful process of participatory preference aggregation and negotiation that exists all the way from the neighborhoods to the citywide level.

Case #2: Mexico’s Federal Electoral Institute
Mexico’s Federal Electoral Institute (IFE) stands out as one of the most successful efforts in the LCR to construct an autonomous agency for public accountability. It sets an important precedent in a region where most such agencies, from independent corruption bureaus to human rights commissions and public prosecutors, tend to be heavily politicized, ineffective and lack the necessary autonomy.5 Mexico’s IFE is particularly successful because, like the participatory budgeting process in Porto Alegre, it breaks with the traditional division between horizontal and vertical accountability and institutionalizes the diagonal participation of civil society.

Combating Corruption, Clientelism and Capture
Electoral administration involves essentially the same problems of corruption, clientelism and capture as does the government provision of any basic good or service like public security, education, health care or infrastructure. For most of the 20th century the electoral process in Mexico has been riddled with corrupt practices. Public funds (particularly funds for “social development”) have been systematically diverted to finance the activities of the dominant Party of the Institutional Revolution (PRI). These funds were subsequently used both to win elections and to maintain a solid clientelistic network throughout the country. Such misuse of public funds led to a situation in which public monies were distributed almost exclusively based on political alliances and without consideration of need or development potential, a situation that has had a profoundly negative effect on economic growth. This occurred because the agencies in charge of running elections and monitoring the activities of political parties had become almost entirely captured by the interests of the dominant party itself.
4. “If 1988 is made equal to 100, in real terms the total participation of all municipal taxes increased to 307 and housing and land taxation alone increased to 285 in 1996” (Navarro, 1998). This increase in tax collection is at least partly due to the increased trust in the effectiveness of government that the PB process itself brought about. 5. The United States is no exception to this rule. The U.S. Federal Electoral Commission, for instance, is notorious for its ineffectiveness and inefficiency as well as for its profound level of politicization (see Laforge, 1996; Jackson, 1990).



These problems have been aggressively confronted in recent years with the construction of an autonomous electoral institute. The IFE’s principal activities include organizing federal elections, distributing public funds to the political parties, monitoring the use of both public and private funds by the parties, checking for media bias in the coverage of political campaigns, putting together and cleaning up the official electoral roll, and running public education campaigns (IFE, 2000a). What makes the IFE special is the way in which its core administrative structure of 2,500 civil servants is simultaneously accountable to and holds accountable an extensive web of government and societal actors above, below, within, and on both sides of it. The IFE institutionalizes multiple external eyes in a broad diversity of forms by allowing society to monitor government, government to monitor society, government to monitor government, and society to monitor society. Its accountability structure includes five distinct elements. First, societal actors are directly involved in the direction and monitoring of the Institute’s activities. Within the IFE there is an independent, “citizen-run” General Council that serves as both a special horizontal accountability agency for electoral affairs and as the IFE’s principal directive body. The General Council is a permanent body made up of nine members who are responsible for monitoring the operations of the administrative staff of the IFE, resolving disputes concerning the interpretation and application of the Federal Electoral Code (COFIPE), and giving vision and direction to the institute as a whole. The Council is made up of council members who are each appointed by two-thirds of the lower house of Congress to serve seven-year terms. The two-thirds rule is significant because no councilor can be elected unless he or she is backed by a multiparty consensus. This excludes council members who are closely affiliated with a single party and assures the professional “citizen” nature of the Council. The General Council also seeks to maintain a close link to society. First of all, the meetings of the General Council are public. Although the population at large cannot attend or directly participate in the meetings, the minutes and decisions are widely publicized, reported on by the media, and are available via the Internet. This is very different from the modus operandi of most government agencies both inside and outside of the LCR and permits civil society to fulfill the crucial task of second order monitoring by overseeing the monitoring activities of their representatives on the General Council. In addition, one representative from each registered political party and one Legislative Councilor from each political party that has representation in Congress sits on the General Council. These party representatives can fully participate in the discussions of the General Council and have access to all of the same information as the council members but do not have the power to vote on initiatives or decisions. It is said that they have the power of voice but not of the vote. The presence of all of the political parties assures the non-partisan behavior of the council members and means that any misstep by the IFE will be immediately publicized. Meanwhile, the politicians’ lack of full authority guarantees the citizen nature of the council and legitimates its claim to be above the interests of individual political parties. In general, all of the above elements of institutional design are intended to guarantee that the General Council represents the interests of society as a whole against the dominant “partyocracy” (Cárdenas Gracia, 2000). Second, in addition to empowering societal actors in the direction and monitoring of the activities of the institute, the IFE’s accountability structure also places it under the monitoring eye of the three branches of government. All of the decisions of the General Council can be appealed to and reversed by a special Federal Electoral Tribunal (TRIFE) within the judicial branch. Also, the IFE’s budget must be included as a part of the President’s budget initiative and approved by Congress every year. In addition, all of the activities of the IFE are monitored by the executive branch through the Secretary of Finance and the Secretary of the Comptroller and Administrative Reform (SECODAM) and by the legislature through the Superior Federal Auditor (ASF). Third, the IFE has been a leader in civil service reform. In Mexico, the Foreign Service, teachers in the public school system, the Office of the Attorney for Agricultural Affairs, the National Statistics, Geography, and Information Institute (INEGI), the Judiciary, and the Tax Administrative System have all implemented civil service reforms. Nevertheless, as David Arellano




and Juan Pablo Guerrero (2000) have pointed out, “the most systematic, open and meritocratic system is undoubtedly that created in the independent agency that organizes elections, the IFE.” The IFE has had a strong Civil Service Statute in place since 1992, but the new 1999 statute definitively places the IFE at the vanguard of civil service reform in Mexico. The new statute clearly codifies each one of the five central elements of the civil service: hiring, training, evaluation, promotion, and sanctions (IFE, 2003). In addition, one of the IFE’s six Executive Directorates is exclusively dedicated to applying the civil service code. Finally, in order to prevent corruption the council members are paid a very high salary. The law mandates that the council members receive the same remuneration as Supreme Court judges (over $100,000 USD a year). Fourth, during its most important moment of service delivery, the organization of the federal elections, the IFE puts in place a structure of multiple surveillance by recruiting a huge army of citizen volunteers. During the months leading up to the 2000 elections the IFE trained over 800,000 volunteer citizens to run 113,423 polling sites (Woldenberg, 2001). Each polling site consists of seven members: a president, a secretary, two counters, and three general substitutes. These members are unpaid and are selected by a double lottery from the full electoral roll, in order to assure impartiality (IFE, 2000a). All of the participants receive two training courses that are designed and implemented by the IFE. In addition, the IFE trains both national and international observers in the basics of electoral law so that they can serve as additional eyes at the voting booths. Finally, each political party is permitted to send one representative to each voting booth on election day. Each one of the citizens present at a voting booth can file complaints concerning how the voting process was organized and how the votes were counted (IFE, 2000a). In total, between 10 and 15 citizens participate at each voting booth in the monitoring of the electoral process, meaning that in the year 2000 more than one million citizens were mobilized in order to assure the realization of free and fair elections. Fifth, in addition to (a) empowering societal actors in its leadership, its internal monitoring and its operations, (b) holding itself accountable to the three branches of government, and (c) institutionalizing accountability within its internal structure, the IFE also (d) carries out important horizontal accountability activities towards external actors. The IFE distributes and supervises the use of large quantities of funds by the political parties (US$377 million in the year 2000 alone; Woldenberg, 2001) as well as monitors the source, size and use of private donations (each party is allowed to collect individually up to 5 percent of the total public funding given to all of the parties). Every year each political party is required to report all of its private donations and all of its expenses to the General Council. The General Council is authorized to review this information, apply sanctions for the misuse of funds and audit the accounts of the political parties. Each year the IFE issues penalties for hundreds of thousands of pesos to the political parties (Lujambio, 2002). In addition, the IFE is also responsible for monitoring media coverage of election campaigns as well as the sale of political propaganda in the media. During the months leading up to the federal election, the IFE publishes reports evaluating bias in media coverage. It also assures that the prices charged for political advertising are not superior to those charged for commercial advertising and that each political party is offered the same price for equivalent time spots and has the same opportunity to occupy the most competitive spots (IFE, 2000a). Overall, the IFE has been remarkably successful. In a short ten years of life, it has successfully consolidated its internal structure, earned respect and recognition from all major political actors, gained widespread popular legitimacy, and perfected its operational effectiveness. It is the IFE that guaranteed the fairness of the groundbreaking 1997 and 2000 federal elections that finally inaugurated a new era towards a more democratic Mexico. The lack of significant post-electoral protests and mobilizations in the year 2000 was unprecedented for a presidential election in Mexico and attests to the success of the IFE in terms of both legal accountability and performance accountability. The Mexican political system has always been unique in the LCR for its highly institutionalized form of governance. Today, instead of standing out for its institutionalized authoritarianism,



Mexico distinguishes itself through its successful institutionalization of democratic accountability through the construction of the IFE.

For decades, electoral reform was one of the most important ways in which the authoritarian state-party regime of the PRI was able to accommodate opposition while simultaneously maintaining its control over the central levers of power. Since the founding of the PRI’s precursor, the Party of the National Revolution (PNR) in 1929, the electoral laws have been modified 22 times (Molinar Horcasitas, 1996). At times, reforms have been made in order to stimulate the participation of small, loyal opposition parties, as when proportional representation was introduced in the 1962-1963 reform and then expanded during subsequent reforms. Other times reforms have been designed in order to assure the dominance of the PRI in the face of important threats to its hegemony.6 The constitutional amendment that created the IFE in 1990 followed in this tradition of a double-edged electoral reform. In response to the crisis of political legitimacy that arose out of the fraud surrounding the 1988 presidential elections7 and the massive social mobilization demanding democratization that followed, the state-party regime had to demonstrate to both national and international civil society its seriousness about combating electoral fraud. Nevertheless, it was simultaneously interested in maintaining its power over the electoral process. As a result, the 1990 reform did not grant the IFE significant autonomy from the executive branch or the ruling party. The Secretary of Government (Secretario de Gobernación) was named the president of the institute, and the voting members included a disproportionate representation of council members whose appointments were closely controlled by or directly represented the PRI and the government (Molinar Horcasitas, 1996). In 1994, the political crisis created by the Zapatista uprising, the mass mobilization against the North American Free Trade Agreement (NAFTA) and for democracy, and the assassination of the PRI’s presidential candidate, led to a new reform that significantly changed the character of the IFE. The reform formally established the figure of “Citizen Councilor,” giving the Chamber of Deputies the right to independently appoint six of the eleven council members. In addition, the reform gave an official role to both national and international electoral observers in the electoral process and created the special prosecutor for electoral crimes (FEPADE) (Pozas Horcasitas, 1996). One of the most important influences on the 1994 electoral reform, as well as on the later 1996 reform, was the activism of non-profit electoral watchdog groups. The leading group during this period was Alianza Cívica. For the 1994 elections, this group mobilized over 12,000 national electoral observers and 400 international observers, carried out its own parallel “quick count” of the electoral results, published a report on bias in media coverage of the campaigns as well as a guide for electoral observers and a final evaluation of the election as a whole (Olvera, 2001). The 1996 reform finally allowed Mexico to begin to turn the corner from authoritarianism to democracy. The 1996 reform was the first one negotiated, designed and implemented by all of the important actors from all three of the main parties from the left, right and center. The reform finally put the IFE under the exclusive control of the Citizen Council members (removing the

6. An example is the notorious “governance clause” first established in the 1986 reform that required the party that won the majority of direct election seats to receive enough proportional seats in order to control over 50 percent of the Chamber of Deputies. This was then expanded in the 1990 reform by requiring that the dominant party only win 42 percent of the direct election seats in order to be assured of an absolute majority in the lower house. 7. The computer system that was counting the votes suddenly “crashed” on election night as the opposition candidate was taking the lead and then mysteriously showed the official party’s candidate as the victor with 51 percent of the votes a week later.




right to vote from the representatives from the executive, the legislature and the political parties), created a fully independent Federal Electoral Tribunal, vastly increased public funding for political parties, improved fairness in access to the media, controlled private campaign contributions, and took the crucial step of granting political independence and elections for the leadership of the Mexico City Government (Becerra, Salazar and Woldenberg, 2000). Nevertheless, the PRI’s continued control over the political system allowed it to build various important flaws into the structure of the IFE even in 1996. Regardless, the fact that there has not been a new electoral reform since 1996 is a testament to the great breakthrough this reform represented, and the legitimacy that it continues to enjoy through today. From this brief history, it is evident that both state and society actors have participated actively in the construction of the IFE. Social mobilization from below has consistently pushed for and stimulated electoral reform. Because the authoritarian regime needed to legitimize its power, it was forced to respond to the protests through negotiation instead of violence. In addition, the active participation of “political society” (that is, political parties) from the full ideological spectrum has allowed for development of constructive and consensus-based solutions. Finally, as the role of Alianza Cívica demonstrates, many of the practices that were finally institutionalized by the state have their roots in civil society itself.

Continuing Challenges
Experts, nevertheless, have pointed out some important areas in which the operation of the IFE is far less than perfect. These include: 1. The fact that the General Council is simultaneously responsible for directing and monitoring the activities of the core administrative structure leads to organizational confusion. As a result of this structure, each branch, or Executive Directorate, of the IFE is both accountable in vertical fashion to the Executive Secretary and through him/her to the President of the General Council and finally to the General Council as a whole and in a horizontal fashion to one of the special commissions formed by the General Council to monitor the activities of each directorate. This double authority structure has led to many inefficiencies and ineffectiveness as council members and administrative officers are simultaneously pulled in two directions at once (Schedler, 1999b). 2. The IFE lacks sufficient autonomy from the forces of partisan politics. The budget is not independently set by the law but has to pass through the same channels as any other budget item each year. This forces the leadership of the IFE to lobby the executive and the legislative branch each year in order to defend its budget, a position that opens it up to the world of quid pro quo politics that can seriously harm its independence (Cárdenas Gracia, 2000). 3. The IFE does not have enough power to conduct full audits of the political parties. For instance, as has been widely publicized over the last two years, the Law of Bank Secrecy has been interpreted by the courts and the federal government in a way that prevents the IFE from having direct access to the bank accounts of the political parties. In addition, there is no law that regulates the financing of primary elections or the financing of propaganda or promotion activities used before the beginning of the official campaign period (Woldenberg, 2002). 4. The IFE does not have sufficient powers to control and punish the illegal use of government funds for electoral purposes or prevent the purchasing and manipulation of voters. Although the IFE can force political parties to pay heavy fines and cancel party registration, it cannot directly prosecute government bureaucrats, party members or normal citizens for wrongdoing. The most it can do in this area is refer cases to the Special Attorney for Electoral Affairs (FEPADE) in the Attorney General’s office, an agency that is heavily underfunded and highly vulnerable to political manipulation as a result of its location within the executive branch (Cárdenas Gracia, 2000).



Although scholars disagree as to whether the IFE needs minor tinkering or a full institutional overhaul, most agree that these four problems are the most important. In addition, it does not appear that these problems exist simply because of sloppy institutional design. To the contrary, the historical record shows that they are the result of an explicit intention to undermine the accountability process by the authoritarian regime that has been in power throughout most of the IFE’s institutional life.

Lessons Learned
This case study offers a number of important lessons for state reformers. First, it confirms the willingness and capacity of poor people to participate in the activities of governance. The mobilization of over one million ordinary citizens as organizers and observers of polling booths in the last federal election is a clear testament to the fact that the poor are interested in more than immediate material survival, government handouts and clientelistic politics. Poor people are also passionate about issues of justice, equality and accountability, and are extraordinarily willing and able to participate bringing about change, if given a chance. Nevertheless, the important second lesson is that this capacity of civil society does not arise magically or spontaneously on its own but depends on how government itself behaves. Normal citizens will only participate at such massive levels in positive “synergistic” ways, as opposed to using more confrontational strategies, if the policies being implemented are seen to respond to demands that have originated in civil society, are designed with the participation of a broad range of actors, and actively incorporate citizens into the process of implementation itself. The rule of equal and opposite reaction applies here once again. Governments who want to earn the trust and participation of civil society in pro-accountability activities need to start by themselves trusting civil society and actively opening up state institutions to participation. In addition, as in the Porto Alegre case, it is important for the government to involve the population at large by reaching out and offering training workshops in order to avoid getting stuck with the already existing societal actors as their only interlocutors. Third, none of the achievements of the IFE would have been possible without a significant amount of resources dedicated to the reform and operation of the IFE itself. Societal participation is best stimulated when it is perceived as a complement rather than as a replacement for government action. Without the core group of 2,500 civil servants, significant salaries for the General Council and a large operating budget (US$480 million in the year 2000; IFE, 2000b), the IFE would not have been able to successfully carry out its tasks nor stimulate the popular legitimacy it needed in order to involve the active participation of civil society. Fourth, this case shows us the importance of independent agencies, super majority rules,8 constitutionally mandated budgets, no reelection provisions, and civil service reforms for maintaining the neutral, nonpartisan nature of pro-accountability institutions. We cannot depend on an abstract idea of professionalism alone to do the job nor allow ourselves to be blinded by the traditional tripartite way of thinking about the division of powers. There are many creative ways to design new institutions in order to defend fundamental elements of the democratic process beyond the tosses and turns of electoral politics. The experience of the IFE suggests various positive, as well as negative, lessons for future reformers. Fifth, the case of the IFE forces us to question the commonly accepted idea that neutrality arises exclusively out of the absence of partisanship. Although some of the IFE’s effectiveness does indeed arise out of the professionalization and nonpartisanship of its staff, a great deal of its legitimacy

8. An example is the notorious “governance clause” first established in the 1986 reform that required the party that won the majority of direct election seats to receive enough proportional seats in order to control over 50 percent of the Chamber of Deputies. This was then expanded in the 1990 reform by requiring that the dominant party only win 42 percent of the direct election seats in order to be assured of an absolute majority in the lower house.




also arises out of the saturation of partisanship or the radical plurality of those who participate in the decision making processes of the IFE. The General Council is made up of nine citizen council members, but also surrounded by a whirlwind of party representatives and media “intrusions.” Each voting booth is staffed by trained members of civil society, but also intensively watched by representatives from each political party. One of the principal reasons why the electoral reform of 1996 was more effective than the reforms of 1990 and 1994 is because a greater diversity of political positions were taken into account at the negotiating table in 1996 than during the other two reforms. Democracy and accountability can and do feed off each other in a positive feedback cycle. There is no need to separate the two with an artificial divide between participation and professionalism or conflict and consensus. These lessons are useful far beyond the realm of electoral administration. The model of an independent institute that effectively balances the mobilization of civil society with a highly professional bureaucratic corps while institutionalizing a vast array of horizontal, vertical, and diagonal accountability relationships has a wide range of applications. In recent years, numerous independent corruption control institutes, autonomous auditing agencies, public prosecutors, and human rights ombudsmen have sprung up throughout the LCR and the rest of the developing world (see Pope, 2000; Cárdenas Gracia, 2000). Reformers working on designing and implementing these institutions have much to learn from the case of the IFE. In addition, when the electoral administration is understood as another example of government service delivery, the case of the IFE gains even wider relevance. There is no reason to believe that the struggle to make the delivery of other goods and services in the LCR equally “free and fair” cannot follow a similar path as that followed to clean up Mexico’s elections.

Case #3: Police and School Reform in Chicago
Like many cities in the LCR, Chicago has a “tradition of machine politics, insular administrative bureaucracies installed in reaction to political manipulations, a vibrant tradition of neighborhood activism [and] extreme socioeconomic inequality” (Fung, 2001). A recent study by Archon Fung explores how the Chicago city government has improved the performance of its schools and police forces by actively incorporating the participation of civil society. As in both of the previous two cases, the Chicago government has gone far beyond simple consultative or advisory methods of participation to open itself up to diagonal participation by normal citizens.

Chicago School Reform
This is particularly true in the case of school reform. In 1988 the city assembly passed the Chicago School Reform Act which created a “local school council” (LSC), comprised of six parents, two community representatives, two teachers, the school’s principal, and an additional nonvoting students for high schools, for each of the Chicago Public School’s (CPS) 530 elementary and high schools. The LSC’s have four principal tasks.
First, LSCs are responsible for hiring, firing, evaluating, and determining the job definitions of the principals of each school. Second, they approve school budgets. [Third,] LSCs also develop a required document called the School Improvement Plan (SIP). SIPs are three-year, long-term plans that articulate improvement goals (attendance, graduation rates, achievement levels, school environment) and steps necessary to reach those goals. The principal has primary responsibility for implementing the plan, while the council is charged with monitoring progress. Finally, reform legislation shifted control of “Chapter 1” funds, discretionary state monies allocated to schools on the basis of economic disadvantage, to the LSCs. (Fung, 2001)

As Fung points out, these reforms made the Chicago school system one of the most open to participation in the entire United States. As in Porto Alegre and with the IFE, normal citizens are endowed with government authority, thus breaking the clean split between vertical and horizontal accountability mechanisms.


Chicago Police Reform
Police reform is also interesting even though it did not as clearly empower citizens. The 1995 reform of the Chicago Police Department (CPD) organized police officers into 279 “beat teams” that are required to hold open community meetings each month in order for “the officers serving the area and its residents to jointly engage in problem identification and resolution efforts” (Fung, 2001). Here the mode of participation is more akin to what has been defined as societydriven horizontal accountability. Citizens are not actually given any direct legal power over the operations of the police. They simply provide information and can try to pressure the officers to attend to specific problems. Nevertheless, the close citizen oversight of the police activities does serve as a powerful accountability mechanism, because citizens’ complaints can trigger existing internal mechanisms of supervision and control. Statistical evaluations are difficult given the number of different factors that can have an impact on school performance and crime rates. Nevertheless, there is significant evidence that both services have greatly improved as a result of the reforms. For instance, Fung (2001) points out that between 1994 and 1998 the murder rate declined 24 percent, robbery fell 31 percent and sexual assault fell 21 percent in Chicago, results that are comparable to radically different “zero-tolerance” strategies like those imposed by Rudolph Giuliani in New York. In addition, school performance as measured by a specially developed “metric of school productivity” shows that between 1987 and 1997 “while students entering the system have become increasingly disadvantaged and less well prepared, the majority of schools have become more effective in educating them.” One of the most important elements of institutional design that Fung emphasizes is the concept of “accountable autonomy.” The central idea here is to break the false dichotomy between centralization and decentralization. The important issue is not whether the center or the periphery should have more power, but what sort of power each actor can most effectively wield. According to Fung (2001), in both cases “the role of central power shifts fundamentally from that of directing local units (in the previous hierarchical system) to that of supporting local units in their own problem-solving endeavors and holding them accountable to the norms of deliberation and achievement of demanding but feasible public outcomes.” For example, while local school councils in Chicago are responsible for drawing up budgets and sanctioning principals they are also simultaneously monitored and evaluated by central agencies. This adds an interesting new twist to the theoretical discussion of accountability because local participative bodies are accountable to centralized bureaucratic agencies. Instead of civil society holding government accountable, it is now government that is holding civil society accountable.

The origins of the two reforms are quite distinct. School reform arose out of conflict and was driven by social participation.
In the Chicago schools, reform resulted from a pitched battle that pitted a diverse social movement composed of parent organizations, “good government” civic groups, educational reform activists, and a coalition of business groups against traditional school insiders such as the Chicago Teacher’s Union and the Board of Education. (Fung, 2001)

Police reform arose out of consensus and was principally directed by reformers within the state. “Absent the street heat and legislative pressure that drove school reform, [the reform] discussions at the intersection of professional, political and civic interests led quietly to the formulation of a participatory variant of community policing…” (Fung, 2001). The important point here is that in general, “far from the result of masterful design, these institutions arose haphazardlythemselves the result of fitful informal deliberations—as reformers inside city offices and activists outside of it groped toward more effective ways of organizing their police departments and schools” (Fung, 2001).




Once again, the importance of both state and society actors in the development of the participation mechanisms is evident. Indeed, perhaps the relative lack of social participation in the design of the police reform is one of the principal reasons why this reform did not end up empowering citizens as much as the school reform. Regardless, it is clear that neither reform arose solely out of the action of individual enlightened bureaucrats, but each was the result of complex social processes and state-society synergies.

Lessons Learned
These Chicago cases reinforce the above lessons concerning pro-accountability reforms. First, as in Porto Alegre and with the IFE, the most active participants in Chicago are the poor and uneducated. “Contrary to skeptical expectations that reforms demanding active participation will further disadvantage less well-off areas, residents of poor neighborhoods participate at rates equal to or greater than those from wealthy ones” (Fung, 2001). In addition, Fung points out that “whereas previous studies have found that African Americans and people of Hispanic backgrounds [in Chicago] are somewhat less likely to vote than others, higher proportions of black and Hispanic students in a school correlated with slightly higher parental turnout rates in the 1996 LSC elections.” Second, Chicago reformers also made a concerted effort to open the process far beyond already organized civil society and, especially in the case of police reform, used government employed community organizers to stimulate participation and facilitate community decision making. Nevertheless, the accountability mechanisms in Chicago were clearly not as open and participatory as those in place in the previous two cases. The local school councils are elected bodies that do not bring a clear popular mandate arising out of popular assemblies like the COP in Porto Alegre and, as discussed above, the police community meetings do not have any direct legal authority over police behavior. Perhaps this is why the level of citizen participation in Chicago is also much lower than it is in Porto Alegre and Mexico. Fung documents that an average of only 20-25 people participate in each beat meeting per month and there are only an average of 1.5 candidates in the elections for each open spot in the school councils. Third, civil society participation in the design phase of participatory structures proved to be crucial. Neither of the Chicago reforms arose purely out of the minds of social planners and their relative success depended on the ability of the government to involve social actors from the very beginning. Fourth, as in Porto Alegre and with the IFE, the supply side of the equation is crucial. Without a capable and well-financed state apparatus that can actually respond to popular demands and participation, such accountability mechanisms would create more disenchantment than hope. Indeed, this point raises important questions as to the exportability of the police reform case to Latin America. How effective would community meetings be in a situation of widespread police corruption and an almost total lack of trust in local police forces? Fifth, these cases push us further towards the conclusion that the supposed either/or choice between centralization and decentralization is a false dichotomy that needs to be reanalyzed. Although devolving power is important, there is an equal need to strengthen the center, at least in its coordinating and monitoring capacities.

Case #4: Decentralization and Rural Development in Mexico
As seen above, decentralization is just as likely to strengthen entrenched local networks of corruption, clientelism and capture as it is to promote participation and accountability. More money for local government alone will not make governments more responsive or accountable. Good results depend on the institutional and social structure of the local units that are empowered by devolution. This is the central lesson of Jonathan Fox’s research on the use of World Bank funds for municipal development projects in rural Mexico.



Horizontal Social Capital
For Fox, the most important element of a pro-accountability “enabling environment” is the existence of strong social practices of mutual trust and cooperation (or “horizontal social capital”; Fox, 2002). Such social networks are key under both authoritarian and democratic arrangements. Under authoritarian situations, in which local officials are imposed from the outside, strong horizontal social capital allows communities to pressure for the honest and effective use of public funds. Under democratic situations, horizontal social capital allows communities to better channel investments towards more productive, long-term development projects that benefit the most poor, as opposed to short-term porkbarrel or politically-oriented populist projects (Fox and Aranda, 1996). The level of horizontal social capital in a particular community or region is dependent on local history, cultural traditions and political institutions. For Fox, social capital is eminently “constructible” (Evans, 1996). This contrasts with Putnam’s (1993) classic study of Italy where the relationship between good government and social capital is almost entirely one way, with stocks of social capital that have accumulated for over 500 years explaining different levels of government effectiveness. In contrast, Fox (2002) argues that institutional design has a clear and direct impact on the development of social capital in the relatively short-term. The relationship between social capital and good government is a two-way street. For instance, “widespread community participation can improve governance, promoting improved transparency and public accountability, while a government in the hands of local elites, who govern from above, can inhibit the participative potential of civil society” (my translation). A series of studies by Fox on Mexico’s Municipal Funds Program give ample evidence of the impact of social capital and institutional design on state-society synergy for accountability. Although issues of legal accountability are an implicit subtext throughout the studies, the focus is fundamentally on performance accountability. The Mexican Municipal Funds program has been almost entirely financed by two large loans received from the World Bank, one for US$350 million for the period from 1991-1994 and a second for US$500 million for the period from 19951999. This money was targeted for use in basic infrastructure improvements (for example, pavement, schools, potable water, sanitation) for the poorest communities in the rural areas of the poorest states. The program was implemented through municipal governments and, particularly during the first period, accountability mechanisms were developed to allow citizens to directly participate in the planning and the implementation of the projects. Specifically, autonomous “solidarity committees” were organized in each community in close coordination with the municipal government. These committees were responsible for deciding which projects would be funded and for contributing the necessary labor power. Because the Solidarity Committees did not have any legal standing or formal authority over the Municipal Funds program itself, the actual level of participation and the real autonomy of the committees from the municipal, state, or federal government depended entirely on the way in which local leaders and citizens approached the task. Needless to say, a great number of committees only had a significant participation in the implementation phase of the projects and were entirely excluded from the planning phase.

The Oaxaca Case
This participatory mechanism was particularly successful in the state of Oaxaca. In this state “the community assemblies made the project selection decisions in almost two-thirds of the cases (63 percent of projects observed were chosen by community assemblies)” (Fox and Aranda, 1996). One of the principal reasons for such high levels of participation is because Oaxaca is endowed with a very high level of horizontal social capital due to a long and rich indigenous tradition of community collaboration, trust and self-governance. For instance,
most rural Oaxaca communities already had active local public works committees, as part of their ethnically based system of rotating community responsibilities. In most of rural Oaxaca, these positions are chosen through community consensus and are unpaid, full-time responsibilities…In the




smaller villages, most Municipal Funds projects seemed to be taken on by these preexisting committees led by municipal authorities, such as the town council or the local agente municipal.” (Fox and Aranda, 1996)

This also speaks well of the government side of the equation in Oaxaca. Instead of imposing a new organizational structure on society, a healthy amount of mixing between state and social forms was permitted. As the Porto Alegre and the IFE cases suggest, state-society synergy is particularly successful when the governmental design of participatory institutions corresponds to preexisting practices within civil society. In Oaxaca, this tolerance of autonomous social forms goes back much further than the Municipal Funds program. The state’s municipal structure itself, with 570 different municipalities based in local organizational forms, demonstrates the government’s long-standing commitment to accommodate legal forms to traditional practices. Significantly, the communities that had higher and more authentic levels of participation had more effective development projects. “Community assembly decision making produced disproportionately better projects, while project selection by mayors and external actors tended to produce insignificant projects”(Fox and Aranda, 1996). When the community was directly involved, it tended to monitor the use of funds more closely and to pick projects that were more clearly useful for the population as a whole. When the selection process was manipulated from outside, investment tended to be shifted towards highly visible, although not always useful, projects. For instance, although mayors and external actors tended to be partial to projects like basketball courts and paved roads, assemblies were more likely to opt for more useful investments like potable water systems or corn mills. After the beginning of the program, the government also explicitly intervened to increase community participation and to make the distribution of resources more fair. The formulas used for poverty measurement and funds distribution were improved and, even more importantly, made public. The amount of funds that could be spent in the municipal capital was limited to 25 percent, thus requiring municipalities to channel funds to the most needy, isolated areas. Finally, the required amount of community contributions was made variable, depending on the impact on poverty the selected project would have. High impact projects required less community contribution than low impact projects, thus encouraging investment in true “public goods” (Fox and Aranda, 1996; Fox, 2002). These changes stimulated community participation and strengthened social capital because they made communities aware of their right to a precise amount of funds, actively involved the poorest areas, and empowered those actors who looked beyond their particular interests and towards the development of the community as a whole. This demonstrates the direct effect institutional reform can have on trust, fairness and participation. As Fox writes,
The maintenance of trust requires trustworthy local institutions that can give strength to relations of mutual responsibility…[For example,] if the local governments utilize transparent and consistent criteria in the selection of more consolidated projects first and give priority to high levels of participation, then this will work as a material incentive for the least cohesive communities to reach agreements about their proposals for future projects. Here, it is once again important to recognize that the logic of citizen participation, its justification, depends in part on the possibility that there is public accountability and transparency, and vice versa. (Fox, 2001, my translation)

Nevertheless, one important element that sets this case apart from the first three cases considered is that the origins of this particular scheme of state-society synergy for accountability was entirely “top-down.” Instead of arising out of intense negotiations between social actors and government reformers, the participation scheme was thought up and designed by the federal government in consultation with World Bank staff. This may go a long way in explaining why community participation has not been more dynamic in the Municipal Funds Program, and why the case of Oaxaca is more of an exception than the rule. Indeed, the “top-down” nature of the entire National Solidarity Program (PRONASOL), of which the Municipal Funds program was

only a part, has led many scholars to disqualify it entirely as an attempt at social manipulation intended to help the PRI and the powerful interests it defends remain in control.
By reinstating the PRI’s role as a welfare machine, PRONASOL has enabled the PRI to regain its position as centerpiece of the party system…Neoliberalism demands an aggressive political and ideological remaking, not simply the implementation of economic measures. PRONASOL entails both. The Salinas administration is presenting neoliberalism as a hegemonic project, and at the same time it is using Solidarity to create a durable base of support for the project in civil society. (Dresser, 1994) PRONASOL is neither an attempt to alleviate the poverty afflicting well over half of the population in Mexico nor a move toward constructing a stronger civil society in a top-down fashion. Instead it is a reflection of new forms of ideological and political domination targeted at preserving the hegemony of the ruling classes while excluding the majority of Mexicans from participating in the formulation of state policy. (Soederberg, 2001)

This evaluation of the program as a whole is solidly supported by the extreme level of progovernment propaganda that accompanied almost every step of the Solidarity Program. Moreover, the distribution of solidarity funds corresponded much more closely to political criteria than to need-based criteria (see Cornelius, Craig and Fox, 1994). Nevertheless, these well documented facts should not lead us to ignore exceptional cases like Oaxaca that support the prospect of statesociety synergy for accountability even under relatively unfriendly conditions.

Lessons Learned
There are various lessons to be learned from this case study. First, as seen in the previous three case studies, the direct involvement of social actors from the initial design stage is absolutely crucial for the success of accountability mechanisms that depend on active participation from civil society. Second, this case also confirms the importance of the formal, legal empowerment of participatory bodies. Without a clear institutionalized location in the decision making process, these bodies are left open to the winds of manipulation and are quickly bypassed by unwilling or authoritarian public officials. Third, this case suggests the crucial importance of horizontal social capital for the success of participatory pro-accountability arrangements. Without some level of community trust and cooperation it is difficult to foment participation that is in the interests of all as opposed to only the most powerful in a particular community. Finally, this case also shows us the crucial impact that government transparency and institutional design have on the development of this type of social capital. Only when government actors respect social actors enough to fully inform them as to the design and implementations of development programs and design the participatory institutions so as to assure the active involvement of the most marginal actors, will state-society synergy be able to truly fulfill its promise of making government more accountable.

Case #5: Toxics Release Inventory in the United States
The case of the Toxics Release Inventory (TRI) in the United States is particularly fascinating because it seems to break with many of the above lessons on how successfully to promote statesociety synergy for accountability. As Archon Fung and Dara O’Rourke have written,
TRI is an accidental success story of regulatory reform. Designed to do much less, the program has proven to be one of the most successful programs to reduce toxics in EPA history. In the past 10 years, the small program has eclipsed its initial goals of providing information for community planners, igniting a “right-to-know” movement in the US, and supporting both industry and environmentalists’ efforts to reduce toxics. (Fung and O’Rourke, 1998)




The TRI is essentially only a pollution accounting system. Developed by the government in 19869 in a top-down fashion in order to help community planning and in the aftermath of the Bhopal chemical disaster, the TRI requires manufacturing firms to report their annual emissions of 651 toxic chemicals to the Environmental Protection Agency (EPA). This information is reported by the firms themselves with very little follow up by the EPA (only about 3 percent of firms are inspected each year) and then made widely available via print and internet media by the government. In total, the government only invests US$23 million in the TRI each year, much less than the tens of billions spent under the Clean Water and Clean Air Acts.

A Self-Perpetuating Virtuous Circle
The TRI has been incredibly successful. Between 1988 and 1995 total releases and transfers of 330 of the chemicals on the TRI list decreased by over 45 percent (U.S. Environmental Protection Agency, 1995). This means that roughly 1.3 billion pounds less of toxic chemicals were emitted in 1995 than in 1988 (Fung and O’Rourke, 1998). These numbers compare quite favorably to other EPA activities. “Though data are difficult to compare and particular causes of reductions highly contested among experts, it is arguable that the TRI has dramatically out-performed all other EPA regulations over the last 10 years in terms of overall toxics reductions, and that it has done so at a fraction of the cost of those other programs” (Fung and O’Rourke, 1998). This overstates things somewhat since the many statutory requirements on the books are also clearly responsible for the reduction of toxic releases. Nevertheless, the TRI can most likely claim at least a part of this success as its own. According to Fung and O’Rourke, the effectiveness of the TRI is principally due to its capacity to stimulate the participation of civil society in the reduction of toxic chemical releases. The way it does this is by fomenting what the authors call “Populist Maxi-Min Regulation.” This is a kind of “environmental blacklisting” by which the TRI database allows groups in society (for example, environmentalists, businessmen, community activists, and journalists) to clearly see which corporations pollute more than others.
Many of these users—especially journalists and environmentalists—use this information to identify and target the most egregious polluters…Public pressure-sometimes in the form of community mobilization and sometimes not often induces these firms to reduce their toxics releases. Others who are not specifically blacklisted improve their environmental performance to avoid the potential negative consequences of being identified as worst in their class…. The mechanism thus induces continual reduction of toxic releases by pressuring whoever happens to fall at the bottom of the list. (Fung and O’Rourke, 1998)

This is called “populist” because civil society, not government, does most of the enforcement. It is “maxi-min” because the maximum pressure is placed on the worst polluters. The TRI database works both directly and indirectly to reduce pollution through the participation of civil society. It functions directly by generating a great deal of negative publicity and public pressure against those corporations that are the worst in their class. This creates immediate problems in terms of operations and sales, and can also seriously affect stock-prices as conscientious investors pull out their support and more pragmatic ones follow. It also functions indirectly by inspiring fear in all firms that are not at the bottom of the list. It is clearly in the interests of managers to invest relatively small amounts of money in making sure that they are not the worst polluters rather then spending huge amounts to defend themselves against a public relations nightmare. Because this is true for all corporations, the tendency is for emissions to go down across the board. As a result, the overall level of “worst performance”
9. The TRI was included as a part of the Superfund Amendments and Reauthorization Act passed by congress in 1986.



will improve, leading to a situation in which the pressure only increases on all corporations to even further improve their levels of pollution in order not to be caught at the end of the list. This creates a seemingly endless virtuous cycle of environmental improvement. This apparent limitlessness is reinforced by the fact that normal citizens usually have very different criteria for evaluating “acceptable” levels of pollution than do government bureaucrats. As O’Rourke has written elsewhere,
Community members in general are much more interested in results—that is, pollution reduction— than in inspections, reports, EIAs [Environmental Impact Assessments], or even agreements to build treatment plants. Mobilized communities thus serve as an expansive team of monitors to follow-up on inspections and promises of improvement. (O’Rourke, 2001)

Because citizens directly feel the impact of pollution, they are interested in its immediate and complete elimination. Their tolerable level of pollution is usually much lower than the acceptable level for a government regulator who is distanced from the real human impacts of environmental destruction.

Recommendations for the Future
As Fung and O’Rourke point out, in spite of the success of the TRI, there is still much more that needs to be done. In spite of the success of the society-driven aspect of the TRI, the government still needs to clamp down more on the reporting requirements and should set clearer minimum standards. On the other hand, the government could also do much more to directly enable the participation of social actors in holding corporations accountable for their toxic emissions. For instance, Fung and O’Rourke (1998) propose that the EPA should consider setting up “Offices of Community Assistance” that “supply technical support to citizen groups that must live with highly polluting facilities” and supporting “Community Based Organizations or other Non-Governmental Organizations (NGOs) to run community environmental assistance programs.” Indeed, as Don Sherman Grant has recently argued, even the present day effectiveness of the TRI depends a great deal on more pro-active state efforts to stimulate societal participation. Sherman Grant conducts a statistical analysis of TRI numbers for states that have implemented the “right-to-sue” provision of the 1986 Superfund Act and those that have dedicated significant funding to “right-to-know” programs. He concludes that when states are actively engaged in empowering societal actors the success of society-driven environmental regulation is much higher. For Sherman Grant this confirms the conclusions of “conflict researchers” who emphasize that,
the success of participatory policies is contingent on the resources they mobilize on behalf of citizens. Policies that emphasize consciousness-raising activities and provide citizens perfunctory hearings to act on their new insights will not give citizens any real leverage in their dealings with industrial polluters. Only policies that are conflict oriented, that are supported financially by the state, and that provide citizens with critical resources of information, network support, legal powers, and so on, are likely to motivate businesses to reduce their emissions. (Sherman Grant, 1997)

Even in the case of the TRI, state-society synergy for accountability needs to be conscientiously cultivated by the state in order for it to be effective. Information alone is not enough.

Lessons Learned
This case study leaves us with a few important lessons. First, institutional reform can have unexpected consequences. Even relatively controlled openings from above like the TRI can detonate significant societal participation in the enforcement of accountability. This is particularly the case when the issue directly impacts specific groups of people and when civil society already has consolidated organizations (for example, environmentalist groups, media, and community organizations) that are prepared to apply the necessary pressure. Second, even under such conditions,




the state still has the responsibility to enforce regulations from above and to facilitate the participation of civil society from below. It is not enough to simply publish information or survey data and then sit back to await the response of civil society, especially in countries where civil society organizations are weak. Third, societal actors can often be more effective if, once stimulated and empowered, they are left to enforce regulations on their own. As evidenced above, much of the success of the TRI has been due to the detonation of an autonomous social movement that constantly and mercilessly pressures corporations to reduce their pollution levels.

Case #6: Women’s Police Stations in Brazil
In 1985, the state government of Sao Paulo, Brazil opened the first all-female police station (delegacia de defensa da mulher-DDM) in the world.
The idea behind its creation was that the traditional institutional response to grievances of violence against women was inadequate and even discriminatory. Police, almost always men, routinely ignored and rarely prosecuted cases of physical and sexual abuse of women and often blamed and harassed the victims. (Nelson, 2002)

Here the central accountability issue is equal treatment and fairness. Police forces throughout the world consistently treat violence against women as a private issue to be worked out within the family and therefore refuse to intervene to correct the problem. The creation of separate police stations run by women, with women and for women redresses this problem by extending the rule of law down to the level of the family and inter-gender relations. By 1995, there were 126 DDMs operating in the state of Sao Paulo and over 200 more in Brazil (Nelson, 2002).

DDMs originated in civil society in three different ways. First, various high profile cases of violence against women between 1979 and 1985 brought the issue of judicial unfairness towards women into the spotlight. For instance, the courts initially accepted the defense of “violent emotion” in the cases of Raúl Doca Street and Lindomar Castilho, both of whom had murdered their female partners. Only after widespread protest and publicity were these two men given long jail sentences. Second, many non-profit organizations, like SOS-Mulher, began to offer independent assistance to women who were victims of violence. These organizations set up centers that provided emotional support and legal council to victims. Lack of funds limited the social impact of these activities, but they nonetheless offered an excellent example for what government could do if and when it had the interest in addressing the issue (Nelson, 2002). Third, when the opposition Party of the Brazilian Democratic Movement (PMDB) won the Sao Paulo state government in 1982, it established a State Council on the Status of Women (CECF) comprised of both representatives from women’s organizations and state-level government administrative departments (for example, education and health). This new council was then the central motivating force for the establishment of the DDMs in Brazil. This state-society body helped design the initial proposal for the women’s police stations (Nelson, 2002). In the beginning, there was also close collaboration between the state and civil society groups in the implementation of the DDM model. Before starting work at a DDM, female police officers were required to attend intensive seminars run by members of the women’s movement on gender relations and the specificities of violence against women. “According to the coordinator of the CECF’s commission on violence, this meeting between members of the women’s movement and police personnel dispelled the initial tension and helped establish a positive relationship between the two, as well as opening a channel for future input from the activists”(Nelson, 2002). Nevertheless, as the number of DDMs expanded and the initial excitement for the project wore off within the government, the project quickly fell exclusively under state direction. Activists, community groups and non-profits were soon cut off from the DDRs and the specialized training itself was eliminated.



Mixed Results/Lessons Learned
One of the most important reasons why the initiative has had such mixed results was due precisely to this pullback from civil society. On the one hand, it has been extremely successful in publicizing violence against women and moving the issue of gender equality squarely into the public debate. For instance, during the first half of 1994 alone there were 54,472 incidents of violence against women registered in the state of Sao Paulo alone (Nelson, 2002). Also, the very existence of so many public institutions designed explicitly for women publicizes the need for extra support. On the other hand, the DDMs’ actual performance was less than might have been expected. Of the 54,472 incidents “only 16,219 (roughly one-third) of these cases resulted in a police investigation and far fewer still in prosecution or conviction. The large majority of reported cases are simply archived and forgotten” (Nelson, 2002). In addition, much of the previous mistreatment of women who reported crimes has continued, now at the hands of fellow women.
The fundamental rasion d’etre of the DDMs is to compensate for the sexist practices of the maledominated criminal justice system in which they are located. However, the absence of special training programs to educate officers about issues relevant to gender-specific violence couples with a lack of social work professionals seriously compromises crucial components of the DDMs’ function…In the absence of such training, the DDMs’ performance will predictably be unsatisfactory. Indeed, my interviews with DDM personnel suggest that women are no more “naturally” compassionate and responsive to their sisters’ needs than men. (Nelson, 2002)

When the institutionalization of social demands and practices is not carried out with the active collaboration of society itself, it risks missing the crucial core of the demands themselves. The original demand for more fair treatment for women and with women has ended up being institutionalized only as a demand for treatment by women. A second reason for the spotty performance of the DDMs may arise from issues of unsuccessful institutional design.
The capacity of the DDMs to fulfill many of their original objectives is necessarily limited by their problematic position within the police bureaucracy—problematic because the DDMs were created in resistance to the very male-dominated criminal justice system in which they themselves are located. In order to exist and proliferate, they must succeed at the basic police duties with which they are charged. Yet they must also, in a sense, fail, or otherwise pose a threat to the legitimacy of the police bureaucracy. That is, they must perform their bureaucratic function while concealing the feminist premises upon which they were founded. (Nelson, 2002)

As seen previously with the case of the IFE, the institutional location of new initiatives of statesociety synergy is absolutely crucial. Such creative new ideas not only need positive enabling environments in society, they also need such environments within the state itself. Any new institution, and especially participative ones, also need time as well as support from within the state to develop and flourish. Finally, the party basis of the governments that try to implement pro-accountability reforms is also important. For instance, the fact that the PMDB is not as closely linked to social movements and community activists as the PT may go a long way in explaining why participation was not successfully institutionalized in Sao Paulo while it was in Porto Alegre. Here once again the echoes of the situation with the IFE are seen, in which the authoritarian coalition intentionally undermined its own pro-accountability innovation. Governments that arise out of popular movements tend to be those most willing and able to incorporate these movements in the work of governance. This case leaves us with the crucial lesson that the active participation of civil society in the development of pro-accountability proposals hardly guarantees the institutionalization of this participation or the long-term success of the proposal. Only when participation finds a supportive institutional home and serious government actors who are willing to involve social actors in the




process of development and maintenance of the new project will state-society synergy for accountability be fully successful. It is not enough for the state to simply respond to social demands, it must also work with society to make these real.

Case #7: Grass-Roots Anti-Corruption Initiatives in India
One area of government that seems to be particularly resistant to opening itself up to societal participation is the auditing of government expenditure. This task is usually seen to be far too technically sophisticated and politically delicate to be opened to the average citizen. As Anne Marie Goetz and Rob Jenkins argue,
there is almost nowhere on earth that citizens or their associations have either been given access and information on, let alone a more substantive role in, formal auditing processes. Indeed, even in the far less sensitive area of expenditure planning, there is just a handful of experimental cases world-wide encouraging citizen involvement. Citizen auditing strikes at the heart of practices that preserve the powers of bureaucrats and politicians: the secrecy in public accounts that can mask the use of public funds for personal advantage. (Goetz and Jenkins, 2001)

Freedom-of-information acts have recently started to sprout up around the world, and citizens are frequently encouraged to use public information to pressure corporations or governments from the outside to comply with their duties or to decide their votes. Yet, it is difficult to find examples in which normal citizens are as directly involved in the activity of auditing government expenditure as they are, for example, in the activity of budget planning in Porto Alegre.

Independent Grassroots Initiatives
Nevertheless, the cases of the Mazdoor Kisan Shakti Sangathan (MKSS) movement in Rajasthan, India and the Rationing Kruti Samiti (RKS), or Action Committee for Rationing, movement based in Mumbai, India, show that when reformist bureaucrats are faced with an active proaccountability movement in civil society, it is possible to make important inroads into the area of social auditing. As Goetz and Jenkins document, the central accountability problem that both of these organizations face is the widespread corruption in the provision of government services to the poor. Wages for public works projects are frequently skimmed off by public managers and the materials used in these projects are often artificially over-priced and of bad quality so as to allow the maximum room for kickbacks. In addition, the nation’s Public Distribution System (PDS), which is in charge of channeling basic food items and other fundamental household goods like kerosene to the poorest households, is rife with corruption. One of the principal problems here is the selling of these goods by owners of “ration shops” for personal profit. Most communities in India already have local “participatory” institutions that are supposedly responsible for monitoring the performance of government programs. Nevertheless, these “Vigilance Committees” and “Village Assemblies” are often captured by actors who are implicated in the process of corruption itself. For instance, the Vigilance Committees are usually chaired by the representative of the municipal ward and their members are appointed in a top-down fashion. As Goetz and Jenkins (2001) argue, “appointment to [vigilance] committees through a process of official selection increases the probability of capture by the very political organizations that are central to the operation of the systems of leakage.” In addition, “since many ration shops are actually owned or controlled by politicians, who are themselves on the vigilance committees, committee members have little incentive to remain eternally vigilant.” Finally, often the shopkeepers themselves sit on the Vigilance Committees, thus setting up a situation in which the auditor and the “auditee” are the same person. The situation is not much better with respect to the Village Assemblies. Goetz and Jenkins (1999a) write that although these are recognized in the constitution, in most parts of Rajasthan where the MKSS works they are “moribund political institutions whose democratic functioning is impaired by the continued existence of constraining social institutions.”



As a result of the failure of these state run participatory mechanisms, movements like MKSS and RKS have found it necessary to create their own autonomous society-driven mechanisms for auditing public projects. The MKSS has developed a methodology through which they independently investigate government spending practices and then expose and compare this information to reality through public hearings (jan sun wai).
These hearings are the culmination of a methodology for reviewing local government accounts and determining whether funds were expended on the development works in the manner indicated in official records. Meticulous research is conducted in the weeks prior to the public hearing. The first step is to procure government expenditure accounts, including receipts for building materials purchased and employment-wage registers. Sometimes this information is given willingly by sympathetic senior bureaucrats; at other times, filching by low-level clerks connected with the MKSS has proven effective. The implied threat of agitation and protest provides a background inducement for co-operation by officials. (Goetz and Jenkins, 2001)

In the hearings themselves obvious discrepancies and missing accounts are presented and the public is given the opportunity to check their own personal experience as public employees or providers with the accounts. Public officials themselves often attend and many cases exist in which this process has worked to directly shame them into returning large amounts of “misdirected” funds. The process has been extremely difficult because of the lack of a strong freedom-of-information act in India. As a result, the MKSS and its allies have been forced to develop a second, complementary strategy.
Although successful in exposing corruption in a number of localities, jan sunwais have been relatively rare because of the difficulty in obtaining certified copies of government accounts from reluctant officials. In response, the MKSS and its allies in Rajasthan’s large and diverse voluntary sector developed a parallel strategy involving large-scale public protests extending over several weeks. The objective: legislative and regulatory reforms to provide a legal basis for local efforts to obtain official documents. (Jenkins and Goetz, 1999a)

These mobilizations have been quite successful. A state-level Right to Information Act was passed by the Rajasthan state assemble in May, 2000, and the state’s Local Government Law has been modified to include “mandatory legal procedures for the investigation of corruption and to institutionalize officially the public-hearing audit method at the village assembly (Gram Sabha) level”(Goetz and Jenkins, 2001). The work of the MKSS is only starting to have an impact on government accountability in one small part of India. Nevertheless, it gives us a fascinating alternative model to so-called “participative” mechanisms like the Bangalore Scorecard, which is limited to simply surveying and reporting on the opinion of the public concerning the performance of government services. As Jenkins and Goetz (1999a) argue, such initiatives are grounded in a fundamentally naïve view of politics and bureaucratic inefficiency. According to the authors, the findings of the Bangalore Scorecards,
can be considered “weapons” only if the politicians and bureaucrats in question are ignorant of the service-delivery problems in the first place. Most, in fact, are already aware of the dismal state of public amenities in India’s slums. The MKSS approach begins from the assumption that what would motivate officials to take remedial action is concrete evidence of their complicity in misappropriating funds intended for addressing these problems. A right to information makes this possible, though not inevitable. It requires associations of people willing to confront authority.

Bureaucrats need to be made directly accountable to the citizenry, and the best way to do that is to allow citizens to get involved in the activity of auditing from the inside and to directly confront bureaucrats with their complicity in the lack of performance or the corruption that exists in the delivery of public goods and services.




Goetz and Jenkins present the case of the RKS in Mumbai as another example of such diagonal accountability. The RKS usually only advocates for policy reform from the outside, but recently they have decided to get involved in directly monitoring the operation of the PDS. Because the official Vigilance Committees are ineffective, the RKS has developed its own parallel system of informal vigilance committees. For each ration shop, five local women who are clients of the shop come together to monitor and evaluate the quality and prices of the goods being sold. This activity has been facilitated by the RKS citywide campaign to oblige shop owners to publicly display prices as well as samples of the goods on sale. The reports of the informal committees are then put together and presented both to the user community and to the central coordinating bureaucracy of the PDS in the city. This process was particularly successful during the period immediately following the 1992 riots in Mumbai, after which the city government was very interested in being perceived as being responsive to the poor. In addition, during this period an important reform-minded bureaucrat held the job of Regional Controller of Rationing. Nevertheless, in the end,
this joint civil society-state monitoring initiative was undermined by politics. State politicians were infuriated that their control over the PDS had been undermined by a bureaucrat and a group of CBOs [Community Based Organizations]. Without state support for its work, the RKS has had to return to more conflictual but rather ad hoc tactics (such as city-wide protest action and sustained community pressure on individual shopkeepers) to pressure local shopkeepers to leak smaller amounts to the open market. (Goetz and Jenkins, 2001)

The authors therefore claim that the RKS’s experience with diagonal accountability has been only a “limited success story.” As seen with the case of the MKSS, society-driven pro-accountability initiatives that confront the state and demand inclusion in the basic activities of government can be highly effective. Nevertheless, the RKS experience also shows us that ultimately the success of these movements often also depends on constructing alliances with progressive government officials as well.

Lessons Learned
From this pair of cases, various lessons can be gleaned. First, state-society synergy for accountability does not need to begin with reformist or progressive government politicians or bureaucrats or be based in consensus between government officials and social actors. Success can also arise out of the action of independent organizations and social movements that press their demands on the state and push their way into the auditing of government programs. Second, it seems that at some point in the process these movements do need allies within the government. Without state support or at least tolerance, such movements will most likely be repressed or rendered ineffective by state action. Third, both of these cases show that public auditing is not beyond the capacity of poor, illiterate citizens. “Both cases are notable for engaging the very poor in scrutiny of official accountability processes, challenging assumptions that the socially marginal and illiterate may lack the human capital or long-term vision to invest in efforts to improve the quality of governance, as opposed to efforts to improve their immediate survival prospects” (Goetz and Jenkins, 2001).





The image of the good bureaucrat—carefully insulated from constituents-has its usefulness, but openness to the role of the “co-producer”… may be the best way to increase effectiveness and ultimately the best way to preserve the integrity of increasingly besieged public institutions —Evans (1996b)


n the introduction the question was posed as to how the relationship between state and society can be best transformed from a tradition of particularistic concessions and manipulations. How can a new partnership of healthy engagement be forged that produces a solid bureaucratic apparatus and policy outcomes that are in the interest of the public as a whole? The discussion above shows that in order to answer this question, two dangerous myths about the poor and marginalized people of the developing world need to be broken: (a) the assumption that they are apathetic, easily manipulated and incapable of participating in the central activities of governance; and, (b) the belief that they are raucous, inherently conflict oriented and untrustworthy. Both of these myths are especially dangerous because they are self-perpetuating. When government institutions act on these assumptions they tend to reinforce the idea that they might be true. For instance, if public officials hide information from the people, the population does indeed become less informed and more apathetic, but through no fault of its own. When the authorities respond to street marches with violence, the almost inevitable outcome is an increase in protest and the expanded use of confrontational strategies on the part of civil society. Fortunately, both of these myths about the comportment and capacities of common citizens are entirely false. As each one of the above case studies show, poor people are exceptionally willing and able to work with government in constructive ways once they perceive that their participation can make a difference. In addition, effective societal participation is by no means limited to the provision of basic services. The poor care about much more than simple survival and local issues. The above case studies demonstrate how normal citizens can get passionately involved in issues as abstract and technical as financial accounting and school administration and as broad and universal




as environmental protection and free and fair elections. It is a grave mistake to think that the poor are incapable of mobilizing themselves in the pursuit of larger social goals. Therefore, the very first step for government reformers looking to construct state-society synergy for accountability is to trust and actively involve societal actors from the very beginning of the process. Reformers should not wait for civil society to start trusting government nor should they wait to involve society until after the government has already designed a new participatory mechanism from above. As each of the above case studies show, the earlier societal actors are involved in the design process, the more effective the participatory measures. There is also a significant difference between a full and partial opening of the state to societal participation. The three most effective examples of state-society synergy summarized above (Porto Alegre’s participatory budget, Mexico’s electoral institute and Chicago’s school councils) are all cases in which the government encouraged societal participation to reach far beyond direct vertical accountability and society-driven horizontal accountability to achieve full diagonal accountability in which society directly carries out activities that normally correspond to the state. Also, the government can only get out of society as much as it puts into it. In addition to opening up new spaces for participation, the government should actively recruit and train new actors in civil society. Although it is important to work with groups that are already organized, it is equally crucial to reach out to the unorganized majority and to show them that their participation matters as well. Nevertheless, it is also important to point out that those state reformers who find themselves with limited resources or staff to implement full participatory schemes should also take heart from the above case studies. The cases of the Municipal Funds program in Mexico, the Toxics Release Inventory in the United States and the grassroots anti-corruption initiatives in India show that even a small opening from above can sometimes stimulate the action of civil society in productive ways. Small efforts are much better than nothing. State-society synergy is not an “all or nothing” endeavor. Both those reformers who pursue limited schemes and those who engage with society more fully should consider granting effective participatory structures official, legal status as soon as possible. As seen in the case of the Municipal Funds program, the formalization of even limited top-down participatory schemes allowed for the development of much fuller participation, especially in those areas that already had important participatory resources like horizontal social capital. The cases of the RKS in Mumbai, India and the women’s police stations in Brazil provide us with important negative examples of this same point. Here the absence of a clear legal framework left participation up to the whims of individual bureaucrats, leading to the eventual overturning of participatory schemes once there was a change of heart on the part of government. Finally, the difference between the two Chicago cases also reveals the importance of formalizing participatory procedures. One of the major reasons why the school reform has been more effective than the police reform is because the former institutionalized the involvement of civil society in the formal legal structure much more clearly and explicitly than the latter. In addition, professionalism and well-financed independent bureaucracies are central elements of successful state-society synergies. Societal actors are much more willing to participate if they see that they are not replacing but complementing the activities of government and if they are able to dialogue with government representatives who are serious about their jobs. Also, societal actors are easily disappointed if the government is not able to deliver the goods at the end of the day— a situation that could leave state-society relations even worse than if no attempt to reach out had been undertaken in the first place. The above case studies also force the question as to the commonly accepted idea that the absence of partisanship and political conflict is the only fertile ground for neutrality and accountability. Professionalism and independence is necessary but by no means sufficient to assure the long-term survival of accountability. In order to survive, pro-accountability structures need to be legitimated by society both at their founding moment and during their everyday operations. This requires the multiplication, not the reduction, of external eyes and the diversification, not unification, of political and ideological perspectives. The history of Mexico’s electoral institute and the experiences of




the Toxics Release Inventory in the United States and the grassroots anti-corruption initiatives in India provide particularly good examples of this. Indeed, as the last two examples show, sometimes the most effective strategy for state reformers might be to stimulate dynamic social movements and social protest and let them take the lead in pressuring and undermining the power of recalcitrant elements of the state. Another pair of lessons involves the importance of institutional location. It makes a significant difference whether the structures of state-society synergy are developed within already existing public agencies/programs or within newly created ones. As seen in the cases of Mexico’s electoral institute, Porto Alegre’s participatory budgeting process and Mexico’s Municipal Funds program, it is much easier to institute innovative reforms within new agencies and programs that have not yet consolidated a tradition of top-down bureaucratic politics. The case of the women’s police stations offers a negative example of the same issue. One of the major reasons for the limited success of these police stations has been the fact that they remain as part of the police bureaucracy that they were designed to question in the first place. At the same time, state-society synergy seems to be more easily stimulated within autonomous agencies than within the massive bureaucratic apparatus of the executive branch. Both Porto Alegre and the IFE provide fascinating examples of the promise of constructing independent entities responsible for aspects of public management grounded in state-society synergy. These case studies also show that decentralization alone does not automatically facilitate the increased participation of civil society or an improvement in the accountability of government. Although devolution and decentralization are important because they bring government closer to the people, if carried out blindly, they tend to reinforce inequalities both within the newly autonomous local units as well as between them. Local power holders are allowed to run free, and underprivileged localities are abandoned to their own devices. Decentralization is only productive if the center remains responsible for the supervision and coordination of activities in the local units. This is demonstrated above by the experiences of school and police reform in Chicago, the Municipal Funds program in Mexico and the participatory budgeting process in Porto Alegre. In sum, this text has revealed that active societal participation can be an absolutely crucial resource in the strengthening of bureaucracies throughout the developing world. This is an important conclusion, but it also brings up a series of new questions: What concrete tools can be offered to state reformers? What are good entry points for the application of such strategies? How can we assure that the participatory mechanisms implemented are sustainable and go beyond the occasional peaks in popular participation that characterize normal politics? The best way to find answers to these and other such follow-up questions is to speak directly with those government innovators and civil society leaders and organizations who have participated in the design and implementation of successful experiences of state-society synergy. Hands-on experience is the best source for practical lessons. Nevertheless, from the above discussion, some general rules of thumb for World Bank personnel can be extracted: 1. Marketization is not the only way to tap into the energy of society. Privatization and subcontracting may be highly effective for some policy areas at specific times and places. Nevertheless, it is often even more fruitful to invite society into the state instead of sending sections of the state off to society. The participative school of the New Public Management (NPM) frequently can be even more effective than the market school (Peters, 2001). The specific advantages of state-society synergy over marketization are: a. It retains the comparative advantage that the state has over the market in the provision of public goods, natural monopolies, basic necessities, and goods that require long term planning and development. b. It keeps transaction costs to the minimum by permitting the focused coordination of multiple programs with parallel goals.


WORLD BANK WORKING PAPER c. It allows citizens to ground their demands for effective service delivery in rights-based discourse (as “citizens”), instead of exclusively in the language of consumer protection (as “consumers”). d. It avoids the inequality producing effects of market based service delivery. State reformers should think twice before assuming that marketization is the best and only way to apply the NPM. Careful attention needs to be put on the type of good or service being provided, the increase in transaction costs marketization might provoke, the possible loss of strength in the accountability signal when citizens are replaced with consumers, and the potential for increases in inequality that can arise from marketization. 2. Transparency is key, but not enough on its own. Opening up the dark chambers of the state to the eyes of the public is a major move forward, but it is only a first step. Governments cannot expect information provision to single handedly and spontaneously generate the positive feedback loops between state and society outlined in the above case studies. Governments need to be encouraged to directly stimulate the participation of society and to institutionalize mechanisms of state-society relations. Otherwise, the only actors who will actually put to use the new information are journalists, academics and non-profit organizations. Although these groups are indeed crucial in maintaining accountability, the above case studies show that there is a qualitative forward leap when the population at large, and the poor in particular, are directly involved in enforcing accountability as well. 3. The best entry points are those where there are previously existing social demands and practices surrounding a particular accountability issue. The participatory budgeting initiative in Porto Alegre was initially proposed by the Union of Residents’ Association, the system of electoral observers in Mexico was first put into practice by Alianza Cívica, the Local School Councils in Chicago were actively demanded by poor communities throughout Chicago, the successful implementation of the Municipal Funds Program in Oaxaca depended on its ability to tap into local cooperative traditions and horizontal social capital, the Toxics Release Inventory was so successful because of the active environmentalist community in the United States, the initial successful stage of the women’s police stations in Brazil arose out of social pressure and was modeled on already existing societal practices, and the actions of RKS and MKSS in India have been fundamental in increasing accountability in the subcontinent. This does not mean that state-society synergy is only effective where there is a highly developed middle class, a lack of social conflict or in relatively simple policy areas. The above cases reveal that normal citizens are extraordinarily capable of participating in highly complex tasks and that there is no need for civil society already to be well behaved or unified prior to the implementation of innovative participative mechanisms. What this does say is that such mechanisms are most effective when state reformers respond to demands articulated by society and actively work with society to design and implement the participatory schemes. 4. Once initiated, the best way to assure the sustainability of a participatory framework is through its full institutionalization. There are three different levels at which participatory mechanisms can be institutionalized. First, participatory mechanisms can be built into the strategic plans of government agencies and rules and procedures can be mandated that require street-level bureaucrats to consult or otherwise engage with societal actors. Second, specific government agencies can be created that have the goal of assuring societal participation in government activities or act as a liaison in change of building links with societal actors. Third, participatory mechanisms can be inscribed in law, requiring individual agencies or the government as a whole to involve societal actors at specific moments of the public policy process. Although the first level of institutionalization is more or less widespread in LCR and the second level is relatively common, the third level is extremely rare. There are some




exceptions, including some of the above cases as well as Bolivia’s Law of Popular Participation (Oxhorn, 2001) and Mexico City’s Law of Citizen Participation (Mellado Hernández, 2001). Nevertheless, these exceptions only prove the rule that participatory mechanisms are usually vastly under-institutionalized, depending too much on the ingenuity and good will of individual bureaucrats. Why this is the case is more or less evident. Law making under democratic conditions involves the messy process of legislative bargaining and a full role for political parties. State reformers and multilateral agencies tend to shy away from such arenas, especially when they are dominated by opposing parties or factions. Therefore, reformers usually settle for executive procedures, special agencies or innovative individual bureaucrats to carry out their participative strategies. This is a mistake. As the above case studies shows, if dealt with in a creative fashion partisanship can be just as effective as isolation in the search for effective accountability mechanisms. It is absolutely crucial to involve political parties and the legislature in order to fully institutionalize participative mechanisms through the law. 5. Legal institutionalization is important but it is not enough. Both the executive branch as a whole as well as the project directors and street-level bureaucrats need to be committed to the importance of the participatory mechanisms. Otherwise, the law risks becoming dead letter. This is particularly crucial in situations where the government as a whole has a low level of legitimacy. In such cases the immediate assumption of the population is that participatory mechanisms are only sophisticated new forms of reverse vertical accountability or manipulation of the people by the government. This can easily create a negative feedback loop between state and society that can easily undermine even the most sophisticated and powerful new law. There are a few strategies that state reformers can use in order to prevent such a downward spiral from occurring. First, individuals with highly respected track records, well developed administrative capacities and creative intelligence should be chosen to lead the implementation of the law in the executive branch. Second, civil society organizations should be invited to help train both government bureaucrats and societal actors in the art of participation and group decision making from the very beginning of the implementation phase. Third, a concerted attempt should be made to involve the widest diversity of perspectives possible from organized and unorganized civil and political society.




his paper has argued that the active involvement of civil society and the strengthening of the state apparatus are not mutually exclusive or even contradictory initiatives. This is the central idea of “state-society synergy” as a concept. If institutions are properly designed, a virtuous cycle that reinforces both state and society is possible. This is particularly important to emphasize today given the thrust of much of the new public management literature that proposes the devolution of state responsibilities to social actors via the market. The above cases show that the empowerment of society does not have to pass through the weakening or reduction of the size or the capacities of the state (see Annex A—Summary of Case Studies). Indeed, they show that exactly the opposite is true. Both state and society are best strengthened by establishing mechanisms that allow each side to stimulate the other, thus creating a positive feedback loop that can lead to significant improvements in governance in the short, medium and long terms (see Annex B-Summary of Lessons for World Bank Staff). In order to complement the research included in the present study, the World Bank could play a useful role by continuing to contribute to research in this area two different directions. First of all, by contacting the individuals and organizations involved in the design and implementation of the above-mentioned and other successful examples of state-society synergy for accountability, the general lessons included above could be complemented with more specific, technical advice needed in order to reproduce these examples as best practices in other contexts. In addition, as the Bank continues to collect information on the implementation of similar initiatives throughout the globe, a complete database could be created that can be used to test hypotheses using statistical and quantitative methods.








CASES Participatory Budgeting: 10% of budget/Full opening to society. Three levels of participation (neighborhood, district, citywide). Weighted pro-poor voting system. Active involvement of society at the design stage. Facilitates exit from clientelistic groups. Government outreach (paid community organizers, district centers, etc.). Independent Electoral Institute: Autonomous citizen-run General Council. Public meetings and party representation. Active involvement of society at design stage. Monitored by other government accounting agencies. Strong civil service. Training of 1 million volunteers Unprecedented fairness of 2000 elections.



Participatory Budgeting Porto Alegre, Brazil

1) Inefficiency, corruption and lack of equity in the deliverys of infrastructure and social services.

Rapid improvement in infrastructure and public services in poor areas. Promotes development of new social actors. Active participation of the poorest. Need to balance decentralization and centralized supervision


2) Clientelism within societal organizations.

Federal Electoral Institute Mexico

1) Capture of bureaucracy by a single party/interest group.

2) Widespread corruption during service delivery.

High levels of trust and legitimacy in the new institution. Active participation of the poorest. Value of the saturation of partisanship. New independent agencies as strategic levers for state reform.

Police and School Reform Chicago Active participation of the poorest. Crucial importance of institutionalization. Need to balance decentralization and centralized supervision

Ineffective service delivery Rapid improvement in service delivery.

Local School Councils: Autonomous citizen-run supervisory body. Open elections of members. Highly institutionalized power (budget, hiring/firing of principal, school improvement plan, etc.) Result of public pressure Community Policing: Involvement of citizens in identification of problems and solutions. Not fully institutionalized. Solidarity Committees in Oaxaca: Grounded in pre-existing social practices Publicity of distributive formula. Requirements of within municipality resource distribution. Toxics Release Inventory: “Populist Maxi-Min Regulation.” Pro-active state efforts to stimulate participation. Active environmentalist community. Importance of “horizontal social capital” Importance of transparency Active participation of the poorest. Problem of “top-down” planning Problem of lack of institutionalization.

Decentralization and Rural Development Mexico

1) Inefficiency, corruption and lack of equity in the delivery of infrastructure.


2) Clientelism within societal organizations.

Toxics Release Inventory United States


Legal loopholes and inefficient “top-down” enforcement of law and government regulations

Information can be powerful when put into action by societal actors and interest groups. Synergy between societal mobilization and government regulation. (continued)


Continued 48


Women’s Police Stations

Brazil Grounded in previous social practice and stimulated by social mobilization. Societal groups train public servants. Social Auditing (MKSS): Independent investigation of government expenditures. Public hearings. Public protests. Informal “Vigilance Committees” (RKS): Autonomous supervision of ration shops. Publicity of prices and samples.

Unfair treatment of a marginalized group by a government agency.

Women’s Police Stations: Separate police stations run by women, with women and for women.

Decay of link between civil society and the state led to return of old practices. Importance of institutional location.


Grass-Roots Anti-Corruption Initiatives


1) Inefficiency, corruption and lack of equity in the delivery of infrastructure and social services.

Active participation of the poorest in highly abstract activities. More participative alternative to relatively tame measures like the “Bangalore Scorecard.” Importance of institutionalization of participatory mechanisms. Society driven accountability does not necessarily depend on the presence of reformist bureaucrats.

2) Captured previously existing participatory mechanisms.






Starting Point Design 1) Information and transparency should be complemented by outreach and active stimulation of societal participation. 2) Encourage training of public officials by societal actors. 3) Reform is easier in new and independent agencies. 4) Decentralization should be complemented by centralized supervision. 5) Select highly capable and reputable individuals to lead state-society synergy initiatives. Implementation

Other 1) Participative mechanisms should be seen as complements, not replacements, for professional, wellfinanced bureaucracies. 2) The multiplication of plurality and partisanship is often more effective than isolation.

1) Consider comparative advantages of participation versus marketization.

1) Work closely with societal actors in the design of participative mechanisms.

2) Respond to societal protest and demands.

3) Search for previously existing societal practices.

2) Fully institutionalize mechanisms through the law (engage with the legislature and political parties).

3) Full opening (“diagonal accountability”) holds much more potential than partial opening (“society-driven horizontal accountability,” “direct vertical accountability”).


Abers, Rebecca. 1998. “From Clientelism to Cooperation: Local Government, Participatory Policy, and Civic Organizing in Porto Alegre, Brazil.” Politics & Society 26(4):511–538. Almond, Gabriel and Sidney Verba. 1963. The Civic Culture. Princeton: Princeton University Press. Arellano, David and Juan Pablo Guerrero. 2000. “Stalled Administrative Reforms of the Mexican State.” Working Paper #88, División de Administración Pública, Centro de Investigación y Docencia Económicas (CIDE), Mexico City. Avritzer, Leonardo. 2000. “Civil Society, Public Space and Local Power: A Study of the Participatory Budget in Belo Horizonte and Porto Alegre.” IDS-Ford Web Site www.ids.ac.uk/ids/civsoc/index.html Baiocchi, Gianpaolo. 2001. “Participation, Activism, and Politics: The Porto Alegre Experiment and Deliberative Democratic Theory.” Politics & Society 29(1):43-72. Becerra, Ricardo, Pedro Salazar, and José Woldenberg. 2000. La Mecánica del Cambio Político en México. Mexico City: Cal y Arena. Behn, Robert. 2001. Rethinking Democratic Accountability. Washington: Brookings Institution. Cárdenas Gracia, Jaime. 2000. Estudios Jurídicos en Torno al Instituto Federal Electoral. Mexico City: Universidad Nacional Autónoma de México. Chhibber, Anthony. 1997. “The State in a Changing World.” Finance and Development 34(3):17-20. Cornelius, Wayne, Ann Craig, and Jonathan Fox, eds. 1994. Transforming State-Society Relations in Mexico: The National Solidarity Strategy. La Jolla: Center for U.S.-Mexican Studies. Dresser, Denise. 1994. “Bringing the Poor Back In: National Solidarity as a Strategy of Regime Legitimation.” In Wayne Cornelius, Ann Craig, and Jonathan Fox, eds. Transforming State-Society Relations in Mexico: The National Solidarity Strategy. La Jolla: Center for U.S.-Mexican Studies. Evans, Peter. 1995. Embedded Autonomy, Berkeley: University of California Press.




______. 1996a. “Introduction: Development Strategies Across the Public-Private Divide.” World Development 24(6):1033-1037. ______. 1996b. “Government Action, Social Capital and Development: Reviewing the Evidence on Synergy.” World Development 24(6):1119-1132. Fearon, James. 1999. “Electoral Accountability and the Control of Politicians: Selecting Good Types versus Sanctioning Poor Performance.” In Adam Przeworski, Susan Stokes, and Bernard Manin, eds., Democracy, Accountability and Representation. New York: Cambridge University Press. Fox, Jonathan. 1994. “The Difficult Transition from Clientelism to Citizenship: Lessons from Mexico.” World Politics 46(2):151-84. ______. 2000. “Civil Society and Political Accountability: Propositions for Discussion.” Presented at the conference “Institutions, Accountability and Democratic Governance in Latin America,” The Helen Kellog Institute for International Studies, Notre Dame University, May 8th. ______. 2002. “La Relación Recíproca entre la Participación Ciudadana y la Rendición de Cuentas” Política y Gobierno 9(1):95-133. Fox, Jonathan and Josefina Aranda. 1996. Decentralization and Rural Development in Mexico: Community Participation in Oaxaca’s Municipal Funds. La Jolla: Center for U.S.-Mexico Studies. Fung, Archon. 2001. “Accountable Autonomy: Toward Empowered Deliberation in Chicago Schools and Policing.” Politics & Society 29(1):73-103. Fung, Archon and Dara O’Rourke. 1998. “Reinventing Environmental Regulation From the Grassroots Up: Explaining and Expanding the Success of the Toxics Release Inventory.” web.mit.edu/dorourke/www/pub.html. Edited version published in Environmental Management 25(2):115-127. Fung, Archon and Eric Olin Wright. 2001. “Deepening Democracy: Innovations in Empowered Participatory Governance.” Politics & Society 29(1):5-41. Goetz, Anne Marie and Rob Jenkins. 2001. “Hybrid Forms of Accountability: Citizen Engagement in Institutions of Public-Sector Oversight in India.” Public Management Review 3(3):363-383. Grindle, Merilee. 1996. Challenging the State: Crisis and Innovation in Latin America and Africa. New York: Cambridge University Press. Haggard, Stephan and Robert Kaufman. 1995. The Political Economy of Democratic Transitions. Princeton: Princeton University Press. IFE (Mexico’s Federal Electoral Institute). 2000a. The Mexican Electoral Regime and the Federal Elections of the Year 2000. Mexico City: IFE. ______. 2000b. Proceso Electoral Federal 2000: Datos y Numeralia Más Importantes. Mexico City: IFE. ______. 2000c. “Características del Servicio Profesional Electoral.” Mexico City, IFE. Available at http://www.ife.org.mx/wwwife/servici1.htm. Jackson, Brooks. 1990. Broken Promise: Why the Federal Election Commission Failed. New York: Priority Press. Jenkins, Rob and Anne Marie Goetz. 1999a. “Accounts and Accountability: Theoretical Implications of the Right-to-Information Movement in India.” Third World Quarterly 20(3): 603-622. ______. 1999b. “Constraints on Civil Society’s Capacity to Curb Corruption: Lessons from the Indian Experience.” IDS Bulletin 30(4):39-49. LaForge, Amanda. 1996. “The Toothless Tiger-Structural, Political and Legal Barriers to Effective FEC Enforcement: An Overview and Recommendations.” Administrative Law Journal 10(351). Lujambio, Alonso. 2002. “Dinero y Democratización: El Financiamiento y la Fiscalización de los Partidos Políticos en la Transición Mexicana a la Democracia, 1988-2000.” Mexico City: IFE. See http://www.ife.org.mx/wwwcai/seminario/ponencias/al.html




Mellado Hernández, Roberto. 2001. Participación Ciudadana Institucionalizada y Gobernabilidad en la Ciudad de México. Mexico City: Plaza y Valdés. Molinar Horcasitas, Juan. 1996. “Renegotiating the Rules of the Game: The State and Political Parties.” In M. Serrano and V. Bulmer-Thomas, eds. Rebuilding the State: Mexico After Salinas. London: Institute of Latin American Studies. University of London. Navarro, Zander. 1998. “Participation, Democratizing Practices and the Formation of a Modern Polity—the Case of ‘Participatory Budgeting’ in Porto Alegre, Brazil (1989-1998).” Development 41(3):68-71. Nelson, Sara. 2002. “Constructing and Negotiating Gender in Women’s Police Stations in Brazil.” In Jennifer Abbassi and Sheryl L. Lutjens, eds., Rereading Women in Latin America and the Caribbean: The Political Economy of Gender. Rowman and Littlefield. Nylen, William. 2002. “Testing the Empowerment Thesis: The Participatory Budget in Belo Horizonte and Betim, Brazil.” Comparative Politics 34(2):127-145. McCubbins, Mathew and Schwartz, Thomas. 1984. “Congressional Oversight Overlooked: Police Patrols versus Fire Alarms.” American Journal of Political Science 28(1):165-179. O’Donnell, Guillermo. 1994. “Delegative Democracy,” Journal of Democracy 5(1). ______. 1999. “Horizontal Accountability in New Democracies.” In Andreas Schedler, Larry Diamond, and Marc Plattner, eds., The Self-Restraining State: Power and Accountability in New Democracies. Boulder: Lynne Rienner. Olvera, Alberto. 2001. Movimientos Sociales Prodemocraticos, Democratización y Esfera Pública en México: El Caso de Alianza Cívica. Veracruz: Universidad Veracruzana. O’Rourke, Dara. 2001. “Community-Driven Regulation: Towards an Improved Model of Environmental Regulation in Vietnam.” <web.mit.edu/dorourke/www/pub.html>. Forthcoming in Peter Evans, ed., Livable Cities: The Politics of Urban Livelihood and Sustainability. University of California. Oxhorn, Phillip. 2001. La Construcción del Estado por la Sociedad Civil: La Ley de Participación Popular de Bolivia y el Desafío de la Democracia Local. Documentos de Trabajo del INDES, Washington: IDB. Paul, Samuel. 1992. “Accountability in Public Services: Exit, Voice and Control.” World Development 20(7):1047-1060. Pope, Jeremy. 2000. Confronting Corruption: The Elements of a National Integrity System. Berlin: Transparency International. Pozas Horcasitas, Ricardo. 1996. “La Observación Electoral: Una Modalidad de la Militancia Ciudadana.” Revista Mexicana de Sociología 59(2):23-40. Przeworski, Adam, Susan Stokes, and Bernard Manin, eds. 1999. Democracy, Accountability and Representation. New York: Cambridge University Press. Putnam, Robert. 1993. Making Democracy Work. Princeton: Princeton University Press. Rose-Ackerman, Susan. 1999. Corruption and Government: Causes, Consequences and Reform. New York: Cambridge University Press. Sadek, Maria Teresa and Rosangela Batista Cavalcanti. 2000. “The Public Prosecution and the Efficacy of Accountability Mechanisms.” Presente`d at the conference “Institutions, Accountability and Democratic Governance in Latin America,” The Helen Kellog Institute for International Studies, Notre Dame University, May 8th. Santistevan de Noriega, Jorge. 2000. “The Ombudsman Institution and Accountability in Societies in Transition.” Presented at the conference “Institutions, Accountability and Democratic Governance in Latin America,” The Helen Kellog Institute for International Studies, Notre Dame University, May 8th. Schedler, Andreas. 1999a. “Conceptualizing Accountability.” In Andreas Schedler, Larry Diamond, and Marc Plattner, eds., The Self-Restraining State: Power and Accountability in New Democracies. Boulder: Lynne Rienner.



______. 1999b. “Las Comisiones y la Pirámide: Notas Sobre la Conflictiva Recentralización del Poder en el IFE.” Política y Gobierno 6(1):187-222. Schedler, Andreas, Larry Diamond, and Marc Plattner, eds. 1999. The Self-Restraining State: Power and Accountability in New Democracies. Boulder: Lynne Rienner. Sherman Grant II, Don. 1997. “Allowing Citizen Participation in Environmental Regulation: An Empirical Analysis of the Effects of Right-to-Sue and Right-to-Know Provisions on Industry’s Toxic Emissions.” Social Science Quarterly 78(4):849-873. Smulovitz, Catalina. 2000. “How Can the Rule of Law Rule? Cost Imposition Through Decentralized Mechanisms.” Forthcoming in José María Maravall and Adam Przeworski, eds., Democracy and the Rule of Law. Cambridge University Press. Smulovitz, Catalina and Enrique Peruzzotti. 2000a. “Societal and Horizontal Controls: Two Cases About a Fruitful Relationship.” Presented at the conference “Institutions, Accountability and Democratic Governance in Latin America,” The Helen Kellog Institute for International Studies, Notre Dame University, May 8th. ______. 2000b. “Societal Accountability in Latin America.” Journal of Democracy 11(4):147-158. Soederberg, Susanne. 2001. “From Neoliberalism to Social Liberalism: Situating the National Solidarity Program Within Mexico’s Passive Revolutions.” Latin American Perspectives 28(3):104-123. Stigler, George. 1971. “The Theory of Economic Regulation.” Bell Journal of Economics and Management Science 2:3-21. Tendler, Judith. 1997. Good Government in the Tropics. Baltimore: Johns Hopkins. U.S. Environmental Protection Agency. 1995. “National Air Quality and Emissions Trends Report, 1995.” Varshney, Ashutosh. 1999. “Democracy and Poverty.” Paper for the Conference on World Development Report-2000, organized by the U.K. Department for International Development, and the Institute of Development Studies, Sussex, Caste Donnington, England, August 15-16. Vellinga, Menno, ed. 1998. The Changing Role of the State in Latin America. Boulder: Westview. Wallis, Joe and Brian Dollery. 2001. “Government Failure, Social Capital and the Appropriateness of the New Zealand Model for Public Sector Reform in Developing Countries.” World Development 29(2):245263. Weingast, Barry. 1997. “The Political Foundations of Democracy and the Rule of Law,” American Political Science Review 91(2). Woldenberg, José. 2001. “Lessons from Mexico.” Journal of Democracy 12(2):151-156. ______. 2002. La Construcción de la Democracia. Mexico City: Plaza y Janés. World Bank. 1997. World Development Report: The State in a Changing World. Oxford University Press.

State-Society Synergy for Accountability is part of the World Bank Working Paper series. These papers are published to communicate the results of the Bank’s ongoing research and to stimulate public discussion. The central question guiding this study is: How can the relationship between the state and society be transformed from a process of particularistic demands into a healthier engagement that produces policy outcomes serving the public interest? This paper first surveys the literature on accountability and establishes a categorization of the different ways by which civil society can interact with the state. It then explores in detail seven case studies of successful experiences of statesociety synergy for accountability. The studies are drawn from a wide range of different contexts (Brazil, India, Mexico, the United States) and from a variety of different areas of government activity including corruption control, environmental regulation, poverty reduction, election monitoring, infrastructure provision, school reform, and police reform. The paper concludes with a series of lessons for World Bank staff on how best to initiate, design, and implement successful accountability mechanisms grounded in state-society synergy. World Bank Working Papers are available individually or by subscription, both in print and online.

ISBN 0-8213-5831-6 THE WORLD BANK
1818 H Street, NW Washington, DC 20433 USA Telephone: 202 473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank.org