CIUDADANÍA Y RENDICIÓN DE CUENTAS: APROXIMACIONES PARA UNA CONCEPTUALIZACIÓN

CIUDADANÍA Y RENDICIÓN DE CUENTAS: APROXIMACIONES PARA UNA CONCEPTUALIZACIÓN Marcus André Melo (UFPE) Seminario Internacional Hacia una política de rendición de cuentas en México Red por la Rendición de Cuentas RCC 22 y 23 de Agosto, 2011 Red Hotel Meliá Ciudad de México

Outline of presentation Theoretical preliminaries: the three pillars of accountability Political competition & Pluralism Information Institutional capacity Empirical evidence from Brazil and Latin America Does accountability matter?

Outline of presentation Theoretical preliminaries: the three pillars of accountability Empirical evidence from Brazil and Latin America Does accountability matter?

Three pillars of accountability: a Madisonian approach The three determinants of the effectiveness of accountability institutions and arrangements Political competition & pluralism Information Institutional capacity

Three pillars of accountability: a Madisonian approach The three determinants of the effectiveness of accountability institutions and arrangements Political pluralism & Incentives Information Institutional capacity

Three pillars of accountability: a Madisonian approach The three determinants of the effectiveness of accountability institutions and arrangements Political pluralism & incentives Information Institutional capacity

The Madisonian approach to institutional design But the great security against a gradual concentration of the several powers in the same department, consists in giving to those who administer each department the necessary constitutional means and personal motives to resist encroachments of the others. Ambition must be made to counteract ambition. La gran seguridad contra una concentración gradual de varios poderes en el mismo departamento consiste en dar a quienes administran cada departamento los necesarios medios constitucionales y motivos personales para resistir las invasiones de los otros Las ambiciones deben contrarrestar a las ambiciones.. If men were angels, no government would be necessary. If angels were to govern men, neither external nor internal controls on government would be necessary. In framing a government which is to be administered by men over men, the great difficulty lies in this: you must first enable the government to control the governed; and in the next place oblige it to control itself. A dependence on the people is, no doubt, the primary control on the government; but experience has taught mankind the necessity of auxiliary precautions. The Federalist 51 Si los hombres fueran ángeles, no sería necesario ningún gobierno. Si los ángeles gobernaran a los hombres, no sería necesario ningún control externo ni interno sobre el gobierno. Al enmarcar un gobierno que va a ser administrado por hombres sobre hombres, la gran dificultad reside aquí: debes primero permitir al gobierno controlar a los gobernados y después obligarle a controlarse a sí mismo. Una dependencia del pueblo es, sin duda, el control principal del gobierno, pero la experiencia ha enseñado a la humanidad la necesidad de precauciones auxiliares

Madison s Therapy If men were angels, no government would be necessary. Si los hombres fueran ángeles, no sería necesario ningún gobierno. If angels were to govern men, neither external nor internal controls on government would be necessary. Si los ángeles gobernaran a los hombres, no sería necesario ningún control externo ni interno sobre el gobierno

1. Political Competion & pluralism In modern language Madison suggested that the structure of incentives is crucial Pluralism is accountability enhancing: The more plural and competive an institutional environment the better the accountability outcomes Market analogy: just as economic monopolies, political monopolies are bad Competition generates political Incentives to hold public agents accountable, to denounce corruption and for credit claiming 11

2. Information Information assymetries are the engine of non-accountable governments Citizens become unable to sanction misbehavior and reward good performance Principal agents relationships Moral hazard post contractual opportunism Adverse selection we know little about politicians Information is more than transparency!

Institutional capacity Institutional capacity consists of several dimensions: Political Independence Resources Organizational autonomy But Information and Institutional Capacity without Incentives is useless!

Institutional capacity Capacity consists of several dimensions: Political Independence Resources Organizational autonomy But Information and Institutional Capacity without Incentives are useless!

Types of accountability mechanisms Vertical accountability Electoral sanctions and rewards Parliamentary Accountability Horizontal Accountability Supreme Audit Institutions (Contralorías and Tribunales de Cuentas) Fiscalías and Ministérios Públicos Ombudsman Independent judicial institutions 15

Types of accountability mechanisms Vertical Accountability (VA) Electoral sanctions and rewards Parliamentary Accountability Horizontal Accountability (HA) Supreme Audit Institutions (Contralorías and Tribunales de Cuentas) Fiscalías and Ministérios Públicos Ombudsman Independent judicial institutions But HA depends onva! 17

The Parliamentary Accountability Cycle Government responds to recomendations and takes actions Auditor geral examines general accoutns and make recomendations Legislative (PAC) publicizes misdeeds and decides about sanctions 18

The Judicial Accountability Cycle Government responds to recomendations and takes actions Tribunal de Cuentas examines general accounts and make recomendations Legislative publicizes misdeeds and decides about sanctions

EMPIRICAL ILLUSTRATIONS 1 Political pluralism and competition

Strength of Audit institutions in Latin America. 90,00 80,00 70,00 60,00 50,00 40,00 30,00 20,00 10,00 0,00 ARGENTINA BOLIVIA BRAZIL CHILE COSTA RICA MEXICO PERU VENEZUELA Source: International Budget Partnership (IBP)

.3.4.5.6.7 Effectiveness Effectiveness of Audit Institutions and Political Freedom Chile Brazil Colombia Costa Rica Nicaragua El Salvador Mexico Peru Argentina Ecuador 1 2 3 4 pol_freedom_ Source: Santiso (2009) and Freedom House.

0 5 10 15 months Celerity of Audit Reports and Political Freedom Mexico Paraguay Bolivia Ecuador Uruguay Chile Argentina Venezuela Colombia Costa Rica Peru Brazil Guatemala 1 2 3 4 pol_freedom_ Source: own elaboration; OECD database

Budget transparency selected countries Budget transparency and external control (selected countries) Indice de Transparência Orçamentária e Qualidade Institucional das ISCs 100 80 60 40 20 0 africa do sul brasil peru colombia mexico botswana costa rica kenia egito zambia guatemala bolivia honduras el salvador namibia argentina burkina faso nicaragua argelia tanzania gana angola equador nigeria marrocos

Note: The IBP index of strength of supreme audit institutions is based on a survey with the following questions: 111 Timeframe for release of the audited annual expenditures of national departments 114 Power to remove head of SAI from office 116 SAI discretion in law to undertake those audits it may wish to undertake 117 Power to determine the budget of the SAI 118 Availability in SAI of designated staff for undertaking audits of security sector 119 SAI mechanisms of communication to receive complaints and suggestions from public 120 Existence of legislative committee for scrutinizing audit reports 121 Availability of a report from executive on steps it has taken to address audit recommendations 122 Publication of report by SAI or legislature tracking actions taken by executive to address audit recommendations 123 Audit reports on security sector and other secret programs provided to the legislature (or relevant committee)

Institutional Veto Players over Budget in Latin America More than one Chamber in the Legislature has power over the budget Argentina Brazil Chile Unicameral or Just one Chamber has power over the budget Costa Rica Peru Meixo Venezuela

Audit Institutions and Budget Transparency in Latin America. 90,00 80,00 70,00 60,00 50,00 40,00 30,00 20,00 10,00 0,00 Audit Institutions Budget Transparency Source: IBP. Correlation: 66.61

Legislative Strength and Budget Transparency in Latin America. 80,00 70,00 60,00 50,00 40,00 30,00 20,00 10,00 0,00 Legislative Strength Budget Transparency Source: Data from IBP. Correlation: 40.36

. Performance of Contralorias and Tribunales de Cuentas in Latin America Country Score Independence Credibility Promptness Enforcement Brazil 0.63 0.88 0.42 0.24 1.00 Colombia 0.61 0.75 0.46 0.21 1.00 Chile 0.59 0.78 0.40 0.18 1.00 Costa Rica 0.49 0.66 0.48 0.16 0.67 Media regional 0.44 0.68 0.29 0.11 0.67 Nicaragua 0.42 0.78 0.20 0.03 0.67 El Salvador 0.40 0.53 0.08 0.00 1.00 México 0.36 0.59 0.38 0.12 0.33 Peru 0.32 0.78 0.12 0.04 0.33 Ecuador 0.28 0.66 0.14 0.00 0.33 Argentina 0.28 0.44 0.22 0.13 0.33

Budget Transparency (2006-2010) VENEZUELA PERU MEXICO COSTA RICA CHILE BRAZIL BOLIVIA ARGENTINA 0,00 10,00 20,00 30,00 40,00 50,00 60,00 70,00 80,00 Source: International Budget Partnership (IBP)

Country LA Index Budget Transparency OBI - Open Budget Index Global Integrity 2010 Global Integrity 2010 - Budget Process Oversight & Transparency TI- Corruption Perceptions Index Year 2003 2010 Various Various 2010 Argentina 44.1 56 87 77 2.9(105th) Bolivia 13 69 69 2.8(110th) Brazil 50.3 71 76 79 3.7(69th) Chile 61.7 72 77 63 7.2(21st) Colombia 44.3 61 68 71 3.5(78th) Costa Rica 48.8 47 79 68 5.3(41st) Ecuador 30.6 31 60 54 2.5(127th) El Salvador 40.3 37 3.6(76th) Guatemala 50 65 71 3.2(91th) Haiti 2.2(146th) Honduras 11 2.4(134th) Mexico 50.4 52 72 67 3.1(98th) Nicaragua 46.4 37 58 69 2.5(127th) Panama 3.6(73th) Paraguay Peru 44.6 65 81 88 3.5(78th) Republica Dominicana 3.0(101st) Uruguay 6.9(24th) Venezuela 34 61 56 2.0(164th)

0 5 10 15 20 25 30 35 40 Corruption increases in political monopolies. Mean no. irregularities 0 20 40 60 80 100 Diferença (%) entre os 2 cand.prefeito + votados (1.t), 2000 predicted number of events Fitted values 32

0 5 Median spline 10 15 0 20 40 60 80 100 Diferença (%) entre os 2 cand.prefeito + votados (1.t), 2000 33

Judicial Independence in LA Uruguay Chile Brazil Costa Rica Dom. Rep Mexico Colombia Guatemala El Salvador Panama Peru Honduras Ecuador Argentina Bolivia Nicaragua Paraguay Venezuela De jure Judicial independence 0,00 1,00 2,00 3,00 4,00 5,00 6,00

0 Density.1.2.3 Density 0 2 4 6 0 5 10 15 Número de Partidos Efetivos, 2000 0.2.4.6.8 1 % de Cadeiras do Partido do Prefeito, 2000 35

EMPIRICAL ILLUSTRATIONS 2 Information and corruption

. % Media owned by politicians in Brazilian states Roraima Rio Grande do Norte Maranhão Goiás Tocantins Alagoas Piauí Bahia Ceará Sergipe Paraíba Amapá Amazonas Paraná Pará Acre Rondônia Mato Grosso Minas Gerais Rio de Janeiro Santa Catarina São Paulo Pernambuco Espírito Santo Distrito Federal Mato Grosso do Sul Rio Grande do Sul 0.00 0.20 0.40 0.60 0.80 1.00 1.20

Econometric evidence that... The more the media is controlled by state elites, the more accumulation of wealth by politicians Corrupt politicians reduce their chances of reelection by 19% when the audit offices release information about corruption

EMPIRICAL ILLUSTRATIONS 3 Institutional Capacity and Corruption

Econometric evidence for independence Tribunales de Cuentas with senior auditors recruited meritocratically in their board sanction irregularities at a much higher rate (20%) than TCs without such auditors across the Bazilian states.

Institutional capacity to sanction Num. Name Description Source.1 Regulatory Agencies Regulatory Governance Index. Measures governance of state and federal reg. agencies in Brazil based on survey data. States with no agency at the time of the studied were set at 0.53 (avg. of other states). Data for 2004/2005. 2. Judiciary * Index composed of three variables using principal component analysis: i) an efficiency index calculated through nonparametric efficiency frontiers; ii) ratio of number of cases tried over cases opened. iii) number of new cases opened per 100,000 inhabitants. 3 Public Prosecutors * Index composed of three variables using principal component analysis: i) Expenditures with public prosecutors per resident; ii) Number of prosecutors per 100,000 residents. iii) Number of staff per 100,000 residents.. 4 Audit Office An index of the level of activity in each state s Audit Office (TCE). 5 National Justice Council (CNJ) Number of procedures initiated in each state by the CNJ (agency that serves as a watchdog over the Judiciary) divided by state GDP (divided by 100,000). Data for 2006. Correa, Melo, Mueller and Pereira (2006). i) Swengberger, 2006, pg 79. ii) Ministério da Justiça. 2004. Diagnóstico do Poder Judiciário. iii) CNJ Indicadores Estat. da Justiça Estadual 2005, pg.278, 2004. Sadek and Lima (2006). Melo and Pereira (2006). Corregedoria Nacional de Justiça. 2006 6 Media Percent of all media concession in each state are not in the hands of politicians. [1] Santos, S. S. e Capparelli. 2005 7 Civic Community index An index of Civic Community in the states constructed by principal component analysis using (i) voter turnout (1990-2006), (ii) voto de legenda (1990-2006), and (iii) nonprofit sectors workers per capita (ABONG-IBGE 2002 study). Timothy Powers. 41

Accountability and C & B in the Brazilian States 1998-2002

Accountability and C & B in Brazilian states 2003-2006

Econometric evidence that accountability matters... The stronger the institutional checks on the state governments the higher the share of education and health expenditures in state budgets the less inneficiency in state administration The smaller the fiscal deficits

Summing up Pluralism generates incentives for control over politicians The more information is produced the stronger the vertical accountability Here civil society plays a key role Generating institutional capacity is a precondition for accountability but the incentives for capacity only exist where there is political pluralism and competition

Summing up Generating institutional capacity is a precondition for accountability but the incentives for capacity only exist where there is political pluralism and competition Horizontal Accountability depends crucially on Vertical Accountability!

Thank you very much!

Findings Research Questions Methodological issues Lit Review Context Empirical strategy Findings A non-linear effect of political competition: at low and high levels it generates incentives for corruption Corruption Political competition 48

Num. Name Description Source 1 Regulatory Agencies Regulatory Governance Index. Measures governance of state and federal reg. agencies in Brazil based on survey data. States with no agency at the time of the studied were set at 0.53 (avg. of other states). Data for 2004/2005. 2 Judiciary * Index composed of three variables using principal component analysis: i) an efficiency index calculated through nonparametric efficiency frontiers; ii) ratio of number of cases tried over cases opened. iii) number of new cases opened per 100,000 inhabitants. 3 Public Prosecutors * Index composed of three variables using principal component analysis: i) Expenditures with public prosecutors per resident; ii) Number of prosecutors per 100,000 residents. iii) Number of staff per 100,000 residents. 4 Audit Office An index of the level of activity in each state s Audit Office (TCE). 5 National Justice Council (CNJ) Number of procedures initiated in each state by the CNJ (agency that serves as a watchdog over the Judiciary) divided by state GDP (divided by 100,000). Data for 2006. Correa, Melo, Mueller and Pereira (2006). i) Swengberger, 2006, pg 79. ii) Ministério da Justiça. 2004. Diagnóstico do Poder Judiciário. iii) CNJ Indicadores Estat. da Justiça Estadual 2005, pg.278, 2004. Sadek and Lima (2006). Melo and Pereira (2006). Corregedoria Nacional de Justiça. 2006 6 Media Percent of all media concession in each state are not in the hands of politicians. [1] Santos, S. S. e Capparelli. 2005 7 Civic Community index Checks & Balances Index An index of Civic Community in the states constructed by principal component analysis using (i) voter turnout (1990-2006), (ii) voto de legenda (1990-2006), and (iii) nonprofit sectors workers per capita (ABONG-IBGE 2002 study). Timothy Powers.

Methodological issues Research Questions Methodological issues Lit Review Context Empirical strategy Findings Unobservable nature of corruption data Limitations of perception-based data CPI -Transparency International Potential endogeneity in data generation process By focusing on a single country we are able to control for a number of institutional features that are constant electoral rules unobservable heterogeneity The advantages of randomized audits 51

Operationalization of variables Research Questions Methodological issues Lit Review Context Empirical strategy Findings Dependent variable number of serious irregularities detected in CGU s municipal audits for 2000-2004 Independent variables Political competition: Margin of victory over second most voted candidate Margin of victory squared to capture non-linear effects % seats held by mayor s coalition % seats held by mayor s party Incidence fo divided government Institutional Checks Checks and balance index for states (Alston, Melo, Mueller and Pereira 2010) Access to the judicial system (municipality is the seat of a judicial district (comarca)? Controls Size of government budget as % of municipal GDP Federal capital transfers - Log population Municipal expenditure in education percapita Municipal Human Development Index Municipal Percapita income Vertical fiscal imbalance transfers as % of expenditures Partisan affiliation of mayors Mayor same party as presidential party 52

0 0 5 10 20 30 40 Quantiles of predicted number of events 10 15 20 0 0.2.4.6.8 Quantiles of predicted number of events 10 20 30 40 1 0.25.5.75 1 Fraction of the data 0.25.5.75 1 Fraction of the data 0.25.5.75 1 Fraction of the data 0.25.5.75 1 Fraction of the data 53

0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 40 0.25.5.75 1 Fraction of the data 0.25.5.75 1 Fraction of the data 54

Findings Research Questions Methodological issues Lit Review Context Empirical strategy Findings Variables pertaining to political competition are statistically significant and have the expected significant in most models Fiscal and economic variables behave as expected in most models They are robust to different model specifications They also have real social and economic importance A non-linear effect of political competition: at low and high levels it generates incentives for corruption 55

Size of Public Sector The determinants of Corruption Dependent variable: number of irregularities in education expenditures Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 1.400*** 1.37*** 1.157*** 2.155*** 1.157*** (.424) (.42) (.225 ) (.334) (.225) Percapita income Federal transfers percapita HDI2000-5.97*** (.56) % seats coalition % seats.69*** mayor s party (.32) -4.02e-09 ( 9.47e-08).008*** (.003) 6.70e-10 (9.61e-08) -1.88e-07*** (7.71e-08).009*** (.002) -1.75e-07** (8.06e-08).008*** (.002) -1.87e-07*** (7.66e-08).009*** (.002) -.004*** (.0005) 1.29e-08 (8.93e-08) Margin of Victory -.015** (.007) -.017** (.008) -.014* (.008) -.0151** (.007) -.0175*** (.008) -.015 ** (.007) -.018** (.008) Margin of Victory 2 0.0001 (0.0001).0001.0001.0001 (.0001) Divided Govern -.22*** (.10) Town with District Court Checks and -.96*** -.92*** Balances (.26) (.26).00014* (.0001) -1.270*** (.225).0002 *q (.00012) -.223* (.118).0001** (.0001) -.170 (.115) -1.240*** (.225).0002* (.0001) -.170* (.100) Log Population.32*** (.04) Edu expendit.0007**.0003.26*** (.05).0008* (.0003).25*** (.05) 0007** (.0003).296*** (.051).253***.064.243*** (.062).33* (.052).0008** (.0003) 56

The Madisonian solution But the great security against a gradual concentration of the several powers in the same department, consists in giving to those who administer each department the necessary constitutional means and personal motives to resist encroachments of the others. The provision for defense must in this, as in all other cases, be made commensurate to the danger of attack. Ambition must be made to counteract ambition. It may be a reflection on human nature, that such devices should be necessary to control the abuses of government. If men were angels, no government would be necessary. If angels were to govern men, neither external nor internal controls on government would be necessary. In framing a government which is to be administered by men over men, the great difficulty lies in this: you must first enable the government to control the governed; and in the next place oblige it to control itself. A dependence on the people is, no doubt, the primary control on the government; but experience has taught mankind the necessity of auxiliary precautions.