Co-governance for Accountability

8/11/2019 Co-governance for Accountability

1/17

Co-Governance for Accountability: Beyond

Exit and Voice

JOHN ACKERMAN *University of California, Santa Cruz, CA, USA

Summary. This article explores the theory and practice of societys participation in strengtheninggovernment accountability. It argues that both exit solutions based in marketization and voicesolutions grounded in coproduction, social protest or consultation are insufficient. The best wayto tap into the energy of society is through co-governance, which involves inviting social actorsto participate in the core activities of the state. By way of demonstration, the article examines casestudies from a wide range of contexts (Brazil, Mexico, the United States, India) and policy areas(poverty reduction, infrastructure provision, school reform, electoral administration, policereform). 2003 Elsevier Ltd. All rights reserved.

Key words accountability, governance, participation, civil society, americas

1. INTRODUCTION

In the contemporary world of second-wavereforms and democratic transitions, the con-struction of an honest, efficient and effective

government apparatus has moved to the top ofthe international policy agenda. Academics andpolicymakers alike now readily accept thatgood governance and accountability are nec-essary preconditions for successful economicdevelopment (Bresser & Spink, 1999; Evans,1995; Grindle, 1996; Manzetti, 2003; Pope,2000; Rose-Ackerman, 1999; Schedler, Dia-mond, & Plattner, 1999; Vellinga, 1998; WorldBank, 1997).

Unfortunately, most contemporary pro-accountability reforms exclude the voice of

societal actors. On the one hand, old publicmanagement strategies such as civil-servicereform and strict procedural monitoring areexplicitly designed to insulate the state fromsociety. On the other hand, New Public Man-agement (NPM) policies such as managedcompetition and performance contracts alsokeep society far away from the core activities ofthe state. Although NPM does have a partici-pative or social control current within it(Bresser & Cunill, 1999; Peters, 2001), this isusually marginalized in favor of marketization

strategies. Marketization itself allows citizensto let their opinions be known through exitoptions, but it prohibits their active participa-

tion in government. Indeed, recent studies haveshown that such policies may even underminecommunity organization and social capital inthe developing world (Cunill, 2000; Wallis &Dollery, 2001).

The voices of the poor (Narayan & Pet-esch, 2002) usually find their way back in,but they are often left speaking into a void.Governments and international developmentagencies have recently moved participatorydevelopment up their discursive agendas (c.f.UNDP, 2002; World Bank, 2003), but actualpractice has lagged far behind. Participation isusually seen to be important only insofar as itreduces government costs and responsibilities.It suddenly appears to be practical andattractive when governments can offload ser-

vice delivery to nongovernmental organizations(NGOs) and community groups or convincelocal residents to donate volunteer labor ormaterials. The direct involvement of citizens

World Development Vol. 32, No. 3, pp. 447463, 2004 2003 Elsevier Ltd. All rights reserved

Printed in Great Britain0305-750X/$ - see front matter

doi:10.1016/j.worlddev.2003.06.015www.elsevier.com/locate/worlddev

* I would like to thank Jonathan Fox, Guillermo

ODonnell, Walter Goldfrank, Susan Rose-Ackerman,

Catalina Smulovitz, Erendira Sandoval, Ernesto Isunza,

Kathy Bain and Patti Petesch for their extremely helpful

comments and support at different stages of the research

on this article. I am also indebted to the insightful cri-

tiques and suggestions made by the two anonymous

reviewers from World Development. Final revision

accepted: 18 June 2003.

447


2/17

and societal groups in the core functions ofgovernment continues to be extremely rare.

This article argues that the opening up of thecore activities of the state to societal partici-pation is one of the most effective ways to

improve accountability and governance.Through an exploration of case studies from awide variety of contexts (Brazil, Mexico, theUnited States, India) and policy areas (povertyreduction, infrastructure provision, schoolreform, electoral administration, police reform)the article shows that state reformers shouldmove beyond strategies based on exit andeven voice (Hirschman, 1970; Paul, 1992) toestablish spaces of full co-governance withsociety. Instead of sending sections of the stateoff to society it is often more fruitful to invite

society into the inner chambers of the state.Section 2 below gives an overview of the lit-

erature on accountability and society. It beginsby defining accountability and discussing thevarious ways it can be enforced. It then focuseson a wave of fascinating recent works on soci-etys pro-accountability role by authors such asAvritzer, Cunill, Evans, Fox, Goetz, Jenkins,Isunza, Paul, Peruzzotti, and Smulovitz.Finally, it defines and proposes the concept ofco-governance for accountability.

Section 3 includes various case studies of

successful examples of co-governance foraccountability. The emphasis is on successfulcases since, as Judith Tendler has pointed out,the mainstream donor communitys adviceabout public-sector reform arises from a liter-ature that looked mainly at poor perfor-mance. . . This means that countries and theexperts that advise them have few models ofgood government (Tendler, 1997, p. 2,emphasis in original). Development profes-sionals are acutely aware of the ways thatgovernments fail. There is a need for the sus-

tained study of successful government innova-tions in order to inspire and direct positiveaction.

The cases are organized according to the levelat which state actors have opened themselves upto and encouraged the participation of civilsociety in the structuring of accountabilityarrangements. The section begins with the caseof participatory budgeting in Porto Alegre andmoves on to the case of Mexicos Federal Elec-toral Institute. It then turns to the case studies ofpolice and school reform in Chicago and decen-

tralization and rural development in Mexico. Itends by discussing a pair of examples of socialauditing from India. Finally, the article con-

cludes with a summary of the principal lessonsfor institutional reformers and developmentprofessionals interested in strengthening govern-ment accountability through the involvementsociety.

2. ACCOUNTABILITY AND SOCIETY

Good government does not emerge sponta-neously or naturally out of the good heartsof individual bureaucrats and politicians. It isthe result of a tough, and often conflict-ridden,process of institutional design. The principleelement that assures good government is theaccountability of public officials. This involvesboth answerability, or the obligation of public

officials to inform about and to explain whatthey are doing (Schedler, 1999a, p. 14) andenforcement, or the capacity of accountingagencies to impose sanctions on powerholderswho have violated their public duties(Sche-dler, 1999a, p. 14). Although some individualofficials may never need institutional structuresto assure their commitment to the public good,most do need it at least some of the time. Theonly way to guarantee good government is byinstitutionalizing powerful accountabilitymechanisms that hold every public official

responsible for his/her actions as a public ser-vant.

The celebration of free and fair elections isone of the most powerful pro-accountabilitymechanisms in existence. Through periodicelections, political leaders who work for thecommon good are supposed to be reelected,and leaders who use public office for particu-laristic ends are supposed to be removed fromoffice. Nevertheless, there are both structuraland contextual problems with elections in thecontemporary world.

There are at least three different structuralproblems with elections as accountabilitymechanisms. First, elections only hold electedofficials accountable. The vast majority ofpublic officials are appointed bureaucrats whoare not directly accountable to the publicthrough the electoral process. Second, becauseelections only occur once every few years andforce an incredible diversity of opinions andevaluations together into a single ballot, it isvirtually impossible for elections to give clearaccountability signals to individual office hold-

ers (Przeworski, Stokes, & Manin, 1999).Third, even if the accountability signal weresomehow clearly discernible, the fact that most

WORLD DEVELOPMENT448


3/17

politicians are elected by only a small portionof the population often forces politiciansto favor patronage, pork or corruption overinitiatives that would bring long-term benefitto the public as a whole (Varshney, 1999).

The situation appears even worse when wetake into account the empirical context. Dem-ocratic institutions are extremely weak in thecontemporary world. The effectiveness of elec-tions as mechanisms of sanction and control isweakened by the distance between political andcivil society, the clientelistic nature of manypolitical parties, the excess private funding forcandidates, and the lack of public informationabout the general workings of government andeven less information about the specificbehavior of individual office holders.

As a result, vertical accountability mech-anisms, such elections, that require governmentofficials to appeal downwards to the peopleat large have been complemented by hori-zontal accountability mechanisms that requirepublic officials and agencies to report side-ways to other officials and agencies within thestate itself. Guillermo ODonnell has definedhorizontal accountability in the followingmanner.

The existence of state agencies that are legally enabledand empowered, and factually willing and able, totake actions that span from routine oversight to crim-inal sanctions or impeachment in relation to actionsor omissions by other agents or agencies of the statethat may be qualified as unlawful (ODonnell, 1999,p. 38).

Examples of horizontal accountabilitymechanisms include human rights ombuds-man, corruption control agencies, legislativeinvestigative commissions and administrativecourts.

There has been a phenomenal growth of suchinstitutions throughout the world in recentyears. 1 Unfortunately, as with elections, thesemany new agencies of horizontal accountabilityare plagued by both structural and contextualproblems. Structural difficulties include theimpossibility of monitoring the almost infinitenumber of government actions (and inactions)as well as the political isolation that resultsfrom these agencies statutory or constitutionalindependence (Maor, 2004). Contextual diffi-culties include the lack of adequate funding,

limited enforcement capacity, the absence ofsecond order accountability (i.e., holdingaccounting agencies accountable) and the

overall weakness of the rule of law needed toenforce agency sanctions.

Fortunately, there is a third way to holdgovernment accountable. In addition to elec-tions and horizontal accountability agencies,

societal actors can directly oblige governmentactors to answer for their actions and sanctionthem for wrongdoing. Samuel Pauls (1992)article in World Development on Account-ability in public services: exit, voice and con-trol was one of the first to put forth such anagenda.

The traditional public accountability mechanismssuch as expenditure audits and legislative reviews seemunequal to the task of ensuring accountability forpublic services at the micro level. . . Public service

accountability will be sustained only when the hierar-chical control (HC) over service providers is rein-forced by the publics willingness and ability to exit[i.e., marketization] or to use voice [i.e., direct parti-cipation] (Paul, 1992, pp. 10471048).

Paul here simultaneously articulates theaccountability function of marketization strat-egies and, even more importantly, makes thecrucial argument that direct societal participa-tion is often even more effective than strategiesbased on exit.

The 1996 symposium on DevelopmentStrategies Across the PublicPrivate Divide,also published in World Development, thenexpanded and filled out this initial discussion ofsocietys pro-accountability role. This series ofarticles argued that state-society synergy(Evans, 1996a, 1996b) is one of the best ways tostrengthen government accountability. In sum,

the image of the good bureaucratcarefully insulatedfrom constituentshas its usefulness, but openness tothe role of the coproducer . . .may be the best wayto increase effectiveness and ultimately the best way

to preserve the integrity of increasingly besieged pub-lic institutions (Evans, 1996b, p. 1131).

For instance, in her contribution to thesymposium Elinor Ostrom documented howthe involvement of citizens in the planning andimplementation of water and sanitation pro-

jects greatly improved their effectiveness andreduced corruption in urban Brazil (Ostrom,1996). Wai Lams contribution showed howcommunity participation in irrigation programsin Taiwan has made service delivery much more

efficient and effective (Lam, 1996). This is alsoconsistent with Tendlers path-breaking workinGood Government in the Tropics (1997) which

CO-GOVERNANCE FOR ACCOUNTABILITY 449


4/17

demonstrated the salutary effects of the co-production of services by street-level bureau-crats and societal actors.

This literature performed a great service in sofar as it pushed academics and development

professionals to take societal participationseriously. No longer was society viewed as abother, a contaminant or as the source ofbureaucratic capture. States and societiescould be strengthened simultaneously. Never-theless, this first wave of writings was alsolimited in scope insofar as it tended toemphasize depoliticized forms of participation,circumscribed societal action to specific localservices and to the implementation phase ofgovernment projects, and left out the importantdiscussion of the legal institutionalization of

participative mechanisms.In recent years, two different currents of

research have arisen that expand on this earlierliterature. First, authors such as CatalinaSmulovitz, Enrique Peruzzotti, Nuria Cunilland Sylvio Waisbord have argued that morepolitical forms of societal participation such asmass mobilization, media exposes and the useof the courts are also effective ways for societyto improve government accountability. Smu-lovitz and Peruzzotti distinguish this form ofaccountability from the electoral and the hori-

zontal forms by calling it societal account-ability. They define this as,

a nonelectoral, yet vertical mechanism of control thatrests on the actions of a multiple array of citizensassociations and movements and on the media, ac-tions that aim at exposing governmental wrongdoing,bringing new issues into the public agenda, or activat-ing the operation of horizontal agencies (Peruzzotti &Smulovitz, 2000b, p. 150; 2002, p. 32).

For example, in their analysis of the social

response to two extra-judicial killings inArgentina, the authors have documented howthe combination of mobilization, legal actionand media exposure can effectively guaranteethat the judicial system operate impartially,even when the perpetrators are well connectedor even part of the government apparatus itself(Peruzzotti & Smulovitz, 2000a, 2000b, 2002).Waisbord has complemented this analysis byfocusing on the role of investigative journalistsand media scandals in obliging public servantsand politicians to be more accountable (Wais-

bord, 2000).Cunill (1997, 2000) follows this same line of

research but focuses more on the action of

citizens in general than on that of organizedcivil society. For her, the most important soci-ety driven pro-accountability mechanisms arelegal reforms such as popular referendum laws,administrative procedure acts that require

public consultations, amparo laws, andfreedom of information acts (Cunill, 2000, pp.2539). Since such laws open up the state to theaction of the common citizen they create spacefor the active enforcement of accountability bythe public.

This literature is a welcome addition to theaccountability debate since it obliges us to lookbeyond well-behaved local participation inspecific government projects to a more openlypolitical and even confrontational engagementwith the government apparatus as a whole.

Nevertheless, these writings still envision anddefend an armslength relationship betweenstate and society. As Cunill has written, co-management is irreconcilable with control. Theefficacy of [social control] is directly dependenton the independence and the autonomy thatsocietal actors maintain with respect to stateactors (Cunill, 2000, p. 9-my translation).

The second alternative current of research ismore transgressive insofar as it explicitlyviolates the separation between state and soci-ety. For instance, Ernesto Isunza has recently

written about transversal accountability inwhich societal actors participate directly in theleadership and operation of state pro-account-ability agencies (Isunza, 2003). This parallelsAnne Marie Goetz and Robert Jenkinsdescription of the The New AccountabilityAgenda which emphasizes hybrid ordiagonal forms of accountability (Goetz &Jenkins, 2001, 2002a) in which vertical actorscarry out intrastate horizontal accountabilityfunctions. In a similar spirit, Leonardo Avritzerhas put forth the idea of participatory pub-

lics which occur when societal participatorypractices are taken up by and embedded withinthe state (Avritzer, 2002). Archon Fung andEric Olan Wright have also followed this line ofresearch in arguing for empowered participa-tory governance which expands the sphere ofdemocratic participation beyond formal elec-toral politics to involve society at large indeliberation over the design and operation offundamental government services such asschooling, policing, environmental protectionand urban infrastructure (Fung & Wright,

2001). In addition, Jonathan Fox has arguedfor an interactive approach to state-societyrelations which envisions the improvement of



5/17

accountability through the participation ofsociety in the core functions of government(Fox, 2000).

This group of authors goes beyond thecircumscribed participation implicit in the co-

production literature as well as the armslength action of the societal accountabilityliterature to posit a full co-governance foraccountability which confuses the boundarybetween state and society. In addition to co-producing specific services and pressuringgovernment from the outside, societal actorscan also participate directly in the core func-tions of government itself. This form of civilsociety participation is special because, asGoetz and Jenkins have written, it representsa shift towards augmenting the limited effec-

tiveness of civil societys watchdog function bybreaking the states monopoly over theresponsibility for official executive oversight(Goetz & Jenkins, 2001, p. 365). The presentarticle looks both to bring together these vari-ous texts into a coherent literature and todemonstrate the salience of this buddingtransgressive school of thought by offeringsome examples of how it works in practice.

3. CASE STUDIES

(a) Participatory budgeting in Porto Alegre,Brazil

The Porto Alegre city government representsone of the most effective schemes of state-society collaboration for accountability in thedeveloping world. Since 1989, when theWorkers Party (PT) first won the city govern-ment, Porto Alegre has placed spending deci-sions for over 10% of its annual budget in thehands of the people. Every year, more than

14,000 citizens in this city of 1.3 million par-ticipate in neighborhood meetings as well as 16regional and five thematic assemblies to setpriorities for government investment in infra-structure and basic social services. Eachassembly then elects two councilors to serve ona citywide Council of Participatory Budgeting(COP), the organ responsible for putting toge-ther the final citywide budget plan. At eachlevel of the process (neighborhood, district,citywide) decisions are made through intensenegotiation and the use of sophisticated

weighted voting systems designed to assure afair distribution of resources. At the end of theprocess, the proposed budget is then submitted

to the local legislature for final approval andpromulgation. During the following year, theregional and thematic assemblies, councilorsand neighborhood groups evaluate the previousyears negotiation process and monitor the

implementation process of the previous yearsbudget.

The participatory budgeting (PB) process isan excellent example of co-governance foraccountability. Normal citizens are involveddirectly in the planning and supervision ofpublic spending, activities normally under theexclusive purview of public officials. Thisarrangement is clearly a step beyond both theco-production and the societal account-ability models of civil society participation.Instead of trying to influence policy from the

outside or only at the local community level,the citizens of Porto Alegre are invited insidethe governmental apparatus itself.

This arrangement has had an importantimpact on accountability. First, it has drasti-cally reduced the possibilities and incentives forcorrupt behavior on behalf of bureaucrats.Each neighborhood and region is informed asto the exact amount of funds that will beinvested in which products and services in itsarea and, even more importantly, since thecitizens themselves participate in designing the

budget, they feel they have a personal stake inmaking sure the government complies with itscommitments (Navarro, 1998, pp. 7071).

Second, the budgeting process reduces thepolitical use of public funds by opening upalternative channels for the participation ofcivil society. The crucial element is the entirelyopen and public nature of the budget assem-blies. Any adult can attend, speak and vote inthe assemblies (Avritzer, 2000, p. 18). More-over, it is easy to form a new group and therebygain access to special organizational represen-

tation. This leads to easy exit options formembers of clientelistic groups where voiceis not an effective form of protest.

Third, PB limits the capture of state institu-tions by wealthy interests. Popular participa-tion itself does this by replacing the power ofmoney with the power of voice. In addition, thespecial design of Porto Alegres system rein-forces this tendency even further. The algo-rithm used for determining budget prioritiesintentionally tilts investments towards poorerneighborhoods. Due to this built in pro-poor

bias, the same need presented by two neigh-borhoods is much more likely to be imple-mented in the poorer one than the wealthier



6/17

one (Baiocchi, 2001, p. 48). Marquetti (2003)has recently empirically demonstrated the sig-nificant redistributive impact of the PB.

The origins of this successful pro-account-ability arrangement can be found in society.

First, the idea of instituting a participatorybudget had its origins within civil society. Itwas the Union of Residents Associations ofPorto Alegre (UAMPA) that first advocatedthe introduction of such a mechanism in thecity in 1986 (Avritzer, 2002, p. 145). Second,Avritzer documents how the expression par-ticipatory budget did not exist in the PTselectoral platform for city government in 1988.The design of todays PB arrangement onlyarose after a period of intense negotiation andparticipation between the new government and

civil society groups (Avritzer, 2000, p. 9).Third, the particular institutional form devel-oped by the Porto Alegre government waslargely modeled on already existing practices ofdeliberation and negotiation in civil society(Navarro, 2002).

The Porto Alegre experience offers manylessons for pro-accountability state reformers.First, poor, uneducated people can and doeffectively participate in the core activities ofgovernance. Abers (1998) documents that whilein 1991 29% of Porto Alegres residents earned

three times the minimum wage or less, 45% ofthe budget participants fit this profile. Theunderprivileged not only actively participate,but they even participatemore, relative to theirsize in the population, than better off groups.

Second, governments can only get back asmuch as they put in to efforts to activate civilsociety participation for accountability. InPorto Alegre, citizens are taken out of theirusual role as only advisors or informationproviders to government projects and thrustdirectly into the decision-making process itself.

In addition, the government actively encour-ages the participation of unorganized citizensthrough the use of government employedcommunity organizers (Abers, 1998, p. 514). Ashas been shown to be the case in other citiesthat have tried participatory budgets, withoutsuch full involvement by the government,participation schemes can easily end up onlystrengthening previously existing clientelisticnetworks and unbalanced intra-communitypower relations (Goldfrank, 2002; Nylen,2002). 2

Third, governments need to take civil societyinto account in the design of the participativemechanisms themselves. The PB did not spon-

taneously arise out of the minds of enlightenedbureaucrats. It originated in civil society, waspushed forward by social actors and was ulti-mately modeled on previously existing practicesin civil society by a new government that itself

consisted mostly of individuals who had madetheir careers as community and social activists.Participatory mechanisms usually hold themark of their birth.

Fourth, according to Fung and Wright(2001), the Porto Alegre experience is anexcellent example of how a healthy balance canbe struck between devolution and central-ized supervision and coordination. Althoughdevolution and decentralization are importantbecause they bring government closer to thepeople, if carried out blindly, they tend to

reinforce inequalities both within the newlyautonomous local units as well as betweenthem. Decentralization is only productive if thecenter remains responsible for the supervisionand coordination of the activities in the localunits.

(b) Mexicos federal electoral institute

Mexicos Federal Electoral Institute (IFE)stands out as another example of successfulco-governance for accountability. 3 The

principal activities of the IFE include organiz-ing federal elections, distributing public fundsto the political parties, monitoring the use ofboth public and private funds by the parties,checking for media bias in the coverage ofpolitical campaigns, putting together andcleaning up the official electoral roll, and run-ning public education campaigns (IFE, 2000a).The IFE actively involves societal actors at fivedifferent levels.

First, the IFE is run by an independent, ninemember citizen-run 4 General Council that

serves as both a special horizontal account-ability agency for electoral affairs and as theIFEs principal directive body (Schedler,1999b). Second, the meetings of the GeneralCouncil are public. The minutes and decisionsare widely publicized, reported on by themedia, and are available via the Internet. Third,one representative from each registered politi-cal party sits on the General Council. Theseparty representatives can fully participate in thediscussions of the General Council and haveaccess to most of the same information as the

councilors but do not have the power to voteon initiatives or decisions. Fourth, the IFEcouncils that are responsible for organizing and



7/17

supervising the federal elections at the statelevel are also citizen-run in so far as they areappointed by the General Council without anyformal interference from local or state govern-ments (Isunza, 2003).

Fifth, during its most important moment ofservice delivery, the organization of the fed-eral elections, the IFE recruits a huge army ofcitizen volunteers. During the months leadingup to the 2000 elections the IFE trained over800,000 citizen volunteers to run 113,423 poll-ing sites (Woldenberg, 2001). Each one of theparticipants receive two training coursesdesigned and implemented by the IFE. Inaddition, the IFE trains both national andinternational observers in the basics of electorallaw (Pozas, 1996). Finally, each political party

is permitted to send one representative to eachvoting booth on election day. In total, morethan one million citizens were mobilized in 2000to assure the realization of free and fair elec-tions.

Overall, the IFE has been remarkably suc-cessful. The lack of significant post-electoralprotests and mobilizations in the year 2000 wasunprecedented for a presidential election inMexico. In addition, the fact that there has notbeen a new electoral reform since 1996 is atestament both to the great breakthrough of

this reform and to the legitimacy that theinstitution continues to enjoy up through thepresent. Other than the 193342 and 196370periods, the seven years during 19962003marks the longest period the Mexican politicalsystem has gone without an electoral reformsince the promulgation of the Mexican consti-tution of 1917 (Molinar, 1996). Finally, theIFEs recent historic US$100 million fine of theParty of the Institutional Revolution (PRI) andits aggressive investigation of the irregularfinancing of the campaign of sitting president

Vicente Fox demonstrates its ability to stand upto even the most powerful interests.

The stimulant for the 1996 reform was thewidespread social unrest and demands fordemocracy that arose out of the economicmeltdown of 199495 as well as the Zapatistauprising. In addition, one of the most impor-tant influences on this electoral reform was theactivism of nonprofit electoral watchdoggroups. The leading group during this periodwasAlianza Cvica. For the 1994 elections, thisgroup mobilized over 12,000 national electoral

observers and 400 international observers,carried out its own parallel quick count ofthe electoral results, published a report on bias

in media coverage of the campaigns as well as aguide for electoral observers and a final evalu-ation of the election as a whole (Olvera, 2003).This organized civic activity motivated the 1996reform and many of the activities that the IFE

carries out today (e.g., the quick counts, thetraining of electoral observers and the analysisof the media) are based in practices that Ali-anza Cvica first initiated (Olvera, 2003).

In addition, the successful 1996 reform wasthe first electoral reform negotiated, designedand implemented by all of the important actorsin political society. The 1990 reform which firstcreated the IFE was pushed through by theruling Party of the Institutional Revolution(PRI) with the support of only a part of therightist opposition party, the Party of National

Action (PAN). The 1994 reform gave the IFEan increased level of autonomy and was passedby the PRI and the PAN along with a smallfraction of the leftist opposition, the Party ofthe Democratic Revolution (PRD). Finally, the1996 reform was negotiated, designed andpassed by all three of the main parties from left,right and center (Prudhomme, 1996).

This case study offers a number of importantlessons for state reformers. First, it confirms thewillingness and capacity of poor people toparticipate in the core activities of governance.

Second, the rule of equal and opposite reactionapplies here once again. Normal citizens willonly participate at such massive levels if thepolicies being implemented are seen to respondto demands that have originated in civil society,are designed with the participation of a broadrange of actors, and actively incorporate citi-zens into the process of implementation itself.

Third, none of the achievements of the IFEwould have been possible without a significantamount of resources dedicated to the reformand operation of the IFE itself. Societal par-

ticipation is best stimulated when it is perceivedas a complement rather than as a replacementfor government action. Without a core group of2,500 civil servants, significant salaries for theGeneral Council and a large operating budgetUS$480 million in the year 2000 (IFE, 2000b),the IFE would not have been able to success-fully carry out its tasks nor stimulate the pop-ular legitimacy it needed in order to involve theactive participation of civil society.

Fourth, the case of the IFE forces us toquestion the commonly accepted idea that

neutrality arises exclusively out of the absenceof partisanship. Although some of the effec-tiveness of the IFE does indeed arise out of the



8/17

professionalization and nonpartisanship of itsstaff, a great deal of its legitimacy also arisesout of the saturation of partisanship or theradical plurality of those who participate inthe decision making processes of the IFE. The

General Council is made up of nine citizencouncilors, but also surrounded by a whirlwindof party representatives and media intru-sions. Each voting booth is staffed by trainedmembers of civil society, but also intensivelywatched by representatives from each politicalparty. One of the principal reasons why theelectoral reform of 1996 was more effective thanthe reforms of 1990 and 1994 is because agreater diversity of political positions weretaken into account at the negotiating table in1996 than during the other two reforms.

(c) Police and school reform in Chicago

Like many cities in the developing world,Chicago has a tradition of machine politics,insular administrative bureaucracies installed inreaction to political manipulations, a vibranttradition of neighborhood activism [and]extreme socioeconomic inequality (Fung,2001, p. 73). Research by Archon Fung showshow the Chicago city government has improvedthe performance of its schools and police forces

by actively incorporating the participation ofcivil society. As in Porto Alegre and withMexicos IFE, the Chicago government hasgone far beyond methods of consultation, co-production and protest to open itself up to fullco-governance with the citizenry at large.

This is particularly true in the case of schoolreform. In 1988 the city assembly passed theChicago School Reform Act which created alocal school council (LSC) 5 for each of theChicago Public Schools (CPS) 530 elementaryand high schools. The LSCs principal tasks are

hiring and firing school principals, approvingschool budgets, developing long-term strategicplanning documents called School Improve-ment Plans (SIP) and dispersing all Chapter 1funds 6 (Fung, 2001, p. 77). These reforms havemade the Chicago school system one of themost open to participation in the entire UnitedStates.

Chicagos police reform also involved a sig-nificant increase in citizen participation. The1995 reform of the Chicago Police Department(CPD) organized police officers into 279 beat

teams that are required to hold open com-munity meetings each month in which policeofficers and citizens work together to identify

problems and plan solutions (Fung, 1999;Skogan & Hartnett, 1997). Here the mode ofparticipation is more akin to societalaccountability. Citizens are not given anydirect legal power over the operations of the

police. They simply provide information andpressure the officers to attend to specific prob-lems. Nevertheless, the close citizen oversight ofpolice activities does serve as a powerfulaccountability mechanism since citizens com-plaints can trigger existing internal mechanismsof supervision and control (Walker, 2001).

There is evidence that both school and policeservices have greatly improved as a result of thereforms. During 199498 the murder ratedeclined 24%, robbery fell 31% and sexualassault fell 21% in Chicago, results that are

comparable to radically different zero-toler-ance strategies like those imposed by RudolphGiuliani in New York (Fung, 1999; Skogan &Hartnett, 1997). In addition, school perfor-mance as measured by a specially developedmetric of school productivity shows thatduring 198797 while students entering thesystem have become increasingly disadvan-taged and less well prepared, the majority ofschools have become more effective in educat-ing them (Fung, 2001, p. 99).

The origins of the two reforms are quite

distinct. School reform arose out of conflictbetween state and society and was driven bysocial protest.

In the Chicago schools, reform resulted from apitched battle that pitted a diverse social movementcomposed of parent organizations, good governmentcivic groups, educational reform activists, and a coali-tion of business groups against traditional schoolinsiders such as the Chicago Teachers Union andthe Board of Education (Fung, 2001, p. 77).

In contrast, police reform arose out of con-sensus between government and civil societyand was principally directed by reformerswithin the state.

Absent the street heat and legislative pressure thatdrove school reform, [the reform] discussions at theintersection of professional, political and civic inter-ests led quietly to the formulation of a participatoryvariant of community policing (Fung, 2001, p. 78).

Nevertheless, neither of these reforms werethe independent creation of far-sighted

bureaucrats. Both state and society actors werecrucial in the development of each participationmechanism, and the more active civil society



9/17

was involved in the development of the reformproposals the more complete was the openingup of the state to society.

These institutions are excellent examples ofwhat Fung calls accountable autonomy. For

both cases,

the role of central power shifts fundamentally fromthat of directing local units (in the previous hierarchi-cal system) to that ofsupporting local units in theirown problem-solving endeavors and holding themaccountableto the norms of deliberation and achieve-ment of demanding but feasible public outcomes(Fung, 2001, p. 87, emphasis in the original).

For example, while local school councils inChicago are responsible for drawing up budgets

and sanctioning principals they are also simul-taneously monitored and evaluated by centralagencies. This adds an interesting new twist toour theoretical discussion of accountabilitybecause here local participative bodies areaccountable to centralized bureaucratic agen-cies. Instead of civil society holding govern-ment accountable it is now government that isholding civil society accountable.

These Chicago cases reinforce the above les-sons. First, the most active participants inChicago are once again the poor and unedu-

cated. In addition, Fung documents thatminority dominant areas tend to have higherparticipation rates than white dominant areas(Fung, 1999). Second, the success of theseChicago cases also depended on the govern-ment opening the process beyond alreadyorganized civil society organizations andemploying community organizers to stimulateparticipation and facilitate community decisionmaking.

Third, civil society participation in the designphase of participatory structures proved to be

crucial here as well. Neither of the Chicagoreforms arose purely out of the minds of socialplanners and their relative success depended onthe ability of the government to involve socialactors from the very beginning. Fourth, as withPorto Alegre and the IFE, the supply side ofthe equation is crucial. Without a capable andwell-financed state apparatus that can actuallyrespond to popular demands and participation,such accountability mechanisms would createmore disenchantment than hope.

Finally, these cases push us further toward

the conclusion that the supposed either/orchoice between centralization and decentral-ization is a false dichotomy that needs to be

reanalyzed. Although devolving power isimportant, there is an equal need to strengthenthe center, at least in its coordinating andmonitoring capacities.

Nevertheless, the accountability mechanisms

in Chicago are clearly not as open and partic-ipatory as those in place in Porto Alegre or theIFE. On the one hand, the local school councilsare elected bodies that do not bring a clearpopular mandate arising out of popularassemblies such as the COP in Porto Alegre. Onthe other hand, the police community meet-ings do not have any direct legal authorityover police behavior as does the IFEs GeneralCouncil. Indeed, this may be why the level ofcitizen participation in Chicago is also muchlower than it is in Porto Alegre and with the

IFE. An average of only 2025 people partici-pate in each beat meeting per month and thereare only an average of 1.5 candidates in theelections for each open spot in the schoolcouncils (Fung, 1999).

(d) Decentralization and rural development inMexico

Decentralization on its own is just as likely tostrengthen corrupt local networks as it is topromote participation and accountability. Pro-

accountability arrangements cannot be expec-ted to arise spontaneously from devolution, butneed to be intentionally structured. This is thecentral lesson of Jonathan Foxs research onthe use of World Bank funds for municipaldevelopment projects in rural Mexico.

The Mexican Municipal Funds Program hasbeen almost entirely financed by two largeloans received from the World Bank, one forUS$350 million for 199194 and a second forUS$500 million for 199599. This money wastargeted for use in basic infrastructural

improvements for the poorest communities inthe rural areas of the poorest states and wasimplemented through municipal governments.Autonomous solidarity committees were tobe organized in each community in order tosupervise government spending, decide whichprojects would be funded and contribute thenecessary labor power.

Unfortunately, since the solidarity commit-tees did not have any legal standing or formalauthority over the Municipal Funds programitself, the actual level of participation and the

effective autonomy of the committees from themunicipal, state and federal governmentdepended entirely on the whims of local



10/17

bureaucrats. Many committees were thereforeentirely ignored or allowed to participate onlyin the implementation phase of the projects.

Nevertheless, this participatory mechanismwas actually relatively successful in the state of

Oaxaca. In this state the community assembliesmade the project selection decisions in 63% ofthe cases (Fox & Aranda, 1996, p. 37). Fox andAranda argue that one of the principal reasonsfor this high level of participation is that Oax-aca is an area that is endowed with a very highlevel of horizontal social capital 7 due to along and rich indigenous tradition of commu-nity collaboration and self-governance. Equallyimportant, the government of the State ofOaxaca was flexible and open to working withthese traditions. Instead of imposing a new

organizational structure on society, a healthymixing between state and social forms waspermitted (Fox, 1994). Indeed, in Oaxaca thistolerance of autonomous social forms goesback much further than the Municipal Fundsprogram. The states municipal structure itself,with 570 municipalities based in local organi-zational forms, demonstrates the governmentslong-standing commitment to accommodatelegal forms to traditional practices.

The communities that had higher levels ofparticipation had more effective development

projects. When the community was directlyinvolved, it tended to monitor the use of fundsmore closely and to pick projects that weremore useful for the population as a whole. Incontrast, when the selection process wasmanipulated from the outside, investmenttended to be shifted toward highly visiblealthough not always useful projects (Fox &Aranda, 1996, p. 37).

A few years into the program the governmentintervened in order to increase communityparticipation and make the distribution of

resources fairer. The formulas used for povertymeasurement and funds distribution wereimproved and, even more importantly, madepublic. Moreover, the amount of funds thatcould be spent in the municipal capital waslimited to 25%, thus requiring municipalities tochannel funds to the most needy, isolated areas.Finally, the required amount of communitycontributions was made variable depending onthe impact on poverty the selected projectwould have. High-impact projects required lesscommunity contribution than low impact pro-

jects, thus encouraging investment in truepublic goods (Fox & Aranda, 1996, p. 12;Fox, 2002, pp. 104105).

These changes stimulated community par-ticipation and strengthened social capital. Thisoccurred because the reforms made communi-ties aware of their right to a precise amount offunds, actively involved the poorest areas, and

empowered those actors who looked beyondtheir particular interests and towards thedevelopment of the community as a whole.Here institutional reform had a direct impacton trust, fairness and participation (Fox, 2002).

Nevertheless, the origins of this particularscheme of state-society synergy for account-ability was entirely top-down. Instead ofarising out of intense negotiations betweensocial actors and government reformers theparticipation scheme was thought up anddesigned by the federal government in consul-

tation with World Bank staff. This may go along way in explaining why community par-ticipation has not been more dynamic in theMunicipal Funds Program, and why the case ofOaxaca is more of an exception than the rule.Indeed, the top-down nature of the entireNational Solidarity Program (PRONASOL),of which the Municipal Funds program wasonly a part, has led many scholars to disqualifyit entirely as an attempt at social manipulationintended to help the former ruling party, theParty of the Institutional Revolution (PRI),

and the powerful interests it defends remain incontrol (Dresser, 1994, p. 144; Soederberg,2001, p. 104).

Such evaluations of the program as a wholeare solidly supported by the extreme level ofpro-government propaganda that accompaniedalmost every step of the Solidarity Program.Moreover, the distribution of solidarity fundscorresponded much more closely to politicalcriteria than to need-based criteria (see Corne-lius, Craig, & Fox, 1994). Nevertheless, thesewell-documented facts should not lead us to

ignore exceptional cases like those of the moreparticipative communities of Oaxaca that sup-port the prospect of successful co-governancefor accountability even under difficult condi-tions.

There are various lessons to be learned fromthis case study. First, as we have already seenabove, the direct involvement of social actorsand practices from the design stage greatlycontributes to the success of accountabilitymechanisms that depend on active participa-tion from civil society. Second, this case also

confirms the importance of the formal, legalempowerment of participatory bodies. Withouta clear institutionalized location in the decision



11/17

making process, these bodies are left open tothe winds of manipulation and are quicklybypassed by unwilling or authoritarian publicofficials.

Third, government transparency and insti-

tutional design have an important impact oncommunity participation. Co-governance foraccountability stands a much better chance atsuccess when government actors respect socialactors enough to fully inform about thedetails development programs and designparticipatory institutions so as to assure theactive involvement of the most marginalactors. Finally, this case also demonstrates thevalue of what Norman Long has calledinterface analysis. This type of analysispushes us to

focus upon intervention practices as shaped by theinteractions among the various participants, ratherthan simply on intervention models, by which ismeant the ideal-typical constructions that planners,implementers or their clients have about the process(Long, 1999, p. 4). 8

The best way to evaluate experiences ofsocietal participation is to delve into how state-society relations work themselves out on theground in specific contexts, as here in the State

of Oaxaca.

(e) Grass-roots anti-corruption initiatives inIndia

One area of government that seems to beparticularly resistant to societal participation isthe auditing of government expenditure. Thistask is usually thought to be far too technicallysophisticated and politically delicate for theaverage citizen. Freedom-of-information actshave recently started to sprout up around the

world, and citizens are encouraged to usepublic information to pressure corporations orgovernments from the outside to comply withtheir duties or to decide their votes (e.g., Fung& ORourke, 2000). But it is difficult to findexamples in which normal citizens are asdirectly involved in the activity of auditinggovernment expenditure as they are, forexample, in the activity of budget design inPorto Alegre.

Nevertheless, as Anne Marie Goetz and RobJenkins have cogently argued, the cases of the

Mazdoor Kisan Shakti Sangathan (MKSS)movement in Rajasthan, India and theRationing Kruti Samiti (RKS), or Action

Committee for Rationing, movement inMumbai, India show that when reformistbureaucrats are faced with an active pro-accountability movement in civil society it ispossible to make important inroads into the

area of social auditing. The central account-ability problem that both of these organizationsface is the widespread corruption in the provi-sion of government services to the poor. Wagesfor public works projects are frequently skim-med off by public managers and the materialsused in these projects are often artificiallyoverpriced and of bad quality so as to allow themaximum room for kickbacks. In addition, thecountrys Public Distribution System (PDS),which is in charge of channeling basic fooditems and other fundamental household goods

like kerosene to the poorest households, is rifewith corruption. One of the principal problemshere is the selling of these goods by owners ofration shops for personal profit (Goetz &Jenkins, 2002b).

Most communities in India already havelocal participatory institutions that are sup-posedly responsible for monitoring the perfor-mance of government programs. Nevertheless,these Vigilance Committees and VillageAssemblies are often captured by actors whoare implicated in the process of corruption

itself. For instance, they are frequently chairedby the representative of the municipal ward andtheir members are appointed in a top-downfashion. In addition, many government rationshops are owned or controlled by the very samepoliticians who are on the committees that aresupposed to supervise them (Goetz & Jenkins,2001, p. 371).

As a result of the failure of these state runparticipatory mechanisms, movements such asMKSS and RKS have found it necessary tocreate their own autonomous society-driven

mechanisms for auditing public projects. TheMKSS has developed a methodology throughwhich it independently investigates governmentspending practices and then exposes and com-pares this information to reality through publichearings (jan sun wai) (Goetz & Jenkins, 2002a,pp. 4142). In the hearings obvious discrepan-cies and missing accounts are presented and thepublic is given the opportunity to check theirown personal experience as public employees orsuppliers with the accounts. Public officialsoften attend and many cases exist in which this

process has worked to directly shame them intoreturning large amounts of misdirectedfunds.



12/17

Goetz and Jenkins present the case of theRKS in Mumbai as another example of whatthey call diagonal accountability, or theparticipation of vertical actors in horizon-tal enforcement activities. Since the official

Vigilance Committees are ineffective, theRKS has developed its own parallel system ofinformal vigilance committees. For each rationshop, five local women who are clients of theshop monitor and evaluate the quality andprices of the goods being sold. This activity hasbeen facilitated by the RKS citywide campaignto oblige shopowners to display prices publiclyas well as samples of the goods on sale. Thereports of the informal committees are then puttogether and presented both to the user com-munity and to the central coordinating

bureaucracy of the PDS in the city (Goetz &Jenkins, 2002b).

This process was particularly successful dur-ing the period immediately following the 1992riots in Mumbai, after which the city govern-ment was very interested in being perceived asbeing responsive to the poor. In addition,during this period an important reform mindedbureaucrat held the job of Regional Controllerof Rationing. Nevertheless, once this reformistleft his post the process became much lesseffective. The authors therefore claim that the

RKSs experience with diagonal accountabil-ity has been only a limited success story(Goetz & Jenkins, 2001). As we saw with thecase of the MKSS, society-driven pro-accountability initiatives that confront the stateand demand inclusion in the basic activities ofgovernment can be highly effective. Neverthe-less, the RKS experience also shows us thatultimately the success of these movements oftenalso depends on constructing alliances withprogressive government officials as well.

From this pair of cases we can learn various

lessons. First, they give us a fascinating alter-native to participative mechanisms like theBangalore Scorecard, which are limited tosimply surveying and reporting on the opinionof the public concerning the performance ofgovernment services. As Goetz and Jenkinsargue, such initiatives are grounded in a fun-damentally naive view of politics and bureau-cratic inefficiency because they assume thatbureaucrats are simply ignorant of the prob-lems with government (Jenkins & Goetz, 1999).Bureaucrats need to be made directly account-

able to the citizenry and the best way to do thisis to allow citizens to get involved in the activityof auditing from the inside and to confront

bureaucrats face-to-face with their complicityin bad performance or corruption.

Second, co-governance for accountabilitydoes not need to begin with reformist or pro-gressive governments. Success can also arise

out of the action of independent organizationsand social movements that press their demandson the state and push their way into theauditing of government programs. Third, itseems that at some point in the process thesemovements do need allies within the govern-ment. Without state support or at least toler-ance, such movements will most likely berepressed or rendered ineffective by stateaction. Finally, both of these cases confirm thatsensitive and complex activities such as publicauditing are not beyond the capacity of poor,

illiterate citizens.

4. LESSONS FOR INSTITUTIONALREFORMERS

This article has argued that the activeinvolvement of civil society and the strength-ening of the state apparatus are not mutuallyexclusive or even contradictory initiatives. Thisis the central idea of co-governance as a

concept. If institutions are properly designed, avirtuous cycle that reinforces both state andsociety is possible. This is particularly impor-tant to emphasize today given the thrust ofmuch of the NPM literature that proposes thedevolution of state responsibilities to socialactors via the market.

In addition, this article questions thosestrands of the old public management liter-ature that emphasize the insulation of bureau-cracy from societal actors. As Kaufman hasrecently argued,

The implication of accountability reform is different,however, when it refers to the establishment of popu-lar assemblies and other forms of direct grassrootsparticipation in administrative decisions. Althoughsome forms of inclusion, such as partnerships withnon-governmental organizations (NGOs) may en-hance capacity, others, such as popular assemblies,may be a step backward in terms of the efficiency,effectiveness, and even the accountability of stateorganizations (Kaufman, 2003, p. 284).

The above case studies challenge this sort of

circumscription of societal participation towell-behaved or enlightened actors such asNGOs and argues for the full inclusion of the



13/17

citizenry as a whole in the core activities ofgovernment.

This article has shown that the first step forgovernment reformers looking to construct co-governance for accountability should be to

trust and actively involve societal actors fromthe very beginning of the process. Reformersshould not wait for civil society to start trustinggovernment nor should they wait to involvesociety until after the government has alreadydesigned a new participatory mechanism fromabove. As the above case studies show, theearlier societal actors are involved in the designprocess the more effective participatory mea-sures tend to be. The best entry points aretherefore those issues and locations where thereare previously existing social demands and

practices surrounding a specific accountabilityissue.

In addition, when designing participatorymechanisms government reformers should beaware that transparency is not enough. Gov-ernments cannot expect information provisionto generate single-handedly the positive feed-back loops between state and society outlinedin the above case studies. Governments shoulddirectly stimulate the participation of society.Otherwise, the only actors who will put to usethe new information are journalists, academics,

nonprofit organizations and already existingcommunity organizations. Although thesegroups are indeed crucial in maintainingaccountability, the cases above show that thereis a qualitative forward leap when the popula-tion at large and the poor in particular aredirectly involved in enforcing accountability.

Once initiated, the best way to assure thesustainability of a participatory framework isthrough its full institutionalization. As we sawin the case of the Municipal funds program, theformalization of even limited top-down par-

ticipatory schemes allowed for the developmentof much fuller participation. The case of theRKS in Mumbai, India provides us withimportant negative examples of this samepoint. Here the absence of a clear legal frame-work left participation up to the whims ofindividual bureaucrats, leading to the eventualoverturning of participatory schemes oncethere was a change of heart on the part ofgovernment. The difference between the twoChicago cases also reveals the importance offormalizing participatory procedures. One of

the major reasons why the school reformhas been more effective than the police reformis because the former institutionalized the

involvement of civil society in the formal legalstructure much more clearly and explicitly thanthe latter.

There are three different levels at which par-ticipatory mechanisms can be institutionalized.

First, participatory mechanisms can be builtinto the strategic plans of existing governmentagencies. Second, new agencies can be createdwhose goal is to assure societal participation ingovernment activities. Third, participatorymechanisms can be inscribed in law.

Although the first level of institutionalizationis more or less widespread in the developingworld and the second level is relatively com-mon, the third level is extremely rare. 9 Whythis is the case is more or less evident. Law-making under democratic conditions involves

the messy process of legislative bargaining anda full role for political parties. State reformersand multilateral agencies tend to shy away fromsuch arenas, especially when they are domi-nated by opposing parties or factions. There-fore, reformers usually settle for executiveprocedures, special agencies or innovativeindividual bureaucrats to carry out their par-ticipative strategies.

This is a mistake. As the above case studiesshow, if dealt with in a creative fashion, parti-sanship can be just as effective as isolation in

the search for effective accountability mecha-nisms. It is necessary to involve political partiesand the legislature in order to institutionalizefully participative mechanisms through the law.

In general, professionalism and independenceare necessary but by no means sufficient toassure the long-term survival of accountability.Effective pro-accountability structures need tobe legitimated by society both at their foundingmoment and during their everyday operations.This requires the multiplication, not thereduction, of external eyes (Smulovitz, 2003)

and the diversification, not unification, ofpolitical and ideological perspectives.

Finally, these case studies also show us thatdecentralization alone does not automaticallylead to an increase in societal participation oran improvement of government accountability.Although devolution and decentralization areimportant because they bring governmentcloser to the people, if carried out blindly theytend to reinforce inequalities both within thenewly autonomous local units and betweenthem. Decentralization is only productive if

the center remains responsible for the super-vision and coordination of activities in thelocal units.



14/17

If carefully applied, co-governance can bemuch more rewarding than alternatives suchas marketization, bureaucratic insulation,co-production or societal accountability.Co-governance for accountability is usually

more difficult to implement, but it is well

worth the effort. By transgressing theboundaries between state and society institu-tional reformers can unleash invaluable pro-accountability processes which are almostimpossible to tap into through less ambitious

strategies.

NOTES

1. In Latin America, Belize, Brazil, Columbia, Costa

Rica, Chile, Peru and Mexico have all recently created or

revived such institutions. This trend is also present in

Asia, Africa, Australia and Eastern Europe. Some recent

examples include the new Ombudsmen in Poland

(founded in 1987), the Philippines (founded 1989) and

South Korea (founded in 1994), the National CounterCorruption Commission in Thailand (founded in 1998),

the Independent Commission Against Corruption in

New South Wales, Australia (founded in 1988), the

Public Protector in South Africa (founded in 1994), and

the Inspector-General of Government in Uganda

(founded in 1996) (Pope, 2000; Schedler et al., 1999).

Over 80 countries currently have a nationalOmbudsman

(Bennett, 1997).

2. Navarro (2003) has recently argued that such intense

involvement of the government is one of the principal

problems with the PB scheme. This is because it violatesthe autonomy of civil society and opens up the

process to manipulation by party and governmental

interests. Following the societal accountability line of

thought, Navarro therefore advocates for what he calls

the full institutionalization of the PB process by

allowing societal actors to decide on the allocation of

public funds without any interference from the govern-

ment (Navarro, 2003, pp. 124125). Needless to say,

from the point of view of the present article the full

interpenetration of state and society present in the Porto

Alegre PB process is one of its most important strengths,

not one of its weaknesses.

3. I am not the first to draw a parallel between Porto

Alegres PB and Mexicos IFE. In his recent text on

Democracy and the Public Space in Latin America,

Avritzer claims that these experiences are two of the best

examples of the institutionalization of what he calls

participatory publics (Avritzer, 2002, pp. 135164) or

the embedding of societal practices within the state. My

approach differs from Avritzers in so far as I give equal

credit to government and party entrepreneurs, while

he tilts the balance toward the role of societal actors.

4. The councilors are elected by two-thirds of thelegislature for seven-year terms and are chosen among

the most well-known scholars and leaders of civil

society.

5. LSCs are formed by six parents, two community

representatives, two teachers, the schools principal and

an additional nonvoting student for high schools.

6. Special state funds allocated to schools on the basis

of the economic disadvantage of their student body.

7. Fox defines horizontal social capital as the

existence of social practices of mutual trust and cooper-

ation within a given community (Fox, 2002, p. 113).

8. I would like to thank Ernesto Isunza for bringing

this extremely valuable text to my attention.

9. There are some exceptions, including the above

Porto Alegre and IFE cases as well as Bolivias Law of

Popular Participation (Cunill, 2000; Oxhorn, 2001),

Mexico Citys Law of Citizen Participation (Mellado,

2001) and Brazils 1995 administrative reform thatformalizes social control through the legal category of

Social Organizations (Barreto, 1998; Bresser, 1999;

Nassuno, 1998). Nevertheless, these exceptions only

prove the rule that participatory mechanisms are usually

vastly underinstitutionalized and depend too much on

the ingenuity and good will of individual bureaucrats.

REFERENCES

Abers, R. (1998). From clientelism to cooperation: local

government, participatory policy, and civic organiz-ing in Porto Alegre, Brazil.Politics and society, 26(4),511538.

Avritzer, L. (2000). Civil society, public space and local

power: a study of the participatory budget in BeloHorizonte and Porto Alegre. IDS-Ford Web Site.http://www.ids.ac.uk/ids/civsoc/index.html.

http://www.ids.ac.uk/ids/civsoc/index.htmlhttp://www.ids.ac.uk/ids/civsoc/index.htmlhttp://www.ids.ac.uk/ids/civsoc/index.html


15/17

Avritzer, L. (2002). Democracy and the public space inLatin America. Princeton: Princeton UniversityPress.

Baiocchi, G. (2001). Participation, activism, and politics:the porto alegre experiment and deliberative demo-cratic theory. Politics and Society, 29(1), 4372.

Barreto, M. I. (1998). Las organizaciones sociales en lareforma del estado. In L. C. Bresser & N. Cunill(Eds.), Lo Publico No Estatal en la Reforma delEstado (pp. 115156). Buenos Aires: Paidos.

Bennett, C. (1997). Understanding ripple effects: thecross-national adoption of policy instruments forbureaucratic accountability. Governance: An Inter-national Journal of Policy and Administration, 10(3),213233.

Bresser, L. C. (1999). Managerial public administration:strategy and structure for a new state. In L. C.Bresser & P. Spink (Eds.), Reforming the State:Managerial Public Administration in Latin America(pp. 114). Boulder: Lynne Rienner.

Bresser, L. C., & Cunill, N. (1999). Entre el estado y elmercado: lo publico no estatal. In L. C. Bresser & N.Cunill (Eds.), Lo Publico No Estatal en la Reformadel Estado (pp. 2558). Buenos Aires: Paidos.

Bresser, L. C.& Spink, P. (Eds.). (1999). Reforming thestate: managerial public administration in LatinAmerica. Boulder: Lynne Rienner.

Cornelius, W., Craig, A., & Fox, J. (Eds.). (1994).Transforming state-society relations in Mexico: thenational solidarity strategy. La Jolla: Center for US-Mexican Studies.

Cunill, N. (1997). Repensando lo publico a traves de lasociedad: nuevas formas de gestion publica y repre-

sentacion social. Caracas: Nueva Sociedad.Cunill, N. (2000). Responsabilizacion por el control

social. United Nations Online Network in PublicAdministration and Finance (UNPAN). http://un-pan1.un.org/intradoc/groups/public/documents/clad/unpan000183.pdf.

Dresser, D. (1994). Bringing the poor back in: nationalsolidarity as a strategy of regime legitimation. In W.Cornelius, A. Craig, & J. Fox (Eds.), Transformingstate-society relations in Mexico: the national solidar-ity strategy (pp. 143166). La Jolla: Center for US-Mexican Studies.

Evans, P. (1995). Embedded autonomy. Berkeley: Uni-versity of California Press.

Evans, P. (1996a). Introduction: development strategiesacross the publicprivate divide.World Development,

24(6), 10331037.Evans, P. (1996b). Government action, social capital

and development: reviewing the evidence on synergy.World Development, 24(6), 11191132.

Fox, J. (1994). The difficult transition from clientelism tocitizenship: lessons from Mexico. World Politics,46(2), 151184.

Fox, J. (2000). Civil society and political accountability:propositions for discussion. Presented at the confer-ence Institutions, Accountability and DemocraticGovernance in Latin America, The Helen Kellog

Institute for International Studies, Notre DameUniversity, May 8.

Fox, J. (2002). La relacion recproca entre la partici-pacion Ciudadana y la Rendicion de Cuentas.Poltica y Gobierno, 9(1), 95133.

Fox, J., & Aranda, J. (1996). Decentralization and ruraldevelopment in Mexico: community participation inoaxacas municipal funds. La Jolla: Center for US-

Mexican Studies.Fung, A. (1999). Street level democracy: pragmatic

popular sovereignty in chicago schools and policing.Paper presented at the American Political ScienceAssociation Annual Meeting, Atlanta, September 25.

Fung, A. (2001). Accountable autonomy: towardempowered deliberation in chicago schools andpolicing.Politics and Society, 29(1), 73103.

Fung, A., & ORourke, D. (2000). Reinventing environ-mental regulation from the grassroots up: explainingand expanding the success of the toxics releaseinventory. Environmental Management, 25(2), 115127.

Fung, A., & Wright, E. (2001). Deepening democracy:innovations in empowered participatory governance.Politics and Society, 29(1), 541.

Goetz, A. M., & Jenkins, R. (2001). Hybrid forms ofaccountability: citizen engagement in institutions ofpublic-sector oversight in India. Public ManagementReview, 3(3), 363383.

Goetz, A. M., & Jenkins, R. (2002a). Voice, account-ability and human development: the emergence of anew agenda. Background paper for Human Devel-opment Report 2002. New York: United NationsDevelopment Programme.

Goetz, A. M., & Jenkins, R. (2002b). Civil society

engagement and indias public distribution system:lessons from the rationing Kruti Samiti in Mumbai.Paper presented at Making Services Work for PoorPeople, World Development Report Workshop(WDR) 2003/04, Oxford, November 45.

Goldfrank, B. (2002). The fragile flower of localdemocracy: a case study of decentralization/partici-pation in Montevideo. Politics and Society, 30(1),5183.

Grindle, M. (1996). Challenging the state: crisis andinnovation in Latin America and Africa. New York:Cambridge University Press.

Hirschman, A. (1970).Exit, voice, and loyalty: responsesto decline in firms organizations, and states. Cam-bridge: Harvard University Press.

IFE (2000a). The Mexican electoral regime and thefederal elections of the year 2000. Mexico City: IFE.

IFE (2000b) Proceso electoral federal 2000: datos ynumeralia mas importantes, Mexico City: IFE.

Isunza, E. (2003). Construccion de la democracia yrendicion de cuentas: una mirada regional de nuevasinterfaces socio-estatales en el contexto de la tran-sicion Poltica Mexicana. Presented at 2003 meetingof the Latin American Studies Association, Dallas,March 2729.

Jenkins, R., & Goetz, A. M. (1999). Accounts andaccountability: theoretical implications of the right-

to-information movement in India. Third WorldQuarterly, 20(3), 603622.

http://unpan1.un.org/intradoc/groups/public/documents/clad/unpan000183.pdfhttp://unpan1.un.org/intradoc/groups/public/documents/clad/unpan000183.pdfhttp://unpan1.un.org/intradoc/groups/public/documents/clad/unpan000183.pdfhttp://unpan1.un.org/intradoc/groups/public/documents/clad/unpan000183.pdfhttp://unpan1.un.org/intradoc/groups/public/documents/clad/unpan000183.pdfhttp://unpan1.un.org/intradoc/groups/public/documents/clad/unpan000183.pdf


16/17

Kaufman, R. (2003). The comparative politics ofadministrative reform: some implications for theoryand policy. In B. Schneider & B. Heredia (Eds.),Reinventing leviathan: the politics of administrativereform in developing countries. Miami: North-SouthCenter Press.

Lam, W. F. (1996). Institutional design of publicagencies and coproduction: a study of irrigationassociations in Taiwan. World Development, 24(6),10391054.

Long, N. (1999). The multiple optic of interface analysis.Background Paper on Interface Analysis. Paris:UNESCO.

Manzetti, L. (2003). Political manipulations and marketreform failures. World Politics, 55(3), 315360.

Maor, M. (2004). Feeling the heat. anti-corruptionmechanisms in comparative perspective.Governance:An International Journal of Policy, Administration,and Institutions, 17(1), 128.

Marquetti, A. (2003). Participac~ao e redistribuic~ao: oorcamento participativo em Porto Alegre. In L.Avritzer, & Z. Navarro (Eds.), A inovac~ao de-mocratica no brasil: o orcamento participativo (pp.129156). S~ao Paulo: Cortez Editora.

Mellado, R. (2001). Participacion ciudadana institucio-nalizada y gobernabilidad en la ciudad de Mexico.Mexico City: Plaza y Valdes.

Molinar, J. (1996). Renegotiating the rules of the game:the state and political parties. In M. Serrano & V.Bulmer-Thomas (Eds.),Rebuilding the state: Mexicoafter Salinas (pp. 2440). London: Institute of LatinAmerican Studies, University of London.

Narayan, D.& Petesch, P. (Eds.). (2002). Voices of the

poor: from many lands. New York: Oxford Univer-sity Press.

Nassuno, M. (1998). El control social en las organizac-iones sociales en el Brasil. In L. C. Bresser & N.Cunill (Eds.), Lo publico no estatal en la reforma delEstado (pp. 333360). Buenos Aires: Paidos.

Navarro, Z. (1998). Participation, democratizing prac-tices and the formation of a modern politythe caseof participatory budgeting in Porto Alegre, Brazil(19891998). Development, 41(3), 6871.

Navarro, Z. (2002). Decentralization, participation andsocial control of public resources: participatorybudgeting in Porto Alegre (Brazil). Presented atworkshop on Citizen Participation in the Context ofFiscal Decentralization: Best Practices in MunicipalAdministration, Inter-American Development Bankand Asian Development Bank, September 26,Tokyo and Kobe, Japan.

Navarro, Z. (2003). O orcamento participativo dePorto Alegre (19892002): um conciso comentaricocrtico. In L. Avritzer & Z. Navarro (Eds.), Ainovac~ao democratica no Brasil: o orcamento partic-ipativo(pp. 89128). S~ao Paulo: Cortez Editora.

Nylen, W. (2002). Testing the empowerment thesis: theparticipatory budget in Belo Horizonte and Betim,Brazil. Comparative Politics, 34(2), 127145.

ODonnell, G. (1999). Horizontal accountability in new

democracies. In A. Schedler, L. Diamond, & M. F.Plattner (Eds.), The self-restraining state: power and

accountability in New Democracias (pp. 2952).Boulder: Lynne Rienner.

Olvera, A. (2003). Movimientos sociales prodemocrati-cos, democratizacion y esfera publica en mexico: elcaso de Alianza Cvica. In A. Olvera (Ed.),Sociedadcivil, esfera publica y democratizacion en America

Latina: Mexico (pp. 351409). Mexico City: Fondode Cultura Economica.

Ostrom, E. (1996). Crossing the great divide: coproduc-tion, synergy, and development. World Development,

24(6), 10731088.Oxhorn, P. (2001). La construccion del estado por la

sociedad civil: la ley de participacion popular deBolivia y el desafo de la democracia local. Docu-mentos de Trabajo del INDES. Washington, DC:Inter-American Development Bank.

Paul, S. (1992). Accountability in public services: exit,voice and control. World Development, 20(7), 10471060.

Peruzzotti, E., Smulovitz, C. (2000a). Societal andhorizontal controls: two cases about a fruitfulrelationship. Presented at the conference Institu-tions, Accountability and Democratic Governance inLatin America, The Helen Kellog Institute forInternational Studies, Notre Dame University, May8.

Peruzzotti, E., & Smulovitz, C. (2000b). Societalaccountability in Latin America. Journal of Democ-racy, 11(4), 147158.

Peruzzotti, E., & Smulovitz, C. (2002). Accountabilitysocial: la otra cara del control. In E. Peruzzotti & C.Smulovitz (Eds.),Controlando la poltica: ciudadanos

y medios en las nuevas democracias latinoamericanas.

Buenos Aires: Editorial Temas.Peters, G. (2001). The future of governing (2nd edn.

revised). Lawrence, KS: University Press of Kansas.Pope, J. (2000).Confronting corruption: the elements of a

national integrity system. Berlin: Transparency Inter-national.

Pozas, R. (1996). La observacion electoral: una modal-idad de la militancia ciudadana.Revista Mexicana deSociologa, 59(2), 2340.

Prudhomme, J. F. (1996). La negociacion de las reglasdel juego: tres reformas electorales (19881994).Poltica y Gobierno, 3(1), 93126.

Przeworski, A., Stokes, S., & Manin, B. (Eds.). (1999).Democracy, accountability and representation. NewYork: Cambridge University Press.

Rose-Ackerman, S. (1999). Corruption and government:causes, consequences and reform. New York: Cam-bridge University Press.

Schedler, A. (1999a). Conceptualizing accountability. InA. Schedler, L. Diamond, & M. F. Plattner (Eds.),The self-restraning state: power and accountability innew democracies (pp. 1328). Boulder: Lynne Rien-ner.

Schedler, A. (1999b). Las comisiones y la piramide:notas sobre la conflictiva recentralizacion del poderen el IFE. Poltica y Gobierno, 6(1), 187222.

Schedler, A., Diamond, L., & Plattner, M. F. (Eds.).

(1999).The self-restraning state: power and account-ability in new democracies. Boulder: Lynne Rienner.



17/17

Skogan, W., & Hartnett, S. (1997). Community policing:Chicago style. New York: Oxford University Press.

Smulovitz, C. (2003). How can the rule of law rule? Costimposition through decentralized mechanisms. In J. M.Maravall & A. Przeworski (Eds.), Democracy and therule of law. New York: Cambridge University Press.

Soederberg, S. (2001). From neoliberalism to socialliberalism: situating the national solidarity programwithin Mexicos passive revolutions. Latin AmericanPerspectives, 28(3), 104123.

Tendler, J. (1997). Good government in the tropics.Baltimore: Johns Hopkins University Press.

UNDP (2002). Human development report 2002: deep-ening democracy in a fragmented world. New York:Oxford University Press.

Varshney, A. (1999). Democracy and poverty. Presentedat Conference on 2000 World Development Report,UK Department for International Development &Sussex Institute of Development Studies, CasteDonnington, England, August 1516.

Vellinga, M. (Ed.). (1998). The changing role of the statein Latin America. Boulder: Westview Press.

Waisbord, S. (2000). Watchdog journalism in SouthAmerica: news, accountability, and democracy. NewYork: Columbia University Press.

Walker, S. (2001). Police accountability: the role of

citizen oversight. Belmont: Wadsworth/ThompsonLearning.

Wallis, J., & Dollery, B. (2001). Government failure,social capital and the appropriateness of the NewZealand model for public sector reform in developingcountries. World Development, 29(2), 245263.

Woldenberg, J. (2001). Lessons from Mexico.Journal ofDemocracy, 12(2), 151156.

World Bank (1997). World development report 1997: thestate in a changing world. New York: OxfordUniversity Press.

World Bank (2003). World development report 2004:making services work for poor people. New York:Oxford University Press.


Co-governance for Accountability

Documents

Transcript of Co-governance for Accountability