Report No. 40144-BR Brazil Toward a More Inclusive and Effective Participatory Budget in Porto Alegre (In Two Volumes) Volume I: Main Report January 29, 2008 Brazil - Country Management Unit Sustainable Development Unit Latin America and Caribbean Region Document of the World Bank Table of Contents ............................................................................................... Acknowledgements ................................................................................................................................... vm List o fAcronyms and Currency Equivalents vi ... Executive Summary ...................................................................................................................................... 1 Background ............................................................................................................................................... 1 Conceptual Framework and Methodology ................................................................................................ 1 Main Findings.,.......................................................................................................................................... 2 Main Recommendations............................................................................................................................ 6 Introduction................................................................................................................................................. 11 A Demand-driven Study.......................................................................................................................... 11 Historical Evolution of the ParticipatoryBudget .................................................................................... 12 The OP Cycle and Process ...................................................................................................................... 14 Conceptual Framework ........................................................................................................................... 16 Methodology ........................................................................................................................................... 20 Chapter 1.Quality of Participation inthe OP ............................................................................................. 22 1.1. Who participates inthe OP? ...................................................................................................... 22 1.2. H o w Legitimate i s the OP?........................................................................................................ 31 1.3. Main Challenges and Recommendations................................................................................... 39 Chapter 2.The Interface between OP and MunicipalFiscal Management................................................. 2.1 Cycles o f Municipal Budget Planning and the OP: Convergence or Divergence? ...................42 -44 2.2 Consistency o f Budget Planning and Execution: Impacts on the OP ....................................... 46 2.3 The Impact o f OP on Capital Investments: Efficient Allocation or Fragmentation?.................51 2.4 Current Expenditure and the OP ................................................................................................ 54 2.5 Fiscal Equilibriumand the OP: Genuine Capacity for Influencing Sound Fiscal Policies........57 2.6 The Relevance o f Cost-Benefit Analysis and Impact Assessment inthe OP Process ...............59 2.7 Main Challenges and Recommendations................................................................................... 60 Chapter 3.The OP and the Accountability of Public Expenditure.............................................................. 64 3.1 Budget Transparency and Public Expenditure Oversight inPorto Alegre ................................ 65 3.2 Limitations inBudget Transparency, Oversight, and Service Performance Monitoring..........68 3.3 Main Challenges and Recommendations................................................................................... 72 Chapter 4. Towards a System o f Social Accountability inPorto Alegre ................................................... 75 4.1 The Multiple Spaces o f Public ParticipationinPorto Alegre .................................................... 75 4.2 The Interface between the OP and Other Spaces o f Political Participation............................... 79 4.3 Main Challenges and Recommendations................................................................................... 83 Chapter 5. Evaluating Participatory Budgeting........................................................................................... 86 5.1 ExpectedResults of OP and Earlier Econometric Impact Studies............................................. 87 5.2 Methodological Approach ........................................................................................................ -89 11 5.3 Results for OP andNon-OP Municipalities ............................................................................... 91 5.4 The Case ofPorto Alegre........................................................................................................... 93 5.5 Conclusions................................................................................................................................ 98 Chapter 6.MainConclusions andRecommendations............................................................................... 100 6.1 MainConclusions............................................................................................................................ 100 6.2 MainRecommendations.................................................................................................................. 102 References................................................................................................................................................. 106 ... 111 Index of Figures Figure 1.The policy and expenditure cycle and participatory fora inPorto Alegre ..................................... 5 Figure2 Participatorybudgetingcycle inPorto Alegre............................................................................ . 15 Figure 3. Key relationships o f power .......................................................................................................... 16 Figure 4.Mechanisms for social accountability throughout the public expenditure cycle ......................... 17 Figure 5.Number o f OP participantsinRegional and Thematic Assemblies, 1990-2006 ........................ -23 Figure6 OP participationby sex inRegionaland Thematic Assemblies, 1993-2005............................... . 24 Figure 7. Legitimacy o frepresentation inthe OP process ......................................................................... 31 Figure 8. Reelectiono f COP delegates, 1991-2007 .................................................................................... 35 Figure9.Ratings o frepresentation by councilors and delegates ................................................................ 36 Figure 10.Ratings o faccountabilityo f councilors anddelegates............................................................... 36 Figure 11. Public perception o f the OP by percentage o f population, 2006................................................ 37 Figure 12.Comparison o f OP and municipalbudgeting cycles.................................................................. 45 Figure 13. Share o f executed investments, 1999-2006 .............................................................................. 47 Figure 14.Percentage o f total investments prioritized by the OP, 1997-2006 ........................................... 48 Figure 15 Accuracy o fmunicipal income forecasts, 1995-2005 ............................................................... . 49 Figure 16.Distribution o f resourcesbetweenregional and thematic demands, 1995-2007 ....................... 52 Figure 17.Ratio o f current expenditure to total income, 1980-2005 .......................................................... 55 Figure 18.Expenditureindicators o f the municipal prefecture o f Porto Alegre, 1980-1999 ..................... 56 Figure 19. Investment as a percentage o f total expenditure, 1988-2005 ................................................... 57 Figure 20.The policy and expenditure cycle and civil society-local government institutions ..................-82 Figure 21.Average wages inPorto Alegre ................................................................................................. 94 Figure 22.Infant mortality rates inPorto Alegre ........................................................................................ 94 Figure 23 .Current revenues inPorto Alegre .............................................................................................. 95 Figure24. Total municipalrevenues inPorto Alegre ................................................................................. 96 Figure 25.Current expenses inPorto Alegre .............................................................................................. 96 Figure 26.Investments inPorto Alegre....................................................................................................... 97 Figure 27.Municipal operational expenses inPorto Alegre ....................................................................... 97 1v Index of Tables Table 1. Porto A1egI.e at a glance ................................................................................................................ 11 Table 2. OP Participation according to age, 1995 -2005 (percent) ............................................................. 25 Table 3.Time devoted to OP by councilors, 2006 ...................................................................................... 25 Table 4.OP participation by level o f education, 1995-2005....................................................................... 27 Table 5. Knowledge o f OP rules and criteria according to duration o fparticipation.................................. 34 Table 6. Financial indicators o f Porto Alegre municipality, compared to FRL limits................................ 43 Table 7. Thematic demands: Ratio o f executed value/ forecast value (inR$) ........................................... 50 Table 8.Regional demands: Ratio o f executed value/ forecast value (inR$)........................................... -51 Table 9.Distribution o f the number o f works by value, 1995-2006 (inR$) .............................................. 53 Table 10. OP demands per sector, 1990-2007 ............................................................................................ 54 Table 11. Capital/current expenditure ratios o f Porto Alegre, 1998-2005 ................................................. 56 Table 12. Sample analysis o f COP meeting minutes .................................................................................. 58 Table 13. H o w Porto Alegre citizens rate their knowledge o f the municipal budget.................................. 69 Table 14. Survey question: Would you know how to say inyour own words the meaning o f...................71 Table 15. Evolution of state-society interactions inPorto Alegre............................................................... 76 Table 16. Public Municipal Councils o f Porto Alegre, as o f 2004............................................................. 77 Table 17. Component sectors o fthe Public Municipal Councils, as o f 2004.............................................. 77 Table 18. Expectedoutcomes o fparticipatory budgeting.......................................................................... -88 Table 19.Impacts usingonly municipalities that implemented OP before 1996........................................ 91 Table 20. Impacts usingmunicipalities that implementedOP before and after 1996................................. 92 Table 21. Impacts on fiscal performance (IPTU/GDP) ............................................................................... 92 Table 22. Impacts on fiscal performance (investment /operational expenditure) ....................................... 93 V List of Acronyms and Currency Equivalents CAR Centro Administrativo Regional (Regional Administration Center) CMAA Conselho Municipalde Agricultura e Abastecimento (Municipal Council on Agriculture and Supply) CMAS Conselho Municipal de Assistencia Social (Municipal Council on Social Assistance CMC Conselho Municipal de Cultura (Municipal Council on Culture) CMD Conselho Municipalde Desporto (Municipal Council on Sports) CMDCA Conselho Municipaldos Direitos da Crianqa e do Adolescente (Municipal Council on Children's Rights) CMDH Conselho Municipaldos Direitos Humanos (Municipal Council on Human Rights) CMDUA Conselho Municipalde Desenvolcimento Urbano e Ambiental (Municipal Council on Urbanand Environmental Development) CME Conselho Municipalde EducaqBo(Municipal Council on Education) CMS Conselho Municipalde Saude (Municipal Council onHealth) COMAM Conselho Municipalde Meio Ambiente (Municipal Council on Environment) COMATHAB Conselho Municipalde Acesso a Terra e HabitaqBo (Municipal Council on Access to Land and Housing) COMPHAC Conselho Municipalde Patrimanio Hist6rico e Cultural (Municipal Council on Historical, Artistic and Cultural Heritage) COMTU Conselho Municipalde Transportes Urbano (Municipal Council on Urban Transportation) COMUI Conselho Municipaldo Idoso (Municipal Council on Senior Citizenry) CONCET Conselho Municipalde Ciencia e Tecnologia (Municipal Council on Science and Technology) CONCONT Conselho dos Contribuintes (Municipal Council on Taxpayers) CONDIM Conselho Municipal dos Direitos da Mulher (Municipal Council on Women's Rights) CONEN Conselho Municipal de Entorpecentes (Municipal Council on Drugs) CONTUR Conselho Municipal de Turismo (Municipal Council on Tourism) COP Conselho do Orqamento Participativo (Participatory Budgeting Council) CROP Coordenador Regional do Orqamento Participativo (Regional Coordinator of Participatory Budgeting) DMAE Departamento Municipal de Agua e Esgotos (Municipal Department of Water and Sewage) DEMHAB Departamento Municipal de HabitaqBo (Municipal Department of Housing) DEP Departamento de Esgotos Pluviais (Municipal Department of Sewage) DMLU Departamento Municipalde LimpezaUrbana (Municipal Department of Urban Sanitation) ESW Economic Sector Work, World Bank FEU Finance Economics and Urban Department, World Bank FASC FundagBode Assist@nciaSocial e Cidadania (Foundation for Social and Citizen Assistance) FRACAB FederaqBo Rio-Grandense de AssociaqGes Comunitarias e de Moradores de Bairros (Federation of Community and NeighborhoodAssociations of Rio Grande do Sul) GLP Gross LocalProduct GPO Gabinete de Planejamento Orqamenthrio (Cabinet of Budgetary Planning) IBGE Instituto Brasileiro de Geografia e Estatistica (Brazilian Institute of Geography and Statistics) ICMS Impost0 Sobre a CirculaqBo de Mercadorias e Serviqos (Tax on the Circulation of Merchandise and Services), a Brazilian state tax IDB Inter-AmericanDevelopmentBank vi IGP-DI hdice Geral de Preqos - Disponibilidade Interna (General Price Index-Internal Liquidity) IP Investment Plan (Plano de Investimentos) IPI Imposto sobre Produto Industrializado(Tax on Industrial Production, a Brazilian federal tax) IPTU Imposto Predial e Territorial Urbano (Brazilian local taxes on real state) IPVA Imposto sobre a Propriedade de Veiculos Automotores Imposto (Brazilian state vehicle property tax) L A C LatinAmerica and Caribbean Region, World Bank LCCSC World Bank Office inBrasilia, Brazil, World Bank LCSPE Economic Policy Sector, World Bank LCSPS Public Sector, World Bank LCSSD Sustainable Development Department, World Bank LCSSO Social Development Unit o f the Latin America and Caribbean Region, World Bank LDO Lei de Diretrizes Orqamentirias (Law on Budgetary Guidelines) LOA Lei Orqamentiria Anual (Annual Budget Law) NA NeighborhoodAssembly NCR Net Current Revenue ObservaPoA Observatbrio de Porto Alegre (Observatory o fPorto Alegre) OP Orqamento Participativo (Participatory Budget) PDDUA Plano Diretor de DesenvolvimentoUrbano e Ambiental (Urban and Environmental Development Directive Plan, the Master Plan o f Porto Alegre) PLOA Projeto de Lei do Orqamento (Draft Budget Law) PPA Plano Plurianual (Multi-Annual Plan or Multi-Year Plan) PPS Partido Popular Socialista (Popular Socialist Party) PT Partido dos Trabalhadores (Workers' Party) R $ Real (Brazil's currency) RGF Relatbrio de Gestiio de Fundos (Fiscal Management Reports) RREO Relat6rio Resumido da Execuqgo Orqamentiria (Summary Budget Execution Report) SDV Social Development Department, World Bank S M A Secretaria Municipal de Administraqgo (Municipal Secretariat o f Administration) SMF Secretaria Municipal da Fazenda (Municipal Secretariat o f Finance) SMCPGL Secretaria Municipal de Coordenaqgo Politica e Governanqa Local (Municipal Secretariat o f Political Coordination and Local Governance) SMOV Secretaria Municipal de Obras e Viaqiio (Municipal Secretariat o fWorks and Roadways) SMS Secretaria Municipal de Saude (Municipal Secretariat o f Health) SMT Secretaria Municipal de Transporte (Municipal Secretariat o f Transport) SPM Secretaria do Planejamento Municipal (Secretariat o f Municipal Planning) U P Unidade de Planejamento (Planning Unit) CurrencyEquivalents: US$l.OO = R$1.76, effective as o fJanuary 14,2008 Government Fiscal Year: January 1-December 31 Weights and Measures: Metric system vii Acknowledgements This study was a joint initiative between the Social Development Unit o f the Latin America and Caribbean Region (LCSSO) and the Social Development Department (SDV) o f the World Bank. The report team i s grateful for the permanent support and encouragement o f McDonald Benjamin (Sector Manager, LCSSO), Steen Lau Jorgensen (Sector Director, SDV) and Caroline Kende-Robb (Sector Manager, SDV). The team would also like to thank the Japanese Consultant Trust Fund for its financial support. The team for this report was led by William Reuben (Senior Social Scientist, LCSSO) and consisted o f Andre Herzog (SDV), Tito Yepes (Infrastructure Economist, LCSSD), Alessandra Heinemann (LCSSO), EriWatanabe (LCSSO), and Irina Ghobrial (LCSSO) o f the World Bank. A backgroundreport by Marco Kamiya (PADECO, Ltd.) served as an input for the study. The team that compiled the PADECO background report relied on inputs from Brazilian consultants Tarson NCiiez, Luciano Fedozzi, Milton Andrt Stella, Ciro Biderman, Glauco Peres da Silva, Itamar Spanhol, Andrey Sgorla, and Leticia Almeida. Peggy McInerny edited the report. The report has benefited greatly from the close cooperation and support o f Mayor JosC Fogaqa and his team in Porto Alegre, including Municipal Secretaries CCzar Busatto, C16vis Magalhiies, Jotio Portella, and Cristiano Tastch, as well as the team at the Observatory o f Porto Alegre. Finally, the report also benefited fkom the feedback and suggestions o f representatives o f the Participatory Budgeting Council (COP), as well as those o f the former Mayor o f Porto Alegre, Jog0 Verle, and the non-governmental organization Cidade. The report team gratefully acknowledges excellent inputs, comments, and suggestions from McDonald Benjamin, Mark Lundell (Sector Leader, LCSSD), Darwin Marcel0 (Consultant, LCSSD), Fernando Blanco (Economist, LCSPE) and John Nash (Lead Economist, LCSSD), as well as from the peer reviewers for the study: Ana Bellver (Public Sector Management Specialist, LCSPS), Gianpaolo Baiocchi (Professor, University o f Massachusetts), Rob Chase (Lead Social Development Specialist, SDV), and Mila Freire (Senior Adviser, FEU). Suggestions made by Ethan Weisman (Lead Economist and Sector Leader, LCCSC), Jennifer Sara (Sector Leader, LCSSD), and Deborah Wetzel (Lead Public Sector Specialist, LCSPS) at the Decision Meeting greatly contributed to improving the quality o f this report. viii Towards a More Inclusive and Eflective Participatory Budget in Porto Alegre Executive Summary Background The Municipality of Porto Alegre, capital city of the southernBrazilian state of Rio Grande do Sul, requested World Bank assistance to improve its Participatory Budget (Orpzrnento Purticipah'vo,or OP) process. Participatorybudgeting i s a process by which citizens andor civil society organizations participate directly in different phases o f the preparation and monitoring of public budgets. More specifically, the municipality asked the Bank to focus on five key concerns: H o w can the "quality o f participation" inthe OP be improved? H o w can the interface between the OP and municipal fiscal management be strengthened? H o w can the capacity o f the OP to monitor budget executionbe improved? H o w can a more coherent system o f social accountability in the city be established, improvingthe articulation between the OP and different participatory fora? What are the social and fiscal impacts o f the OP? This report examines these five critical concerns andproposes recommendations to address them. Participatory budgetingin Porto Alegre originatedin an experimental, consultativeprocess driven by the municipal government and new social movements in the late 1980s. The process was initiated during the early years o f re-democratization and decentralization in Brazil, following the end o f the military dictatorship in 1985. The 1988 constitution initiated a process o f decentralization and tax reforms that created the fiscal space for municipalities to make more significant public investment decisions. A plethora o f participatory governance institutions have since proliferated in Brazil, most importantly, municipal policy councils and participatory budgeting bodies. The OP was formally introduced in Porto Alegre under the leadership o f the Workers' Party (Partido dos Trabalhadores, or PT) in coalition with pro-democracy social movements. Although initially fraught with difficulties due to fiscal constraints, the OP in Porto Alegre became gradually more systematized over time. Today, the OP has a complex methodology for organizing participation in a city o f over 1.4 million inhabitants, as well as for prioritizing public investments based on three main criteria: unmet basic needs, population, and citizen preferences. However, the OP constitutes one element in a broader complex system o f participatory governance inPorto Alegre. ConceptualFrameworkandMethodology This study draws on the conceptualunderpinningsof the 2004 World DevelopmentReport, which identifies participation as a central element in the practice of social accountability, definedas the range o fefforts by which citizens express their voice, engage indirect management or monitoring o f public services, and/ or participate in service delivery (Goetz 2003). Several interrelated arguments underscore the importance o f voice in achieving better accountability and 1 Economic Sector Work governance: voice i s a central element o f control o f bureaucratic performance and contributes to ownership, empowerment, civic learning, and democratic legitimacy. Furthermore, participatory mechanisms (like the OP) can break inefficient equilibria inthe structure o f a state and potentially bringabout efficiency gains interms of fiscal and social outcomes. The assumption is that the OP may contribute to municipal revenue enhancement because where citizens have voice in budgetary decisions they will be more willing to pay local taxes. A related assumption i s that participation by the poor may bring about Pareto-efficient reallocations, which governments would not be able to identify without "directly asking" citizens. Inthe context o f mechanisms for social accountability and voice, the OP i s a means to inform budget formulation. Its effectiveness will depend on the quality of voice. The success o f all participatory accountability mechanisms, however, hinges on an enabling environment for civic engagement. The methodology of this report draws on recent work on social accountability and governance to focus on five key dimensions of the OP: (i) the participatory process and conditions that determine the quality and inclusiveness o f citizen voice; (ii)the upstream interface between the OP and fiscal management; (iii) the downstream capacity o f OP as an oversight mechanism inbudget execution; (iv) the limits and potential of the OP within the broader system o f social accountability and participatory governance; and finally, (v) its social and fiscal impacts. The report uses a combination o f qualitative and quantitative methods, including econometric modeling, data analysis, representative surveys o f Porto Alegre households, semi-structured interviews, organized focus groups, and an examination of the annual city budgets and minutes o f OP meetings at which public works investments are decided. The research for this Economic Sector Work (ESW) has benefited from an iterative process o f dialogue with key stakeholders, who have been involved from the design phase through the presentation o f research findings. MainFindings What is the quality ofparticipation in the OP? Participation in the OP is substantialin Porto Alegre, but certain social groups suffer from a lack of incentives to participate. Around one-fifth o f the population reports having participated in the OP at some point in their lives. Women and low-income groups show high percentages o f representation. Ethnic minorities and people with disabilities are also well represented. Under-represented groups include the extremely poor, youth, middle- and high- income groups, as well as certain professional groups, such as entrepreneurs. Time requirements, the use o f complex language, and a limited understanding o f the OP process present a barrier to participation, especially at higher levels o f representation. For the very poor, the opportunity cost and expense o f participation i s quite high. Higher-income groups lack incentives to participate because the OP has a narrow focus on infiastructure provision to the poor and participation i s time consuming. The OP enjoys credibility among participants and the population of Porto Alegre as a whole, despite certain shortcomings regarding its internal upstream and downstream accountability.There is an unevenplaying field for OP participants in delegating to upper levels of representation. Participants with more years o f experience, available time, and knowledge o f the complex procedural rules o f the OP are more likely to be elected as regional delegates and councilors, which has led to low levels o f rotation among delegates. Additionally, reporting mechanisms between delegates and their constituencies are unsystematic and characterized by the 2 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre absence o f records. The prevalence o f a verbal culture among OP constituencies, which demands and expects less accuracy, and the fact that the OP i s itself perceived as a net gain in public accountability, may explain the existing mismatch between OP credibility and the shortcomings inits internalaccountability. What is the quality of the interface between the OP and the municipalfiscal management? Despite enjoying high levels of participation and credibility, analysis of the interface betweenthe OP and municipalfiscal management yields mixed results. Contrary to popular belief, the OP in Porto Alegre does not lead to suboptimal resource allocation or investment fragmentation. When it comes to the risk o f investment fragmentation (posed by regional pressures to allocate resources to small public works), the OP performs quite well: although the number o f demands for small investments has increased in recent years, it has not affected the overall share o f resources allocated to satisfir those demands. There i s no evidence, however, that the OP has played a positive role in contributing to fiscal balance in Porto Alegre. Indifferent years, expenditure and revenue forecasts did not match actual values. This fiscal gap was usually covered by generating current deficits and/or postponing capital investments. Inconsistencies between the OP cycle and the actual budget calendar have also contributed to the formulation o f capital investment plans based on weak fiscal data. These plans ended up creating a backlog o f non-implementedOP demands and consequently generated strong public discontent. The OP in PortoAlegre has worked mainlyas an instrumentfor interested citizens to voice their preferencesregardingcapitalinvestments, rather than a space of public debatewhere municipal fiscal policy and revenue and expenditure management as a whole can be discussed. An analysis o f the minutes of key meetings o f the Participatory Budgeting Council (COP) showed that it has not undertaken substantive assessment o f municipal fiscal policies and revenue capacity, particularly at critical moments. This behavior o f the COP can be attributed to the absence of clear motivation among its members to give fiscal oversight priority over ensuring that the demands o f their respective communities are included inthe current InvestmentPlan. The lack of regular training for COP members can be also a possible explanation, as the Law on Budgetary Guidelines (LDO), Law on Federal Budget (LOA),and Multi-Annual Plan or Multi- Year Plan (PPA) are o f a technical complexity that laymen can find difficult to understand. What is the capacity of the OP to monitor budget execution? There is room to improve the potential of the OP to contribute to downstream accountability of public expenditure. Although the OP in Porto Alegre has contributed to a greater understanding o f pro-poor public spending, and to directly engaging ordinary citizens in voicing their priorities for capital expenditure planning, it has not fulfilled its potential to improve budget literacy. It has also paid scant attention to budget oversight. The municipality offers positive conditions for budget oversight: it presents accounts through varied processes, has a sophisticated e-government system, and provides updated information related to budget planning, public expenditure, local government programs, and detailed socioeconomic data on its Website. In 2007, it launched a new publication, "Caderno de Prestaqio de Contas," reporting on key fiscal and budgetary issues and presenting for the first time the implementation o f OP demands in a consolidated manner. However, oversight i s still neither well codified nor supported by 3 Economic Sector Work sustained efforts on the part o f COP members and municipality officials, and thus takes place mainly in an ad-hoc manner. What is the relationship of the OP to other social accountability initiatives? Although participatory budgeting in Porto Alegre is probably the most well-known mechanismfor direct participation inBrazil, there are many other spaces for participation in localgovernance.As is the case inmany other cities, the multiplicity o f spheres o f interaction between the local government and civil society lends itself to overlapping roles, competition, and undesirable tension between different venues. At the same time, the absence o f an overarching framework leads to functional gaps inthe current constellation o f participatory mechanisms, such as revenue and expenditure tracking, procurement oversight, and performance monitoring and evaluation. I t is important to recognize, therefore, the limits of the OP within the complex network of citizen participation in Porto Alegre. Since its inception, the main objective o f the OP has been the inclusion o f citizen input into capital investment planning. Over time, and as a result o f its success and permanence, but also as a response to its need to understand and scrutinize the municipal budget, the OP has developed some capacity to monitor the execution o f prioritized public works. However, this capacity i s weak and needs to be strengthened. On the other hand, other spaces o f participation were created as major spaces for civic engagement before and after the establishment o f the OP. Although these other channels have specific purposes, such as the Public Policy Councils (which focus mainly on sectoral policy making) and the Programa de GovernanqaSolidaria Local (Solidarity inLocal Governance Program, which has a more holistic approach to local governance), there are clearly many overlaps and conflicts. It must therefore be recognizedthat the OP in Porto Alegre may be limited in its ability to expand into other areas o f participatory local governance, given that it needs to focus on strengthening its current role. However, the municipality and different non-governmental stakeholders that have an active role inother various channels ofparticipationhaveyet to come upwith aharmonizingframework that would create a more effective and efficient system o f civic engagement and social accountability inPorto Alegre (see Figure 1below). 4 Towardsa More Inclusive and Effective ParticipatoryBudget in PortoAlegre Figure 1. The policy and expenditure cycle and participatory fora inPorto Alegre POLICY & EXPENDITURE CIVIL SOCIETY- CYCLE LOCAL Gov. \ j I I I Lack of overarching `> framework to articulate the different civil society-local Planmnn Fora I I governanceinstitutions I Pamcipatoty ................... ........ .......:............., :: : Mlrring :: ...............ii ..sa What are the OP's social andfiscal impacts? The econometric analysis conducted for this study isolated the fiscal and social outcomes that can be attributed to participatory budgeting as opposed to other factors.. The econometric analysis inthis report compares Brazilian municipalities that adopted OP to a group o f non-OP municipalities that had the same opportunity to make improvements inpoverty, access to basic services, and fiscal performance. The analysis controls for the fact that in most municipalities where OP was implemented, the Partido dos Trabalhadores (PT, or Workers' Party) was a key political actor, either because the mayor belonged to the party or the party was the second most important political force in city. The analysis accordingly accounts for whether estimated impacts are the result o f OP, and not the result o f a political process (including strong participation through informal channels o f voice) associated with substantial support to the PT. The analysis thus ensuredthat OP was the instrument that causedthe impacts. The econometric analysis suggests that as a mechanism for improving pro-poor capital investments, participatory budgeting has contributed to a reduction of poverty rates in municipalities where it was adopted prior to 1996. It i s worth noting that this poverty impact occurred despite a reduction in GDP per capita inthese municipalities, suggesting that the OP can contribute to a redistributive impact in the long run. To date, it seems that the impact on better access to water and sanitation i s positive for all OP municipalities. Current findings do not indicate consistent positive impacts on fiscal performance. The second hypothesis o f the econometric analysis tested the impact o f OP on fiscal performance, assuming that it empowers the poor and, therefore, leads to more transparency in the way the public budget i s allocated, as well as to increased budget literacy. Participatory budgeting was not, however, found to have a consistent impact on fiscal performance. In fact, its impact appeared to be negative, when measured by the ratio o f operational expenditure to total expenditure. The absence o f positive impacts on fiscal performance associated with the OP in 5 Economic Sector Work Porto Alegre can be explained by the exclusive focus o f the OP on investment capital allocations and its weak connection with other fiscal management decisions (e.g., current expenditure allocations and execution, revenue collection, etc.). A more simple analysis consistent with the logic of treatment and controlevaluation sheds light on the specificsof OP inPortoAlegre. This analysis comparedPorto Alegre with all state capitals and with only non-OP state capitals. Using available data, it could not be concluded that Porto Alegre performed better than other cities in terms o f living conditions. On the fiscal performance side, the figures suggested that it i s not very likely that the recent fiscal problem in Porto Alegre was actually connected to the OP. It is more likely that Porto Alegre increased its operational expenses after 1996 for some other reason, and that it was not able to solve this problem even with the (relative) increase inrevenues observed after 2000. Overall Conclusions The abovementioned findings help to identify the potential and limitations of the OP. Findings on the poverty and fiscal impacts o f OP in Brazil suggest that OP is a participatory mechanism that has significant potential for pro-poor distributive impacts that lead to poverty reduction outcomes in the long run. Its ability to have a positive impact on fiscal performance is less evident. This apparent limitation has important consequences for the sustainability o f pro- poor investment. The results of the impact analysis confirm the overall findings: the OP in Porto Alegre constitutes a representative and credible conduit o f voice for the poor to inform the allocation o f public investments, but it has limitations with respect to its ability to enhance fiscal performance. The cycle mismatch between the OP and the municipal budget and the narrow scope o f OP investment allocations prevent it from enhancing budget literacy and accountability or enhancing the quality o f expenditures beyond investment allocation, to include revenue collection and overall quality fiscal and service delivery performanceinPorto Alegre. MainRecommendations Changes and improvementsin OP methodology can help overcome some of its limitations. Nevertheless, the structural limits o f the OP must be acknowledged by social actors and municipal authorities. There are other participatory mechanisms that are better suited to improving the accountability o f certain stages o f the public expenditure cycle, such as sectoral councils (which hold sectoral policies accountable for the delivery o f services). The following recommendations are intended to help improve the performance and impacts o f the OP in Porto Alegre. They also propose involving other organizations and mechanisms inthe establishment o f a system o f social accountability inwhich the OP would play a leadingrole. 6 Towardsa More Inclusive and Effective Participatory Budget in Porto Alegre How can the "quality ofparticipation in the OP be improved? Creating incentives and removing barriers to make the OP more inclusive is a critical challenge. Measures to address these challenges could include expanding the OP's focus to issues o f strategic importance for the city as a whole (including budgetary oversight, municipalrevenues and current expenditures) that affect the space for OP investments and would attract interest from the middle class and professional groups, such as entrepreneurs. Furthermore, participation could be made more convenient to attract those who currently do not participate, for instance, by making available online voting on investment priorities or reimbursing the financial cost o f participation borne by the poorest groups (e.g., the cost o f bus tickets for participants below a certain income level). Improving the effectiveness o f OP council meetings by reforming meeting procedures would save participants' time. Finally, improving the training and education o f participants i s essential for fully realizing the potential benefits o f the OP to enhance citizen understanding o f the municipal budget process. Training needs should be identified as an integral part o f the OP process at each level o f participation (e.g., general, delegates, and councilors) and for each stage o f the OP cycle. Making OP representation more legitimate and accountable is crucial to maintaining its credibility. Simplifying the OP's procedural rules and making them more accessible to OP participants would level the playing field among participants and prevent a situation in which only a handful o f OP representatives have knowledge o f the rules. Furthermore, establishing simple guidelines and formats for presentations by councilors and delegates would enhance the system o f reporting to constituencies and enhance downstreamaccountability. How can the interface between the OP and municipal fiscal management be strengthened? Tackling the time mismatch between the OP prioritization process and the municipal budget cycle is crucial. Movingthe OP cycle one month forward to ensure consistency with the Organic L a w o f the Municipality would product the best results in terms o f the quality o f information available during the prioritization process and consequently enhance the effectiveness o f OP demands. An alternative solution would be to ensure that the executive ann o f the municipality provides a reasonably accurate revenue estimate for the period for which OP investment demands are being prioritized. Between M a y and July, then, each region and sector should receive a more precise estimate o f the budgeted caps for capital expenditure inthe coming year. The articulation between municipal fiscal management and the OP could be further enhancedby expandingthe OP cycle for complexpublic works to 2 years, while maintaining an annual time frame for those OP investments that can be executed in a shorter time period. This step would increase the capacity o f OP participants to engage inthe oversight o f public works by reducing the time pressure that the one-year limit currently creates. Yet, it must be acknowledged making the OP bi-annual may result inthe possible tradeoff o f reducing the interface between the population and local government. Furthermore, introducing a mechanism in the OP cycle to allow for adjustments to the Investment Plan during budget execution would prevent the accumulation of a backlog of investments in cases of unforeseen decreases in revenue.This measure can be complemented 7 Economic Sector Work with measures to strengthen municipal capacity in budget planning, such as collecting and publishingindicators on the accuracy o frevenues and expenditures for eachbudget cycle, as well as creating an indicator to measure the accuracy o f disbursement estimates for each item in the Investment Plan. These indicators can then be used as benchmarks that are tracked by the OP Council. Aditionally, the municipal government could define a budget cap for the investments prioritized by the OP so that they do not exceed the municipal budget envelope. Improving fiscal and budget literacy, as well as creating spaces for overall fiscal policy dialogue inthe OP cycle, are essential for maximizingthe OP's potential positiveimpact on fiscal management. These goals can be achieved through a set o f measures, including adding a stage at the beginning o f the OP cycle for more in-depth public discussion o f the overall fiscal policy o f the municipality and the execution o f the budget. This should go hand in hand with basic training in municipal finance for OP councilors, including familiarizing them with Multi- Annual Plans or Multi-Year Plans (PPA), the Law on Budgetary Guidelines (LDO), the Annual Budget Law (LOA), bimonthly Summary Budget Execution Reports (RREO), and Fiscal Management Reports (RGF). Partnering with local universities and think tanks that could assist the COP (Participatory Budgeting Council) to undertake independent fiscal policy and budget analysis would complement these measures, as would developing the ObservaPoA Website into a system that offers statistical information to assess budget priorities and monitor the impacts o f investments. How can the capacity of the OP to monitor budget execution be improved? Strengthening budget literacy in Porto Alegre is a precondition for enhancing the OP's capacity to monitor budget execution. Currently, budget literacy is limited and uneven among the citizens o f Porto Alegre. Although the OP has contributed to a greater understanding o f the need for pro-poor public spending, it has not fulfilled its potential to improve budget literacy. To raise awareness o f basic municipal finances and budgeting, the OP could make more frequent use o f local radio, emulating experiences inPeru, Ecuador, and Guatemala, as well as in Anglophone African countries that have piloted this approach with the support o f the World Bank. Porto Alegre's OP could also include a supplement on fiscal and budget literacy inits Statute, as well as detailed information on municipal revenue and expenditures in the OP Notebook o f the Investment Plan. Moreover, OP Regional Journals could include detailed technical and financial information on each project being implemented in the respective regions, allowing for comparisonbetween planned activities and their execution, together with an overall evaluation o f the implementation o f all municipal programs in the Booklet o f Municipal Accounts, using the performance indicators adopted by the new management system. Furthermore, the scope of the OP could be expanded to systematically include budget oversight in its cycle of activities. The OP Council can use and publishon its Website simple indicators that compare overall budget planning data (e.g., amounts, time of implementation, basic technical details) with budget execution data for different investments. These indicators can be developed to enable cross-regional comparisons and benchmarking. It would be advisable that the municipality make available on its Website regular budget execution reports (Le., the bimonthly Summary Budget Execution Reports, or RREO) in a citizen-friendly manner. Aditionally, the municipal government could define a budget cap for the investments prioritized by the OP so that they do not exceedthe municipalbudget envelope. 8 TowardsaMore Inclusive and Eflective Participatory Budget in Porto Alegre Finally,strengthening mechanismsfor civil society monitoring and evaluation of municipal services would contribute to overall social accountability. These functions may well be beyond the scope o f the OP, but they can be performed by other civil society organizations and spaces for dialogue that involve user organizations, civil society, and the Municipality o f Porto Alegre, such as sector-based municipal public councils and the Forum o f Services. It would thus be advisable to strengthen the capacity o f these groups to monitor and evaluate service delivery. Citizen feedback mechanisms, such as community score cards and citizen report cards, can be introduced to strengthen client power in monitoring service performance. By establishing a citizens' charter o f municipal services that sets citizen-friendly performancebenchmarks for each service provided by the city, together with appropriate feedback channels, the city can create powerful tools for enhancing the quality o f service delivery. How can the system of social accountability be made more coherent and articulation between different participatory governancefora be improved? First, it is essential to re-examine the relationship betweenthe OP and MunicipalCouncils to ensure maximum synergies. The interface that currently exists between institutions o f direct participation i s weak, undermining the possibility o f a coherent system o f social accountability. Options to achieve better synergies could include the establishment o f an overarching framework that would encompass the different municipal councils and the COP; the framework would clarify formal and informal functions o f these bodies and remove overlaps in budget planning and oversight. At the same time, the interface between the different municipal councils and thematic assemblies o f the OP should be strengthened. This could be complemented by better integration o f the Urban Fora (public debate spaces on municipal master planning) with OP investment plan prioritization via a specific OP thematic group. Finally, it is essential to strengthen the alignment of long-term sector development objectives with OP annual capital investment planning for purposes of coherence and efficiency. This goal could be achieved by using the City Conference to implement a well- structured strategic planning process among existing civil society and local government institutions. Such a process would establish clear long-term goals that would align policy malung and budget planning. Moreover, the development o f a system o f social accountability in Porto Alegre would benefit from the establishment o f a permanent governance structure to oversee the formulation, implementation, and revision o f the city's strategic planning process. In addition to the City Conference, this structure should include a multi-stakeholder coordination committee, an executive secretariat, a technical team, and thematic task forces, comprising representatives o f the municipal public councils, thematic areas o f the OP, urban fora, and the Solidarity Local Governance program, together with representatives o f the municipality (from both direct and indirect administration), the local legislature, and the private sector. How can the understanding of OP impacts be refmed? The econometric analysis conductedfor this report suggeststhat further analyticalwork is necessaryto better understandthe impactsof OP and optimize its results. It is advisable that universities and think tanks in Brazil continue to assess the social and fiscal impacts o f the OP. The model and database developed for the impact analysis conducted by this study have the potential to be further explored and expanded to shed light on impact causalities surrounding the 9 Economic Sector Work OP and the value of the participatory process for bringing about poverty reduction and development . 10 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre Introduction A Demand-drivenStudy The Municipality of Porto Alegre, capital city of the southernBrazilian state of Rio Grande do Sul, requested analysis and advice to improve its Participatory Budget process (OrGamento Participativo, or OP, in Portuguese). Participatory budgeting i s a process through which citizens and/or civil society organizations participate directly in different phases o f the preparation and monitoring of public budgets. OPs vary in terms o f the level and type o f participation, as well as the technical criteria and instruments used to allocate resources. Inmost cases, participatory budgeting focuses on expenditures and involves participatory decision making in an annual or multi-year public budget plan.' The Municipality o f Porto Alegre was one of the first municipalities in Brazil to introduce this practice, which brought international attention in the early 1990s. Since that time, over 200 municipalities in Brazil have introduced participatory budgets. A fiscal crisis in 2002, however, generated concerns about the fiscal performance o f the municipality and the sustainability o f the OP. Porto Alegre accordingly sought support to improve its performance, inpart by strengthening its OP. Table 1 Porto Alegre at a glance Population 1,453,076 GDP (2003) R$14,655,093 GDPper capita (2003) R$10,437 Municipal budget (2006) R$7,330,270 Municipal budget surplus (2006) R$1,634,339 Staffing costs/ revenues (2006) 46.95% Debt/ revenues (2006) 20.91% Credit operations/ revenues (2006) 1.07% Debt servicing/ revenues (2005) 2.02% Life expectancy at birth(2000) 71.5 years Literacy rate ages 15+ (2000) 96.6% Human Development Index (2000) 0.87 Gini Index (2000) 0.61 Indigence (2000) 4.28% Poverty (2000) 11.33% Sources: IBGE(2000), ObservaPoA,Municipalityof Porto Alegre, Muninet. The Municipality requestedthat the World Bank focus on five key concerns related to OP participation, public management,accountability,and impactx2 The first concern relates to the quality of participation in the OP. This concerns the extent to which particular social groups are excluded from the process; the representativeness o f the participatory process and the effect this representativeness may 'Annex Definitionadopted fromthe WorldBank's SocialAccountability Sourcebook. 1 presents a synopsisof the mainstudies about the OP and its implementationinPorto Alegre. 11 Economic Sector Work have on its outcomes; as well as whether and how information on the OP process and citizens' understanding thereof could be improved. The second concern relates to the quality of the interface between the Porto Alegre OP and municipalfiscal management. One of the first issues raised i s whether the OP and the legal framework that regulates the municipal budget cycle are coherently interrelated. Furthermore, the municipality wanted to better understand the impact o f the OP on the quality o f public expenditure and how to strengthen OP capacity to contribute to sound financial management. The third concern relates to the potential of the OP to influence other stages of thepublic expenditure cycle where social accountability could be enh~nced.~ Appropriate operational guidance from the Bank team was thus requested. Specifically, the municipality was interested in finding out how user and citizen participation could contribute to holding the government accountable for actual expenditures and service provision, so as to improve the coverage and quality o fpublic services. Fourth, the municipality requested an analysis o f how the OP process interacted with other spaces for dialogue and participation in Porto Alegre, as well as an assessment o f the potential for building a system of social accountability that could overcome actual fragmentation among the different arrangements between civil society and local government . And finally, the municipality sought a better understanding o f the impacts o f the OP on fiscal and social outcomes. To date, Porto Alegre has not instituted rigorous analysis of the fiscal and social impacts o f the OP process, in spite o f pioneering participatory budgeting inBrazil. Historical Evolution of the Participatory Budget A mechanismfor increasedparticipationin the municipalbudget process first began to be discussedin Porto Alegre in the late 1980s. This discussiontook place in the context of the re-democratizationand decentralizationof the country following the end of the military dictatorship. Today's OP can be traced back to an experimental, unstructured consultative process that was set up to define the city's investment priorities by a newly elected municipal government in the late 1980s in response to highly mobilized social movements in the city. Although initially fraught with difficulties due to fiscal constraints, these processes became more institutionalized over time and in 1991, specific criteria and guidelines were defined for determining investment priorities and the distribution o f resources in Porto Alegre. Since that time, the methodology has been refined to include the prioritization o f investments by territory and sector. Social accountability is defined here as an approach toward building accountabilitythat relies on civic engagement, Le., an approach in which ordinary citizens and/or civil society organizations participate directly or indirectly in exacting accountability. In a public sector context, social accountability refers to a broad range of actions and mechanisms that citizens, communities, independent media, and civil society organizations can use to hold public officials and public servants accountable. These include, among others, participatory budgeting, public expenditure tracking, monitoring the delivery o f public services, investigativejournalism, public commissions, and citizen advisory boards(Carmen, Foster, and Singh, 2004). 12 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre Constitutional reforms gave local governments more power to make decisions on local investments, while stipulating that participatory mechanisms be established. The de- concentration o f funds encouraged by the Constitution o f 1988 and the approval o f local tax reform (Reforma Tributdria Local) created the necessary fiscal space for public investments to be determined at the municipal level. At the same time, the Constitution created a hybrid political system that i s fairly unique in that it combines forms o f representation and participation at both the federal and local levels (Avritzer 2006). Political participation follows two tracks: (i) direct participation as an expression o f sovereignty in the shape o f plebiscites and referenda, and (ii) institutionalized participatory mechanisms, such as sector-based public councils and participatory budget processes. The principal participatory institutions that have proliferated in Brazil after 1988 are municipal policy councils (Conselhos de Politicus) and participatory budgeting (Orpunento Purticiputivo). Municipal policy councils resulted from the constitutional articles regulating health (Lei Orgdnica da Sazide),social assistance (Lei Orgdnica da Assistkncia Social), and child and adolescent issues (Estatuto da Crianqa e do Adolescente). While the parameters o f these policy councils are determined in the Constitution, their institutional format i s defined by local legislation, which makes for a substantial degree o f variation between them. Unlike policy councils, participatory budgets are not mandated by the 1988 Constitution, but they also strike a characteristic balance between representation and the broad participation o f citizens+lected officials share their decision-making power on budgetary priorities with a host o f regional and thematic assemblies that operate on the principle o f direct and voluntary participation. Box 1. A briefhistory of the OP Systematic calls for innovation in urban governance emerged with the rise o f new social movements in Brazil inthe 1970s. The first experiments with participatory schemes at the municipal level date back to the 1980s. These movements emphasized the need for greater participation in decision making, better accountability, democratic access to services, and the end o f patronage schemes. These demands coincided with a rise in local organizations and networks, such as neighborhood associations, as well as national movements, such as the Housing Movement and the Collective Transports Movement, which stressed the right to certain services. Experimentation with participatory schemes to achieve greater local control over resource allocation decisions began in a number o f municipalities governed by opposition parties while the military regime was still inpower. After the military regime ended, democratization and decentralizationcreated the political environment and fiscal space for policy innovations like the OP, many o f which were implementedwhen the Workers' Party (PT) was inpower. The Constitution o f 1988 and subsequent decision by most state capitals to reformtheir finances created the necessary fiscal space at the sub-national level to allow municipalities to adopt innovative policies like the OP. The rise o f PT mayors and left-wing coalitions from 1988 onwards contributed to the widespread adoption o f participatory programs, which varied broadly in scope and methodology. InPorto Alegre and a number of other Brazilian municipalities, the PT administration instituted an early version o f the OP in the late 1980s. The process has been institutionalized more formally since that time, despite changes in municipal leadership. The early success o f OP is credited in part for having kept opposition at bay and maintaining the PT in power until 2004 (Baiocchi et al. 2004). The onset of a financial crisis in 2002 and the resulting reduction o f the municipality's investment capacity created a backlog o f investments that were prioritizedby the OP but not executed. In2004, a coalition o f opposition parties was elected, led by mayor JosC Fogaqa o f the Popular Socialist Party (Partido Popular Socialista, or 13 Economic Sector Work PPS). The current municipal administration won the elections on a platform of rebalancing municipal finances while maintaining the OP as a mechanismto informinvestment decisions. Today, the OP in Porto Alegre functions within a complex system of civil society- governmentarrangements.These arrangements include municipal policy councils, participatory urban planning fora, and a citywide strategic planning conference (Congress0da Cidade),among others. The proliferation o f participatory fora has sometimes created challenges for coordination and articulation between these institutions, especially at the stage o f policy formulation. Inrecent years, the municipality o f Porto Alegre has begun to implement measures to modernize its public administration and local governance schemes. These measures include expansion o f the scope o f participatory governance by improving social monitoring and information mechanisms, such as the Obsewatdrio de Porto Alegre (Observatory o f Porto Alegre), and fostering partnerships between civil society and the private sector to address local community challenges through the Programade GovernanCaSolidciriaLocal (Solidarity inLocal Governance Program). The OP is viewed as a centralelement inthe participatorygovernancearchitectureof Porto Alegre. This report proposes recommendations to address some o f the critical challenges that the municipality currently faces inmaking the OP more inclusive and effective. The OP Cycleand Process The first step inthe OP processis a review of the implementationof the investmentplanfor the preceding year (Prestaqio de Contas, or presentation o f accounts). This review takes place inMarch-April of every year and focuses primarily on capital investments of the previous year, rather than on municipal budget expenditures as a whole. The first step in the cycle also includes mobilization and awareness-raising activities that explain the functions o f the OP, as well as its rules and regulations. The second step in the process is establishing thematic priorities. From April until May, regional and thematic assemblies take place with the objectives o f (i)establishing thematic priorities by voting, (ii) electing councilors for each region, (iii) defining the number o f delegates, and (iv) repeating the budget review for the preceding year at the local level. These meetings are open to all citizens and constitute the central and most inclusive component o f the process. Porto Alegre i s divided into 16 regions, which are again subdivided into several micro-regions where regional assemblies are held to ensure close proximity to the grassroots level. The five thematic areas discussed inthe OP at city-level assemblies are: transport and traffic; education, leisure, and culture; health and social welfare; economic development and taxation; and organization o f the city, urban and environmentaldevelopment. The third step is the prioritization of specific investment demands. From May until July, regional and thematic assemblies elect delegates and the prioritization o f demands takes place (votes are tallied and aggregated to provide regional and overall rankings). For each theme, a point score i s calculated by multiplication, using a coefficient and the total number o f votes cast per priority ranka4Delegates from all regions then meet to review the funding requests; visits to proposed project sites are used to inform their discussions. A pre-agreed formula and indicators These coefficients are 4, 3, 2 for the first 3 priorities, and 1for priorities ranking from4 to 13. 14 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre are used to allocate resources among sectors and regions on the basis o f population, unmet services and infi-astructure, and ranking o f sector preferences. In July, new councilors officially take over their mandates. Their main function is to harmonize competing demands and prepare the draft investment plan in coordination with the municipal administration, which undertakes a technical feasibility and financial analysis o f the proposed projects. In August and September the investment plan (IP) i s incorporated into the city's budget matrix and submittedto the local legislature for revision and approval. Once the prioritization process is finalized, investmentplans are presented at regionaland thematic forums. These meetings are attended by representatives o f the Cabinet for Budgetary Planning (Gabinete de Programaqio Orqamenthria, or GPO) and the Municipal Secretariat o f Political Coordination and Local Governance (Secretaria Municipal de CoordenaqEo Politica e Governanqa Local, or SMCPGL) and other sector secretariats, as necessary. The final step in the OP is a revision of its statute. This step takes place in November and December. Amendments to and modifications o f OP rules can be proposed and voted on before the new cycle begins the following March. Figure 2. Participatory budgetingcycle inPorto Alegre 15 Economic Sector Work ConceptualFramework This study on the Participatory Budget of Porto Alegre draws on the conceptual underpinnings of the 2004 World Development Report (WDR04).' The WDR04 discusses the role o f citizens and service beneficiaries in the framework o f accountability for service delivery. It identifies two ways in which citizens participate and demand accountability: by expressing voice (i.e., through voting, direct participation in policymaking, or influencing public officials) and by participating and demanding accountability as "clients" o f public services (implementing policy and participating directly in service delivery, management, or monitoring). For example, parents who participate inparent-school associations to monitor the quality o f education provided by their local schools engage inthis second form o f participation, referred to as clientpower (see Figure 3). Voice is an indispensablefactor that balances the power relationshipsbetweengovernment, service providers, and users. After an extensive review o f service delivery experiences all over the world, the WDR04 concluded that citizens' voice, expressed through electoral processes or other forms o f political participation, creates important incentives to improve both social policy design and the oversight and regulatory role o f the state over service providers. In other words, the report finds that the existence o f strong accountability relationships between citizens and politicians has a positive impact on the rules o f engagement (the compact) that governs the relationshipbetween the state and its service provision agents. ";(a Figure3. Key relationshipsof power Long route of accountability The WDR04 explorestwo routes through which citizens or clients can exert accountability. Expressing client power i s considered the short route to achieving accountability because beneficiaries exert direct influence and power over services, affecting service delivery without going through the state. Expressing voice i s considered the long route to achieving accountability because it involves citizens trying to influence service delivery via the state and its social policy. WorldBank (2004). 16 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre Both types o f participation are important parts o f the practice o f social accountability, defined as the range o f efforts by which citizens express their voice, engage in direct management or monitoring o f public services, and/or participate in service delivery (Goetz 2003). This study focuses primarily on voice, since the OP i s a mechanism o f voice through which citizens engage inaparticipatoryprocess aimedat influencingpublic expenditure. The World Bank has identifieda set of social accountability mechanismsand practices that are implementedby civil society at different stages of the public expenditure cycle.6 These mechanisms can be tailored to each stage of the budget cycle and provide an entry point for voice inmunicipal decision making. Civil society organizations, independently or in conjunction with local and national governments, have engaged in participatory processes to influence decision making in planning and budget preparation, budget analysis and resource allocation, and expenditure tracking and the monitoring o f service delivery (see Figure 4). The OP was designed as a mechanism to inform the first step inthe budget cycle. However, the process can potentially be expanded to cover other stages o f the cycle, such as analysis o f the government budget as a whole and oversight o f expenditure quality. Figure4. Mechanismsfor social accountabilitythroughoutthe publicexpenditurecycle STAGESINPOLICY AND BZ'DGZT SOCIAL ACCOUXTAB~ZTY CYCLE APPLICATIOXSAMl TOOLS Participatory poLicy/ budgat Iormulatlon The effectiveness of participatory accountability mechanisms to improve public decision making depends on inherent factors relatedto the quality of voice. If citizens' voice is not representative or badly informed, it will not make a positive contribution to improving accountability and development effectiveness. The success o f participatory accountability mechanisms hinges on an enabling environment for civic engagement.' This study examines the OP in Porto Alegre as a mechanism for the articulation o f voice and attempts to assess its contribution to enhanced accountability and better governance throughout the public expenditure 'SeeWorldBank(2003). Carmen, Foster, andSingh(2004). See 17 Economic Sector Work cycle. The framework applied here identifies three analytical dimensions o f the quality of voice in the Port0 Alegre OP: Inclusiveness. D o a range o f social actors participate in the process, especially groups that have traditionally suffered from exclusion? Are there barriers to the participation o f certain groups? What are the incentives to participate? What i s the "cost" o f participation and are the necessary minimumresources available for participation? Legitimacy. I s the OP process itself considered fair, credible, and lawful? Are participants autonomous or are there risks o f patronage, capture by internal leadership, or cooptationby the municipality and/or political parties? What i s the internal accountability o f the process itself? Does the OP actually influence budgetary decision making? Information. D o citizens have easy access to timely, relevant, and reliable information in order to participate constructively in the process? D o deliberation and debate in the context o f the OP improve the quality o f projects and their prioritization? In addition to inherent factors, the ability of voice to influence decision making and impact development outcomes depends on broader political economy factors and the insertion of voice within the existing governance scheme? Representative democracy i s in itself not a guarantee for good governance. Acemouglu, Ticchi and Vindigni (2007) have modeled the emergence and persistence o f inefficient states based on a theory o f how patronage can affect the transition to democracy. Under certain circumstances, by choosing inefficient state structures an established elite may be able to use patronage and capture democratic politics. This enables them to reduce the amount o f redistribution and the provision o f public goods in a democracy. Moreover, an inefficient state creates its own constituency, which contributes to its persistence over time. This governance scheme leads to a political equilibrium that explains the survival o f inefficient and captured states informal democratic frameworks. The de jure and de facto exercise of power by a new political coalition is essential for breaking an old equilibrium. When a new political coalition comes to power ina democracy, its incentive i s to break the old political equilibrium, reform the state structure, and make it more efficient, thereby enabling the state to collect higher taxes at a lower cost, lower bureaucratic rents, and allocate public investment to meet the priority needs o f coalition constituents. Acemoglu and Robinson (2006) suggest that this type o f reform can only be undertaken when there i s reasonable consistency between the exercise o f dejure and defacto political power. Their paper offers a rich conceptual basis for situating OP in the context o f the political economy o f a transition from a system o f elite capture to one o f more broad-based democratic politics. As the authors argue, changes inthe dejure institutional setup may not translate into an altered defacto balance o fpower, since established elites are in a better position to maintain defacto power. Even when an electoral process brings about changes inthe dejure balance o fpower, established elites can attempt to maintain their defacto power by intensive lobbying. Broad-based participation can 8 InWilliamson's (2000) categorization, governance represents the thirdlevel o f institutions. The first level consists o f basic social customs, norms, and religion, and takes centuries to change. The second level encompasses the cornerstones o f the institutional environment, such as constitutions, and can change more rapidly than the institutions comprising the f i s t level. The third level refers to the actual rules that govern transactions and organizations and make up the governance scheme. Finally, the fourth level comprises routine economic transactions. 18 Towards a More Inclusiveand Effective Participatory Budget in Porto Alegre serve to help newly elected officials to strengthen de facto power and break instances o f elite capture. Participation mechanisms that encourage collective action from the constituent base and channel this voice to guide decision making can be a powerful tool for legitimizing reform programs. These participation mechanisms can take different forms and have varying degrees o f formalization, ranging from loose consultative processes to structured co-management arrangements. The Worker's Party (or Partido dos Trabalhadores, PT), which was newly elected to a number o f municipalities in Brazil in the 1980s, experimented with different degrees o f formalization for participatory mechanisms. In some municipalities, the chosen method to encourage collective action included participatory planning schemes and different types o f consultations. In Porto Alegre and other Brazilian municipalities the PT chose to formalize participationto a greater extent, institutingan OP upon coming to power. The creation of the OP in Porto Alegre provides an example of a case where municipal governance structures formalized participative processes in order to upset the old political equilibrium. This formalized participation helped ensure that the feeble dejure power obtained by the new political coalition through elections was complemented by a collective action mechanism such as the OP, which strengthened its defacto power. Through the OP the executive branch o f the municipality opened up the budgetary process to formalized participation from the population. As a result, Porto Alegre instituted a mechanism whereby investment priorities decided by OP participants are incorporated into the official budget that the mayor presents to the municipal council for its approval. Formalized vs. ad-hoc types of participation. In the majority o f municipalities participation occurs in an ad-hoc fashion and i s not subjected to pre-established rules. The effectiveness o f each type o f participation i s debatable. A model developed by Dixit (2003), which examines the benefits o f moving from an informal governance system to one that i s based on rules, can help explain when each o f the two models works more effectively to meet expected governance results. Although Dixit's paper focuses on trade transactions, its conclusions offer insight into public sector governance. The paper concludes that informal governance arrangements work for small communities, whereas a (costly) external system o f governance i s necessary in larger ones, even though the payoff of such a system may not be larger than that o f a smaller one. Applying these conclusions to the OP, one can infer that a small community i s in lesser need o f a rules- based system to determine resource allocation since everyone has relatively equal access to decision makers and the latter are, in turn, aware o f citizen preferences (i.e., their investment needs). In a large city, however, resource allocation decisions are more difficult and there i s a benefit to creating a rules-based system o f participation that ensures voice i s more evenly distributed, althoughthis system comes at a cost. Better social outcomes and higher fiscal efficiency are the expected results of OP. Assuming that participation inthe OP process inBrazil brings decision making closer to each neighborhood through a rules-based process, it i s possible to expect, using this framework, an increase in income distribution and an improvement in fiscal management, unless some form o f "capture" occurs in the participatory process. The underlying assumption o f this study i s that the introduction o f OP will change municipal governance by increasing budget transparency and accountability and empowering the poor to voice their public investment preferences through a process regulated by pre-established rules o f engagement. The argument i s that formalized 19 Economic Sector Work popular participation in budgetary process in large cities will lead to: (a) better social outcomes, since opening a channel o f participation to all citizens (not just the elites, where lobbying power i s concentrated) can ensure that government spending corresponds more closely to the actual investment needs o f the poor; (b) more efficient fiscal policy, as citizens should be more willing to pay local taxes when they have more control over allocation decisions, and the incentives for elected officials to improve tax collection are also greater when a broad base o f citizens has concrete public investment expectations; and (c) lowering the incidence o f corruption, as local government expenditures will become more transparent This study aims to investigate how these assumptions have played out in Porto Alegre by responding to the five key concerns about OP performance presented at the beginning o f this Introduction. Methodology The methodology of this report draws on recent work on social accountability and governance. It involves an analysis o f five key dimensions o f the OP: (i) the participatory process and conditions that determine the quality and inclusiveness o f citizens' voice; (ii) the upstream interface between OP and fiscal management; (iii) the downstream effects o f OP on the entire policy and public expenditure management cycle; (iv) the potential and limits o f articulating a system o f social accountability, together with other participatory spaces; and, finally, (v) the correlationo f OP with both fiscal and social outcomes. The report uses a combination of qualitative and quantitative methods and has benefited from an iterative process of dialogue with key stakeholders. The stakeholders were involved from the design phase through the presentation o f research findings; their input has been incorporated either as part o f the analytical inquiry or as evidence. More specifically, the chapters use the following methods: Chapter 1 examines the extent and quality o f OP participation by drawing on a representative survey o f Porto Alegre households, as well as semi-structured interviews and organized focus groups with OP participants and non-participants alike. The survey data took into account regional characteristics in order to ensure fair representation o f players within each stakeholder group and a proportional number o fplayers across all stakeholder groups. Semi-structured interviews and focus groups were instrumental in establishing sound stakeholder analysis for the study, as well as uncovering and discussing issues o f inclusion, governance, and management constraints o f the OP model. A new survey was conducted in Porto Alegre in December 2006 for the purpose o f this report with a representative sample o f 1,000 people and a control group o f 300 OP participants. The survey yielded valuable information about the OP and disaggregated OP participants by region, sex, age, and income group. Chapter 2 analyzes the interface between OP and municipal fiscal management by reviewing data on municipal fiscal and management reforms, municipal finances (i.e., fiscal discipline, resource allocation efficiency, etc.), and public accountability systems (i.e., strategic budget planning, budget formulation, resource management, internal checks and balances, accounting and reporting, external audits, etc.). This longitudinal analysis o f 15 years i s limited only to data relevant to the evolution o f 20 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre the OP process in Porto Alegre, especially data from budget headings and items that have had greater leverage inthe participatory budgeting process and are critical to the delivery o f public services. Chapter 3 examines the downstream effect o f the OP on the broader system o f social accountability by using institutional analysis, as well as quantitative and qualitative methods employed inChapter 2. Chapter 4 analyzes the OP within the broader framework o f participation and demand for good governance. It also looks at the potential o f the OP to strengthen the system o f social accountability in Porto Alegre, using the same methods employed in chapters 1and 2. Finally, chapter 5 uses econometric tools to undertake a counterfactual analysis o f the impacts o f OP on fiscal and social outcomes in a group o f more than 150 Brazilian municipalities that have implemented OP, as well as specifically inPorto Alegre. 21 Chapter 1. Quality of Participationinthe OP This chapter focuses on how to improve the quality ofvoice inthe OP, that is, how to reduce the extent to which particular social groups are excluded from the process and the effects this exclusion has on OP outcomes. Analysis o f the quality o f participation in the OP process i s a decisive element for verifying its sustainability. Furthermore, the chapter analyzes the information available to citizens on the OP and whether and how that information and citizen understanding thereof could be improved. Specifically, the chapter examines: who participates in the OP at different levels o f representation and the obstacles to their participation; the legitimacy o f the OP, i.e., the autonomy o f its agents and the internal accountability o f the process; the quality o f information provided to OP participants; and the extent to which the OP has been able to influence decision making. Finally, the chapter offers specific recommendations for improving the quality o f voice and participation inthe OP. This chapter relies on quantitative and qualitative research methods, including a number of surveys conducted on the OP, including one on OP participation completed in December 2006 specifically for this study. It also draws on focus group interviews conducted with groups o f OP participants and non-participants alike, as well as interviews with individual^.^ 1.1. Who participatesinthe OP? This sectionfocuses on the degreeof OP inclusivenessby examining the OP participationrates o f different groups in society over time (including their level o f representation) and the main obstacles to their participation, including measures taken by the municipality to enhance OP inclusiveness." The number of OP participants grew rapidly during the early years of implementation. More recently, participant numbers have declined and stabilized at around 11,000 participants per year. Participation grew from 628 people in 1990 to over 6,000 in 1992 (see Figure 5). From 1993 to 1996, the number o fparticipants stabilized at around 7,000, before a new wave o f expansion started in 1997, reaching an apex o f 17,000 in 2002. Since 2002, participation has fallen to the levels o f 1997, declining to 11,500 in 2006. The financial crisis o f 2002 and the resulting backlog o f approved OP investments are plausible explanations for the reduction in participation since 2002. 9See Annex 2 for Guidelines for the Focus Group. loFor a comprehensive discussion ofparticipation inthe OP, see Fedozzi (2007). 22 Towardsa More Inclusiveand Effective Participatory Budget in Porto Alegre Figure 5. Number of OP participantsinRegionaland Thematic Assemblies, 1990-2006 Evolution of the number of OP participants 1990 to 2006 Numberof OP parlicipants +-General Total 20000 7 18000 16000 14000 12000 10000 8000 6000 4000 2000 0 Year Source: Municipality o f PortoAlegre. Although not all social groups display the same level of participation in the OP, overall participation is significant. In general, OP participants tend to come fiom lower-income households, be older, and have low-to-medium levels o f education. However, despite the under- representation o f certain groups (such as the very poor), the involvement o f large numbers o f citizens in discussions o f the public budget reveals an interest in participation. The survey conducted for this report found that 19.8 percent o f Porto Alegre residents had participatedin the OP at some point intheir lives through 2006. Men and women participate equally in the OP and in recent years, women have reversed the trend of under-representation at higher levels of representation. Historically, public attendees o f OP community assemblies have consisted o f a fairly equal number o f women and men, but females tended to be under-represented in the COP (Conselho do OrGarnento Participativo, or Participatory Budgeting Council) and the delegates' forum. However, the situation began to change in 1998. By 2005, women participants for the first time slightly outnumbered men participants (52.8 percent), as did women councilors (53.5 percent) and women delegates (52.4 percent). Women also outnumber men in the regional fora (54.5 percent), while men outnumber women in the thematic councils (53.2 percent). Of note, participation rates vary by topic-women make up the majority of participants in the thematic council on health and social assistance (80 percent), while men make up the bulk of participants in the council on economic and tributary development (70 percent). Overall, OP representation rates reflect Porto Alegre's demographic structure, as shown inFigure 6 (IBGE 2000)." l1Since 2003, the minimumshare o f 50 percent female participation inthe COP and Forum o f Representatives has been treated as a recommendation, not as a compulsory rule. 23 Economic Sector Work Figure 6. OP participation by sex inRegionaland Thematic Assemblies, 1993-2005 PercentageParticipationin the OPby Sex (1993-2005) 70 60 50 40 - - =Masculine 1993 1995 1998 2000 2002 Year -Feminine Source: Fedozzi (2007). Participation among women varies according to marital status; married women are rarely elected councilors or delegates. A survey conducted with the participating public in 2005 indicatedthat female councilors are predominantly single (62 percent), while male councilors are mostly married (71.4 percent). Among delegates, a considerable share of women are widowed or separated, while males are predominantly married (57.7 percent). This indicates that while women as a whole are represented in the OP, married women are rarely elected delegates or councilors, perhaps due to time constraints or a patriarchal culture. Some of the concrete barriers to greater participation that were identified by married women included the late evening time of the meetings, childcare obligations and the perception that the OP does not provide a space for the discussion of gender issues.'* Data about the ethnicity of OP participants indicates higher rates of participation among Afro-Brazilians, indigenous people, and Brazilians of Asian descent, while those who identify themselves as white are under-represented. While Afro-Brazilian participation was low in 1995, their reported participation rates have doubled since that year, a finding that could signify an increase intheir participation or a change inethnic consciousness.13Surveys conducted in 2000 and 2002 indicated that indigenous people and Asian-Brazilians were over-represented relative to their share o f the population in the 2000 census. Those who identified themselves as white were under-represented at both the assembly level and as councilors and delegates. Higher participationrates among ethnic minorities are a positive indication o f social inclusion, since their participation i s assumed to counterbalance power inequalities present in Brazilian society. However, it should be noted that the OP attracts participants who are more inclined to identify themselves as belonging to a minority group. ''See Annex 5 for complete survey tabulation. l3See Annex 5. 24 Towards a More Inclusive and Efective Participatoly Budget in Porto Alegre 42 to 49 18.8 19.4 18.6 18.6 21.7 35.3 26.9 13.8 50 or + 22.4 25.2 25.9 24.5 27.6 44.0 37.4 29.3 NiR 0.9 0.3 1.1 0.1 0.0 Total* 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 On average, youth are underrepresentedin the OP, although in recent years the share of youth participation has increasedover time. Historical data indicatethat the participation share o f youth aged 16-25 has risen moderately over the last decade, from 15.8 percent in 1995 to 19 percent in 2005. Similarly, the participation share o f those aged 50 and older has slightly increased, while participation by the middle age bracket (ages 34-41) has fallen. Time requirements are a likely reason for the under-representation of younger citizens, especially among councilors and delegates. A survey conducted by this study in 2006 of 44 current and former councilors showed that participation in the OP i s time consuming. The majority of delegates devote between 4 and 8 hours per week to the OP; 21.4 percent spend more than 12 hours per week. The amount o f time required to participate inthe OP, especially at higher levels o f representation, is likely a barrier to participation for under-represented groups such as younger people who are infull-timeemployment. Table 3. Time devotedto OP by councilors, 2006 Time spent I Yo Up to 4 hoursper week 23.4 Between 4 and 8 hoursper week 40.4 Between8 and 12hoursper week 12.8 More than 12 hours per week 21.4 Although the middle-aged and elderly make up a significant share of OP participants, especially councilors, their presence has not translated into specific demands for their age group. The relative over-representation o f older participants is even more pronounced when the age distribution of delegates and councilors i s analyzed: people aged 50 years or older represent 25 Economic Sector Work almost half o f elected council members (43.96 percent).I4 Despite their relative over- representation, the OP has not focused on issues o f special interest to this group, as the Municipal Council for the Elderly (COMUI) maintains. Participants seem to represent localized demands rather than sectoral issues, such as health. In fact, the methodology o f the OP appears to disadvantage sectoral issues, as one community member observed during a focus group meeting, "We aren't able to gather enough elderly to vote; each community gathers a lot of people due to the demands of the location, but not specijkally elderly people. " Data on working hours presents two clusters of participation, suggesting that councilors have long, albeit flexible, workdays. A considerable share o f OP participants reported being without remunerated work (over 30 percent), while a comparable share reported working 4 0 4 8 hours per week (over 35 percent). Among those who had participated for 5 years or more, a significant share worked 40-48 hours per week (41 percent). Workday data did not reveal differences between delegates and councilors, although councilors tended to have long, if flexible, work days. The under-representation of youth i s especially severe at delegate and councilor levels, where only 1.7 percent o f youth ages 16-25 have been elected councilors and only 8.1 percent as delegates. Those aged 33 and younger have comparable representation levels. Participating youth are predominantly male and their participation i s higher in regional (20.3 percent) than thematic (13.9 percent) assemblies. Student participation in the OP i s around 4 percent overall, with 0.9 percent o f this group elected councilors and 1.2 percent, delegates (Fedozzi 2007). The OP has been criticized for not involving youth more actively. Inthe words o f one interviewee, "the OP doesn't involve enoughyouths and they 'renot encouraged to bepart of it."" Youth also identify the content and format of meetings as disincentive to broader youth involvement. Some youth complained about the lack o f a systematic discussion o f public policies relating to youth in the OP as a reason for their non-participation. While youth groups have created a Youth Forum and have held sub-meetings in the thematic assemblies, these strategies appear to have yielded only limitedresults. The hip-hop movement, which i s primarily comprised o f poor youth fiom the suburbs, i s one of the rare exceptions o f a youth group that participates in the OP. Young people are most actively involved inthe thematic assembly on culture, where they represent 30 percent o f all participants. In terms of educational level, the majority of OP participants have completed primary education. A survey conducted in 2000 found that the majority o f participants (55.1 percent) had either completed primary education or more, which was close to 53.35 percent inthe 2000 census (IBGE 2000). The most recent survey o f 2005 showed that the share o f participants with primary education or more now accounts for more than 63 percent o f OP participants (see Table 4). A comparison o f participant education levels in2000 and 2005 with demographic data (IBGE 2000) shows that people with higher education and people without education or with incomplete primary education were under-represented in the OP. Data also indicates differences in educational background across OP fora: in2005, more than 60 percent o f participants inthematic fora either had completed secondary education (36.8 percent) or had a complete or incomplete higher education (27.7 percent); the figures for regional assemblies were significantly lower (Fedozzi 2007). l4Fedozzi (2007). Interview with A l e d o Guara, leader of the Hip-Hop Movement and OP participant. 26 Towards a MoreInclusive and Effective Participatory Budget in Porto Alegre Table 4. OP participation by levelof education, 1995-200516 LevelsofEducation(%) 1995 1998 12000 2002 1 1Council members Population ( 2 0 0 5 ) - ~ ~ ~ ~ 2005 (2005) Yo - HigherEducation 14.2 16.1 20.1 12.0 15.5 26.7 18.0 21.38 complete and incomplete NIR - 2.0 3.8 0.1 0.4 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Complexity of language used and a limited understanding of the OP process are perhaps the most severe barriers to participation for youth and those with low educational attainment. Producing materials that are sufficiently accessible to these groups has been a significant challenge, given the technical complexity o f the budget. These groups also show limited understanding o f how the OP functions. Hence, the lack o f a comprehensive learning methodology and capacity-building initiatives are also severe barriers to expanding OP participation and enhancing its quality. The percentage o f people who have been participating in the OP for 2 to 4 years and do not have knowledge o f OP rules remains high (40 percent). The share o f participants who know only a few rules i s even higher (45 percent). Those with secondary and higher education, however, tend to understand the rules (over 50 percent). Conversely, those with incomplete primary education tend to declare that they do not know the rules. These figures point to an absence o f an effective pedagogic methodology for ensuring full mastery o f the rules, which i s a serious concern for the sustainability and legitimacy o f the process. Delegates and councilors tend to have higher levels of education and income compared to other participants.The series o f abovementioned OP participant surveys reveals that contingents with lower education and income are not adequately represented at delegate and councilor levels. According to a survey conducted in 2005, for example, 47.7 percent o f participants had a basic education, but only 23.3 percent o f this group was elected to the COP (Fedozzi 2005). Representation of those with lower levels o f schooling i s higher among delegates. This finding has significant implications for the learning opportunities offered by the OP process. l6Size of the sample surveys was 10percent of the total number of participants. The sampling error was *2.4. 27 Economic Sector Work Overall, OP participants report low household incomes; higher-income groups are proportionally under-represented and their participation continues to fall. While 31.1 percent o f participants reported household incomes valued at up to 2 minimum salaries (MS) in the 2000 census, 63.2 percent o f OP participants fell into this income bracket in 2005. Conversely, data indicate disproportionately low levels o f participation among higher-income groups. The share o f OP participation among groups earning more than 10 MS i s 10.9 percent. Comparisons with historical data on the household income o f OP participants indicate a relative increase o f participants with up to two MS (from 39.4 percent in 2003 to 63.2 percent in 2005). Conversely, the share o f participants with incomes over 12 MS has fallen by half inrecent years, from 7.2 percent in 2002 to 3.4 percent in2005. As with householdincomelevels, differences exist accordingto the forum. Participants inthe regional assemblies reported lower incomes than those who participated in thematic assemblies, with 42.5 versus 53.6 percent reporting incomes o f up to two MS. Similarly, 15.5 percent o f participants reported having a household income o f over 8 MS in the thematic assemblies, compared to 9.3 percent inthe regional assemblies. The observed drop inthe participation rate o f high-income groups raises questions about the continued support o f the OP process by different sectors o f the population, as well as the capacity o f the OP to express the needs o f the city as a whole. The poorest sectors of the population identified the opportunity cost and expense of participation as obstacles to their participation.Overall, the consensus is that the OP is a model that does not present barriers to entry either for the poor or those who are not politically organized. However, the extremely poor" identified transportation costs, such as bus fares and forgone earnings, as barriers to their representation o f their community as councilors or delegates. One community leader from the Santa Terezinha neighborhood, who collects paper for recycling, stated: "Iwas invited to be a councilor, but I turned it down. When I go to the meetings, I'm not able to work so much. Somebodyfrom the village pulls my cart to help me, but it is never the same thing. However, Idon't want to abandon theprocess." This explanation likely applies not only to collectors of recyclable waste, but to all of the poorest populationsectors, which may find transport and opportunitycosts an obstacle to participation at higher levels of representation.If one assumes that the poor earn one minimumsalary (R$360) per month for 200 hours worked, 8 hours o fparticipation inthe OP per week would translate into a cost o f R$58 per month, or one-sixth o f the income o f a household living on one minimum salary. Representing one's community as a delegate or councilor therefore increases the opportunity cost o f the very poor. Those who belongto lower-incomefamilies (incomes of up to 2 MS) are less represented at the councilor level. Those elected to councilor functions tend to fall into the income ranges o f 2- 4 and 4-8 minimum salaries, indicating that the very poor are rarely elected to these positions, perhaps due to the opportunity cost o f participation. Interestingly, there are fewer differences among delegate income levels and the distribution among them i s more in line with general OP participation. 17About 7 percent of the city's population has a monthly householdincome ofbetween 0.5 to 1minimum salaries (Le., R$160-R$320, or US$80-160). IBGE(2000). 28 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre Lower participation by wealthier income groups is likely due to the fact that (i) the economic incentives for their participation are weak and (ii)they perceive the OP as a mechanismfor the poor to demand basic services, rather than a mechanismfor addressing complex issues such as publicsecurity. One middle-class focus group interviewee stated, `low- income neighborhoods participate much more, becausepeople with open sewers are going to try to obtain public works. " Participation in the OP i s thus inextricably connected to basic needs: "They [the poor] have many more needs; they believe that their voice is really going to bring about improvements. But the middle and upper classes do not think so. It is not worth going there to argue. Residents o f wealthier neighborhoods appear to doubt whether the OP i s a viable " mechanism for addressing complex issues that go beyond basic services: "The lower-class people participate because the matters that are resolvedfaster are basic sanitation and infrastructure. . . Iimagine that whatwein themiddlewouldlikeispublic safety-more complicated things-from [the] OP." The non-participationof higher-income groups can also be attributed to their perception that the poor are in greater need, as well as the challenges of mobilizingthe heterogeneous interests of the better-off. Higher-income groups recognize that basic service provision is a priority in low-income neighborhoods: "People do not consider the claims of [wealthy] neighborhoods to be a priority. I don't think they have the same urgency as the poorer neighborhoods do. . . They have open sewers. . . .therefore I think they have a greaterpriority." They also acknowledge the difficulties o f mobilizing higher-incomeneighborhoods: "The middle class does not organize itself to participate. That is cultural. Or, `Yt's not even worth it for I' people to go. What is going to happen is that there is going to be halfa dozen middle-class people wanting something such as safety, health, or something like that and about 800,000people wanting basic sanitation. " Expandingthe scope of discussiontopics to include issues of strategic importancefor the city, as well as oversightof public expenditures,would capturethe interest of higher-income non-participants.As one interviewee noted, Yt would be very interesting . . people could participate in the sense of discussing public policies that would,for example, generate income, because this moves the economy, promotes development, and would end up affecting our companies.,,18 Greater emphasis on the supervision of public expenditures would also stimulate greater participation o f middle-income sectors. Another intervieweenoted, "Everyone is united in wanting to supervise. Everyone is united. I t is not me wanting something for my neighborhood, you for your neighborhood. Everyone (is) wanting things to be done correctly, so as to save moneyfor us. " Although the creation of thematic assembliesin the mid-1990swas intendedto broaden OP discussionsto includestrategic issues andthus attractthe participationof middle- andhigh- income groups, this strategy has failed.lgThematic assemblies were created as a forum to discuss strategic issues and facilitate discussions less driven by localized demands. However, analysis o f the investment discussions over the past two years shows little variation between the content o f the discussions in the two types o f assemblies. Thematic assemblies do not serve as a space for discussion o f strategic municipal sector policies and have thus failed to attract groups that feel the OP does not concern them, such as entrepreneurs. 18 Interview with a localbusiness leader, January 2007. l 9The thematic assemblies discuss: (i)transport and traffic; (ii) and social assistance; (iii) health education, sports, and leisure; (iv) economic development, taxes, and tourism; (v) organization o f the city, urban and environmental development; and (vi) culture. 29 Economic Sector Work Other strategies to attract these groups have also failed to yield substantial results. Initiatives by the municipality to stimulate the participation o f these groups have focused on advertising campaigns at the beginning o f the annual OP cycle, inviting the population to participate via local newspapers, the radio, and television spots. In 2002, the municipal government also reached out to specific groups and sectors, visiting leading business entities and universities to formally invite them to participate.20However, these efforts have not been very successful. The location of regional assemblies has been identified by higher-income groups as a barrier to participation; some also suggest giving meetings a more attractive format. One focus group interviewee summarized what i s needed to attract the middle class as follows: "A bigger venue, parking, somewhere safe. Because the neighborhood associations are generally located near poorer areas, wealthier people are probably intimidated to go. A more neutral location would certainly attract more people. Linking OP meetings with cultural activities was " also suggested to attract a broader set o f participants. A young music student suggested that a different format "can attract more people. . . Maybe it can be a concert of a local artist where more is achieved when something happens afterwards. " Among middle- and high-incomegroups, some peopleperceive the OP as an overly political process, although these perceptions seem to have softened recently. Some interviewees identified political and ideological considerations as a reason for their non-participation, viewing the OP as a mechanism inextricably associated with the Worker's Party. The leader o f the Union o f the Civil Construction Industry summarized this view: "In my opinion and in the opinion of thosepeople that I know and who did not and do not participate in the OP, the latter became a much morepolitical instrument than an instrumentfor societal participation. Then thesepeople, mostly the middle class, distanced themselves and ended up not going. Since it was a political instrument, it ended up being used politically, distancing many peoplefrom it. "" Although this view appears to have softened since the new administration took over, the participation rates o f these income groups remains unchanged. In the words o f one entrepreneur, the OP "now, in this current administration, seems to have changed a little. , . I t seems that the current administration treats it.. . like an instrument of society.''22 Given the preponderance of lower-income participants, the bulk of OP participants are unskilled workers-certain professional groups, such as entrepreneurs, are under- represented.About a quarter o f participants describe their occupation as unskilled manual labor. Students and teachers make up almost 10 percent o f participants. There has been a significant increase among those who report working in the informal sector over the years, from 1.4 percent in 1998 to 4.7 percent in2005. Yet the participation rates of micro entrepreneurs fell from 1.7 percent in 1998 to 0.27 percent in 2005, while larger entrepreneurs made up less than half o f one percent (0.47 percent) o f participants in 2005. Professional groups that have shown a decline in participation over the same period include civil construction workers (from 5.3 to 2 percent) and maids and nannies (from 12.8 percent to 6.6 percent). The percentage o f participants who are unemployed or public servants, on the other hand, more than doubled between 1995 and 2005. The participation ofretirees also remains significant-nearly 10percent in2005. 20Interview with Jog0 Verle. Februarv2007. 21Interview with the leader of a local entrepreneurial organization, January 2007. 22hid. 30 Towardsa More Inclusive and Effective Participatoly Budget in Porto Alegre The involvementof differentmunicipal secretariats in the OP processfluctuates, reflecting the nature of the public work projects being planned and executed. The degree o f involvement o f different municipal bodies in the OP depends on the priorities established by the communities. However, certain bodies (e.g., the Municipal Secretariats o f Finance [SMF] and Administration [SMA]), do not have a systematic relationship with the OP due to the nature o f their activities and the specific characteristics o f the OP process. Data indicate that the presence o f the municipal secretariats at OP meetings peaked in the period 1997-2000 and declined thereafter. 1.2.How Legitimateis the OP? This section discusses the legitimacy of representation in the OP process. This legitimacy involves three main dimensions: the authenticity o f the process o f formation o f voice and its transmission upstream; the downstream process o f accountability to the broad base o f OP participants and the city as a whole; and the perception o f credibility by the population and different social actors (see Figure 7). The following paragraphs examine how these dimensions play out inthe OP, based on the survey, interviews, and focus groups conducted for this study, as well on existing literature and analytical contributions. Figure7. Legitimacy of representationinthe OP process 31 Formation and Transmission of Voice The participation model of the OP favors the autonomy of participating agents in the formation and upstream transmission of voice. One o f the main controversies in the debate that led to the design o f the OP inthe years 1989-1992 was the choice between: (i) a participation model characterized by direct and open participation o f individuals, combined with representation o f organized expressions o f civil society, and (ii) a corporate model limited to the involvement o f representatives o f social entities, such as neighborhood assemblies (NAs). The prevalence o f the first model i s considered a positive factor by most studies o f the OP (Abers 2000; Baierle 2000; Baiocchi 2001; Goldfrank 2001; Gret and Sintomer 2002; Fedozzi 1997; Menegat 1995; Silva 2003) for the following reasons: The participation model put an end to the monopoly o f civil society representation exercised by the NAs, reduced the influence that political parties could have on the OP (via its social base), and diminished the control that traditional leaders could assert over delegation processes (i.e., through the NAs and other pre-existing forms o f social organization). People who share this perspective opposed attempts to ensure a predominant role for the NAs, which was defended at the time by the Union o f Neighborhood Assemblies (UAMPA),23to the detriment o f the plurality o f individual representation and the emergence of new expressions o f civil society, which surged markedly at the end o f the 1970s. These civil society movements included such organizations as Mother's Clubs, Popular Councils, Street Commissions, leisure, sport and/or recreation organizations, and entities organized by specific subjects (e.g., children, the handicapped, ethnicities, etc.). Inshort, the representation model selected expressed the real complexity o f Brazilian society. The participation model also diminished "barriers to entry" to the OP and, therefore, to an institutionalized relationship between citizens and the local government, by including groups that normally did not have the capacity or capital to organize formal collective action. Individual and open participation allowed direct interaction among people who lived in different neighborhoods, and between people and their local governments, reducing the intermediation that would occur when roles are delegated to leaders o f social entities. This aspect o f participation is associated with learning experiences that result from bringingtogether people who do not h o w one another and, therefore, possibly increasing their social capital. Despitethe fact that a direct participationstructure was adoptedfor the OP at its founding, the controversy over this structure has not come to an end. Community leaders still consider the real impact o f the OP on the associative realm o f the city to be ambiguous. For some, the OP reduced the power o f patronage exercised bothby City Council members and representatives o f a number o f formal civil-society organizations. For others, open access to the OP weakened the associations because it stimulated direct participation outside these entities. These two positions were bluntly expressed by civil society leaders who participated in a focus group conducted for this study (see Box 2). 23UniZo das Associaqaes de Moradores de Port0 Alegre, founded in 1983. 32 Towards a More Inclusiveand Effective Participatory Budget in Porto Alenre Box 2. Focusgroup views onthe OP participationmodel "It [the OP] . . tookthe power o f patronage from the memberso f the City Council. . . It put an end to all that. Ithink the OP is a wonderful way to democratize and include people in decisions regarding how public funds should be spent. Another thing it did was demystifya number of entrenched associations.The OP doesn't necessarily need to go through an associationthat isn't even involvedin its community-onethat is just entrenchedthere and doesn't even go beyond the little structure it has anymore. For me, this cycle has been broken. The OP has a new criterion: any organized group can call an assembly, put forth a demand, fight for it, and succeed. Ithink this is a very positive point made by the OP because we began to realize that we could become a movement, as long as we're organized and not tied down to a structure." -Nelsa, Former OP Councilwoman, and Director, UNIVENS Cooperative Social Movements Focus Group "[Therewere] entrenched associationsin my own community [but]when the OP began and Ifirst became aware of [it], Ientered directly into the budget meetings without being from an association or anything-I became a delegateto the OP. Ienteredthe OP to find out what it was really all about, because until then, people said the OP was an institution using a particular acronym to manipulate the people and to some extent Iagree with that, because ifNelsa can go into the community, just like Laura can go and get 30-40 people and get them to participate. ..thenit establishes a certain [claim]...I[n]my region, we had our share o f victories, but it was within the organized form o f our association, neighborhood groups, the church-all o f us united for the sake o f a demand that was obtained in 1996 and came out in the year 2000. It was won, but there had to be continuity in the struggle, keepup the pressure; that's where the associative question comes in, of us having to go there and do it." -Ivani, Social Movements Focus Group "Ibelieve the OP . . . emptied the community associations, that the community association leaders had the role of addressing the demands of their communities. The OP did not require that demands go through an institutional route. So anyone could go there as a community resident and present a demand. This causedrelations between the presidents o f neighborhood associations and the executive power [of the municipality] to cool. . . That was the first thing the OP caused." -Horacia, Current President o f UAMPA I f it is true that the OP has promotedthe autonomousparticipationof individualsand new social movements in the discussion of public policy and resource allocation, it is also true that the OP tends to blur the realms of civil society and government. The evolution o f co- management functions in the OP has led to situations where there i s a lack o f differentiation between the roles o f political agents and government officials (e.g., CROPS,'~council members, secretaries, etc.) and those o f the OP and its civil society representatives. This risk o f "contamination" o f the independent voice o f citizens and civil society actors, along with the dependence of the OP on government initiatives, was recognized and addressed by the municipal government during the discussion of OP reform in 2001 (Verle and Brunet 2002, 19). Actually, the relationship i s considered to have the "difficult practical solution" o f assuring the political- organizational autonomy o f communities with respect to the municipal administration, while 24 CROPS are Regional Coordinators of the OP. Council members were established by the first PT administration, which integrated the FASCOM (Forum of Community Council members of the Municipal Secretariats), an internal body of the government that is involved incoordinating the OP. 33 Economic Sector Work recognizing the need for practical intervention by the government in the regions for the sake of agile communication (e.g., information dissemination, clarification, mobilization). The new structure was intended to put an end to attempts to exclude groups and to mediate conflicts that emerge indisputes over limitedresources. Transparent rules of engagement are a precondition for strengthening the authenticity of the upstream process of formation and transmission of voice. The autonomous participation o f participants requires knowledge o f the rules o f engagement. It i s evident that the monopoly o f knowledge of these rules by a limited number o f participants (usually those having more years o f engagement) neither favors a level playing field, nor a more equitable distribution o f power to influence decisions. Participation in the OP i s often spontaneous and chaotic. In many cases, participants are limited to mobilizing themselves around their particular demands without greater knowledge o f the process or rules o f engagement, or a broader interest inthe municipal budget as a whole. In2005, the majority o f surveyed OP participants who had worked with the OP for less than two years claimed that they had poor or no knowledge o f the rules. Table 5. Knowledge of OP rules and criteria according to duration of participation Knowledge of OP rules and Duration of OP participation working criteria 1year I 2 to4years I 5 to 7 years I 8 years or more n 22 98 94 153 Most of the rules % 11.1 25.9 47.7 64.0 Only some %n I 66 144 60 58 33.2 38.0 30.5 24.3 Few n 46 81 30 23 % 23.1 21.4 15.2 9.6 Doesn't know n 65 56 13 5 % 32.7 14.8 6.6 2.1 Total n 199 379 197 239 % 100 100 100 100 Lack of knowledge of the rules of engagement increases the risk of capture and/or monopolization of the OP. According to an evaluation conducted by the municipality on the occasion o f the debates to improve the OP in 2001, there i s a trend o f increasing "appropriation of the process by an elite that impedes theformation of a broader base of participants." Today, the Internal Rules o f the OP are compiled in a long, complex document of more than 60 pages. This favors the capture ofthe process by a group o f leaders that havemore experience andtime to get to know the "secrets" o f how the system functions (Verle and Brunet 2002, 22). The control that results from knowing the rules could be one element contributing to the trend towards decreasing levels o f rotation o f elected COP members over the last five years, as shown in Figure 8. 34 Towardsa More Inclusive and Effective Participatory Budget in Porto Alegre Figure 8. Reelection of COP delegates, 1991-2007 50 Yo 45 40 43 40 A -/ - 35 40 f33 30 30 25 20 Y 15 23 22 yo v 21 10 19 17 5 Reelection 0 Source: Fedozzi (2007). To summarize, the OP process represents a trade-off between autonomous participation of civil society and government influence over civil society actors. The OP representation model favors autonomous engagement o f individual citizens and new civil society organizations, but tends to blur government and civil society realms, reinforcing the dependency o f civil society on government initiative, which compromises the autonomy o f participating agents. Transparency and knowledge o f the rules o f engagement thus become a critical factor inpromoting autonomous participation and reducing the risk o f capture. Internal Accountability of the OP The downstream accountability of COP delegates and members is a critical legitimacy factor. The presentation o f accounts (Prestaqio de Contas) by delegates and councilors (members o f the COP) to their constituents is basically carried out at OP regional forums, which take place every 15 days. At the regional forums, councilors report back to regional delegates and interested community members on discussions held in the COP. They also review the implementation o f public works in the Investment Plan and the services provided by the municipality, as well as collect community demands and requests. Not as visible as regional or thematic assemblies, these periodic meetings constitute privileged spaces o f downstream internal accountability within the OP process. Presentations by councilors and delegates are predominantly informal. Reports are presented verbally, with very sporadic use o f written documents and hard data. Reports also do not follow a pre-established guideline or a simple communication format. Because they are verbal, the reports are not recorded for eventual checks by regional OP members. One delegate described how he receives information as follows: "It arrives somewhat altered. It depends on the interest of the councilor. If he is interested he will let me know. . . I noticethat the information arrives in pieces, and it is not very clear and it is always late." Limited skills and lack of incentives explain the absence of a more formal and systematic presentation of COP discussions to regional assemblies. There is no training program to help 35 Economic Sector Work COP members prepare more accurate and relevant reports, nor does the OP have user-friendly forms and procedures to facilitate a better presentation o f accounts. Yet COP members and OP delegates have little incentive to adopt a more rigorous accountability process, given that survey data shows they enjoy highlevels o f credibility among participants in spite o f the manner o f their accountability. Inother words, the current internal accountability process i s perceivedas effective and legitimate by citizens who participate inthe process (see Figures 9 andl0). The accountability of the OP to the population at large suffers from similar problems. Most o f the information about the performance o f the OP flows through the media ina casual way and via inaccurate formats. There are little to no flows o f information between regional delegates (or the regional forums themselves) and their respective communities. As the survey conducted for this study revealed, 23 percent o f respondents claimed they were not informed about the OP and 25 percent respondedthat media was their main source o f information. Figure 9. Ratings of representation by councilors and delegates Do you consider that your councilman/delegate Yesl a k a y s represents the interests of your community? IAlmostahays iSometimes n OP SJNey OP survey 2Yo .,- .." 23% .- / / / / 0% 20% 40% 60% 80% 100% Source: Study survey, 2006. Figure 10. Ratings of accountability of councilors and delegates Do the delegateslcouncilmen maintain IYes you informed regarding their activities? ISometimes ON0 OP survey Delegates Contro group1 0% 20% 40% 60% 80% 100% Source: Study survey, 2006. 36 P Pro-poor y ai life? c O/O 20 40 60 80 100 Two factors may ~~~~a~~the ~ i ~betweenathe QP's~weak internal a ~ t ~ c c ~and~ ~ ~ ~ ~ lts positivepublic p ~ ~ ~ e pTlic~firstn factor is that the OP ~ o . I Salready pcrccwe.cf as an i ~ ~ ~ ~ o r ~ a n ~ a c c ~ u ~ ~aac ~h i~~~~ ie ~~it7yeand~of itself. A s has been shown in earlier sections of this chapter, i i the QP has gwen voice to largc sectors of the ~ o ~ u l a ~ ~i opn a ~ ~ cthc ~poor)r lin~ the ~ a p r ~ ~ r I t t z a ~ofo~n ? ~ t~~'es~mentS,e n a b ~ i nthem to i n ~ u e ~budget allocatior~sto some extent, l b I i ~ ~ c e In the a b s c ~ ~ofethe OP, these a l l o c ~ ~ ~ owould be decided by a small goup of elected officials n s and bureauc~atswho would not face the social contra[ created by the OP and therefore m t s l d rttosr. likely engage in clientc~is~~c~ ~ ~Even~if~thcc as .~ p ~ c ~of the C>Pr c~p r ~~s e ~~~ ~ia t i~o ~ ~ ~ ~ ~ y process IS wcak, u ~ ~ ~ decisiondmaking via ~ e ~ e ~ a tisoalready a more d r a ~ ~ i n ~ ~ arid~ ~ c r ~ 37 Economic Sector Work culture based on verbal communication, which demands and expects less accuracy and associates credibility with other factors, such as confidence inpersonal links, networks, and the media. Available Information on the OP There have been significant improvements regardingthe informationmade available about the OP. Although the OP Website offers information on the OP process, according to OP councilors and delegates, as well as municipal technical staff, this information i s not regularly updated. The operations necessary to analyze the situation o f the Investment Plan at a specific point intime are also quite complex, demanding considerable effort and exhaustive repetition o f procedures. The ObservaPoA (the city's online repository o f geographically based census information and socioeconomic data) also provides a wealth o f information on the city, including social indicatorsdisaggregated by region. While detailed informationon the OP and budgetary matters is available, it is not easily accessible to all groups due to its complexity and format. The information available on the Internet i s not easy to navigate, nor i s it appropriate for many groups. Obtaining detailed information presumes a degree o f technical howledge that the average citizen does not have. L o w levels o f understanding o f the OP process are evidence of the existing challenges o f information provision, which i s a precondition for improving the quality o f participation and voice inthe OP. (For a more detailed discussion o f this topic, see Chapter 3). Ability of the OP to Influence Decision Making While the current municipal administration incorporatedthe OP as one of its 21 programs, the municipality has never been legally bound by its recommendations. As previously discussed, the OP has its roots in an experimental consultation process that became more formalized over time. When a new municipal administration took over in 2005, the OP was maintained as a mechanism for the transmission o f voice on budgetary decisions. This decision i s significant evidence o f the sustainability o f the OP process and its capacity to inform municipal decision making. Public perception in Porto Alegre is that the OP has considerable influence on municipal decision making. This influence is not, however, generally interpreted to mean that the OP actually determines city investments. For example, when a general sample o f the population was asked who determined public investment in Porto Alegre, 16.1 percent responded that the population did via the OP. Most respondents (41.6 percent) responded that the mayor and his team decided on investments. Insame survey, however, 75.2 percent o f respondents agreed with the statement that the OP hadyielded results interms o f investments for the population. A survey o f OP participants found that only 57.4 percent believed that demands presented in the OP process were included in the Investment Plan, while 24.5 percent reported that their demands had not been included (the remainder didnot answer the question or didn't know the answer). Participantsperceive the OP as empoweringcommunitiesinthe decision-makingprocess.A survey o f former council members, for example, found that 83.7 percent believed the OP increased the decision-making power o f communities (Beras 2006). Another survey found that over 50 percent o f OP participants believed the community always or nearly always decided issues o f public works and services through the OP, although this perception has decreased slightly over time (Fedozzi 2007). The second survey also found that longer participation inthe 38 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre OP process was associated with a stronger sense o f empowerment, a possible indicator that the OP is a learning process that eventually translates into empowerment. Among people who had participated in the OP for 8 or more years, the majority believed that the population always decided issues o f public works and services. Over the past 10 years, the OP has informed almost half of all municipal investment decisions, which is a significant proportion. As shown in more detail in Chapter 3, the percentage o f total municipal investments represented by the Investment Plan has varied over time and in some instances has exceeded 60 percent. For the period between 1997 and 2006, however, the average IP represented only 46 percent o f total municipal investments. The fact that the OP hasinformed almost halfof municipalinvestment spendingfor over 18 years points to significant decision-making influence, despite the shortcomings of the process. Given its shortcomings, especially the poor representation o f certain social groups and the absence o f an effective forum to discuss more strategic and/or sectoral issues, the OP i s a considerable achievement because it has influenced such a significant share o f municipal investment. The methodological and operational limitations o f the OP, and the fact that it exists within a complex system o f participation, make it unrealistic to expect that a larger share o f investment spending could be resolvedby the process. 1.3.Main ChallengesandRecommendations As this chapter has shown, although a large number of citizens participateinthe OP, not all social groups participate to an equal extent. Under-represented groups include the extremely poor, youth, middle- and high-income groups, as well as certain professional groups, such as entrepreneurs, all o f whom face significant barriers to participation. Certain challenges to the legitimacy o f the OP also need to be overcome to make the participatory budgeting process more accessible, especially the challenges o f transparency and knowledge o f the rules o f engagement. The following paragraphs outline the major challenges to the quality o f participation and propose recommendations to address them. Challenge I: Creating incentives and removing barriers to make the OP more inclusive. Currently, the OP lacks incentives for the participation o f a number o f groups in society and/or presents barriers to their participation. Among these barriers are the large amount o f time required to participate, the location and timing o f meetings, their format, and the complexity o f the subject matter. Actions to address these challenges could include: Create incentivesfor participation.Several people identify the narrow focus ofthe OP (i-e., on the provision o fbasic infrastructure for poor neighborhoods) as a disincentive for participation. Expanding its focus to include issues o f strategic importance for the city as a whole would attract interest from the middle class, professional groups, and entrepreneurs. o Reconsider boththe role o fthematic assemblies inthe OPprocess and their scope o f discussion. o Consider expanding the focus ofthe OP to include issues o fstrategic importance for the city as a whole, including budgetary oversight, particularly the execution o f the InvestmentPlan and other fiscal policies and municipal revenues and current expenditures that affect the space for OP investments, which would 39 Economic Sector Work attract interest from the middle class and professional groups such as entrepreneurs. Make participation more convenient. The time and location o f meetings have been identified as barriers to the participation o f middle- and high-income groups, as well as married women. Alternative means o f participation could open an avenue to including these groups in the process. For instance, online voting on investment priorities would provide a quick and convenient mechanism for their participation. o Consider forms o fonline participation. o Consider reimbursing the financial cost o fparticipationbythe poorest groups (e.g., the cost o fbus tickets for participants below a certain income level). o Improve the effectiveness o fOP council meetings byreformingmeeting procedures, i.e., establishing a time frame, agenda, and considering the possibility o f further decentralizing meetings to micro-regions. Improve the training and education of participants. Ideally, the OP should be a mechanism that contributes to citizen understanding o f the municipal budget process. As it currently functions, however, the OP does not offer specific training to participants to ensure that they understand the process and can make high-quality contributions to it. Training needs should be identified at each level o f participation (e.g., general, delegates, and councilors) for each stage o f the OP cycle. Training modules should then be specifically tailored to each level and stage and made available as an integral part o f the process. The challenge is to identify the appropriate training content as well as effective instructional methods. Specifically: o Level knowledge amongparticipants. Delegate forums are the obvious choice for training, as they are close to the social base and delegates are potential council members. The OP Council should also institute required training. Becoming acquainted with the rules of the game should be the first step toward effective democratization and the establishment o f equal conditions o f participation. o Improve knowledge about thepublic budget. This is a challenge o f considerable complexity, as it involves making technical aspects o f the budget easily accessible by and understandable to OP participants. Realistic training should consider differentiated levels for OP Council members and delegates that take into account their diverse educational backgrounds. o Test training methods tojnd an appropriate model. The two principal challenges will be to overcome the ad-hoc nature o f OP meetings and recognize the heterogeneity o f the people involved (in terms o f professional training, formal schooling, etc.) inorder to offer equal opportunities for learning. Challenge 2: Make OP representation more legitimate and accountable. OP rules are currently overly detailed, making them difficult to manage for participants and newly elected delegates. There i s also no systematic dissemination o f rules among participants, nor do newly elected delegates have access to training programs. Concentration o f knowledge o f the rules o f engagement favors limited rotation o f delegates and COP members. Additionally, downstream accountability o f COP members to delegates (and of delegates to Porto Alegre participants) does not include systematic reporting mechanisms. Communication o f the OP to Porto Alegre communities i s sporadic and depends on the initiative o f the media. Simplify the rules of engagement and make them more accessible to OP participants. Rules o f engagement should consist o f a simple, homogeneous, and 40 Towardsa More Inclusive and Effective Participatory Budget in Porto Alegre well-known set o f guiding principles. Ample dissemination and adoption o f those principles by all levels and regions of the OP would create common ground for all participants and avoid concentrating knowledge o f the rules in only a handful o f OP representatives. Local media and simple dissemination mechanisms could be used to make the rules more accessible to all participants and to communities as a whole. Create simple formats for presentations by councilors and delegates. Establishment of simple guidelines and formats would enhance the system by which delegates and OP council members report back to their constituencies. Expedited creation and circulation o f meeting minutes would also help downstream accountability. The COP and other forums of the OP require minimumequipment to register and disseminate minutes. An OP campaign concerning the right o f communities and regional forums to demand accountability from their elected representatives would have a double positive effect: it would raise awareness o f both communities and OP participants about the importance of monitoring how well their delegates represent and transmit their interests, and it would create an incentive for delegates and COP members to improve their presentations. 41 Chapter 2. The Interface between OP and Municipal Fiscal Management Historically, the focus o f the OP debate has been mainly directed towards popular participation, without much attention to issues o f fiscal management. Most studies o f participatory budgeting have focused on its political aspects, its impacts on social capital, and the inclusion o f disenfranchised groups inthe prioritization ofpublic investments. Resourcerevenue andpublic finance management were mostly subordinate to these issues, as ifreserved for a secondary role, so that the OP could achieve its objective o f empowering the poor and re-directing public investment priorities. This concept, however, has evolved over time. Currently there is growing concern that the financial health o f the government i s a fundamental aspect o f the sustainability o f the OP process and the fulfillment o f its social objectives. The acknowledgement o f this interdependence has revealed a complex relationship between the management o f public finances and the OP participation process. For instance, an increase in personnel expenditure i s often offset by lower capital investment. At the same time, a weak capacity to collect revenues undermines the ability of the local government to make necessary investments. Similarly, frequent errors in over-optimistic revenue forecasts and unrealistic expenditure plans are resolved during budget execution by reducing investments that have already been approved. Since the implementation o f the OP in Porto Alegre, the municipality has passed through moments o f both fiscal equilibrium and imbalance that directly impacted the OP process. During the period 2002-2004, the municipality accumulated a deficit o f R$138 million, which severely deteriorated its ability to fulfill already approved investment demands from the OP process. The inherited stock o f non- implemented OP investment projects then accumulated and the fiscal deficit became a serious challenge to the fulfillment o f new OP demands approved insubsequent years. This reversal inPorto Alegre's fiscal equilibrium raised profound questions about the role o f the OP and the impact o f a fiscal crisis on its economic sustainability. The 2002-2004 period coincided with a moment when many questions were raised about the dynamics o f the OP itself. The dual crisis made the municipality and OP participants increasingly aware o f the importance o f sound financial management. Sound management i s not only vital for ensuring the capacity o f the municipal government to make necessary capital investments and provide adequate services, it i s also fundamental to the effectiveness o fparticipatory budgeting per se. Since the most recent fiscal crisis, the municipality has re-established a sound fiscal policy. As can be seen in Table 6, the primary budget balance of Porto Alegre has shifted from a series o f deficits to a surplus. The city is now well below the limits established by the Fiscal Responsibility Law (FRL) for debt, personnel expenditures, debt servicing, and credit operations. Porto Alegre currently shows a positive trend o f diminishingexpenditures inrelation to net current revenue (NCR). 42 Towardsa MoreInclusive and Efective Participatory Budget in Porto Alegre Table 6. Financialindicatorsof PortoAlegre municipality,comparedto FRLlimits FRL 2003 2004 2005 2006 Limits 2 63% 2.84% 1.25% 1.07% 16.00% PrimarySurplusINCR -1.72% -4.38% 1.81% 3.29% Source:Tesouro Nacional, Tribunal de Contasdo Estadodo Rio Grandedo Sul, and Porto Alegre municipality. Although these key municipal finance indicators suggest that Porto Alegre has recovered sound fiscal management and can potentiallybetter fulfill the demands of the OP process, some critical questionsremainconcerningits abilityto ensure a sound interfacebetweenthe OP processand the municipalfiscal system. While the fiscal surplus effort o f the current administration was necessary to honor short-term debts created by past fiscal deficits and recover the city's long-termborrowing capacity, the annual investment envelope designated for the OP had to be seriously reduced, creating more social discontent. This chapter analyzes the relationship between the Porto Alegre OP and municipal fiscal management in more depth. One o f the first issues addressed i s whether the OP and the legal framework that regulates the municipal budget cycle are coherently interrelated. Constant delays in the execution o f the InvestmentPlan have raised questions about the extent to which OP resource prioritization i s based on accurate revenue forecasts. It has also raised questions about whether overall budget planning and execution are consistent, and ifnot, what impacts this lack o f consistency has on the OP process. Becausea budgeting processshould ultimatelyproduce sound resource allocation, this chapter also looks at the impact of the OP on the quality and quantity of capital investment. First, it analyzes whether the OP favors the fragmentation o f public resources v i s - h i s strategic investments. Second, it examines whether the OP considers the cost-benefit aspects o f proposed investments, as well as whether there i s any correlationbetween the OP and the way inwhich the municipality manages taxes and current expenditures. Finally, the chapter attempts to understand whether the OP has the capacity to contribute to sound financial management, and if so, why it was unable to preventthe recent fiscal crisis. Any analysisof the relationship betweenthe OP inPortoAlegre and municipalfiscalmanagement must startbytakinginto accountthe overallmunicipalbudgetframeworkinBrazil, as well as macroeconomicand politicalconditionsinthe country. The performance o f Porto Alegre municipal finance and its OP program i s not conditionedexclusively on the local context; it i s also affected by budget regulationsimposedby the national government, macroeconomic policies, and structural political changes. Indeed, the 18-year life cycle o f the OP has coincided with profound changes in Brazil. The process o fre-democratization o f the country brought new actors into the political arena, redirecting a great number o f social movements away from the macro democratization agenda to local decision making and service delivery. As a result, the local space became a privileged governance sphere boostedby the decentralization reforms o f the 1988 Constitution. On the other hand, policy reforms at the national level have imposed clear limits on municipal government, particularly its discretionary power to assign revenues and expenditures. Budget rigidity arises on both the revenue and expenditure sides o f the budget. On the revenue side, constitutional, or legal, earmarking mechanisms link certain types o f revenues to specific expenditures. As a result, it is forbidden to shiftresources that are legally earmarked towards other types o fexpenditure. 43 Economic Sector Work On the expenditure side, certain expenditures are considered mandatory, including legal (constitutional) obligations, such as interest payments, personnel salaries, social security entitlements, and assistance benefits (see Box 3). On the macroeconomic side, Brazil has passed through periods of growth and recession, high inflation, and tight monetary policies. Duringthe 18 years o f the OP, the municipality o f Porto Alegre has gone through at least three distinct phases. The beginning o f the OP in 1989 through the first years of the 1990s was a period marked by increased revenue capacity and implementation o f bottom-up governance reforms. A second period, which began in 1993 and ran through 1999, saw the consolidation o f the OP and the revenue capacity of the city. The third phase, which began in 2000, was a period in which the finances of the municipality deteriorated, together with its capacity to execute the investments prioritized by the OP. Box 3. Municipal budgetrigidity Municipalities in Brazil suffer from acute budget rigidity. On the revenue side, constitutional norms establish minimum levels o f expenditure o f municipal revenues in specific sectors (e.g., 15 percent in health and 25 percent in education). On the expenditure side, the increase in personnel payments observed in small and medium-sized municipalities also contributes to expenditure rigidity. Additionally, the weight o f debt servicing inhighly indebtedmunicipalities also contributes to the problem. However budget rigidity i s measured (whether as the ratio o f earmarked revenues to total revenues or the ratio o f mandatory expenditures to total expenditures), it i s clear that the portion o f the budget that a municipal government can allocate "freely" within an annual budget process i s quite limited. Source: World Bank (2006). 2.1 Cyclesof MunicipalBudgetPlanning andthe OP: Convergenceor Divergence? The OP Timeframe and Legal Budget Cycles Althoughthe cycle of the OP in Porto Alegre is widely known, it remainsrelatively unclearwhether this cycle is well articulatedwith legalprovisionsthat regulatethe municipal budget cycle. Figure 12 shows how the cycles o f the OP and LDO (Law on Budget Guidelines) overlap, revealing a mismatch between the time frames for OP budget prioritization and the preparation of annual municipal budget guidelines. The prioritization o f public works in the OP starts in early M a y and continues through July. Meanwhile, the LDO i s elaborated by the executive arm of the municipal government between July and August, then submitted to the municipal legislative chamber by August 15th. As a result, the timingofthe two processes i s not completely synchronized. 44 Economic Sector Work minutes reveal that the analyses o f the PPA and the LDO were always carried out within the time frames established by the municipal executive, the time allocated for discussions with the COP were not always adequate. In some cases, other COP priorities took precedence-to the detriment o f discussions of the LDO or PPA. Inother cases, the executive sent the budget documents to the COP very close to the date by which they hadto be submittedto the legislature. The Timing of Complex Public Works The one-year cycle ofthe OP is problematicfor complex publicworks, which requirelongplanning and implementation periods. Large investments have a planning, procurement, and execution time frame that goes far beyond the annual OP cycle. The Municipal Department o f Housing (DEMHAB) and the Municipal Department o f Water and Sewage (DEMAE) o f Port0 Alegre, for example, usually administer projects that require greater resources, complex technical analyses, and multi-year implementation. On several occasions, large investments have been dismembered into smaller parts to fit into the OP annual resource allocation schedule. Although this appeared to be a creative solution, it runs the risk that a project will not receive resources in the following OP cycle that are needed for its completion. Municipal staffs who were interviewed mentioned, for example, that topographic inventories o f urban development projects have been lost because resources to continue the work were not prioritized within the valid inventorytime frame. To address this issue, some municipalities in Brazil have modified the OP cycle to a bi-annual periodand/or created a separate OP for complex urbaninterventions.Cities such as Belo Horizonte and Santo AndrC, among others, have extended the OP cycle to two years due to the backlog in the implementation o f their respective InvestmentPlans. Large projects, such as housing or transport, usually entail more than one year fiom planning to execution, particularly when they involve land acquisition, resettlement, andor environmental licensing. As a result o f the timing problem, tensions constantly emerge because citizens become deeply disappointed when municipal priorities o f the previous year are not implemented as planned. Inthis regard, Belo Horizonte also created a special OP only for the housing sector, which absorbs larger amounts o fresources and requires complex technical and financial planning. The bi-annualOP cycle of many municipalities also seems to allow more time for communities to engage in the oversight of procurement and public works. The analysis inChapters 1and 3 indicates that OP delegates and councilors already spend considerable time on budget formulation, but do not effectively monitor the execution o f public works. The two-year cycle-theoretically, at least-expands the time available for OP participants to carry out oversight, given that this task would not overlap so greatly with the period o fbudget formulation. 2.2 Consistency of BudgetPlanningandExecution: Impactsonthe OP This section evaluates the consistency between the planned budget and its actual execution. Three basic analytical instruments are used in the analysis: (i) the percentage of execution o f the LP; (ii)a comparisonbetween the budgeted value of a sample o f works and the actual sum spent on the works; and, (iii) consistencybetweenforecastrevenueandcollectedrevenue.Eachoneoftheseinstrumentsis the intendedto approach more specifically a phase o f elaboration or execution o f the budget o f the municipal administration. Additionally, the study analyzed the volume o f municipal investments determined by the OP. 46 Towardsa More Inclusive and Effective Participatory Budget in Porto Alegre Execution of the Investment Plan The percentage of the Investment Plan (IP) that has been implemented has dropped continuously since the early years of the OP process. This percentage determines, in a relatively simple manner, whether budgeted investments were effectively concluded within the forecast time frame and whether all o f the demands included in the IP were effectively carried out. Whereas between 1990 and 1999, approximately 90 percent o f IP projects were finished before the forecast date, since the year 2000, this percentage has fallen significantly. In 2004, it was forecast at little more than 60 percent. The rate o f executed investments dropped even further in2005 and 2006. However, it i s importantto notice that in the last two years a significant number of projects have continuedto be classifiedas "in progress." Inother words, these investments mightbe executed and the rate o f total executed projects for 2005-2006 might be higher after considering those investments in progress. Another important factor to take into account i s that projects approved in a certain IP that are not implemented due to a lack o fresources are usually re-prioritized duringthe subsequent OP cycle. This implies that delays in the conclusion of works included in the IP do not necessary represent an abandonment o f a specific initiative approved through citizen consultation. Figure13. Share of executedinvestments, 1999-2006 100% 90% 80% 70% 60% % 50% 40% 30% 20% 10% 0% 1999 2000 2001 2002 2003 2004 2005 2006 Year Source: Port0 Alegre Municipality. Another important issue to consider is the effect of capital expenditures financed through credit operations,which increase the fiscal deficit when loans are not backedby revenue efficiency. Inthe short term, credit operations can boost capital investments, but in the medium term, they result in a diminished capacity to invest ifthe initial borrowing i s not followed by increased revenue efficiency. For instance, in Porto Alegre a portion o f IPShave been executed with financing from credit operations, creating a fiscal deficit. Without this mechanism, the rate o f IPimplementation would probably have been much lower. After 2005, however, measures taken to recover from fiscal deficit and the impact o f loan maturation limited the capacity o f the municipality to implement already approved investments. During the first years of the new administration, fiscal policy concentrated onrecovering medium- and long-term 47 Economic Sector Work borrowing capacity by achieving more strict fiscal balance. As result, previously approved investments had to be renegotiated. At the same time, credits from loans diminished and services increased. The OP Share of TotalInvestment The share of investment prioritizedthrough the OP in relation to total municipal investmentshas fluctuated significantly. The data provided by Porto Alegre shows that the percentage o f total investments prioritized by the OP increased significantly during the 1997-2002 period, but has since declined considerably. It i s important to note here that, from time to time, large citywide investments are prioritized through the OP thematic plenaries, such us the I11Perimetral (ring-road), which increase the share o f OP investments. Nevertheless, the 46 percent average o f OP investments as a proportion o f total capital costs give a more precise picture o f the OP impact on municipal investments than the literature on the OP often implies. This OP historic average of close to 50 percent reveals the real limits of the share of capital investmentsthat canbe prioritizedthrough the OP.More in-depth analysis o f capital investments that are not prioritized through the OP process could clarify the other investment pressures on the municipal budget. For instance, municipal departments and municipal enterprises need to make regular capital investments that are subject to public debate. Budget re-allocations and pressure from municipal councilors are thus likely sources o f capital investments that are decided outside o f the OP process. Figure14. Percentageoftotalinvestmentsprioritizedby the OP, 1997-2006 60% 50% 40% 30% 20% 10% 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Source; PortoAlegre Municipality. Capacity to Forecast Income The revenueforecasting capacity of the municipality has deteriorated in recent years. Analyzing the revenue estimations o f the city administration against actual collections shows that in the period 1995- 2000, public planners forecast revenues that were less than or equal to those actually collected. After 2001, however, the revenue forecast margin o f error increased substantially, as shown inFigure 15. When added to the OP practice o f overestimating municipal capital funds during the prioritization process, the impact o f weak forecasting can have a strong negative effect on citizen expectations and trust in the OP, particularly given the transparency o f the OP process inPorto Alegre. 48 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre Figure15.Accuracyof municipalincomeforecasts, 1995-2005 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Of the two main sources of current income (taxes and current transfers), the most optimistic forecast was for tax revenues. Current transfers do not represent a behavior pattern related to effective forecasting and collection. This income category was sometimes conservative and other times, optimistic. The inconsistency o f tax revenue forecasts i s a serious concern, given the contribution o f this income source to the municipality's total current income. As Figure 15 shows, over-optimisticforecasts of municipalincomehavebeenstrikinginmagnitude. Current account surplus and borrowing have been the main sources o f capital investments in Porto Alegre. In Brazil as a whole, the budgetary process o f most states and municipalities similarly use expenditure authorizations based on income forecasts. Weaker control o f these two income components thus affectedcapitalexpenditures and ledto current deficits. It is important to highlightthat discrepanciesbetween income forecasts and expenditures are not exclusiveto the municipalityofPortoAlegre. Historically, the Brazilianpublic finance sector has found it difficult to correlate income forecasts with executed expenditures. The first problemhas been inflation. Until 1994, the country was devastated byhighinflation, whichnot only alteredthe price system, butalso the entire forecasting and planning system, whether public or private. The introduction o f new taxes and modifications of how existing taxes could be applied (mainly the reduction o f collection periods and the use o f indexes with daily monetary corrections) gave municipalities the capacity to resist the income erosion caused by inflation. The end of inflation,however,imposeda new conditionon administrators. The private sector quickly adapted itself to stable prices. However, the public sector reacted much less quickly and still suffers from weak forecasting capacity. In light o f controlled inflation and the need for fiscal balance, it became imperative to establish public finance norms aimed at responsible tax administration. In 2000, the National Congress accordingly approved a code o f conduct for public administrators called the "Law o f Social Responsibility." Another matter that complicates the forecasting of municipal income is the complexity of the Braziliantax regime, which distributes tax income between different levels o f government and makes municipalities dependent on resources from the states and/or federal government. After the Constitution o f 1988 was adopted, the municipalities began to receive 22.5 percent o f the federal income tax and 25 49 Economic Sector Work percent o f the federal Industrial Production Tax (PI),in addition to 25 percent of the state service tax (ICMS) and 50 percent of the state vehicle tax (IPVA). Since municipal dependence on these resources i s very large, the legislation of reforms, as well as changes inthe macroeconomic environment that alter the collection o f these resources have a direct repercussion on the income available to municipalities. Because o f this dependency, revenue forecasting becomes highly complex. Capacityfor Investment Forecasting Discrepanciesbetweeninvestment planning and effective capital expenditures have been common. These discrepancies were analyzed for a random sample o f 10 regional demands and 10 thematic demands inthe years 2004 and 2005. The results show values that were very different among themselves, with systematic forecasting errors apparent inthe OP demands. The magnitude o f variation depended on each case. The samples analyzed in Tables 7 and 8 show that DEMHAB (the Municipal Department of Housing) estimates exhibited the least variation between investment planning and execution, and those of the FASC (the Foundation for Social and Citizen Assistance) and DEMAE (the Municipal Department of Water and Sewage), the greatest variation. Table 7. Thematic demands: Ratioof executedvalue/ forecast value (in R%) Forecast Executed Variati IPyear Region Demand Dept. Thematic value value on Work or service Pavement of Diretriz 2004 Center-South 1992 DEMHAB UrbanDevelop. 103,177 121,713 17.9% 4522 2004 Northeast 2054 DEMHAB UrbanDevelop. 146,974. 169,829 15.5% Pavement of Street 6 2004 Citywide 1905 DEMHAB Urban Develop. 150,000 138,706 7.5% Emergencyhouses 2005 Citywide 590 SMC Culture 64,000 44,800 -30.0% Ten dance workshops Two Afro-Brazilian 2005 Citywide 600 SMC Culture 12,800 8,960 -30.0% workshops 2005 Citywide 2658 DMAE UrbanDevelop. 180,000 110,987 -38.4% Reforms ofproperties Water network linking 2004 Citywide 1930 DMAE UrbanDevelop. 300,000 2,343,000 681% system Expansion ofwater 2004 Citywide 1919 DMAE UrbanDevelop. 115,000 147,200 28% networks Topography and 2005 Citywide 2651 DMAE UrbanDevelop. 135,000 301,000 123.9% cadastral inventory Source: Port0 Alegre Municipality. 50 Towardsa More Inclusive and Efective Participatory Budget in Porto Alegre Table 8. Regional demands: Ratio of executed value/ forecast value (in R$) Forecast Executed IPYear Region Demand Department value value Variation Work or service Substitutionof the water 2004 North 1522 DMAE 270,000 878,300 225.3% network 2005 South 1335 DMAE 55,000 53,100 -3.5% Extensionwater network Substitutionof the water 2004 North 1517 DMAE 85,000 149,000 75.3% network 2005 North-east 1474 FASC 40,102 70,866 76.7% SASEAgreement Expansionof60 goals of 2004 Cruzeiro 939 FASC 16,708 43,195 158.5% SASE SASE Agreement Assoc.Vila 2004 Restinga 1019 FASC 33,416 92,254 176.1% Coema SASE Agreement with 2004 Cruzeiro 937 FASC 16,708 44,530 166.5% DwellersAssociation SASE Agreement with 2005 Gloria 270 FASC 40,102 60,465 50.8% DwellersAssociation Source: PortoAlegreMunicipality. Consultations with specific municipaldepartments revealed several technicaljustifications for the discrepancies between planned and executed expenses. These justifications included: (i) changes in evaluation criteria; (ii) differences between commitments and execution; (iii) contingencies, (iv) a cost forecast value insufficient to respond to the demand; and, (v) transfer o f expenditures to the following budgetary year. These inconsistencieshavegeneratedserious risksto the credibilityof the OP, and to itscore values of transparency, participation,and accountability. Since several works had an estimated value below actual cost, the budget ended up inducing a greater demand that could only be partially fulfilled due to lack o f resources. This fact i s often a source o f deep frustration for citizens participating in the process. Moreover, the reconciliation between approved OP demands and real investment capacity takes place during the execution o f the budget, which does not feature the same level o f transparency and participation for which the OP in Porto Alegre i s internationally known. It i s therefore fundamental to OP that the values allocated to public investment projects be consistent with the real expected level o f future municipal capital investment. Only this consistency can ensure that the sophisticated participatory budgeting process that has evolved over the past 18 years does not generate a mere wish list. 2.3 The Impact of OP on CapitalInvestments:EfficientAllocation or Fragmentation? Distribution of Resources among Local and Citywide Investments Contrary to common perception,the OP in PortoAlegre has not undermined the prioritizationof citywide investments.Considering that the OP Statute defines public investments that aim to benefit the entire city as "thematic demands,'' whereas "regional demands" are defined as addressing local needs, the distribution o f resources presented in Figure 16 shows that historically, a much larger share o f resources has been allocated to thematic demands,*' indicating that the OP has in general favored citywide investments, such as the Third Perimetral (ring road) and the Conduto Alvaro Chaves (sanitation 25 The sumallocated to each type ofdemand is chosen spontaneously by the regional assemblies. 51 Economic Sector Work infrastructure). Of note, while the share o f resources allocated to citywide investments increased significantly after 1996, the share of resources allocated to "regional demands" (i.e., small localized investments) decreased. This change might denote a situational trade-off: whereas localized and strategic investments are necessary, the municipality has not had the financial capacity to address both at the same time. Figure 16. Distribution of resourcesbetweenregionaland thematic demands, 1995-2007 1 Regional Thematic 1 I 100%1 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% I 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Source: Municipalityof Port0Alegre. This tendency of greater share of resource being allocated to citywide investmentsseems to have changed directionin 2006. One possible reason behind the new trend could be popular pressure to fulfill investments prioritized in previous years, particularly small neighborhood public works that were not funded due to the municipality's decision to devote expenditures to accumulated thematic demands. OPImpact on Resource Fragmentation A comparison of the number of small and large investments prioritized by the OP reinforces the previous finding that the OP process has not undermined the municipality's capacity to allocate resources to large investments. Analyzing data on regional and sectoral OP investments for the period 1995-2006, it can be verified that a great number o f the investments (77.9 percent) did not surpass R$400,000 (approximately US$200,000).26 This finding indicates that most works prioritized by the OP are relatively small in scope. However, it does not mean that the OP process privileges small-scope works, but rather, that small-scope works are considered in the OP process. Despite the great prevalence o f investments o f small value, the total sum of investments spent on works valued at up to R$400,000 was equivalent only to 15.6 percent o f total disbursements in the 11-year period analyzed here. Investments valued at over R$9 million (approximately US$4.5 million), on the other hand, represented 34.3 percent o f total capital expenditures. 26Correctedby the annual average GeneralPriceIndex(IGP-DI, values as ofDecember2006). 52 Towards a More Inclusive and Effective Participatory Budget in PortoAlegre Table 9. Distributionof the number ofworks byvalue, 1995-2006 (in R$) Value Frequency Cumulative YO 0 - 30,000 911 22.42% 30,001- 150,000 1,295 54.29% 150,OO1-400,000 957 77.85% 40,001 - 1,000,000 547 91.31% 1.000,OOl -2,500,000 200 96.23% 2,500,001 - 6,000,000 87 98.38% 6,000,001 -20,000,000 54 99.70% 20,000,000 12 100.00% Source: Municipality of Porto Alegre. Since 1999, the number of small investment demands has risen, reaching 89.5 percent of the total number of OP demands prioritized in 2006. Yet this figure does not reflect a fragmentation of investment resources because the total amount allocated to small demands decreased in the same period. In 1996, investments valued at up to R$400,000 totaled 28.3 percent of total resources; while in2006, they totaled only 9.8 percent o f the total allocated Investment Plan. These data also suggest that when the budget for large investments increases, the share ofresources allocated to small public works diminishes. Distribution of OPDemands by Sector The distribution of OP demands shows a clear prioritization of projects for roads, water and sewage, and housing. A sector assessment was conducted for the period 1990-2007, based on the number o f works carried out. Approximately 26 percent o f OP investment demands involved road works (SMOV), including paving and maintenance, among others. In second place were demands for the treatment o f water and sewage (DEMAE), with 17.5 percent. Works related to housing, carried out by DEMHAB, were in third place, with 12.7 percent o f total demands. It is also important to notice that projects executed by SMOV, DEMAE, and DEMHAB are usually more complex and expensive than those executed by other municipal departments. 53 Table 10. OP demandsper sector, 1990-2007 YOof total OP AdministrativeUnit Frequency demands SMOV 1,566 26.20% DMAE 1,043 17.50% DEMHAB 760 12.70% SMT 548 9.20% DEP 423 7.10% SMED 313 5.20% SMAM 273 4.60% SMS 231 3.90% FASC 229 3.80% SMIC 206 3.40% SMC 135 2.30% SME 97 1.60% DMLU 47 0.80% SPM 32 0.50% GP 17 0.30% SMA 18 0.30% SGM 9 0.20% SMF 14 0.20% SMDHSU 3 0.10% PREVIMPA 3 0.10% PROCEMPA 3 0.10% PGM 2 0.00% SECAR 2 0.00% GAF'LAN 2 0.00% Total 5,976 100.00% Source: Port0 Alegre Municipality. The number of demands for water, sewage, and housing investments has been increasing since 1998. Between the years 1990 and 1997, OP investments were initially linked to road works (SMOV), followed by demands for DEMAE and DEMHAB works. Since 1998, however, there has been a clear reduction in the demand for SMOV works (overall, this percentage fell from 33 percent o f total demands in 1990to nearly 7 percent in2007) andanincrease indemandfor DEMAEandDEMHAJ3works. These findings show that local communitieshave modified their priorities.Inpart, this change might have occurred because problems linked to road works (SMOV) are gradually being resolved. Finally, the distribution o f OP investment demands has improved. Whereas in 1991, the three types o f projects most indemandcomprised nearly 80 percent ofall OP demands, thispercentagehad fallen to only 47.8 percent in2007. 2.4 CurrentExpenditure andthe OP Effects of Increasing current Expenditures Although there have been recent improvements, historicallythe city has shown a high degree of fiscal commitment to current expenditure. Analyzing the historic period 1980-2005, the average commitment to current cost expenditure (direct administration) exceeds 85 percent o f total income. It i s important to highlightthat since 2000, this ratio has steadily diminished. 54 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre Figure17. Ratioof currentexpenditureto totalincome, 1980-2005 140% 135% 130% 125% 120% 115% 110% 105% 100% 95% 90% 85% 80% 75% Source: Municipalityo f Porto Alegre. The impactof legislativereforms during this period must be considered.Before the implementation o f the Law on Fiscal Responsibility and other follow-up fiscal legislation, deficits o f indirect administrations (i.e., municipal enterprises and foundations) were commonly resolved by transfers from the direct administration. Given this practice, the change in legislation can partially explain the reduction inthe commitment to currentexpenditure. Dueto the impactof nationalpolicies,more coherent resultsare obtainedby separatingthe analysis of Porto Alegre public finances into two chronological periods: the period between 1980 and 1999, and the period between 2000 and 2005. This temporal division i s required for several reasons. The federal allocation for the Municipality Participation Fund (FPM) was increased in 1983 and 1985. In 1988, the new Constitution implemented a decentralization o f government income. In addition to this decentralization, municipalities also received larger shares o f collected service taxes (ICMS) and were given greater participation inthe FPM. Analvsis of the 1980-1 999 ueriod This periodwas characterizedby significant changes in the fiscal situation of the municipality of Porto Alegre. Inorder to evaluate the expenditure structure duringthe period, three indicators were used: (i)current expenditure/total expenditure; (ii) capital expenditure/total expenditure, and (iii) total investmenthotalexpenditure. As seen inFigure 18, the current expenditure/total expenditure ratio shows a growing trajectory over time. The capital expenditure/total expenditure ratio fell throughout the period, with a growing commitment to the payment o f current expenses. When public finances reflect these characteristics, it i s natural that investment-which i s normally an adjustment variable-is reduced. As can be observed, current expenditure grew from 65.6 percent in 1980 to 95.2 percent in 1989. Certainly 1989 represented the peak o f municipal investment; nevertheless, the graph shows that total expenditure was continuously more committed to current expenditure. Between 1990 and 1999, the growth in this commitment slowed, although it seemed to be suspended at a very highlevel o fnearly 90 percent. 55 Economic Sector Work As a direct consequence of current expenditure coming to represent a larger share of total expenditure, the ratio of investment to expenditure was significantly reduced. In 1980, this ratio was equivalent to 29.6 percent and in 1999, 5.9 percent. Investments on the order of 15 percent could be observed inthe periodbetween 1991 and 1994. However, investments returned to levels below 10 percent between 1995 and 1999 and reached a nadir of 3.4 percent in 1997. Figure 18. Expenditure indicators of the municipal prefecture of Porto Alegre, 1980-1999 # CurrentExpenditureflotal flCapitalExpenditure/Total 211 Total Investment!Total Expenditure Expenditure Expenditure 100% , 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Source: Porto Alegre Municipality. Analvsis of the 1998-2005 period The share of investments in relation to total expenditure also experienced continual decline from 1998 to 2005. As shown inTable 11and Figure 19, investments expenditures amounted to 11.2 percent of total expenditure in 1998, but deteriorated to 5.3 percent by 2005. Table 11. CapitaYcurrent expenditure ratios of Porto Alegre, 1998-2005 CurrentExpenditure/ CapitalExpenditure/ Investment/ TotalExpenditure TotalExpenditure TotalExpenditure 1998 87.6% 12.4% 11.2% 1999 89.3% 10.7% 9.7% 2000 90.1% 9.9% 9.0% 2001 90.9% 9.1% 8.1% 2002 89.5% 10.5% 7.7% 2003 91.6% 8.4% 6.3% 2004 89.9% 10.1% 7.3% 2005 91.1% 8.9% 5.3% Source: Municipalityof Porto Alegre. 56 Towards a More Inclusive and Effective Participatoly Budget in Porto Alegre Figure19. Investment as a percentageof totalexpenditure, 1988-2005 Inwstment/ Total Expenditure 12.00% 10.00% 8.00% --tInvestment/ 6.00% Total 4.00% Expend1ture 2.00% 0.00% Source; Porto Alegre Municipality. Duringthe past several years, the municipalityof PortoAlegre has undertakena significant effort to reduce its deficit and recover its investment capacity. The municipality i s currently attempting to modernize its management model to improve control over expenditures and the tax system. These efforts have already demonstrated positive fiscal results in the form o f a better equilibrium o f revenue and expenditure, better tax compliance and the achievement of a surplus after three consecutives years o f deficit (2002-2004). These results have not only enabled the municipality to recover its capacity to invest with its own resources, but have also improved its creditworthiness. According to recent data releasedby Porto Alegre, the level o f investments made in 2006 with its own resources was the highest o f the last six years (R$84.8 million), while local revenues*' increased 7.14 percent without any increase in the individual tax rate.28 2.5 FiscalEquilibriumandthe OP: GenuineCapacity for Influencing SoundFiscal Policies Analysisof the minutesof key meetingsof the ParticipatoryBudgetingCouncil(COP) shows that it has not always undertaken substantive and independent analysis of municipal fiscal policies, particularly at critical moments. The following content analysis examines the behavior o f the OP debate in the face o f concrete fiscal challenges, as well as its interest in and capacity to hold substantive debate on fiscal issues. Among the various meeting minutes analyzed, those o f September 19,2002, stand out, as they correspond to a critical period during the reversal o f Porto Alegre's finances. During the discussion o f the LDO, an increase in expenditure was proposed because there was no other way to finance previous OP demands for priority works. Available data, however, clearly indicated that the budget would runinto a deficit ifexpenditures were increased, given that the budget balance depended on credit operations. ''Thatis, revenues generated by ITBI, ISSQN, and IPTU. 28See Prefeitura Municipal de Porto Alegre (2007). 57 Economic Sector Work Pressure on councilors to make sure that community demands were included in the Investment Plan prevailed over sound fiscal policy debate. Certain councilors questioned whether or not they should accept an LDO that implied a deficit, but the majority o f the council accepted such proposals, stating that ifthe government was proposing an expenditure, it should worry about how to finance it. Table 12. Sample analysis of COP meetingminutes Date Guideline Description Observations 10/01/91 L D O technical criteria Discussionwas basically on technical criteria. The L D O was not Lack of attention to fiscal voted on issues 10/15/91 Tax revenue Discussion on collecting taxes (IPTU, Garbage Tax, and Transport Attention to fiscal issues Tax). 09/12/96 Budget proposal Presentation of appropriate values for each secretariat and agency. Lack o f attention to fiscal issues 09117/96 Budget proposal Discussion of values and doubts. Attention to fiscal issues 04/22/97 Multi-annual plan Beginning o f discussion o f amendments to the PPA; clarification Attention to fiscal issues of what the PPA is. 04/24/97 Multi-annual plan Continuation o f discussion o f the PPA (participants demonstrate Attention to fiscal issues difficulty inunderstanding the scope of the PPA). 05/22/97 L D O Discussion o f the LDO and other subjects. Attention to fiscal issues 05/26/97 L D O Discussion of the LDO (participants demonstrate difficulty in Attention to fiscal issues understanding the scooe of the P P A l Source: Selected COP minutes, 1996-2006. The COP appears not to have undertaken a systematic initiative to analyze fiscal and budgetary policies.The review o f COP minutes clearly revealed its limited capacity to react to fiscal policy. COP members mainly focused on the prioritization of investments, rather than on long-term fiscal policies. Most discussions were limited to the redistribution o f resources betweenthe original priorities o f the LDO without serious consideration o f the sources o f those resources, i.e., whether the municipality was incurring a deficit, whether a deficit would be temporary or permanent, and how it would be reversed. The decisions o f the COP show a certain short-term bias, as participants identified works and services for which there was an immediate need in the communities they represented. Their position thus biased them against arguments for medium- and long-term fiscal adjustments. When they were given the option to increase expenditures to fulfill their demands, they normally accepted this option without considering the future impacts o f such decisions (e.g., the commitment of income to pay interest, the amortizationo f debt, etc.). 58 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre The COP cannot conduct informed, independent fiscal and budgetary analysis without proper training and access to information.The limitations o f OP analysis of public municipal accounts are linked to two basic problems. The first i s the technical limitation o f COP members to perform rigorous financial analysis. Ifit i s possible to identify a systematic demand o f COP members, it i s their request for technical training so that they can work with budget figures on an equal basis with municipal technical staff. This request was made inthe very first minutes o f the OP and continuedto be made through 2006. The second limitation i s the absence o f available financial information that i s easily accessible to all citizens. The availability o f comparable data over time would allow individuals, private companies, and social entities that possess qualified technical personnel to analyze the figures or carry out independent budget analysis, as well as to alert members o f the COP of eventual imbalances that the municipal administration should clarify. An important lesson learned during past financial crises is that OP councilors clearly understand that they needto be well informed regarding budget and fiscal matters. Without a better leveling o f fiscal knowledge, the capacity o f civil society to carry out qualified analysis and oversight o f municipal fiscal management i s limited. 2.6 The Relevanceof Cost-BenefitAnalysis andImpactAssessment inthe OP Process Although some cost-benefit discussionoccurred at COP meetings, an analysis of the minutes of 624 out of 1,120 meetings reveals that there was no systematic approach to this i~sue.2~ In fact, cost- benefit analysis occurred only at specific moments, in specific forums, and at the initiative o f specific participants. It was not a process that was systemized and organized either by the COP or the municipal administration. Only 12 percent (73 o f 624) o f the meetings examined involved discussions o f the budget process for the Multi-Year Plan (PPA), Law o f Budget Guidelines (LDO), Annual Budget Law (LOA), Direct Administration Budget, or Municipal Administration Deficit. A cost-benefit analysis o f the investmentpriorities proposed at COP meetings was rarely conducted. In addition, lack of systematic data collection on the results of previousinvestments hamperedthe ability of the COP and the municipality to conduct impact assessments. Even where statistical information did exist, there was no continuous collection o f the same data points, which i s essential for time series analysis. Furthermore, the evidence shows that there was no institutionally coordinated process for data collection, nor was there a determination o f which variables were most important for tracking and evaluating the benefits generated by OP works. The tasks o f defining indicators, collecting data, and systemizing analytical methods are the responsibility o f different secretariats and entities o f the municipal administration. Some units, such as the Municipal Secretariat o f Health (SMS) and the Secretariat o f Justice and Safety (Secretariat o f State o f RS), collect data that permits cost-benefit analysis in specific areas, such as birth and crime rates, respectively. However, these individual efforts do not constitute a system for monitoring the impact o f OP projects, nor do they inform OP decision making. Some Brazilian municipalities are currently developing innovative systems to better inform local decision making and evaluate the impact of OP projects. Belo Horizonte, for example, has developed a sophisticated system o f indices to monitor how and which resources allocated by the OP are applied and their observed impacts. The city i s using a multi-dimensional quality-of-life index (ISAVC) based on an 29Inaddition to meeting minutes, interviews with public administrators and a sample of municipal reports were analyzed. 59 Economic Sector Work index previously used to redistribute OP resources among different municipal territorial planning units The index used by Belo Horizonteenables the municipality to measure the quantitative impact of the public works prioritized by the OP on populations living or working near the respective affected areas, as well as to assess the qualitative impacts ofprioritized investments with an emphasis on their social aspects. The initiative represents a concerted effort to implement a system that can inform public municipal policies, helping better integrate sector interventions and better respond to complex needs ina more effective and sustainable manner.31 The recently launched Observatory of Porto Alegre (ObservaPoA) has the potentialto strengthen the capacity of the OP to conduct better analyses that can inform decision making and monitor impacts. The ObservaPoA could be usedto enable OP participants to access a vast database o f available statistical data (objective information) to qualify their perception o f their own needs (subjective information). It could also help the municipality better monitor the impact o f its different programs, as the web-based information site allows users to compare special-allocation investments against highly desegregated and territorially located socioeconomic data. While the staff o f the municipality managing the ObservaPoA have undertaken new analysis using this new tool, COP members and the population in general have not utilized it to date in their deliberations. It i s also unclear whether the municipality will use the ObservaPoA as a unified system to monitor and disseminate the impacts o f different programs, as different platforms are currently being used to manage these programs (see Chapter 3 for additional information). 2.7 MainChallengesandRecommendations Despite the fact that the OP enjoys high levels of participation and credibility, analysis of the interface betweenthe OP and municipal fiscal management yields mixed results. On the one hand, contrary to common perception, the OP inPorto Alegre does not lead to suboptimal resource allocation or investment fragmentation. When it comes to the risk o f investment fragmentation posed by regional pressures (i.e., allocatingresources to small public works), the OP also performs quite well. Although the number o f demands for small investments has increased in the last years, it has not affected the overall share o f resources allocated to satisfy those demands. On the other hand, there i s no evidence that the OP has played a positive role in contributing to fiscal balance inPorto Alegre. In different years, expenditure and revenue forecasts performed unsatisfactory vis-a-vis their actual amounts. The city typically covered the fiscal gap by generating current deficits and/or postponing capital investments. The inconsistency between the OP cycle and the real budget calendar also contributed to capital investment plans being based on weak fiscal data. Both problems ended up creating a backlog of non-implemented OP investment demands and consequently generatedincreasing public discontent. The OP in Porto Alegre has worked mainly as an instrument for interested citizens to voice their preferences regarding capital investments, rather than as a space of public debate to discuss municipal fiscal policy and revenue and expenditure management as a whole. An analysis o f the minutes o f key meetings o f the COP showed that the body has not undertaken substantive assessment o f municipal fiscal policies and revenue capacity, particularly at critical moments. This behavior o f the COP can be attributed to the absence o f a clear motivation among its members to give fiscal management a priority over their interest in ensuring that community demands are included in the Investment Plan. The 30 The index, knownby its acronym ISVAC, is calculated on the basis of 75 quality-of-life indicators. 3 *See Prefeitura Municipal de Belo Horizonte (2007). 60 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre lack o f regular training o f COP members i s another possible explanation, as the LDO, LOA, and PPA involve many technical issues that ordinary citizens find difficult to understand. Challenge I: Bridge the temporal mismatch between the OP cycle and the legal cycle of the municipal budget. Since the OP process is part o f the budgetary process o f the municipality, improved coordination between the LDO, LOA, PPA, and the OP cycle would support a more efficient budgetary process. Options to address this problem include: Have the municipality's executive arm provide a reasonably accurate income estimate for the period for which OP investment demands are prioritized. Between M a y and July, each region and sector should receive a more precise estimate o f the budgeted amounts that will be available for capital expenditure inthe coming year. An alternative solutionwould be to move the OP cycle, or at least the thirdstage o fthe cycle, one month forward, restructuringthe terms o f the OP in accordance with those o f the Organic Law o f the Municipality. This seems to be a simpler solution that would offer the best results in terms o f the quality o f information available during the prioritization process. This solution would also enhance the effectiveness o f OP demands. Challenge 2: Amend the OP cycle so that it can handle complex public works that require more than one year of implementation.The annual cycle o f the OP is not conducive to investments in large and complex works, which require several years to complete. This negatively affects the planning and execution time frames o f complex urban operations and the accuracy o f municipal capital investment forecasts. Actions to address this problem could include: Create a mechanism that links the income o f the IP o f the following years to works that have already been selected, ensuring that additional resources will be allocated to fund ensuing stages o f their execution. An alternative option would be to maintain an annual time frame for a certain group o fpriorities, with those that require more complex execution assigned to a bi-annual cycle.32This step would increase the capacity o f OP participants to engage in public works oversight because it reduces the pressure o f one-year time cycles. Challenge 3: Ensurethe implementationof the InvestmentPlaninthe projectedtime frame. Failure to conclude implementation within the foreseen time frame generates serious risks to the credibility o f the OP, undermining participants' confidence in its capacity to influence actual municipal investments. Options to address this problem include: Create indicators and benchmarks for both investment and current account objectives that would clearly show how increased current expenses limit the government's investment capacity. Improve the quality o f public expenses, particularly current expenditures. Challenge 4: Ensurethat OP's InvestmentPlandoes not exceedthe actual ammount assignedto OP investments by the municipal budget envelope, Failure to allocate OP priorities according to the municipalities fiscal capacity leads to pressures to expand beyond what i s fiscally sound and negatively 32 I t mustbe acknowledged that making the OP a bi-annual process involves other important factors and trade-offs, such as a possible reduction o f the interfacebetween the population and the local government. 61 Economic Sector Work affects the credibility o f the OP among the population, when investments prioritized by the OP are not implemented because o f fiscal constraints. Actions to address this problem could include: Eachyear the municipality assigns a budget cap prior to the preparatory meetings of the OP. Include a rule inthe OP Internal Regulation (Regimento Interno) preventing that demands exceed the budget cap. Challenge 5: Ensure revenue and expenditure forecasts are as accurate as possible. The capacity of the municipality to forecast revenues and expenditures has deteriorated in recent years (a problem not exclusive to Porto Alegre), which affects both OP resources as a share o f total municipal investments and the municipality's ability to execute planned investments in the projected time frame. Actions to address thisproblemcould include: Introduce a mechanism in the OP cycle that permits adjustments to the Investment Plan during budget execution, allowing the OP to better address unforeseen falls inrevenue. Strengthen municipal capacity in budget planning. The use o f appropriate financial software could help the municipality evaluate the impact o f capital investments on tax revenue and current expenditures, as well as better incorporate external macroeconomic factors into its forecasts. Software could be also used to measure the impact o f changes in revenue and current expenditures on capital investments. A citizen-friendly version could help better inform OP delegates and councils duringthe prioritization process. Collect indicators on the accuracy o f income and expenditures and make them public duringeach budget cycle. Create an indicator (e.g., number o f demands budgeted correctly/total o f demands) to measure the accuracy o f disbursement estimates for each demand inthe IP. Challenge 6: Balance small localized public works with citywide strategic investments. Although historically the OP has not led to investment fragmentation, a recent trend o f increasing small investments can be discerned, together with the absence o f sound debate on long-term investments. Options to address thisprobleminclude: Use strategic policy objectives that are debated in city conferences and different sectoral public policy councils to guide discussions at the thematic meetings o f the OP. This guidance would help better align the prioritization o f OP resources with strategic policy issues. Incorporate a strategic planning methodology into the OP process to facilitate thematic meeting discussions, thus avoiding discussion o f demands that should be addressed in regional meetings and approved after citywide debates. Challenge 7: Strengthenthe OP's capacity to conduct independent,qualified analysis of municipal fiscalpolicies and monitor the impact of OP projects. Ifthe OP i s to contribute to Porto Alegre's fiscal health, a more systematic and in-depth public debate on the municipal fiscal policies and decisions needs to occur. Options to address this problem include: Incorporate in-depth public discussion o f the overall municipal fiscal policy and budget execution at the beginning of each OP cycle. Offer annual training inmunicipal finance to OP participants-particularly new OP delegates and councilors-that covers such topics as Multi-Year Plans (PPA), the Budgetary 62 Towardsa More Inclusive and Effective Participatory Budget in PortoAlegre Guidelines Law (LDO), the Annual Budget Law (LDO), bimonthly Summary Budget ExecutionReports (RREO), and Fiscal Management Reports (RGF). Establish partnerships with local universities and think tanks that could assist COP to undertake independent fiscal policy and budget analysis. Make the fiscal and budget information available on municipality portals more citizen- friendly, as well as update it more frequently. Develop the ObservaPoA into a system that offers statistical information that enables the OP to better allocate investments and monitor the impacts of different municipal programs. Offer regular training to OP councilors and delegates on how to incorporate the objective data available on the ObservaPoA into the OP process. 63 Chapter 3. The OPandtheAccountability of PublicExpenditure The debate on Participatory Budgeting has focused mainly on the role of citizen voice in capital expenditure planning, not their potential oversight role in monitoring public expenditure and service delivery. In addition to underestimating the role that the OP has played in empowering poor and socially excluded groups indecision making, the role o f the OP inengaging citizens inpublic oversight o f government expenditure and monitoring public works and services deserves discussion. Includingthe prioritiesof the poor in the municipalinvestment planis fundamental. However, it is insufficient to ensure that public resources are actually allocated to the intended priorities in an efficient manner, or that intended beneficiaries gain access to high-quality services. To cite one illustration, the actual cost o f a school construction project prioritized through a participatory process can, at the end o f the day, exceed the estimated cost due to expenditure inefficiencies and leakages. In addition, after building the school, what really matters i s the quality o f the education it imparts, which should be equal to or better than that o f other, similar schools. Therefore, for the OP to become a truly effective mechanism o f social accountability and poverty reduction, it should not only empower citizens to voice their priorities when planning capital expenditures, it should also strengthen their role in both scrutinizing public expenditures and holding accountable and disciplining service providers. Budget transparency and literacyis the first entry point for civil society to hold local governments accountablefor howthey managepublic resources.With adequate andtimely information, citizens can closely monitor policy implementation and budget execution. In several cases, local governments have actively engaged citizens inpublic bidding and procurement processes, as well as tracking flows o f public resources from revenue sources to final outputs o f goods and services. Through this process, local governments attempt to uncover budget waste, leakage, and favoritism-limitations that traditional check-and-balance systems often fail to uncover. Civil society can also play an important role inthis area by analyzing, disseminating, and building awareness o f government policies and budgets in an independent manner. Investigative journalism, local media, and civil-society organizations have an important role in scrutinizing public policy and government expenditures. Inaddition to a direct impact on accountability, independent policy and budget analysis i s a powerful mechanism for social learning, as it increases the literacy and awareness o f civil society on issues that are opaque to people without a technical specialization or who have yet to recognize the relevance o f such issues to their lives. Strengtheningthe power of service users is another a powerfulinstrument of social accountability because it can better discipline providers. Services providers are not always responsive to user needs and preferences, particularly inmonopoly frameworks. As the World Bank's WorldDevelopment Report 2004 (WDR04) explained, engaging users in an active role (e.g., as purchasers, monitors, and co- producers) can improve service performance tremendously. This finding i s particularly relevant to the poor, given that services have a strong impact on their well-being and that they usually have less access and purchasingpower to discipline providers. Strengthening client power means making the interests o f the poor matter more to providers. This can be achieved by increasing their individual purchasing power, tahng collective action to exert power over providers, and building their sense o f capability and empowerment. Experiences with participatory approaches to service performance assessment, such as community score cards and citizen report cards, have not only demonstrated positive changes in service responsiveness and performance, they have also awakened citizens, civil society, and the media to play a more active role inlocal governance. 64 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre Taking these issues into consideration, this chapter attempts to shed light on the largely unknown role the OP in facilitating greater social accountability of public procurement and the delivery of goods and services. It starts by analyzing the level o f budget transparency and literacy, as well as the actual contribution o f the OP to promoting budget transparency and public expenditure oversight in Porto Alegre. The analysis then looks at whether the OP has monitoring and evaluation mechanisms to discipline service providers. Finally, the chapter highlightsthe main shortcomings o f the OP inpromoting budget transparency and oversight o f public works and services; it then concludes by offering a range o f recommendations to address these challenges. 3.1 BudgetTransparencyandPublic ExpenditureOversightinPorto Alegre Brazil has detailed legislation relating to budget transparency and access to municipal fiscal accounts. Law 4320 established the normative framework that regulates the budgetary process. The law stipulates a series o f formal regulations pertaining to the development o f city budgetary proposals inall o f their dimensions, as well as the presentation o f accounts on budgetary execution. Publication o f the budget plan, together with annual balance and budgetary execution audits, are determined by current national legislation. In addition, the national L a w on Fiscal Responsibility, the state constitution, and municipal legislation also set strict guidelines for budget transparency. The Porto Alegre administration not only complies with national Brazilian legislation, it goes even further in budget transparency. The municipality rigorously complies with all legal requirements, to which it has added many other instruments for the presentation o f public fiscal accounts, some o f which are directly related to the OP and some o f which are not. As analyzed in the sections that follow, the municipality carries out different processes o f account presentation within the framework o f the OP, all o f which are aimed at informing participants about the execution o f Investment Plans. Additionally, the municipality has developed a sophisticated e-government system that gives the public access to a vast range of information related not only to budget planning and public expenditure, but also to different local government programs, as well as detailed socioeconomic statistical data. The use of the Internet in Porto Alegre represents one of municipality's major advances in transparency. Through the municipal Website (www.portoalegre.rs.gov.br), it i s possible to access not only all reports that the municipality i s required by law to make accessible to citizens, but also detailed data regarding capital investment planning, execution o f public works, and the implementation o f different municipal programs. As the box below shows, information on Porto Alegre's budget can be accessed through three different web portals, which together represent both a significant innovation and an advance inthe use o f the Internetto promote budget transparency. 65 Economic Sector Work Box 4. Access to budgetinformationonthe PortoAlegre Website (www.portoalegre.rs.gov.br) OP portal (www2.portoalegre.rs,gov.br/op/). The OP portal offers information on how the OP functions, news about the current OP cycle, and information about the implementation o f each OP demand according to the year it was prioritized, the secretariat o f the agency responsible for its execution, the region or thematic forum o f which it i s a part, and its identification number. The portal also provides links to key pages o f the municipal budget Website (e.g., multi-year plan, financial reports, etc.). MunicipalFinance DepartmentWeb page (www.portoa1egre.rs.gov.br). This Web page offers comprehensive information on municipal finance management, tax system, audit reports, procurement processes, and contracts. The Website has a link to the OP Portal. Municipal Management Portal (www2.portoalegre.rs.gov.br/portaldegestao/). This portal allows users to monitor the implementation all municipal programs using different performance indicators. For example, the management portal offers four main indicators to assess OP performance, including unmet investments, executed investments, budget execution, and number o f participants. The Management Portal also explains the rationale behind the new municipal management model, which i s based on programmatic areas rather than administrative units. Observatory of the City of Porto Alegre (www2.portoalegre.rs.gov.br/observatorio/). The Observatory ("ObservaPoa") i s probably the city's greatest contribution to informing its citizens about their municipality. ObservaPoA organizes and makes accessible the most relevant census information about the city, usinga GIS system that allows cross-referencing o f statistical data and geographic areas, such as regions o f the OP. It i s also a portal for accessing quantitative studies on different social and economic groups o f the population. The municipality believes that ObservaPoA can play a fundamental role in better informing public debate in Porto Alegre, particularly by offering quantitative socioeconomic data to help OP participants prioritize investments. Regular publications on municipal performance are another important contribution to transparency and accountability. Booklets are published on a bi-annual-basis that report on local government achievements during a specific period. Most include information on the city's fiscal and budgetary situation, as well as the implementation of different municipal programs.33This material i s not specifically directed at OP participants, since it has a wider distribution (i.e., the general population). Together with the information available on the Internet, these publications ensure a high degree o f transparency and accountability on the part o fthe municipality. OP Contribution to Public Expenditure Transparency and Oversight The OP, which has debated Porto Alegre's budget for the last 18 years, has unquestionably contributedto increasingpublicaccess to budgetinformationand awarenessof pro-poor spending. 33For the purposes of this investigation, material from the previous administrations was analyzed for the years 1995, 1997, 2000, and 2002, and the presentationof accounts of the current administration was analyzed for the years 2005 and 2007. 66 Towards a More Inclusiveand Effective Participatory Budget in Porto Alegre The survey conducted for this study showed that 85 percent o f people interviewed had knowledge of the OP process. Moreover, about 80 percent agreed or partially agreed with the statement that the OP favors poor people. This finding indicates high awareness among the local population about the redistributive nature o f the OP. As discussed below, the methodology o f the OP in Porto Alegre also favors increasing budget transparency by offering citizens different means to access information about local government expenditure. The most directway for citizensto learnabout municipalfinances is most likely at the OP meetings organized in each region of the city, or at the OP meetings organized around specific thematic areas. Duringthese meetings, the local government presents the overall situation o f municipal finances and the stages o f implementation o f the Investment Plan approved the previous year. Inaddition, as part o f the ordinary OP process, the municipality discusses its Multi-Annual Plan, the Budgetary Guidance Law, and the Budget Matrix in more detail at COP meetings. Usually, the government provides overall budgetary numbers and general concepts related to the budgetary process inPower Point presentations at COP meetings. These meetings between the municipality and citizens represent not only an effort to increase budget transparency, but also an opportunity to improve budget literacy. The various OP publications produced by the municipality are another important instrument of budgettransparency.Probably the most important publication is the booklet "OP: Plan0 de Investmento e Serviqos" ("Investment and Service Plans o f Porto Alegre Participatory Budgeting"). This document presents the budgetary matrix (i.e., forecasts o f current and capital expenditure) and the investments by administrative unit (direct and indirect), overall revenue from local taxes, inter-governmental transfers (i.e., SUS), and credit operations. It also presents the list o f works prioritized in each OP region and thematic area, as well as the estimated cost o f each investment. Finally, the OP Notebook also provides general information about the OP process, including a diagram o f the OP cycle, the names o f councilors, and general information on the municipality. The OP Notebook is, therefore, the main written document o f the municipality on overall budget figures and includes a detailed list o f investment priorities approved bythe OP. Regionalnewslettersare another importantmeans of promotingtransparency and public access to information. The 2001-2004 administration launched the OP Regional Journal, which informed the public about OP activities in each o f the city's regions. The current administration replaced this publication with the Regional Bulletin (Jornal da RegiEo), which aims to provide have a broader scope o f information on the city, including information about the new program, Solidarity in Local G ~ v e r n a n c e . ~ ~ Ingeneral, the newsletter is structuredinfour parts. The first pagepresents an actionofthe municipality o f particular interest to the local population or describes projects being carried out in association with civil society. The central part o f the newsletter presents information on matters o f interest to the entire city, for example, the participatory budget calendar, which normally has the same content for all regions. The last page i s used for matters o f interest to a specific region, such as actions carried out in the region by social organizations or the municipal government. Monitoring and Evaluation of Delivery of Public Works and Services Although the overall focus of the OP is on investmentplanning,it also provides a space for citizens to monitor the budgetary execution process. During the past 18 years, the OP in Porto Alegre has created instruments that allow citizens to monitor the execution o f the investments prioritized by them. The "Commissions o f Procurement and Monitoring o f Works" i s the formal oversight instrument o f the OP. Article 47 o f the Internal Regime o f the OP gives, at least in theory, great responsibility to the OP 34 The last number ofJornal da Regico was issued inJanuary 2007. The municipality plans to continue its publication in2008. 67 Economic Sector Work commissions to oversee budget procurement (i.e., tendering, bidding, and contracting) and execution o f the Investment Plan. In addition, OP delegates and councils regularly debate community demands regarding the quantity and quality o f municipal services. An inventory o f COP minutes between 1990and 2006 showed that services were often an agenda topic at these meetings. Out o f the 712 meetings for which minutes were analyzed over the 17-year period, services provided by the municipality were discussed at 14.4percent o f the meetings. Public debate on the quality of the services, however, takes place in the "Forum of Services," a social accountability space that functions in parallel with the OP. These service fora are public hearing mechanisms created as direct communication channels between citizens and the municipality on the quality o f public services. The service fora are not formally related to the OP process, but they function according to the same territorial divisions as the OP. In fact, the "Forum o f Services" is often associated with the OP, given that councilors and delegates see it as a privileged space that can be used to demand better services for their neighborhoods. Although not its primary objective, the recently created Solidarity in Local Governance (Governanqa Soliddria Local) programis another relevant space which provides opportunities for citizens to discuss service delivery.As part o f the SLG, the municipality established local governance committees (Cornit2 Gestor Local), the aim o f which i s to further decentralize and integrate the actions o f municipal departments and agencies in the city's sub-regions. According to the municipality, local governance committees are intended to better respond to the particularities o f each sub-region, facilitate municipal action at the neighborhood level, articulate the OP with the SLG, and promote a culture o f solidarity. This initiative i s providing a space where representatives o f local communities can discuss the quantity and quality o f municipal services with municipal secretariat staff. Inaddition, the program seeks solutions to the problems raised by the community through partnerships with local business and non- governmental organizations. The new modelof programmaticmanagementin the Porto Alegre municipalityhas also established a system for monitoringand evaluating service performance. This system allows the public to access information on the performance o f different municipal programs with respect to specific performance targets. A simple system o f colors (green, yellow, and red) i s used to rate the performance o f each program and its sub-activities. Although it offers greater transparency on the performance o f municipal programs, this model does not allow for citizen feedback. In this respect, the new system i s more an internal monitoring mechanism for the municipal administration. 3.2 LimitationsinBudgetTransparency,Oversight,and Service PerformanceMonitoring Limitations of the OP in Increasing Budget Transparency and Literacy Although the OP in Porto Alegre has made important achievements in terms of transparency, its impact has been very uneven. Findings from the public opinion survey, focus groups, and interviews conducted for this study indicate that the OP has had a limited impact on overall budget transparency and literacy. OP participants generally consider themselves poorly informed on the budget, particularly with respect to procurement and public works execution. Only a minority (4-8 percent) o f people interviewed stated that they were well informed. The majority said that they did not have much information on the budget. Interviews and focus groups corroborated these findings. When asked about their degree o f information on the subject, most people confirmed that "people do not have access, have no idea about how much tax is collected, and so on. " The information level o f citizens is, moreover, precarious and 68 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre obtained in a partial manner: "Some things. . .people hear on the radio. For example, Ilisten to the radio and once in a while theyprovide numbers, but no concrete information." Table 13. How Porto Alegre citizens ratetheir knowledgeof the municipalbudget General OP Non- Sample Control group participants participants Well informed 5% 9% 8% 4% More or less informed 29% 34% 39% 27% Without much information 65% 57% 53% 69% No answer 1% 1% Source: Public opinion survey, 2006. OP participants nevertheless considered themselves better informed about the municipal budget than non-participants. This finding leads to two conclusions. On one hand, after many years of implementation, the OP has apparently been unable to significantly expand the general population's access to budget information. On the other hand, the process has contributed to a certain increase in citizen knowledge o f the municipal budget, at least for those actively involved inthe process. The municipality has not, however, carriedout regular evaluationsof budget awareness and public access to information.Public surveys administered by the administration have historically concentrated on generic evaluations o f administrative performance, focusing more on the public image o f the government than on citizen awareness and perceptions o f budget transparency. Overall, the municipal administration lacks an effective communication strategy to deepen citizens' knowledge o f public expenditures. Limitedpublic knowledgeof the municipalbudget is partially a result of the lack of public interest in the subject. In spite o f efforts made by the municipality to inform the population o f different aspects o f the budgetary process, such as through the recent publication "Caderno de Prestaqgo de Contas," which reported on key fiscal and budgetary issues, ordinary citizens have not often taken the initiative to read municipal publications. As some people who were interviewed stated, "Ido not have information regarding the balance sheet of the municipality and I never tried to find out." This type o f response indicates that a strategy to promote budget transparency and access to infomation must go beyond making budget information publicly available-it must actively engage the municipality inraisingpublic awareness and buildingcivil society capacity to undertake independent budget oversight. One contributing factor to narrowbudget literacyis that the informationpresentedat OP meetings i s mainly focusedon the execution of public works. Analysis o f the material presented at OP meetings reveals important shortcomings. First, the presentation o f accounts during these meetings does not relate to the municipal budget as a whole, only to the Investment Plan (IP).There i s thus no presentation o f what was effectively spent, nor a comparison between what was forecast and what was effectively spent on public works. Although the OP process constitutes a privileged dialogue between local government and civil society, it i s limited mainly to public works chosen by the OP that are subsequently carried out. Meetings do not address the financial situation o f the municipality, nor do they effectively analyze the quality o f public expenditure in general. Consequently, it i s difficult for participants to gain a deep understanding o f the city's fiscal and budgetary situation. At most, they have a sense o f the progress o f IP implementation. Similarly,the OP Notebookof InvestmentPlandoes not allow citizens to effectively evaluatebudget execution. Overall numbers are presented in overwhelming profusion, but without analysis (e.g., the relation between planned and executed expenses or deficits and/or surpluses). Nor does the budgetary 69 Economic Sector Work matrix convey the exact share o f investments prioritized in the OP process in relation to overall capital expenditures. Resources for forecasted investments are also presented in a different manner every time, making it very difficult to analyze whether municipal finances are improving or deteriorating. As discussed inthe previous chapter, the information provided by the municipality thus limitsthe capacity of the OP to monitor overall fiscal policy. Only citizens with specific budgetary knowledge can analyze the informationprovidedby the city and extract the data needed to understand the real budget situation. For example, it is not easy to understand what share o f capital investment i s prioritized by the OP process. Currently, this comparison must be done by comparing the total amount o f investments proposedin the IP for each priority project with the general budget matrix. The complexity o f this operation for common citizens, and even for OP council members, limits the possibility o f the OP serving as a mechanism for budget transparency and independent oversight. In the same way, regionalmunicipalnewslettershave not been usedby local communities as a tool for monitoringpublic expenditures at a more detailed level. The regional newsletters (Le., the new Jornal da Regiio, or the old OP Regional Journal), for example, contain general information on budget execution, the financial situation o f the municipality, and the stages o f IP implementation. The publications thus function mainly to publicize government actions in each region. To a much lesser degree, they support citizens with the budgetary information needed to hold the municipality accountable for its use o f public resources. Although the Municipality of Porto Alegre has instituted significant innovations in using the Internet,the realpotentialof this mediumto increasebudgettransparency and literacy has not yet been fully explored. The information available on the various Websites created by the city is in many cases fragmented and not consistently updated. These problems highlight internal barriers that prevent municipal secretariats fi-om upholding a common transparency goal. Moreover, OP councilors do not regularly use the information available on the e-governance platforms. Although the municipality presented these innovations at OP meetings, interviews and focus groups revealed that only one OP participant had used this instrument to access budget data. This finding cannot be attributed to the fact that many councilors do not have computers connected to the municipal network because management o f the OP requires this equipment and the municipality offers access to computers and the Internet through its "Telecentros." Rather, it indicates that OP participants do not have the knowledge or habit o f using this kind o f tool. The availability o f data on the public network is therefore insufficient to ensure oversight in the absence o f proper awareness and capacity-building programs. The disinclination to access online data i s probably similar among everyday citizens, who are less involved indiscussions o f the public budget. Although the ObservaPoA offers an overwhelming amount of budget information,the platformis not sufficiently easy to beusedby averagecitizens.The mainproblem is that it is difficult to locate and reach budget reports on the portal, a process that involves a series o f successive commands and information criteria that must be searched. These procedures presume a degree o f technical knowledge that the average citizen evidently does not possess. In addition to this data, essential perspectives on budgetary control, which are strategic considerations o f the OP, are not related to the information on the portal. Finally, both councilors and delegates, as well as municipal technical staff, point out that the information on the portal tends not to be updated ina timely manner. TheLimitations of the OP regarding Budget Oversight and Service Performance Evaluation Participatorymonitoringof budgetexecutionis a secondarypriorityof the OP process,the primary focus of which is the elaboration of the annual municipalInvestmentPlan.A review of the 2007 OP Statute ("Regimento do OP 2007") clearly shows that oversight o f budget execution i s given much less 70 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre emphasis than participation inthe budget formulation cycle, Infact, the stages o f budget execution are not even included in the graphic representation o f the OP cycle. While the OP Statute i s very detailed in its definitions o f the methodology used in participatory budget formulation, oversight work has a much lower level o f systematization. The lack o f a proactive, sustained goal among OP participants o f exposing delays in the implementation o f public works or other budget executionproblems corroborates the notion that most participants do not have a clear incentive to scrutinize budget execution once they have ensured that their preferred public works have been included in the approved budget matrix. Thus, while the OP has done a good job in organizing and promoting the process by which citizens can voice their needs for public works and services, it has been unable to significantly advance effective oversight o f public expenditures. The Commissionsof Procurement and Accompaniment of Works have not always functioned in a systematic or efficient manner. The study survey shows that only 1 in 3 OP councilors usually participate in these commissions, which supervise the execution o f public works. Nor i s any methodological approach specified for delegates and councils to use in monitoring procurement for and the execution o f public works. The effectiveness o f oversight largely depends on the interest and knowledge o f the parties involved. The interviews conducted for this study made clear, for example, that certain commissions have undertaken careful oversight o f investment implementation intheir specific OP regions. According to the Secretary o f Budget Programming, OP councilors and delegates who have a better understanding o f budget execution, and have access to municipal technical staff, play a very effective oversight role, helping local authorities manage the process. The knowledge and skills requiredto carry out basic budget oversight work is uneven among OP representatives.A structured questionnaire completedby 44 current and former councilors demonstrated great fragility in their understanding o f the fundamental concepts o f public finance. Although the majority o f people interviewed answered the questions correctly (or approximately correctly), the percentage o f correct answers varied between 65 and 72 percent. The incidence o f wrong and no answers, moreover, indicated significant weaknesses in their basic knowledge o f how fiscal expenditures are monitored. As community representatives who decide certain municipal expenditures, the members o f the COP, particularly OP participants, should at least understandbasic budgetary concepts. Table 14. Survey question:Would you know howto say inyour ownwords the meaningof.. . Subject True More or less Wrong No answer Multi-annualplan 51.1% 17.0% 8.5% 25.4% Law on BudgetaryGuidelines 40.4% 25.5% 19.1% 14.9% Expenditure costs 66.0% 6.4% 17.0% 10.6% Deficit 53.2% 17.0% 17.0% 12.8% Source: Structuredquestionnaire o f44 OP councilors, 2006. Furthermore,the commissions are fragmented in character, with each devoted to monitoring one work in particular,without a global vision of budgetary execution (ie., quantitative data on public expenditure) or efficiency indicators (i.e., quality analysis). To the extent to which each work i s supervised in an isolated manner, no parameters exist that would permit comparisons o f public expenditure across regions. In addition, the monitoring o f a public works project by delegates directly interested in their realization could lead to a less rigorous review o f public expenditure, given that it i s in their the interest that the project be implemented. 71 Economic Sector Work The municipality and the COP mainly interact during the formulation phase of the Investment Plan and after the presentation of accounts at the beginning of the following year. There are no systematic meetings with municipal secretariats during the year to track execution o f the budget. Contacts with the secretariats are sporadic and depend on a decision or request o f the COP, as well as the consequent acceptance o f such by the secretariats, which inmany cases do not respond. Limitedattention to the execution of the municipalbudget may have indirectly contributed to the current difficult situation of the OP, where a largevolume of investments is not being executed as planned. Because o f the narrow focus o f the Porto Alegre OP on capital expenditure planning, it has become a very limited instrument for engaging civil society in actual monitoring and evaluation o f public expenditure management. Indeed, the analysis inthis study shows that the OP has been unable to promote in-depth public scrutiny o f critical fiscal situations or proposals for their concrete solution. Limitations of the OP with Respect to Service Monitoring Despite the various spaces for citizens participation in Porto Alegre, user feedback on public services performance is ad-hoc and lacks a methodology. Discussions between users and service providers currently occur through the OP, the Solidarity in Local Governance program, Municipal Public Councils, and direct contacts between "clients" and "providers." None o f these discussions employ structured methodological tools or systems for user feedback on service provision. Although an analysis o f COP minutes shows that there have been debates on service quality, these debates have taken place only when a specific issue comes to the forefront, not as a result o f a systematic procedure. The new municipalsystemof result-based managementis not directed at social accountability, but at internal control. The city is using management performance indicators to evaluate the quality o f public services mainly for internal purposes. The selection o f indicators has not yet been discussed with the actual users o f the different services. As result, the system has not incorporated indicators that are important from a user perspective. 3.3 MainChallengesandRecommendations There is room to improve the potentialof the OP to contribute to downstream accountability of public expenditure. Although the OP in Porto Alegre has contributed to greater understanding o f pro- poor public spending and has directly engaged ordinary citizens in voicing their priorities for capital expenditure planning, it has not fulfilled its potential to improve budget literacy and has paid scant attention to budget oversight. The municipality offers positive conditions for budget oversight: it carries out different processes o f account presentation; has a sophisticated e-government system; and provides information updates on budget planning, public expenditures, and local government programs, as well as detailed socioeconomic data through its Website. However, oversight on the part o f COP members and municipality officials i s neither well codified nor supported by a sustained effort, and thus takes place mainly in an ad-hoc manner. The following paragraphs identify challenges related to the OP's ability to play a more effective role in monitoring public expenditures and providing feedback on the delivery o f public works and services. Challenge1: StrengthenbudgetliteracyinPortoAlegre. Currently, budget literacy is both limitedand uneven among Porto Alegre citizens. Although the OP has contributed to a greater understanding o f pro- 72 Towards a More Inclusive and Effective Participatory Budget in PortoAlegre poor public spending, it has not fulfilled its potential to improve budget literacy. Actions to addressthis challengecouldinclude: Make more frequent use o f local radio to raise public awareness o f basic municipal finances and budgeting. Similar radio programs have been developed by the World Bank inPeru, Ecuador, and Guatemala, as well as inAnglophone African countries. Include a supplement on fiscal and budget literacy inthe annual OP Statute. Include detailed information on municipal revenue and expenditures in the OP Notebook o f InvestmentPlan. Include detailed technical and financial information on each project being implemented in the various regions inthe regional newsletters, allowing comparisonbetween plan and execution. Include an overall evaluation o f the implementation o f all municipal programs in the Booklet o f Municipal Account, using the performance indicators adopted by the new management system. Programs and activities that are under-performing should offer technical justifications for their performance incitizen-friendly language. Improve the articulation o f different pages o f the Municipal Website, particularly the OP, ManagementPortal, and ObservaPoA pages. Include issues o f municipal finance and budgeting in the curriculum o f Porte Alegre secondary schools (some states o f Brazil have already instituted this step). Regularly conduct surveys and focus groups with users o f these different budget communications tools to identify concrete recommendations that can make the platforms more users friendly. Challenge2: Expandthe scope of the OP to include independent budget oversight in a systematic manner. Although the OP formally enables citizens to monitor the execution o f the Investment Plan, oversight takes place mainly in an ad-hoc manner. The Commissions o f Procurement and Accompaniment o f Public Works function only in regions where councilors and delegates are more interested in oversight work, have prior knowledge o f the projects, and enjoy better contacts with the municipal administration. Actionsto addressthis challengecouldinclude: Incorporate a phase for budget execution oversight and related activities into the OP cycle. This change in the OP Statute could be informed by a series o f debates between the COP and the municipality to identify how the OP cycle could strengthen oversight o f budget execution (e.g., tendering, bidding, and contracts). Representatives from municipalities where this process i s already inplace (e.g., Belo Horizonte) should be invited to share their experience. Create an effective reporting system and incentives to promote citizen engagement inmonitoring the implementation o f public works prioritized by the OP. This reporting system should be linked to incentive mechanisms. For example, OP regions could be rewarded for their efforts to ensure the technical quality and financial soundness o f budget implementation o f projects in their respective regions. Inconsultationwith the COP, the municipality should develop a practical manual for citizens to monitor budget execution. This manual should provide simple tools for citizens to review public tenders and contracts ("licita@es"), as well as to track the execution o f these investments. OP councilors and delegates should receive training from the municipality on budget oversight each year, preferably at the end o f the budget formulation cycle. Training modules should ideally 73 Economic Sector Work be made available online in order to explore the potential o f the e-government platform and the accounting systemto monitor revenues and public expenditures. The OP Website should include simple indicators that compare overall budget planning data (e.g., amounts, time o f implementation, basic technical details) with budget execution data for different investments. These indicators should be developed to enable the use o f benchmarks and cross- regional comparisons. The municipality should make citizen-friendly budget executionreports (e.g., the bimonthly Summary Budget ExecutionReports, or RREO) available on a regular basis. Challenge3: Strengthenmechanismsfor the monitoringand evaluation of municipalservices. Port0 Alegre citizens do not have systematic opportunities to provide feedback on service delivery. Moreover, this feedback is limitedand lacks a structured methodology. It is important to note that, intheory, certain specific social accountability spaces, such as the sector-based municipal public councils and the Forum o f Services, are more suitable for the discussion o f service performance than i s the OP. It i s thus recommendedto: Strengthen the capacity o f sector-based municipal public councils and Forum o f Services to monitor and evaluate service delivery. Citizen feedback mechanisms, such as community and citizen report cards, should be introduced to strengthen client power in monitoring service performance. Engage local civil-society organizations, academia, and research institutions to undertake more technical monitoring and evaluation o f municipal fiscal policies, public expenditures, and services performance. These social actors could be instrumental in providing independent information to stakeholders engaged inthe OP. The municipality should introduce clear indicators and benchmarks for service quality, reliability, and satisfaction into the new municipal program management system. The municipality should consider developing a Citizens' Charter o f Municipal Services that would set citizen-friendly performance benchmarks for each service provided by the city, as well as a mechanismfor user feedback. 74 Chapter4. Towards a System of SocialAccountability inPortoAlegre Duringthe last two decades, decentralizationand democratizationreforms have driven the creation of severalgovernance innovationsaimed at strengtheningthe voice of civil society at the locallevel. The 1988 Constitution, also known as the Citizenry Constitution ("Constituip7o Cidada""), defined clear instruments o f direct democracy, among them, public councils, referendums, and popular petitions. These institutional mechanisms guarantee citizens the exercise o f their constitutional rights to voice their preferences in public decision-makingprocesses and exert social control over the management of public resources and the delivery o f public goods and services. The synergy between collective action, democratization reforms, and the election o f progressive mayors has been decisive in incorporating direct participation into different aspects of municipal management in Brazil (Avritzer 2002). Public municipal councils, participatory budgeting, and many other fora for dialogue have become common practice in most Brazilian cities. Although participatory budgeting in Porto Alegre is probably the most well-known mechanismof direct participation in Brazil, there are many other spaces for participation in localgovernance. As i s the case in many other cities, Porto Alegre has multiple spheres o f interaction between the local government and civil society. Each o f these spheres has a particular focus, ranging fiom expressing citizen voice in policy making, urban planning, and budgeting, to the co-production o f public goods and services. Ultimately, these different spaces should be articulated in a coherent manner to create a system of participation and social accountability that can translate participation and mediation into development outcomes. As the previous chapter discussed, while the OP has become a vibrant practice o f participatory governance, there i s room to improve not only the OP per se, but other spheres o f public participation in the city. This chapter looks at the interface betweenthe OP and other mechanisms of participation inPorto Alegre to identify challenges to and possible recommendationsfor the creationof a coherent system of participation and social accountability. The chapter starts by mapping the main spaces o f public participation in local governance apart from participatory budgeting. Subsequently, it analyzes the nature and quality o f the interfaces between the OP and these venues o f participation, identifying unnecessary transaction costs, information failures, and other signs o f weak coordination. Finally, it makes recommendations on how to strengthen the synergy between the different participatory mechanisms so that they can function as a coherent system. 4.1The Multiple Spaces of PublicParticipationinPorto Alegre Innovative participatory governance and collective action have a longer history in Porto Alegre than does participatory budgeting. As Baiocchi (2005) points out, participation in Porto Alegre has evolved over many decades to create multiple spaces o f interaction between the municipal government and civil society. Baiocchi notes that the relationship between the state and civil society in the period 1945-1977 was marked by the tutelage o f popular associations and limited participation in social programs. This relationship began to change with the increasing social mobilization that occurred as a result o f the relaxation o f the authoritarian regime inthe period 1977-1985, a period followed by the first concrete attempts to institutionalize, albeit in limited form, public participation in local governance (1986-1988). With the election o f a local government supported by a wide spectrum o f social movements, Porto Alegre entered a new phase o f empowered participatory local governance in the late 1980s. 75 Economic Sector Work Table 15. Evolutionof state-societyinteractionsinPorto Alegre 1945-1977 1977-1985 1986-1988 1989to present State-society Tutelage, Tutelage Tutelage with Proliferation o f relations authoritarian initial institutions o f state consultative empowered local governance participatory mechanisms governance Formal state-society Firstsector State-run Self-help Establishment o f arrangements councils community programs, Public Councils, councils consultative fora OP, and city-wide consultative fora Civilsociety and FRACAB FRACAB, UAMPA, Citywide, political institutions UAMPA, and first popular councils, territorial-based popular councils NGOs, and (district) and issue- (conselhos political parties based networks of populares) civil society organizations, political parties Principal agenda "Social pacts" Re-democratization Initial Broad with the state, and certain acceptance o f institutionalization occasional anti- community participation in o fpublic paternalist participation in local governance participation movements social programs Source: Adapted from Baiocchi (2005). Public Municipal Councils One important innovationin social participationin Porto Alegre has been the creation of Public Policy Councils (Conselhos de Politicas Pliblicas). Although these councils already existed in certain sectors and at various levels o f government, the Constitution o f 1988 made them a formal institutional space where civil society could exert its voice and control over policy making and public expenditures. Public Policy Councils can be generically defined as institutionalizedmechanisms that allow civil society, service providers, and governments to discuss public policies at all three levels o f the federation (national, state, and municipal). Their characteristics and powers range from consultative to deliberative. Particularly in those sectors with deliberative power, public councils have impacted municipal management in a profound manner because all policy decisions are first discussed and approved by the council before they are implementedby the municipal administration. The city of Porto Alegre currently has 20 Public Municipal Councils. The first councils were established in the 1930s and subsequently expanded. With the re-democratization o f the country after 1985, other venues with a more participative nature emerged. Most were reformulated to broaden participation or inresponse to new legislation. Current legislation defines these venues as "organizations o f societal participation inpublic management with the finality o fproposing, monitoring, and deliberating over materials referring to each sector o f the admini~tration."~~Among the 20 councils, two could be considered multi-sectoral (the OP and the Urban and Environmental Development Council); the rest are sector-specific. 35Article 101o f the Organic Law o f the City o f Porto Alegre (2004). 76 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre Table 16. PublicMunicipalCouncils of Porto Alegre, as of 2004 Acronym Title CMAA Agriculture and Supply CMAS SocialAssistance CMC Culture CMD sports CMDCA Children'sRights CMDH HumanRights CMDUA UrbanandEnvironmentalDevelopment CME Education CMS Health COMAM Environment COMATHAB Access to LandandHousing COMPHAC Historical, Artistic and Cultural Heritage COMTU UrbanTransportation COMUI Senior Citizenry CONCET Science and Technology CONCONT Taxpayers CONDIM Women's Rights CONEN Drugs CONTUR Tourism COP ParticipativeBudgetCouncil - Source: Nahra (2005) and Alfonsin (2004). The composition of each council varies according to their history of social mobilization. According to Nahra (2005), there are a total o f 512 positions on the councils. Nearly 20.5 percent o f these positions are taken by representatives o f the municipal administration, while civil society holds about 69 percent o f the seats. The 10.5 percent are shared by representatives of the Municipal Chamber and o f the state and federal government. Table 17. Component sectorsofthe PublicMunicipalCouncils, as of 2004 Councilmembers Representatives Number Representativesof the city administration 105 20.50 Representativesof civil society 353 68.94 Others 54 10.56 Total 512 100.0 City Congress(es) The need to engage civil society more evenly in local strategic development planning bec me an importanttopic inPorto Alegre. At one point, previous municipal administrations and key social actors became conscious o f the need to bring sectors o f society into direct political participation in addition to those already involved in the OP. They were also aware that the city lacked a long-term, multisectoral perspective on local development. In order to fill this gap while maintaining direct participation processes, the city administrationlaunched Project Porto Alegre, a discussion space where representatives o f civil society defined action guidelines for the municipal government. This initiative culminated in the 77 Economic Sector Work First City Congress o f December 1993, which used a strong participatory approach to establish major guidelines and strategic goals for the development o fthe city. Since 1993, citywide strategic planningdebateshavebeenorganized periodically and draw growing numbers of participants. The Second City Congress, discussed below, was almost entirely dedicated to discussing the revision o f Porto Alegre's master plan. The Third Congress, held betweenNovember 1999 and M a y 2000, attracted more than 7,000 participants. The strong social mobilization for the congresses has established them as a public reference on the formulation o f strategic goals for the future o f Porto Alegre. With more than 10,000 participants, the Fourth City Congress (entitled "Democracy and Management o f the State"), focused on making changes in the structure and culture o f the municipal administrationto accommodate the expanded number o f spaces for political participation incity affairs. Smaller conferenceshavebeen organized by the MunicipalCouncils to follow up the resolutions of City Congresses in their respective sectors. Organized annually or bi-annually, sector-specific conferences continue to debate public policies and monitor their implementation by municipal administrators. This process o f sector-based conferences i s complemented by district fora that focus on adapting sectoral policies to the realities o f different territorial areas. Urban Fora The revisionof the city's master planled to a new space for politicalparticipation.In 1995, during the Second City Congress, participants agreed on the need to reformulate the Master Plan of Urban Development o f Porto Alegre (PDDUA). Describing the process by which the plan was formulated, the then mayor remarked, "[Tlhe innovation was in the adopted methodology, which had democracy in the urban planning process as its main ingredient, allowing and stimulating the widest possible participation o f the population and its representative entities" (Genro 2005, 3). A variety o f organizations from every field participatedin the discussions o f the Master Plan during the Second City Congress.36Its resolutions touched on the entire range o f matters related to urban planning (e.g., housing, transport, zoning, environmental and municipal management) and explicitly incorporated public participation and decentralizedmunicipal management. Urban Planning fora were then established to continue discussions of the master plan in a decentralizedmanner. These public fora seek to encourage a more active participation o fcivil society in the review o f the city master plan. They are usually attended by councilors and delegates o f the OP regions, technical staff o f the Municipal Secretariat o f Planning (SPM), and representatives o f civil society. Coordinated by the SPM, discussions o f revisions to the PDDUA currently take place in eight different regions o f Porto Alegre. Solidarity in Local Governance Program The current administration of Porto Alegre has introduced a new civil society-local government arrangement in the form of the Solidarity in Local Governance program -(PGSL)?'Theprogram aims to create a new governance system by articulating social networks o f democratic political participation in local areas. The program uses the same territorial structure o f the OP regions, but them decentralizes further to the neighborhood level. 36For 8 months, more than 200 organizations and a total o f 2,600 people participated inthe discussion groups and sessions of this Congress. ''Programa de GovernanGaSolidaria Local inPortuguese. 78 Towardsa More Inclusive and Effective Participatory Budget in Porto Alegre The initiative represents a new effort to address important challenges that key stakeholders in Porto Alegre were already debating in City Congress events3*and other fora. First, it seeks to address the problem o f the city's limited capacity to solve all its problems with limited pubic funds by mobilizing social capital and public-private partnerships. By doing so, it also attempts to broaden social participation to include segments o f the population that have not been active in existing spaces for public participation, particularly the middle class and the private sector. Ultimately, the PGSL aims to establish a holistic local governance platform that can better articulate the various needs o f public, private, and civil society actors incommon local development goals. Although during the research for this paper the PGSL was still in the early stages of implementation, it has already shown tangible results. Launched inlate 2005, all 17 regions o f Porto Alegre have finalized implementation o f the forum and 12 regions have validated their agendas3'. Several investments have already been implementedby the PGSL through partnerships between the municipality, local community organizations, and the private sector. Usually, local businesses provide construction materials, the community provides labor, and the municipality provides machinery and financial resources. 4.2 The Interfacebetweenthe OP and Other Spaces of Political Participation The OP interfaces with the different venues of participation in Porto Alegre in several ways, althoughthese interfacesare often neitherclear nor systematic. These channels o f interaction can be ad hoc or take the form o f a formal institutional structure. Many social activists and civil society organizations are active in more than one venue of participation, as each venue offers different opportunities to voice their interests. On a more formal level, trilateral commissions have been established to bringthe OP, Municipal Councils, and local government representatives together to address the same agenda. However, as explored in Chapter 1, the weak participation o f important social actors (e.g., the middle-class, private sector, local academia, and other professional associations) limited the potential o f the OP to cooperate with other spaces ofparticipation inthe city. The priority given to each space of participationby the municipaladministration also influencesthe ability of the OP to interfacewith other participatoryvenues. Basedonits privilegedrole, the municipal government influences the participation spaces to which it pays more attention. While under previous administrations, the OP was unquestionably the backbone o f the entire structure o f citizen participation, the current administration seems to give a more balanced weight to each participation space. Indeed, previous administrations saw the OP as probably the most important channel for empowering traditionally disenfranchised social groups to participate in local policy making and therefore contributed to a hierarchy in which the OP was more important than other spaces o f participation. The current administration, however, invokes a more "holistic" approach to participation and focuses its efforts on promoting voice in all the city's various participation channels, including previously existing channels and the newly created PGSL. 38For example, the need to establish a new model o f municipal management based on programbudgets, local management committees, public observatories (Le., sources o f information), local governance networks, a municipal capacity-building system, etc. 39 Monitoramento do PGSL(Programa de GovernanGaSolidaria Local), September 2007. 79 Economic Sector Work However, the interfacebetweendifferentspheresofparticipationremainsfragile.Although the topic i s not new? local actors have been unable to establish clear "rules o f the game" to organize the different spheres o f participation in a positive way. Serious concerns about conflicting agendas make it difficult to establish a system that links policy formulation, capital investment planning, budget execution, and service delivery. In order to create synergy among the different participation spaces, Port0 Alegre must establish a way to overcome the current fragmentation o f these spaces, which limits its capacity to capitalize on the potential o f participatory local governance. The OP and Municipal Councils The interfacebetween the OP and the MunicipalCouncils is not always positive. The relationship between the OP and the councils i s marked by both cooperation and conflict. According to studies undertaken at the Fourth City Congress, practically 60 percent o f municipal council participants believed there were conflicts between the councils and the OP. The same study pointed out that there was a positive integration between the OP and the councils on issues o f education and social assistance. On the other hand, it noted clear conflicts between them on issues o f land and housing, urban transport, children's rights, and women rights and health. During the focus group discussions conducted for this study, it became clear that the nature o f these conflicts varies and the problem o f conflicting organizational interests has not yet been resolved. Although the OP has functioned for 18 years, questionsremain about its reallegal power vis-his the municipal councils, whose power is established by law. Most municipal council participants emphasize the legal status o f the councils, a perception that clearly contrasts with the OP, which i s not defined in a legal framework. The difference i s thus a distinction between dejure bodies (the municipal councils) and a defacto body (the OP) over the prioritization o f capital investments. Participantsin the OP and municipal councils clearly diverge on the issue of integration. Several representatives o f the councils feel the municipality should play a more active role in coordinating the bodies. According to one interviewee, "The Municipal Council of Science and Technology has the role of assisting the municipality in formulating policies in the area. Although independent, the council plays a part in the municipal management in [this sector]. Therefore, it is not up to the . . . council to go to the OP meetings to dispute resourcesfor this sector. It's the administration that has to take the proposals from the council to OP meetings." The city government has not, however, consistently tried to coordinate the OP and the councils. There is also a dispute between the OP and the councils over social legitimacy. Quite often, the COP does not recognize municipal councils, which have more limited grassroots mobilization. Councils that have been established more recently, or do not have an active social base, experience greater difficulties inestablishing positive relations with the OP. As one member of the Senior Citizen Council noted, "The Senior Citizen Council is new and therefore has dificulties in becoming integrated with the OP." This means that councils with weaker social bases do not have a strong voice in investment planning deliberations and are not able to approve demands that reflect their distinct needs. The OP and municipal councils rarely reconcile long-term sector policies with short-term investment planning. The logical connection between policy malung and budgeting breaks down when each issue i s debated in a different venue o f participation. According to several municipal council representatives, "The councils focus on sector policies for the city as a whole, while thepeople in the OP have a narrowfocus on localized and short-term demands.` It i s important to highlight that the creation I o f thematic areas inthe OP and the opening o f venues for the OP and municipal council representatives to ~~ 40 This was a key topic during the 4' City Congress. 80 Towardsa More Inclusive and Effective Participatory Budget in Porto Alegre meet are intended to connect sectoral policies to investments. However, as several OP participants recognize, the focus o f thematic meetings on citywide investmenthas been gradually lost. As discussed in Chapter 3, many investment proposals approved in the thematic areas are clearly regional in nature because OP councilors who fail to win approval o f their communities' demands at regional meetings treat thematic discussions as a second chance to do so. The hierarchy between the OP and municipal councils varies according to the decision-making power of the council. Councils that have legal deliberative power, such as the Council on Health, rank higher than the OP, while those with only consultative power rank lower. According to one member o f a council with deliberative power, "[It is] the council that decides the use of R$2.5 millionfrom the social assistance fund, whereas the Budgetary Programming department has defined R$500,000 for social assistance investment through the OP. Another council member reflected, "The council receives a list of " demandsfrom the OP per region and attaches the deliberations of the OP. The CMAS makes theJinal decision considering the OP demands, but it has to go through the CMAS. " The story i s different with councils that have only consultative power. According to a member o f one o f these councils, "The budget proposal doesn`t come to the council directly, itfirst goes through the OP and then we have to chase after it." This situation creates a serious problem, as in some cases, sector-based policy clearly sets priorities for investment. Inother cases, sector investments are not tied to broad policies and policy-making bodies thus lackthe muscleto influencepublic investments. The establishment of "Trilateral Commissions" is an attempt to encourage greater harmony betweenbudgetingand policy. These commissions include representatives from the OP, sector councils, and different agencies o f the municipal government. Their main function i s to discuss specific criteria for OP investments, as well as to promote partnerships among different sectors. For example, in the case o f the Council on Primary Care, Social Assistance, and Health, the commissions are restricted to discussing criteria for partnership agreements between day care centers, family social assistance, and basic health care units. Although the role o f the commissions as an interface between two different dimensions o f participatory governance (budgeting and policy making) i s restricted, they have achieved some results in terms o f downstream inter-sectoral coordination. OP and Urban Fora The interface between the OP and other public fora regarding reformulation of the city master plan is plagued by similar problems of failing to link policies to investments. During the Fourth City Congress o f 2003, there was nearly unanimous agreement among the regional urban fora that the failure to integrate participative planning with participative budgeting was a fundamental problem. One o f the examples cited by representatives o f these fora was a region that had drawn up a regional plan, but did not receive the necessary investment to implement it because the plan wasn't approvedby the OP. Lack of discretionary power over the allocation of resources has undermined participation in urban fora. Civil society evidently loses interest when issues debated in the regions must be taken to the Municipal Council on Urban and Environmental Development (CMDUA) for decisions. For example, regional discussions were held on the multi-annual plan o f the city to generate proposals. The suggestions were not, however, considered by the government nor incorporated into the Multi-Annual Plan, thereby causing the fora to feel excluded from the decision-making process. Their frustration eventually led to a gradual demobilization o f the local community. The foregoinganalysis clearly indicatesthat Porto Alegre lacks a coherent localgovernance system for participation and social accountability. While the city i s a champion o f deepening democracy at the local level, it has ended up creating a number o f different spaces for direct participation without an 81 Economic Sector Work overarching framework (see Figure 20). In fact, in several situations, duplication o f roles, competition, and real tensions exist between different venues. Certain important functional gaps in the current constellation o f participatory venues also exist, such as inrevenue and expenditure tracking, procurement oversight, and performance monitoring and evaluation. Although many o f these problems are known, there has been limited progress towards making the various spaces o f public participation more coherent and effective. Ironically, failing to address this fundamental challenge seriously undermines the basic principles and goals o f Porto Alegre's participatory local governance model. Figure 20. The policy and expenditure cycle and civil society-local government institutions POLICY& EXPENDITURE CIVIL SOCIETY- MAIN CYCLE LOCAL GOV. INSTNS. CHALLENGES / '- Councils Lack o f overarching- framework to articulate different civil society-local I I governance institutions I I I I I I I I I I I I I I I I I ............................ '*. I **., I .': ......................... I : missing i I I :i: i I .......................... .: I: I I /I - 8 The PGSL was, to a large extent, created to cope with the fragmentation of spaces for public participation. As the current local government and many other actors point out, Porto Alegre has developed a network o f participation over the past several decades that significantly deepened local democracy, increased social control over policy making and investment planning, and promoted more socially inclusive public policies. However, the organic evolution o f these different spaces o f participation has created a complex and fragmented network that does not always express the richness o f the city's social fabric or fulfill its potential for cross-sectoral synergies. The PGSL seeks to address these challenges through multi-stakeholder engagement and cross-discipline dialogue, broadeningconsensus on local development goals and actions to achieve them. 82 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre However,the Solidarity in Local Governance programwas initially greeted with misgivings by a significant number of OP councilors. On several occasions during research for this study, COP members expressed their discontent with the new program because they could not understand the need for an additional mechanism for public participation in the city. They were also concerned that the program would increase competition for scarce municipal resources. One important factor contributing to this mistrustwas the perception o f certainOPparticipants that the new political alliance that came to power in the city could potentially weaken the OP process by legitimizing other venues o f participation. The idea o f using non-public resources to address city problems (i.e., through public-private partnerships encouraged by the PGSL) was also perceived as a proposal in line with a "minimum state" model, as opposed to the welfare state concept that inspired social reforms inpast. The relationshipbetweenthe OP and the Solidarity in Local Governanceprogram has improved since the launch of the latter, although certain coordinationproblems remain. As the program is being rolled out and initial results have begun to appear, OP participants have become more receptive to the PGSL and have started to engage in its activities more systematically. However, certain OP participants find it difficult to participate in the Solidarity in Local Governance program due to limited free time. For instance, while most OP meetings take place after business hours and on weekends, meetings o f the governance program are held duringnormal office hours. The highcost o f participation in other spaces o f participation, such as district fora and sector-based public councils, thus creates a critical trade-off between different spaces. 4.3 MainChallengesandRecommendations Although participatorybudgetingin PortoAlegre is probablythe mostwell-knownmechanismfor direct participationinBrazil,there are many other spaces for participationinlocal governance.As i s the case in many other cities, the multiplicity o f spheres o f interaction between the local government and civil society lends itself to overlapping roles, competition, and undesirable tensions between different venues. At the same time, the absence o f an overarching framework allows for functional gaps in the current constellation o f participatory mechanisms (e.g., inrevenue and expenditure tracking, procurement oversight, and performance monitoring and evaluation). I t is importantto recognize, therefore, the limits of the OP within this complex networkof citizen participationin Porto Alegre. Since its inception, the main objective o f the OP has been to include citizen input in capital investment planning. Over time, and as a result o f its success and permanence, but also as a response to its need to understand and scrutinize the municipal budget, the OP developed some capacity to monitor the execution o f prioritized public works. However, this capacity i s weak and requires strengthening. On the other hand, other spaces o f participation were created before and after the OP. Although these other channels have specific purposes, such as the Public Policy Councils (which focus mainly on sectoral policy making) and the PGSL (which has a more holistic approach to local governance), many overlaps and conflicts clearly exist. It i s therefore useful to recall that there may be limits to the ability o f the OP in Porto Alegre to expand its role to include other areas o f participatory local governance, particularly if it i s to strengthen its current role. It i s also clear that the municipality and the different non-governmental stakeholders that play an active role in the various channels o f participation have not yet created a harmonizing framework for a more effective and efficient system o f civic engagement and social accountability for the city. Challenge I: Reconsider the relationship between the OP and Municipal Councils to ensure maximumsynergies.Currently, the interface betweenexisting institutions o f direct participation is weak, 83 Economic Sector Work undermining the possibility o f a coherent system o f social accountability that would support the local policy and expenditure cycle. Options to achieve better synergies could include taking action to: Establish an overarching agreement framework between the different municipal councils and the COP to clarify formal and informal functions and remove overlapping roles in budget planning and oversight. Strengthen the interface between the different municipal councils and the thematic axis o f the OP. This goal couldbe achieved by assigningrepresentativeso fthe municipal councils to relevant OP thematic areas. Better integrate the UrbanFora (i.e., public debate on the city's master plan) with OP investment prioritization through a specific OP thematic group. Challenge 2: Strengthen the alignment of long-term sectoral development objectives with OP annual capital investment planning. Although certain important resolutions have been adopted, City Conferences have not generated sufficient momentum or direction to strategically influence the different spaces o f public participation in the city and hence guide the development o f Porto Alegre. To address these challenges, the following suggestions are offered: Use the City Conference to implement a well-structured strategic planning process among existing civil society-local government institutions. Such a process would establish clear long- term goals that would align policy making and budget planning. Establish a permanent governance structure to oversee the formulation, implementation, and revision o f the city's strategic planning process. Inaddition to the City Conference, this structure should include a multi-stakeholder coordination committee, an executive secretariat, a technical team, and thematic task forces. Task forces should be comprised o f representatives o f the municipal public councils, thematic areas o f the OP, urban fora, and the Solidarity in Local Governance program (according to the specific sector with which they are aligned), together with representatives o f the municipality (direct and indirect administration), the local legislature, and the private sector. Organize public debates on the implementation o f the city development strategy at the beginning o f each OP cycle Create special task forces for (i)local economic development and (ii)municipal financial management and administrative modernization. These two areas are presently undermined by the multiplicity o f participative spaces, yet remain critical for helping Porto Alegre overcome its underperformance in fiscal policy and economic growth. Challenge 3: Support organized civil society to undertake independent oversight of municipalpolicies, public expenditures, and service delivery. Although certain spaces o f participation undertake these oversight activities, they are largely marginal to their core agendas. The experience o f many other countries with independent oversight shows that i s produces clear development outcomes, such as reduction o f budget leakages, better alignment o f policies with public expenditures, and greater responsiveness o f services providers. Recommendations include: Given that Porto Alegre, together with most Brazilian municipalities, has only nascent capacity in independent oversight, it i s recommended that local authorities promote a series o f seminars and public debates where national and international experience could be presented. 84 Towardsa More Inclusive and Effective Participatory Budget in Porto Alegre Engage local professional organizations and academia to establish independent coalitions to analyze municipal policies, scrutinize public procurement, and undertaken routine evaluation o f the performance o f different service providers. 85 Chapter 5. Evaluating ParticipatoryBudgeting This chapter presents an estimation of the impacts that participatory budgets have had on Brazilian municipalities. Since the late 1980s, municipalities in Brazil have experimented with participatory mechanisms in budget preparation. The analysis that follows will use popular econometric techniques to evaluate the impact o f the OP on intended variables. Isolating the causal relationship i s the most important aspect o f econometric analysis, allowing it to make compelling arguments about the usefulness o f a given program. As the OP has a functional objective o f expanding public participation, the focus o f analysis will be whether the ultimate intended variables are impacted by the OP participatory processes. It is importantto underlineon the methodologicallimitationsof an econometric analysisof the OP, given that in addition to operational participation, it promotes accountability, transparency, and other major institutional changes (some o f which have been achieved and others are currently under way). By definition, the breadth o f the OP makes it extremely difficult to isolate its impacts. The overall impact o f the OP on the structure o f governance in Porto Alegre may thus not be fully captured by econometric analyses. In view o f those limitations, this chapter thus adopts a rather simple approach o f explaining changes in final outcomes by changes in determinant variables, which include the OP. Many other important dynamics are consequently not accounted for. Impact evaluation usingeconometric techniques presentsseveral challenges because of the needto isolatethe influenceof OP fromthat of other variableswhich affect fiscal and socialoutcomes. The very decision to adopt OP in some municipalities and not in others can produce a bias inthe identification o f impacts. For example, in Brazil there i s a trend to find OP processes in cities with a stronger presence o f the Workers' Party (PT), which in itself may entail the promotion o f similar social goals and outcomes independently o f the adoption or not o f a OP process (Baiocchi et al. 2006). Inthis chapter, the analysis does not test whether these two political processes-OP and votes for the PT-have common roots, but it does seek to isolate the impact o f OP from other elements o f the political process, thereby permitting greater certainty that OP caused the observed changes in outcomes. The percentage o f votes for PT is, therefore, a permanent control variable. Another, perhaps simpler, control would be to make compare OP municipalities with municipalities that tend to be equally affluent. Inother words, the impact o f OP must take into account differential initial conditions in order to isolate the effects o f participatory budgeting. Moreover, certain impacts are likely to occur over a longer time period than that for which data are available; in other cases, insufficient data exist on given variables to positively confirm expected effects o f OP. As shown below, the findings of this econometric analysis are mixed, suggesting that: (i)expectations of the OP process should be managed; (ii) the OP process could usefully be strengthened in both Porto Alegre and other cities; and (iii) further empirical impact analysis is warranted. The econometric studies undertaken to date point to different results, depending on the specifications and outcome variables considered. They show that the OP has positive impacts on reducing corruption and achieving certain other outcomes (e.g., increasing water access), but less positive impacts with respect to other variables. The chapter begins by discussing the expected results from OP implementation and summarizes the main findings o f earlier econometric studies. It then explains the methodological challenges o f analyzing OP impacts and the approach used inthis study. It then examines fiscal and social outcomes for OP versus non-OP municipalities in Brazil since 1992, using panel data. The chapter closes with concluding remarks. 86 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre 5.1 ExpectedResultsof OP andEarlier EconometricImpact Studies The major underlyingassumption of this chapter is that the introductionof OP changesmunicipal governancestructures and public management and, consequently, municipaloutcomes.The second underlying assumption i s that formalized popular participation in budgetary decision making will cause citizens to be more willing to pay local taxes since they have more control over allocation decisions. Similarly, the incentive o f elected officials to improve tax collection i s larger when a broad base o f citizens has concrete public investment expectations. OP can also be assumed to lead to better social outcomes, as it opens channels o f participation to all citizens (and notjust elites, where lobbyingpower i s concentrated) and thus increases the opportunity for government investments to be targeted to social needs. That is, the local government may not enhance public budget allocation unless it directly consults its citizens. While there is a consensus that the OP in PortoAlegre has been responsiblefor more progressive expenditures and fostering a stronger sense of citizenship, few quantitativestudies have sought to estimate whether these impacts are associated strictly with OP. Analyses to date include Marchetti (2004), Baiocchi et al. (2006), and Zamboni (2007). Marchetti's (2004) evaluation concludes that the OP inPorto Alegre has improved the spatial distribution o fpublic goods. However, his analysis is limited to Porto Alegre and it does not have a counterfactual with which to compare results. This study also takes into consideration the existence o f a technical rule inPorto Alegre that privileges regions o f the city with larger populations and lower average incomes, a rule that may compromise his results, given that his analysis does not make clear whether it i s the rule or the OP that drives observed results. These issues are addressed inthe present analysis by comparing changes in outcomes in OP municipalities with changes in non-OP municipalities, rather than comparing the intra-municipal distributional impacts o f the OP in Porto Alegre alone. One econometric study points to a reduction in the prevalenceof corruptionin municipalitieswith OP. Zamboni (2007) suggests that municipalities which implementedOP were likely have less corruption than municipalities that did not implement the program. Using audit reports, Zamboni analyzes a set o f Brazilian municipalities usinga pair-wise matching approach. H e concludes that in municipalities where OP was implemented, the likelihood o f corruption was lower than in non-OP municipalities. The author also suggests that participation in local public administration i s more relevant to governance than voting inelections. These results also suggest that participatorybudgeting may act as an effective constraint on the executive duringbudget implementati~n.~~ Another study points to positive outcomes in OP versus non-OP municipalities in Brazil, drawing on cross-sectionaldata for municipalities across the country. The study o f Baiocchi et al. (2006), for example, estimates the impact o f implementing OP in the periods 1993-1996 and 1997-2000, usingnon- OP municipalities as the counterfactual. The authors use a cross-sectional data set o f Brazilian municipalities to compute changes in outcome variables between 1991 and 2000, with a dummy variable approach employed to estimate the impact o f implementing OP in either the 1993-1996 or 1997-2000 period. They also compare pairs o f similar OP and non-OP municipalities and find that OP municipalities demonstrated improved outcomes for revenues and the provision o f social services. This study adopted the same methodological approach used in Baiochi et. A1 (2006). However, the sample o f municipalities i s not the same. Furthermore, whereas he uses a set o f variables to make municipalities comparable based on physic characteristics, this study uses the same outcomes in years prior to the baseline so to compare 4' Although Zamboni's (2007) analysis does have certain methodological similarities with the analysis presented in this chapter, the variable o finterest to Zamboni is corruption, while this study focuses on fiscal performance, municipal resource allocation, and welfare changes. 87 Economic Sector Work municipalities that had the same dynamics inthe intended outcomes before the OP was implemented. As such, this study claims that any impact from OP could have had the same probability o f being observed in either group o f municipalities. The evaluation of OP impacts must begin by reviewing the intended results, or outcomes, of the introduction of participatory budgeting. Many intermediate results could have been achieved by a different budget allocation; however, such results may not be sufficient justification for the introduction o f a policy. Instead, the evaluation o f impacts should look at whether implemented policies have contributed to the ultimate goals they intendedto achieve. Table 18 describes the expected effects o f OP. Table 18. Expectedoutcomesof participatory budgeting What effectsare What are the mechanismsthroughwhich it causes What variables are proxies expectedfromOP? the expectedeffects? for outcomes? Strengthenedpro- OP i s a municipal governance approach that aims to Access to water poor capital close the gap between the preferences o f the poor and -- Access to sewage investments decision making o n investments by empowering the poor to voice their needs incapital budget formulation. Inthe short term, OP is expectedto result ina capital budget more consistent with the real needs o f the poor. ~ Reduction inpoverty Bymeans ofredirecting the capital budget to pro-poor - Povertyheadcount investments, it i s expected that OP can have a mid- to long-term positive impact o n reducing poverty. Strengthenedbudget OP i s a municipal governance approach that aims to - Total municipal revenues transparency and strengthen budget transparency and citizen awareness. including the collection of literacy Itis thus expectedto result inincreasedlocal tax property taxes compliance and a better balance between current and -- Municipal investments capital expenditure. The OP may also increase municipal public investment vis-u-vis operational Operational expenditures to expenses. capital investment ratio Living conditions are estimated by examining income poverty, access to potable water, and access to sanitation. The rationale for this analysis lies in the fact that OP opens space for the participation o f the poor, who may traditionally have lower access rates to improved living conditions and less lobbying power and voice than the better-off. Secondly, the OP has the potential to improve the quality o f public spending, since citizen involvement in the municipal budget will increase scrutiny o f government expenditure. An indirect way to check the pro-poor capital investment hypothesis i s to look at the outcome, i.e., to verify whether OP municipalities reduced poverty rates at a faster pace than non-OP municipalities. As discussed above, the channels through which these impacts are achieved are not tested here. The "willingness to pay" argument is tested by examining fiscal impacts. The rationale for fiscal impacts goes back to the assumption that citizen influence over allocation decisions will increase willingness to pay taxes, as well as create pressure on the municipal government to collect taxes to ensure that resources are available for investments. If this argument holds, one can expect OP municipalities to display better fiscal performance than non-OP municipalities, reflected in property tax collection (not service taxes4*), since the focus o f OP i s individual participation, although small and medium enterprises 42The main sources of municipal revenue inBrazil are property taxes and a tax on services (i.e., a sales tax applied solely to companies inthe service industry). 88 Towardsa More Inclusive and Effective Participatory Budget in Porto Alegre may be well representedby their owners in the OP process, Accordingly, the analysis tests if there i s any significant impact on the growth o f total revenue, property taxes, and service taxes. OP may also induce a different allocation of resources in cities that adopt it, since their budget decisions are expected to differ from those of other municipalities.Inparticular, it may be that OP favors municipal capital investment over operational expenses. This hypothesis i s tested comparing the per capita investment o f OP municipalities with that o f non-OP municipalities. Other possible differences in budget allocation are tested for all main expenditure items. The analysis does use an a priori hypothesis, but participation may induce different expenditure emphases. 5.2 MethodologicalApproach A standard method for evaluating social programs uses the outcomes of non-participants to estimate what participantswould have experiencedhad they not participated in the program. The difference between participant and non-participant outcomes i s the gross impact o f the program. The net impact i s obtained using the same approach, but computing the difference in time so unobservable (unmeasured) determinants are cancelled. For instance, changes in an outcome for participants are compared to changes in the same outcome for non-participants. The analysis in this chapter uses this approach to measure the impacts o f OP on selected outcomes, which together account for net impact o f the OP, measured by the time difference in outcomes observed in municipalities that adopted OP versus the time differences for the same outcomes in municipalities that did not adopt OP. Such an analysis i s possible because o f panel data that exists for Brazilian municipalities for the years 1991-1992 and 2000, which permits a comparison o f OP municipalities versus proper control municipalities. Non-OP Municipalities, or the Control Group The key questionbehindthese methods is whether observed impactscan be attributed to the policy under analysis, in this case, participatory budgeting. That is, does a causal relationship exist between the implementation o f OP and the observed changes (expected to be positive) in outcomes? To answer this question, a control group o f non-OP municipalities was constructed. One can selectively pick a control group that performs poorly compared to OP municipalities and attribute the difference in performance to OP. The analysis must, however, prove that both OP and non-OP municipalities had the same opportunity to observe improvements in outcomes prior to the implementation o f OP. In this analysis, OP municipalities are compared to a distribution (kernel) o f non-OP municipalities-the equivalent o f being compared to a theoretical distribution o f the same outcomes, given similar initial conditions. A traditional analytical alternati~e,~~also presented inthe tables below, i s to compare one or a few non-OP municipalities to each OP municipality. This approach i s not, however, the preferred option inthis chapter because there are relatively few OP municipalities (11before 1996, 37 after 1996) among the more than 5,000 Brazilian municipalities. Instead, all non-OP municipalities are included in the control group, but are given different weights depending on how close their initial conditions were to those o f OP municipalities. To do so, a propensity score was obtained using a kernel distribution o f weights for all non-OP municipalities. Variables that define these similar conditions in the baseline are described below. First, the control group is controlled for whether other political processes are taking place at the same time as OP. Such parallelprocesses prevent econometric analysis from separating out the source o f the impact. For instance, as discussed by Baiocchi et al. (2006), there seems to be a high correlation 43For example, through Propensity Score Matching. 89 Economic Sector Work between the percentage o f the population voting for Labor Party (PT) candidates and the existence o f OP in a specific municipality. Therefore, either the improvements produced by channeling voice through a formal participation process (the OP) and/or the political process (including strong participation through informal channels o f voice), both o f which were associated with a substantial adherence to the PT, actually caused the impacts or they did not. From 1989 to 1992, more than half o f the municipalities that adopted the OP process were managedby a P T mayor (during that tenure, the PT hadjust 1percent o f the mayoralties in the country). PT-managed cities continued to be prevalent among municipalities that adopted the OP. Duringthe periods 1993-1996 and 1997-2000, the share o f these cities was similar to the period 1989-1992 (around 50 percent). The proportion diminishes to 30 percent only in the most recent period (2001-2004). Considering that the percentage o f municipalities managed by the PT have also increased considerably from the first to the last period under analysis, the sample i s most even in the last period analyzed, but still highly concentrated. Furthermore, incertainmunicipalities that implementedOP where a PT mayor was not elected, the party nevertheless had a large share o f total votes or was part o f the winning coalition. The specific issue o f votes for the PT i s extensively tested inthis chapter, but there may be other issues that the analysis here not fully addresses. Health sciences, for example, usually deal with potential contaminations by randomizing who receives treatment, which was by no means possible here because OP i s the result o f a political process. The identification o f the causal relationship i s improved by introducing votes for the PT in the construction o f the control group. Both results, with and without votes for the PT, are presented in order to illustrate that failure to model the political process could leadto a bias inestimated impacts. Second, the group of non-OP municipalities is controlled for the outcome variable in the baseline year and its dynamics before OP implementation. For example, when analyzing changes inpoverty, a distribution o f municipalities i s selected that had poverty levels comparable to those o f OP municipalities. (An improvement in poverty must be related to a city's initial poverty level.) The dynamics o f poverty between 1980 and 1991 are also included to determine whether or not a process that began before OP implementation actually helped improve poverty. Conversely, each outcome analyzed has a control group distribution comprisedo f bothbaseline levels and poverty dynamics prior to the introductiono f OP. Third, the non-OP municipalities are controlled for the potential differential effects of the Constitution of 1988. This constitution could have created differential impacts on different sized municipalities because it changed the way federal fiscal resources were allocated. Specifically, small municipalities were granted larger resources relative to the preceding distribution o f resources. Inorder to control for this situation, population size in 1990 i s included inthe construction o f the control group. TheDifference in Diference Equation With the control group at hand, a model is estimated to analyze the determinants of changes in outcomes. The model includes a constant for each municipality and differentiates or cancels the unobservable term, assuming that it remains constant over the period o f analysis. All other variables, dependent and independent, are treated as time differences. A dummy variable i s set to zero for all municipalities for the first data point, or baseline. For the second data point for 2000, the dummy i s switched to 1 for OP municipalities; it therefore captures the dynamic impact o f the implementation o f OP. Other determinants, or independent variables, are included inorder to control for whether the change in the dependent variable was caused by traditional determinants o f improved outcomes. Specifically, economic growth measured by GDP per capita and transportation costs to Sao Paulo are included. Transportation costs measure the relative location o f the municipality within the Brazilian territory, which control for the differential impact that national processes, as opposed to economic growth, can have on municipalities, depending on their spatial locationwithin a country. For instance, the Constitution o f 1988 could have created differential impacts among municipalities beyond those captured in the control group throughpopulation size, thus favoring poor northeast Brazilian municipalities. 90 Towards a More Inclusive and Effective Participatoiy Budget in PortoAlegre 5.3 Resultsfor OP andNon-OPMunicipalities The econometric analysissuggests that participatorybudgeting as a mechanismfor improvingpro- poor capital investments has contributed to ameliorating the living conditions of the poor in the municipalities where it has been adopted. Inthis study, Brazilian municipalities that adopted OP are compared with a group o f non-OP municipalities that had the same opportunity to observe improvements in final outcomes, like poverty and access to basic services, before the advent of OP. The analysis controls for the fact that in most municipalities where OP was implemented, the PT was a key political actor, either by runningthe major's office or being the second most important political force in a city. In this way, the analysis accounts for whether estimated impacts were the results o f OP andnot the result o f the same political process that brought the PT into power. Therefore, the analysis i s able to determine whether it was OP, and not the political process (including the presence o f strong social movements and informal mechanisms o f voice) that enacted OP, which caused the impacts. The impact can be read in Table 19 according to the sign and significance o f the dummy OP. Highlighted columns on the right are models with a control group that were constructed using votes for the PT along with the other abovementioned variables. The OP sign i s expected to be negative because the poverty rate i s reduced, while the sign for access to piped water and improved sanitation should be positive, as they are expected to increase. The size o f the coefficient i s not meaningful here. Note that results are consistent when the control group i s constructed by including votes for the PT. Only municipalities that implemented OP before 1996 were includedinthis analysis. Table 19.Impactsusingonly municipalitiesthat implementedOP before 1996 Note: p < .l; p < .05; * ** *** p <.01. It should be noted, however,that OP helpsreducepoverty rateswhen implementedfor a decade.A second analysis was performed to include the 37 municipalities that implemented OP between 1996 and 2000. A dummy was switched to 1for municipalities that had implemented OP duringthe entire 10 years between 1990 and 2000. A second dummy was switched to 1 for those municipalities that implemented OP after 1996. The objective o f breaking the dummy into two periods was to capture which one better explained the variance in outcomes. It seems that impacts on better access to basic services were positive inboth cases, but positive impacts on poverty reduction were present only inthose municipalities where OP had been inplace longer. 91 Economic Sector Work Table 20. Impacts using municipalitiesthat implementedOP before and after 1996 R2 Note: p c .l; p C .05; * ** *** p C .01. Fiscal Performance Participatory budgeting does not have a consistent impact on fiscal performance. The second hypothesis tested was that OP empowers the poor, creates more transparency, and strengthens budget literacy among the population. The assumption i s that giving citizens influence over allocation decisions will increase their willingness to pay taxes, as well as create pressure on the municipal government to collect taxes to ensure that resources are available for investments. The ratio o f PTU (local taxes on capital) to GDP measures the capacity o f the local municipal administration to collect taxes inrelation to the city's economic potential. When municipalities that implemented OP both before and after 1996 are included in the treatment group, there appears to be an apparent positive impact o f OP on fiscal performance. However, this impact disappears when only municipalities that implemented OP before 1996 are included in the sample. Nevertheless, there seems to be a positive impact on tax collection, but this resultwouldneedto occur amongthe different specifications to ensure that it isnot a spurious result. Table 21. Impacts on fiscal performance(IPTU/GDP) OP before/after 1996 OP before 1996 Note: * p < .1; ** p <.05; *** p < .01. The impact of participatory budgeting on the investment to current expenditure ratio is negative. Regarding the share o f resources expended on the operation o f municipalities, as opposed to resources devoted to capital investments over which the OP has influence, the results o f analysis show a negative impact. A possible explanation of this result i s that the OP i s financing social infrastructure, which creates staffing and maintenance needs that drive up current expenditures. 92 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre OP before/after 1996 OP before 1996 Note: p < -1; * ** p < .05; *** p < .01. 5.4 The Case of Porto Alegre Inorder to analyze the specific impacts o f OP on Porto Alegre, this study compares actual city data with averages for the other comparable municipalities, i.e., the analysis seeks to answer the question: how does Porto Alegre compare to the control group, on average? The first idea i s to compare Porto Alegre with a group o f other state capitals weighted in such a way that they resemble Porto Alegre (in terms o f the observables or measured variables) as much as possible. In other words, weights are larger for more similar municipalities. The weighted sum, that we call "synthetic Porto Alegre," i s the control group. This strategy follows the work o fAbadie and Gardeazabal(2003) andAbadie, Diamond, and Hainmueller (2007). These authors deal with the same methodological problem: how to analyze an event (or policy intervention) that happens at the aggregate level and affect individual entities. So they compared the aggregate effect with effects on individual municipalities. Despite the benefits o f comparing Porto Alegre with a non-OP group, there are two main shortcomings in this type o f analysis. First, it is difficult to choose the municipalities for the comparative group without being arbitrary. Second, given that the comparison i s between one group and one municipality, inference techniques do not apply. Abadie and Gardeazabal (2003) address the methodological shortcomings o f comparative case studies with a data-driven procedure to reduce discretionary bias that the researcher can impose to the comparison. It forces researchers to demonstrate affinities between treated and control units (in our case municipalities). However, it i s usually very difficult to find a single or "best" control group. The synthetic control represents a very similar group (but no identical because nothing can be two units o f observation at the same time), that better resemble the treated unit.The analysis here therefore uses data from 1986 to 1989 for all 27 capitals (and the Federal District) to construct a synthetic Porto Alegre and then compares this synthetic unitto the actual Porto Alegre inthe period 1990to 2004. Living Conditions Since time series data i s required to conduct the analysis described above, available data limits the comparison to one o f average wages and infant mortality as broad proxies for living conditions. Figure 22 below show the difference in average formal wages between Porto Alegre and the synthetic control for two alternative groups o f state capitals. One with all Brazil state capitals (with the exception o f Porto Alegre) and another that includes only capitals that didnot implement OP between 1986 and 2004. 93 Economic Sector Work Figure 21. Averagewages inPorto Alegre (a) All state capitals (b)Non-OPstatecapitals Looking at Figure 21a it i s very difficult to correlate the OP to the dynamics of formal wages in Porto Alegre. The vertical line represents the first year o f OP. Comparing with all municipalities; it does appear that wages in both, the synthetic control and Porto Alegre, followed approximately the same trend. Once the control group i s builtjust usingnon-OP municipalities (Figure 21b), it does not resemble Porto Alegre very well before the OP cut o f 1989, thus it i s not possible to say very much about the performance o f wages in Porto Alegre. Figure 23 compares Porto Alegre with the same two synthetic controls for infant mortality. Figure 22. Infant mortality rates inPorto Alegre and "synthetic" controlfor two groups of state capitals, 1986-2004 (a) All state capitals I(b)Non-OPstatecapitals 1885 I690 1985 2000 2005 -treated unit-----syntheticcontrolunit -treated an0 unit-----syntheticcontrol unit The results for infant mortality are the opposite than what was expected. According to the analysis on Figure 22a, other (weighted) state capitals reduced infant mortality at a faster pace than Porto Alegre. However, the graph also makes it clear that Porto Alegre started at a much lower mortality rate than other, very similar state capitals. Since it i s difficult to reduce infant mortality beyond a certain threshold, this may explain why the impacts on this variable for Porto Alegre are not significant. For instance, certain impacts on this variable could be expected as a result o f sanitation improvements. 94 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre Inconclusion, it cannot be said that Porto Alegre performed better than either synthetic Porto Alegre in terms o f average wages and infant mortality using available data. It i s also difficult to say that the impact on the two variables was worse in Porto Alegre than the two control groups, given that the impact on average wages i s unclear and the impact on infant mortality i s blurred by the fact that Porto Alegre started at a much lower level then the control groups. Fiscal Performance It was expected that analysis o f fiscal performance inPorto Alegre and the two synthetic control groups would find greater differences, since this variable i s more sensitive to changes in the short run and represented the primary target o f the program. Ifwe examine current revenues, however, it seems that the OP had the opposite impact. Right after the program began current revenues in the synthetic control for all state capitals increased faster than inPorto Alegre. However, the result i s unclear when Porto Alegre i s compared with the synthetic control for non-OP state capitals. In any case, this indicator seems to demonstrate a reversion inbehavior inthe years 1995-1996, when Porto Alegre revenues started to grow faster than the synthetic controls. (a) All state capitals (b) Non-OPstate capitals 0 1885 1990 lSQ5 2000 2005 lQ85 1990 1995 2000 2005 ano ano l- treated unit-----svntheticcontrol unit I -treated unit -----syntheticcmtrol unitI Looking at total revenues in Figure 24, no considerable difference i s found between Porto Alegre and the two synthetic control groups. They seem to follow a very similar path, except after 2000, when Porto Alegre seems to perform better than either control. The difference around 2000 is very clear when Porto Alegre i s compared to all other state capitals, but less clear when the city i s compared to non-OP municipalities (where the break seems to occur in 1995). This result i s puzzling. First, it i s difficult to relate the difference inperformance to the OP, given that the break occurredlong after the introduction o f OP. Second, we know that Porto Alegre had a fiscal problem after 2000, but this analysis suggests that its fiscal performance was not worse, but probably better, than other state capitals in this period. It i s thus more likely that the explanationthus lies with expenses and not revenues. 95 Economic Sector Work Figure 24. Total municipalrevenues inPorto Alegre and "synthetic" controlfor two groups of state capitals, 1986-2004 (a) All state capitals I@) Non-OPstate capitals 8- I h 4 el h! 1985 1990 1995 1885 1em 16% 2wo 2ws an0 2000 2005 am -treated unit -----syntheticcontrol unit -treated unit -----syntheoccontrol unit With respect to current expenses, shown in Figure 25 below, both comparison groups suggest that Porto Alegre spent more than the synthetic controls after the years 1995-1996. A jump occurs from 1995 to 1996 that then stabilizes, i.e., Porto Alegre follows the path o f the two synthetic control groups but remains at a higher level. This finding may explain the fiscal problem observed from 2002 on. Porto Alegre was systematically expending more than it should (comparing to other state capitals) and, although it was able to increase its own revenues (especially after 2000), these revenues were insufficient to compensate for the (accumulated) extra expenditure. Given that the OP was implemented in 1989, it i s difficult to accept that these additional expenses were indeed connected to the OP. However, the OP itself could have changed around this time, puttingpressure on the municipal budget. Ifthis i s the case, the data should show an increase ininvestments inPorto Alegre inthe period around 1995-1996. Figure 25. Current expensesinPorto Alegre and "synthetic" controlfor two groups of state capitals, 1986-2004 (a) All state capitals I@) Non-OP state capitals 1'1 I/ 1I / I I I 7 1B85 I& 1ano Be5 2000 2005 2005 treated unit -----syntheticcontrol unit -treated unit -----syntheticcontrolunit Examination o f Figure 26 below, however, shows that the expectation o f greater investments (implied here by the variable "total expenses") in Porto Alegre in 1995-1996 i s incorrect. Right after 1989, investments in the synthetic control group for all state capitals i s much larger than in Porto Alegre, whereas there i s no clear difference in investment dynamics when the city i s compared only to the state capitals that have never implemented the OP. The difference might therefore be explained by another 96 Towardsa More Inclusive and Effective Participatory Budget in PortoAlegre variable, or it may be connected to the debt dynamics that are not observed when using the variable of current expenses. Figure26. InvestmentsinPorto Alegre and "synthetic" control for tw groups of state capitals, 1986-2004 (a) All state capitals (b) Non-OPstate capitals 0 1985 I& 1995 2000 2do5 1985 1960 1995 2m 2w5 an0 ana I- treated unit -----svntheticcontroiunitI -treatedunit-----syntheticmntroi unit When the comparison uses the variable of operational expenses (see Figure 27), the samejump fiom 1995 to 1996 occurs as it does in the comparison o f current expenses. This finding suggests that it i s unlikely that the fiscal problem inPorto Alegre was actually connected to the OP, which i s related to investments. It i s more likely that Porto Alegre increased its operational expenses after 1996 for some other reason and that it was unable to solve the problem o f growing expenses even with the (relative) increase inrevenues observed after 2000. Figure 27. Municipaloperational expensesinPorto Alegre and "synthetic" control for 1 10groups of state capitals, 1986-2004 (a) All state capitals @) Non-OPstate capitals o / , -1 w 0 l9& 1990 1 9 s l9&5 1880 1ms ZOW 2005 am 2000 2WS am -treatedunit-----synmetcmntroi unit I-treatedunit-----svntheticmntmlunit 1 97 Economic Sector Work 5.5 Conclusions The analysis conducted for this study isolated the fiscal and social impacts of OP in order to establish whether they were in fact caused by participatory budgeting. Based on econometric findings to date, it is important for municipalities that are implementing OP to have clear expectations with respect to expected impacts. Since the institutional reforms involved in the application of OP continue to be relatively new, there may also be genuine potential for improving the impacts from OP in Porto Alegre and inother cities o f Brazil. The analysis inthis chapter comparedBrazilianmunicipalities that adopted OP to a group of non- OP municipalities that had the same opportunity to observe improvements in poverty, access to basic services, and fiscal performance before the advent of OP. The analysis controlled for the fact that in most municipalities where OP was implemented, the PT was a key political actor, either by winning the major's office in elections or being the second most important political force in a city. The analysis thus accounted for whether the estimated impacts were the result o f OP and not a political process (including strong participation through informal channels o f voice) associated with substantial support o f the PT. The analysis thus ensured that OP was the instrument that caused the impacts. The econometric analysis suggests that participatorybudgeting, as a mechanism to improve pro- poor capital investments, has contributed to a reduction of poverty rates in the municipalities where it was adopted prior to 1996. It is worth noting that this poverty impact occurred despite a reduction in GDP per capita in these municipalities, suggesting that the OP can contribute to a redistributive impact over the long run. To date, it seems that the impact on better access to water and sanitation i s positive for all OP municipalities. Currentfindings do not indicate consistentpositiveimpactsof OP on municipalfiscal Performance. The second hypothesis tested the impact o f the OP on fiscal performance, assuming that the OP empowers the poor and, therefore, leads to more transparency inthe way the public budget i s allocated, as well as increased budget literacy. Participatory budgeting was not, however, found to have a consistent impact on fiscal performance, although there i s an apparent positive impact when municipalities that implemented OP before and after 1996 are included in the same treatment group. However, this impact disappears when only municipalities that implemented OP before 1996 are included in the sample. The impact o f OP on fiscal performance, as measured by the ratio o f operational expenditure to total expenditure, furthermore appears to be negative. The absence o f positive impacts on fiscal performance associated with the OP can be explained by the exclusive focus o f the OP on investment capital allocations and its weak connection to other fiscal management decisions, such as current expenditure allocations, expenditure execution, and revenue collection. A simpler analysis, but one consistent with the logic of treatment and control group evaluation, shed light on the specificsof OP in Porto Alegre. It compared Porto Alegre with two synthetic control groups, one comprised o f all state capitals and one comprised only o f non-OP state capitals. The findings o f this analysis were, however, inconclusive. Using available data, it cannot be said that Porto Alegre performed better than the two control groups (synthetic Porto Alegre) in terms o f living condition. It i s also difficult to say whether the impact o f OP on average wages in Porto Alegre was worse than in the control groups. Its impact on infant mortality is, moreover, blurred by the fact that Porto Alegre started at a much lower level for this variable. With respect to fiscal performance, comparative analysis suggests that it i s unlikely that the fiscal problem in Porto Alegre was actually connected to the OP, which i s related to investments. More likely, the city increased its operational expenses after 1996 for some other reason and was unable to solve this problem o f growing expenses even with the (relative) increase in 98 Towardsa More Inclusive and Efective Participatory Budget in Porto Alegre revenues observed after 2000. The causes o f the Porto Alegre fiscal situation are most likely not connectedto OP. 99 Economic Sector Work Chapter 6. Main Conclusionsand Recommendations 6.1 MainConclusions Participationin the OP is substantialin Porto Alegre, but certain social groups suffer from a lack of incentives to participate. Around one-fifth o f the population reports having participated in the OP at some point intheir lives. Women and low-income groups show highpercentages o f representation. Ethnic minorities and people with disabilities are also well represented. Under-represented groups include the extremely poor, youth, middle- and high-income groups, as well as certain professional groups, such as entrepreneurs. Time requirements, the use o f complex language, and limited understanding o f the OP process present barriers to participation, especially at higher levels o f representation (COP). For the very poor, the opportunity cost and expense o f participation is very high. Higher-income groups lack incentives to participate because the OP has a narrow focus on infrastructure provision for the poor and participationinthe process i s time-consuming. The OP enjoys credibility among its participantsand the populationof Porto Alegre as a whole, despite certainshortcomings regardingits internalupstreamand downstreamaccountability. There i s an uneven playing field for OP participants in the delegation to upper levels o f representation. Participants with more years o f experience, time availability, and knowledge o f the complex procedural rules o f the OP are more likely to be elected as regional delegates and councilors, which has led to low levels o f rotation among delegates. Additionally, reporting mechanisms between delegates and their constituencies are unsystematic and characterized by the absence o f records. The prevalence o f a verbal culture among OP constituencies, which demands and expects less accuracy than a written culture, and the fact that the OP i s itself perceived as a net gain in public accountability, may explain the existing mismatch between OP credibility and shortcomings regarding its internal accountability. Despite enjoying high levels of participationand credibility, the analysis of the interfacebetween the OP and municipalfinancialmanagementyields mixed results. Contrary to popular belief, the OP in Porto Alegre does not lead to suboptimal resource allocation or investment fragmentation. When it comes to the risk o f investment fragmentation posed by regional pressures, which seek to allocate resources to finance small public works, the OP also performs quite well: although the number o f demands for small investments has increased in the last few years, it has not affected the overall share of resources allocated to satisfy those demands. However, there i s no evidence that the OP has played a positive role in contributing to fiscal balance inPorto Alegre. In different years, estimates o f expenditure and revenue did not match actual values. This fiscal gap was usually covered by generating current deficits andor postponing capital investments. Inconsistencies between the OP cycle and the actual budget calendar have also contributedto the formulation o f capital investment plans based on weak fiscal data. This problem created a backlog o f non-implemented OP demands and consequently generated strong public discontent. The OP in Porto Alegre has worked mainly as an instrumentfor interested citizens to voice their preferences regarding capital investments, rather than as a space of public debate to discuss the municipalfiscalpolicy and revenueand expendituremanagementas awhole. An analysis o fminutes o f key meetings o f the COP (Participatory Budgeting Council) showed that it has not undertaken substantive assessment o f municipal fiscal policies and revenue capacity, particularly at critical moments. This behavior of the COP can be attributed to the absence of a clear motivation among its members to give priority to these issues over including the demands o f their respective communities inthe Investment 100 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre Plan. The lack o f regular training o f COP members i s another possible explanation, as the LDO, LOA, and PPA are o f a technical complexity that laymen find difficult to understand. There is roomfor improvingthe potentialof the OP to contribute to downstream accountabilityof public expenditure.Although the OP inPorto Alegre has contributed to a greater understanding o f pro- poor public spending and to directly engaging ordinary citizens in voicing their priorities in capital expenditure planning, it has not fulfilled its potential to improve budget literacy and has paid scant attention to budget oversight. The municipality offers positive conditions for budget oversight: it offers different types o f budget account presentations, has a sophisticated e-government system, and provides updated information related to budget planning, public expenditure, local government programs, and detailed socioeconomic data through its Website. However, oversight i s not well codified or supported by sustained efforts on the part o f COP members and municipality officials, and thus takes place mainly in an ad-hoc manner. Although participatorybudgetingin PortoAlegre is probablythe most well-knownmechanismfor direct participationin Brazil, there are many other spaces in the fiscal cycle for participationin localgovernance. As is the case inmany other cities, the multiplicity o f spheres o f interactionbetween the local government and civil society lends itself to overlapping roles, competition, and undesirable tension between different venues. At the same time, the absence o f an overarching framework leads to functional gaps among the current constellation o f participatory mechanisms, such as revenue and expenditure tracking, procurement oversight, and performance monitoring and evaluation. I t is important to recognize, therefore, the limits of OP within the complex network of citizen participationinPortoAlegre. Since its inception, the main objective o fthe OP has beenthe inclusiono f citizen input in capital investmentplanning. Over time, and as a result o f its success and permanence, but also as a response to its need to understand and scrutinize the municipal budget, the OP has developed a certain capacity to monitor the execution o f prioritized public works. However, this capacity i s weak and requires strengthening. On the other hand, other venues were created as major spaces o f civic engagement, both, before and after the OP was established. Although these other channels have specific purposes, such as the Public Policy Councils (which focus mainly on sector policy making) and the Programa de Governanqa Solidaria Local (Solidarity in Local Governance Program, which has a more holistic approach to local governance), there are clearly many overlaps and conflicts. It must therefore be recognized that the OP in Porto Alegre may be limited in its ability to expand to other areas o f participatory local governance, if it i s to first strengthen its current role. However, the municipality and various non-governmental stakeholders that have an active role in the city's various channels o f participation have yet to come up with a harmonizing framework for a more effective and efficient system o f civic engagement and social accountability inPorto Alegre. The econometric analysis conducted for this study isolated the fiscal and social impacts of OP in order to establish whether they were in fact caused by participatory budgeting. Based on econometric findings to date, it i s important for municipalities that are implementing OP to manage expectations with respect to anticipated impacts. Since the institutional reforms involved in the application o f OP continue to be relatively new, there may also be genuine potential for improving the application o f OP both inPorto Alegre and other cities inBrazil. Econometric analysis compared Brazilian municipalities that adopted OP to a group of non-OP municipalities that had the same opportunityto observe improvements in poverty, access to basic services, and fiscal performance before the advent of OP. The analysis controlled for the fact that in most municipalities where OP was implemented, the P T was a key political actor, either by winning the major's office in elections or being the second most important political force in a city. The analysis thus 101 Economic Sector Work accounted for whether the estimated impacts were the result o f OP and not the result o f a political process (including strong participation through informal channels o f voice) associated with substantial support of the PT. The analysis thus ensured that OP was the instrument that caused the impacts. The econometric analysis suggests that participatorybudgeting, as a mechanism to improve pro- poor capitalinvestments, has contributed to a reductionof povertyrates inthe municipalities where it was adopted prior to 1996. It is worth noting that this poverty impact occurred despite a reduction in GDP per capita in these municipalities, suggesting that the OP can contribute to a redistributive impact over the long run. To date, it seems that the impact on better access to water and sanitation i s positive for all OP municipalities. Current findings do not indicate consistent positive impacts of OP on fiscal performance. The second hypothesis tested using econometric analysis was the impact of the OP on fiscal performance, assuming that OP empowers the poor and, therefore, leads to more transparency in the way the public budget i s allocated, as well as to increased budget literacy. Participatory budgeting was not, however, found to have a consistent impact on fiscal performance. The impact o f OP on fiscal performance, as measured by the ratio o f operational expenditure to total expenditure, furthermore appeared to be negative. The absence o f positive impacts on fiscal performance associated with the OP can be explained by its exclusive focus on investment capital allocations and its weak connection with other fiscal management decisions, such as current expenditure allocations, expenditure execution, and revenue collection. A simpler analysis, but one consistent with the logic of treatment and control group evaluation, shed light on the specifics of OP in PortoAlegre. It compared Porto Alegre with two synthetic control groups, one comprised o f all state capitals and one comprised only o f non-OP state capitals. The findings o f this analysis were, however, inconclusive. Using available data, it cannot be said that Porto Alegre performedbetter than the two synthetic control groups interms o f livingconditions. With respect to fiscal performance, this comparative analysis suggested that it i s unlikely that the fiscal problem inPorto Alegre was actually connected to the OP. More likely, the city increased its operational expenses after 1996 for some other reason and was unable to solve the problem o f growing expenses even with the (relative) increase inrevenues observed after 2000. The above findings help identifythe potentials andlimitationsof OP.The findings on the poverty and fiscal impacts of OP in Brazil described in chapter 5 suggest that the OP i s a participatory mechanism with significant potential for achievingpro-poor distributive impacts, such as poverty reductionoutcomes, over the long run.Its ability to have a positive impact on fiscal performance i s less evident. This apparent limitation has important consequences for the sustainability o f pro-poor investment. The results o f the impact analysis confirm the overall findings o f this study: the OP in Porto Alegre constitutes a representative and credible conduit o f voice for the poor to inform the allocation o f public investments, but it is limited in its ability to enhance the fiscal performance o f the city. The cycle mismatch between the OP and the municipal budget and the narrow scope o f the OP on investment allocations prevent it from significantly enhancing budget literacy, accountability, the quality o f expenditures beyond investment allocation, revenue collection, or overall fiscal and service delivery performance in Porto Alegre. 6.2 MainRecommendations There are changes and improvements in OP methodology that can help overcome some of its limitations. Yet the OP's structural limits must be acknowledged by social actors and municipal authorities alike. Other participatory mechanisms are better suited to improving the accountability o f 102 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre certain stages o f the public expenditure cycle, such as sectoral councils (which can be used to ensure the accountability o f sectoral policies and service delivery). The following recommendations are intended to contribute to improving the performance and impacts o f the OP in Porto Alegre, as well as to involve other organizations and mechanisms in the establishment a system o f social accountability in which the OP would play a leading role. The recommendations are organized to respond to each o f the questions that originated this study. How can the "quality ofparticipation" in the OP be improved? Create incentives and remove barriers to make the OP more inclusive. Measures to address these challenges could include expanding the OP's focus to address issues o f strategic importance for the city as a whole, includingbudgetary oversight, municipal revenues, and current expenditures that affect the space for OP investments and would attract interest from the middle class and professional groups (such as entrepreneurs). Furthermore, participation could be made more convenient to attract those who currently do not participate, for instance, by making online voting on investment priorities available or reimbursing the financial cost o f participation borne by the poorest groups (e.g., the cost o fbus tickets for participants below a certain income level). Improving the effectiveness o f OP council meetings by reforming meeting procedures would also save participants' time. Finally, improving the training and education o f participants i s essential to fully realize the OP's potential as a mechanism for broadening citizen understanding o f the municipal budget process. Training needs should be identified as an integral part o f the OP process at each level o f participation (e.g., general, delegate, and councilor) and for each stage o f the OP cycle. Make OP representationmore legitimate and accountable to maintainits credibility.Simplifying its procedural rules and making them more accessible to OP participants would level the playing field among participants and prevent knowledge o f the rules becoming concentrated in only a handful o f OP representatives. Furthermore, establishing simple guidelines and formats for presentations by councilors and delegates would enhance the system o f reporting to constituencies and enhance downstream accountability. How can the interface between OP and municipalfiscal management be strengthened? Tackle the temporal mismatch between the OP prioritizationprocess and the municipal budget cycle. Moving the OP cycle one month forward to ensure consistency with the Organic Law of the Municipality would offer the best results in terns o f the quality o f information available during the prioritization process and would consequently enhance the effectiveness o f OP demands. An alternative solution would be to ensure that the executive arm o f the municipality provides a reasonably accurate revenue estimate for the period for which OP investment demands are prioritized. BetweenM a y and July, each region and sector should thus receive a more precise estimate o f the budgeted caps for capital expenditures inthe coming year. The articulationbetween municipal fiscal management and the OP could be further enhanced by expanding the OP cycle to two years for complex public works, while maintaining an annual time frame for OP investments that can be executed in a shorter time period. This step would increase the capacity o f OP participants to engage in the oversight o f public works by reducing the time pressure o f the current one-year cycle. Yet, it must be acknowledged that a possible tradeoff o f a bi-annual OP process i s a reduction inthe interface between the population and the local government. Introduce a mechanism in the OP cycle to allow for adjustments to the Investment Plan during budget execution,which would prevent the accumulation of a backlogof investmentsinthe event of 103 Economic Sector Work unforeseen decreases in revenue. This measure can be complemented with measures to strengthen municipal capacity in budget planning, such as collecting and publishing indicators on the accuracy o f revenues and expenditure for each budget cycle, as well as an indicator to measure the accuracy o f disbursement estimates for each item in the Investment Plan. These indicators can then be used as benchmarks which the OP Council can track. Improve fiscal and budget literacy and create spaces for overall fiscal policy dialogue in the OP cycle. These steps are essential for maximizing the potential of the OP to have a positive impact on fiscal management. This goal can be achieved through a set o f measures, including the incorporation of in-depth public discussion o f the overall fiscal policy o f the municipality and its budget execution at the beginning o f the OP cycle. This change should go hand in hand with providing basic training to OP councilors in various topics o f municipal finance, including Multi-Year Plans (PPA), Budgetary Guidelines Law (LDO), Annual Budget Law (LDO), bimonthly Summary Budget Execution Reports (RREO), and Fiscal Management Reports (RGF). These measures can be complemented by partnering with local universities and think tanks to assist the COP to undertake independent fiscal policy and budget analysis, together with developing the ObservaPoA Website into a system that offers statistical information to assess budget priorities and monitor OP impacts. How can the capacity of the OP to monitor budget execution be improved? Strengthen budget literacy in Porto Alegre, which is the precondition for enhancing the capacity of the OP to monitor budget execution. Currently, budget literacy i s limited and uneven among Porto Alegre citizens. Although the OP has contributed to a greater understanding o f the need for pro-poor public spending, it has not fulfilled its potential to improve budget literacy. T o raise awareness o f basic municipal finances and budgeting, the OP could make more frequent use o f local radio, emulating the experiences o f Peru, Ecuador, and Guatemala, as well as o f Anglophone African countries that have been piloted such dissemination strategies with the support o f the World Bank. The OP in Porto Alegre could also include inits Statute a supplement on fiscal and budget literacy, and ensure that the OP Notebook o f the Investment Planpublishes detailed information on municipal revenue and expenditures. Moreover, OP Regional Journals could include detailed technical and financial information on each project being implemented in the respective regions, allowing for comparison between planned activities and their execution, In addition, an overall evaluation o f the implementation o f all municipal programs could be included in the Booklet o f Municipal Accounts, using the performance indicators adopted by the new municipal management system. Expand the scope of the OP cycle to systematically includebudget oversight. The OP Council can use and publish simple indicators on its Website that compare overall budget planning data (e.g., amounts, time o f implementation, basic technical details) with budget execution data for different investments. These indicators could be developed to enable cross-regional comparisons and benchmarking. It i s also advisable that the municipality make available on its Website regular budget execution reports (i.e., the bimonthly Summary Budget Execution Reports, or RREO) in citizen-friendly formats. Finally, strengthen mechanisms for civil society monitoring and evaluation of municipalservices as a way to enhance overall social accountability. These functions may well be beyond the scope o f the OP, but they can be performed by other civil society organizations and spaces o f dialogue in which user organizations, civil society, and the municipality o f Porto Alegre participate, such as sector-based municipal public councils and the Forum o f Services. It i s thus advisable to strengthen the capacity o f these mechanisms to monitor and evaluate service delivery. Citizen feedback mechanisms, such as community score cards and citizen report cards, can also be introduced to strengthen client power in monitoring service performance. By establishing a citizens' charter o f municipal services that sets citizen- 104 Towards a More Inclusive and Effective Participatovy Budget in Porto Alegre friendly performance benchmarks for each service provided by the city, together with appropriate feedback channels, these mechanisms become powerful tools for enhancing the quality o f service delivery. How can a more coherent system of social accountability be established in the city, improving the articulation betweenparticipatory governancefora? First, it is essential to re-examine the relationship between the OP and Municipal Councils to ensure maximum synergies. Currently, the interface between existing institutions o f direct participation i s weak, undermining the possibility for a coherent system o f social accountability. Options to achieve better synergies could include the establishment o f an overarching framework for the different municipal councils and the COP, which would clarify formal and informal functions, remove overlapping roles in budget planning and oversight, and strengthen the interface between different municipal councils and the thematic assemblies o f the OP. These steps could be complemented by better integration o f the Urban Fora (spaces o f public debate on the city's master plan) with OP investment prioritization via a specific OP thematic group. Finally, strengthen the alignment of long-term sectoral development objectives with OP annual capital investmentplanning. This alignmentis essentialfor coherenceand efficiency. This goal could be achieved by using the City Conference to implement a well-structured strategic planning process among existing civil society-local government institutions. Such a process would establish clear long- term goals that would align policy making and budget planning. Moreover, the development o f a system o f social accountability in Porto Alegre would benefit from the establishment o f a permanent governance structure to oversee the formulation, implementation, and revision o f the city's strategic planningprocess. In addition to the City Conference, this structure should include a multi-stakeholder coordination committee, an executive secretariat, a technical team, and thematic task forces comprised o f representatives o f the municipal public councils, OP thematic groups, urban fora, the Solidarity Local Governance program, the municipal administration (direct and indirect), the local legislature, and the private sector. How can understanding of OP impacts be improved? The econometric analysis conducted for this report suggests that further analytical work is necessary to better understand the impacts of the OP and optimize its results. It is advisable that universities and think tanks inBrazil continue to assess the social and fiscal impacts o f the OP. The model and database developed for the impact analysis conducted by this study have the potential to be further explored and expanded to shed light on impact causalities surrounding the OP, as well as the value that the participatory process for poverty reduction, local governance, and development. 105 References Abers, Rebecca. 1998. From Clientelism to Cooperation: Local Government, Participatory Policy, and Civic OrganizationInPorto Alegre, Brazil. Politics and Society 26, no. 4:511-37. Acemoglu, Daron; Ticchi Davide and Vindigni, Andrea. 2006. Emergence and Persistence of Inefficient States. National Bureau of Economic Research (NBER) Working Paper, no. 12748. NBER, Cambridge, MA. Acemoglu, Daron & Robinson, James A., 2006. Persistence of Power, Elites and Institutions, NBER Working Papers 12108, National Bureau of Economic Research, Inc. Acioly Jr., Claudio; Herzog, AndrC; Sandino, Eduardo; Andrade and Victor Henry. 2003. Participatory Budgeting in Santo AndrC. Institute for Housing and Urban Development Studies IHS, Rotterdam, The Netherlands. Ackerman, John M. 2005. Social Accountability in the Public Sector: A Conceptual Discussion. Social Development Papers.World Bank, Washington, DC. Alfonsin, Betania. 2002. Mecanismos da Democracia Participativa. Grupo de Trabalho I.V Congress0 I da Cidade dePortoAlegre. PrefeituraMunicipalde Porto Alegre (PMPA). Mimeo. Alsop, Ruth, and Nina Heinsohn. 2005. Measuring Empowerment in Practice: Structuring Analysis and FramingIndicators. World Bank Policy ResearchWorking Paper. World Bank, Washington, DC. Arcidiicono, P. and et al. 2003. Sociedad civil y presupuesto participativo. Fondo Editorial Del Congreso del Peni, Lima. Avritzer, Leonardo. 1999. Public Deliberation at the Local Level: Participatory Budgeting in Brazil. http://www.ssc.wise.edd-wrighdavritzer .pdf.UniversidadeFederal de Minas Gerais, Brazil. . 2003. 0 Orqamento Participativo a Teoria Democritica: e Um Balango Critico. In Leonardo Avritzer and Zander Navarro, A Inova@o Democrhticano Brasil, Siio Paulo: Cortez. Avritzer, Leonardo, and Zander Navarro. 2003. A Inovaqiio Democrbtica no Brasil, o Orqamento Participativo. S9o Paulo: Cortez. Azevedo, Sergio de, and Fernandes, Rodrigo Barroso. 2005, Orqamento participativo: Construindo a democracia. Revan: Rio de Janeiro. Baierle, SCrgio. 2002. OP ao termidor? In J. Verle and L. Brunet, eds., Construindo um novo mundo (Porto Alegre: Editorial Guayi). Baierle, S. 2005. Estudo de Caso-Lutas em Porto Alegre: Entre a Revoluq9o Politica e o Transformismo. Rio de Janeiro: BASE. Baiocchi, Gianpaolo. 2005. Participation, Activism and Politics: The Porto Alegre Experiment and Deliberative Democratic Theory. Department o f Sociology. University of Wisconsin. Madison 106 Towards aMore Inclusive and Effective ParticipatoryBudget in Porto Alegre . 2001. From Militance to Citizenship: The Workers' Party, Civil Society, and the Politics of Participatory Governance inPorto Alegre. PhD. dissertation. Madison: Universityof Wisconsin. Baiocchi, G, P. Heller, S. Chauduri, and M.K. Silva. 2004. Evaluating Empowerment: Participatory BudgetinginBrazilian Municipalities.World Bank, Washington, DC. Baiocchi, G., S. Chauduri, P. Heller, and M.K. Silva. 2006. Evaluating Empowerment: Participatory Budgeting in Brazilian Municipalities. University of Massachusetts. http://www.people,umass,eduroaiocchi/papers/B-Report.pdf. Accessed April 15,2007. Beras, CCsar. 2006. Os Processos de Democratizaqzo do Estado: Orqamento Participative-Um Novo Equilibrio de Poder Entre Estado e Sociedade Civil. Ph.D. dissertation. Universidade Federal do Rio Grande do Sul (UFRGS). Biderman, Ciro. 2007. Estimating the Impact of Participatory Budget on Observed Outcomes. Work in Progress.World Bank, Washington, DC. Bonfigliolo, Angelo. 2003. Empowering the Poor: Local Governancefor Poverty Reduction. New York: UnitedNations Capital Development Fund. Brautigan, D.and et al. 2004. ThePeople's Budget? Politics, Power, Popular Participation and Pro-poor Economic Policy. New York: UnitedNations. Busato, Cezar, and Plinio Zalewski Vargas. 2004. Governanqa Solidhria Local: Fundamentos politicos da mudanqa emPorto Alegre. Municipalityof Porto Alegre. Busato, Cezar. 2005. Governanqa Solidhria Local-Desencadeando o Processo. Porto Alegre: PROCEMPA-PMPA. Cabannes, Yves. 2004. Financiamento Local e Orqamento Participativo. Documento base do Seminirio de Lanqamento da RedeUrbalN09.PrefeituraMunicipalPorto AlegrePGU-ALC, Porto Alegre. Cabbannes, Yves. 2004. Participatory Budgeting: A Significant Contribution to Participatory Democracy. Environment and Urbanization 16, no. 1 (April). Carmen, Malena, with Foster Reiner and Janmejay Singh. 2004. Social Accountability: An Introduction and Emerging Practice. Social DevelopmentPaper, no. 76. Washington, DC: World Bank. Chavez Miiioz, Daniel. 2002. A New Sustainable and Replicable Model of Participatory and Democratic Governance? La Hague. Unpublishedpaper presentedat local politics and democracy conference. CIDADE (Centro de Assessoria e Estudos Urbanos). 2002 and 2003. Quem C o pcblico do Orqamento Participativo. Porto Alegre: CIDADE. . 2003a. Quem C o pcblico do Orqamento Participativo.Porto Alegre: CIDADE. . 2003b. Olhar de mulher.Porto Alegre: CIDADE. Coelho, Vera S.P, and Marcos Nobre. 2004. Participaqzo e Deliberaqzo. Teoria Democrhtica e Experiencias Institucionais no B r a d Contemporfineo. SBo Paulo: Editora 34. 107 Economic Sector Work Cortes, Soraya M.V. 2001. Arcabouqo Institucional e ParticipaqBo em Conselhos de Saude e AssistCncia Social. In Desigualdades Socioespaciais nu Regiiio Metropolitana de Porto Alegre, (Porto Alegre: FundaqBode Economia e Estatistica). Dias, Mhrcia. 2002. Sob o sign0 da vontade popular: 0 Orqamento Participativo e o dilema da Cdmara Municipalde Porto Alegre. Belo Horizonte: Editora UFMGe Rio de Janeiro: IUPERJ. Darcy, F.C. dos Santos. 2005. Analise da evoluqBo das finanqas do municipio de Porto Alegre, 1988- 2004. PrefeituraMunicipalde Porto Alegre. Faria, L.A. 2002. A AdministraqBo Popular em Porto Alegre: Uma Experiencia Alternativa de Reforma do Estado na AmCrica Latina. Revista deEconomia Contemporhea, no. 6 (Rio de Janeiro: UFRJ). FaseRS; CoordenaqBo de Relaqdes Comunithrias (CRCRMPA); CIDADE; Abers, Rebecca. (UCLMSA). 2005. Pesquisa sobre a PopulaqBo que Participa da Discussso do Orqamento Publico Junto a PrefeituraMunicipal de Porto Alegre. Porto Alegre. Mimeo. Fedozzi, Luciano. 2000. 0 Poder da Aldeia: G&nese e Historia do Orqamento Participativo de Porto Alegre. Porto Alegre: Tom0 Editorial. . 2001. Orqamento Participativo: Reflexties sobre a Experihcia de Porto Alegre. 3a. Ediqgo. Porto Alegre: Tom0 Editorial . 2002. A InvenqBo Permanente da Democracia: Contribuiqdes A DiscussBo sobre Presente e o o Futuro do OP de Porto Alegre. In J.E. Verle and L. Brunet, Construindo um novo mundo (Porto Alegre: Editorial Guayi, 2002), 165-201. Fedozzi, Luciano. 2005. Perfil Social e Associativo dos Participantes do OP de Porto Alegre em 2005. Mimeo. . 2006a. Orqamento Participativo de Porto Alegre: Elementos para um Debate Conceitual. In Nilton Fischer and Jacqueline Moll, eds., Por uma nova esfera pziblica (Rio de Janeiro: Editorial Vozes). . 2006b. Rede da Democracia Participativa em Porto Alegre. Seminario do Observatorio de Porto Alegre (Novembrer 2006). Mimeo. . 2007. Perfil Social e Associativo: AvaliaqBo e Expectativa sobre o Futuro do Orqamento Participativo de Porto Alegre. Analise Histbrica de Dados. Observatorio D e Porto Alegre, Ufrgs E Tom0 Editorial. Mimeo. Furtado, Adriana; Costa, Beatriz Morem; Macedo, Carlos Eduardo; Germano, Liane Rose; Macerata, Marco Ant8nio; Silva, Marcia Elizabeth da; Rodrigues, Tdnia Regina Quintana and Bassani, ValCria D.S.2004. Mapada InclusBo e da ExclusBo Social de Porto Alegre. Observatorio da Cidade de Porto Alegre. http://www,observapoa.com.br. PrefeituraMunicipal de Porto Alegre. Porto Alegre. Genro, Tarso. 1994. 0 Controle P6blico NBo-estatal e o Orqamento Participativo. DEBATE, caderno do PT.PortoAlegre, no5: 11-13. Genro, T. and U. Souza. 1997. Orqamento Participativo-A Experihcia de Porto Alegre. la. ed. SBo Paulo: Editora FundaqBoPerseuAbramo. 108 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre Genro, Tarso. 2005. Resoluqces do Congress0 das Cidades-0 lugar de Todas as Coisas. Porto Alegre: PrefeituraMunicipalde Porto Alegre. Gret, Marion, and Yves Sintomer. 1994. Porto Alegre: L'espoir d'une autre Dhocratie. Paris: La DCcouverte. Horn. C. 1994.PortoAlegre: 0Desafio da Mudanqa. Porto Alegre: Editorial Ortiz. IBAM - Instituto Brasileiro de AdministraqSio Municipal. 1994. Democracia, Participaqio e Orqamento. Rio de Janeiro: DAM. IBGE (Instituto Brasileiro de Geografia e Estatisticd Brazilian Institute of Geography and Statistics). 2000. Rio de Janeiro: IBGE. Marquetti, Adalmir. 2002. 0 Oreamento Participativo como uma Politica Redistributiva em Porto Alegre. IEncontro de Economia Gaucha,May 16-17,2002, Marquetti, Adalmir. 2003. Democracia, Equidade, e EficiCncia: 0 Caso do OP em Porto Alegre. In Leonard Avritzer and Zander Navarro, eds., A Inovaqio Democrcitica no Brasil: 0 OrGamento Participativo (SHo Paulo: Cortez). Marquetti, Adalmir. 2003, ParticipaqHo e RedistribuiqSio: 0 Orqamento Participativo emPorto Alegre. In: Avritzer, Leonardo; Navarro, Zander. (org). A inovaqgo democratica no Brasil. led. SSio Paulo: Cortez Editora, v. 1; p. 129-156. Monitoramento do PGSL (Programa de Governanqa Solidaria Local), September2007. Internal PGSL management document of the Municipality of Porto Alegre. Nahra, Clicia M.L.2005. Os Conselhos Muncipais Gestores de Politicas Publicas em Porto Alegre: Um panorama. SMGLPMPA. Porto Alegre. www.observapoa.com.br. Navarro, Zander. 2000a. Uma analise do Orqamento Participativo: Sua ImplantaqSio e Desenvolvimento. InTarso Genro, ed.,Portoda Cidadania(Porto Alegre:Artes eOficios, 1997), 179-235. . 2000b. La Democracia Afinnativa y el Desarrollo Redistributivo: El Caso de Presupuesto Participativo de Porto Alegre, Brasil. InE. Jarquin and A. Caldera, Programas Sociales: Pobreza y Participacidn Ciudadana (Banco Interamericano de Desarrollo). Washington, DC. . 2003. 0 Orqamento Participativo de Porto Alegre, 1989-2002: UmConciso Comentario Critic0 InL.E.Avritzer andZ.Navarro,A InovaqioDemocrcitica noBrad (SiloPaulo: Cortez). Nufiez, Tarson. 2006. ParticipaqSio Popular e Combate iPobreza em Porto Alegre: Estudo de Caso do Projeto de Fortalecimento das Capacidades em GestSio Urbana na AmCrica Latina e o Caribe. Porto Alegre. PGU-ALC HabitatmTLNCities Alliance. Mimeo. Nufiez, Tarson, and Lucian0 Fedozzi. 1993. Quem C Quem no Orqamento Participativo: Interpretaqgo de Dados Quantitativos da PopulaqHoque Participa do OP em Porto Alegre. Mimeo. Olegario, Filho, B. Assis, CRCPMPA, CIDADE, and Gianpaolo Baiocchi. 1999. Quem C o Publico do Orqamento Participativo: Seu Perfil, Porque Participae o que Pensa do Processo.Porto Alegre. 109 Economic Sector Work Osbrio, Pedro Luis da Silveira. 2003. Comunicaqgo e Cidadania: A Contribuipgo da Administraqiio Popular de Porto Alegre para um Novo Modelo de Comunicaqgo Politica e Governamental. Dissertapgo de Mestrado emComunicaggo Social. Porto Alegre: UFRGS. Pereira, Brewer. 1981.A SociedadeEstatal e a Tecnoburocracia. S ~ Paulo: Editora Brasiliense. O Plummer, Janelle. 2000. Municipalities and Community Participation: A Source Book for Capacity Building. Virginia, USA: DFID (Department for International Development), GHK International, EartthscanPublications. Pont, Raul. 2000. Um Caminho Seguro para o Futuro da Cidade. In Resoluq6es do 111Congress0 du Cidade. Porto Alegre: PMPA. Prefeitura Municipal de Belo Horizonte. 2007. Manual Metodol6gico Instrumentos de Articulaqgo entre Planejamento Territorial e Orpamento Participativo. Belo Horizonte. http://www.pbh.gov.br/noticias/redeurbal9/manual_metqortugues-.pdf.Accessed June 2007. Prefeitura Municipal de Porto Alegre. 2007. A Cidade Cuidando da Cidade: Primeiro Caderno de Prestaqgo de Contas 2005/2005. Porto Alegre: Secretaria de Coordenaqgo Politica e Governanqa Local. .2004. Organic Law of the City of Porto Alegre. . "Governanpa Solidhria Local: Documento de Referencia." Secretaria Municipal de Coordenapgo Politica e Governanqa Local. Prefeitura Municipal de Porto Alegre. http://www.governancalocal.com.br. Accessed December 12,2007. Santos, Boaventura Souza. 1998. "Participatory Budgeting in Porto Alegre: Toward a Redistributive Democracy." Politics and Society (December). Serageldin, Mona. 2003. Assessment of Participatory Budgeting in Brazil. Graduate School of Design, Center for Urban Development Studies, HarvardUniversity,Boston.. Schneider, Aaron, and Marcelo Baquero. 2006. Get What You Want, Give What You Can: Embedded Public Finance inPorto Alegre. I D S Working Paper, no. 226. IDS, Brighton, England. Silva, Marcelo K. 2003. Associativismo e Participagiio Social no Orpamento Participativo: As Experiencias dos Municipios de Porto Alegre, Alvorada e Gravatai. Porto Alegre: EditoraUFRGS. Silva, G.P. and Carlos E.F. Carvalho. 2006. Referenciais Te6ricos para Desenvolver Instrumentos de Avaliaqgo do Orqamento Participativo. Nova Economiu 16, no. 3 (Belo Horizonte, Set/Dez): 423-5 1. Solidariedade. 2003, Caminhando para umMundo Novo. Petr6polis: EditoraVozes. Souza, Celina. 2005. Participatory Budgeting in Brazilian Cities: Limits and Possibilities in Building Democratic Institutions. University of Birmingham Working Paper, no. 28. University of Birmingham, United Kingdom. http://www.idd.bham.ac .uk/research/Projects/urban- governance/resources.htm. Accessed JadO8. Souza, Marcelo Lopes de. 2000. Orqamentos Participativos e sua Espacialidade: Uma Agenda de Pesquisa. TerraLivre 15 (Sgo Paulo): 39-58. 110 Towards a More Inclusive and Effective Participatory Budget in Porto Alegre Torres, A.R, and G. Grazia. 2003. Experiencias de Orqamento Participativo no Brasil, 1997-2000: Foro Nacional de Participacih Popular. Petrbpolis: Editora Vozes. Utzig, JosC Eduardo. 1996. Participatory Budgeting o f Porto Alegre: A Discussion in the Light of the Principle o f Democratic Legitimacy and of the Criterion of Governance Performance. http://www.worldbank.org/participation/s&pa/spathematic.htm.World Bank, Washington, DC. Waiselfiz et al. 2004. Nos Caminhos da InclusBo Social: A Rede de ParticipaqBo Popular de Porto Alegre. Brasilia: UNESCO. Wampler, Brian. 2004. Brazil's Participatory Budgeting: Decentralization, Participation, and Empowerment. World Bank, Washington, DC. Webb, Steven. 2004. Fiscal Responsibility Laws for Sub-national Discipline: The Latin American Experience. World Bank Policy ResearchWorking Paper 3309. World Bank, Washington, DC. Williamson, 0. 1979. Transaction Costs Economics: The Governance of Contractual Relations. Journal of Law and Economics 22:233-61. Williamson, 0. 2000. The New Institutional Economics: Taking Stock, Looking Ahead. Journal of Economic Literature 38595-613. World Bank. 2003. Enabling Environment for Civic Engagement in PRSCs. Social Development Note, no. 82. World Bank, Washington, DC. . 2004. World Development Report: Making Services Workfor Poor People. Washington, DC: Oxford University Press and The World Bank. . 2006. Brazil: Inputsfor a Strategy for Cities. Report #35749PR. World Bank, Washington, DC. Verle, JoBo E., and Lulciano Brunet.2002. Construindo um Novo Mundo: AvaliaqBo da Experisncia do Orqamento Participativo emPorto Alegre - Brasil. Porto Alegre: Ed. Guayi. Verle, JoZo. 2003. Mensagem do Prefeito. In Resoluq6es do I V Congress0 da Cidade (Porto Alegre: PMPA). Zamboni, Yves. 2007. Participatory Budgetingand Local Governance: An Evidence-Based Evaluationof Participatory Budgeting Experiences in Brazil. Working Paper (May 28, 2007). http://siteresources.worldbank,org/INTRANETSOCIALDEVELOPMENT/Resources/Zamboni.pdf. Accessed Jan/08. 111