Internal transparency: compliance and punishment in the Brazilian municipal budget process

Abstract

Much has been written about external transparency in the public sector. However, internal transparency - the inter and intragovernmental disclosure of information - is commonly overlooked as an object of study. Underestimating the importance of internal transparency is surprising, as it is a determining factor for the coordination of public policies and the quality of democratic governance. This paper analyzes the short-and-long-term costs of failing to comply with internal transparency requirements linked to municipal budgets in general and education spending in particular. In this study, we employ mixed methods, quantitatively analyzing the short-term incentives to comply with transparency requirements for education finance and, qualitatively, the future consequences of noncompliance. The results show that municipalities receiving the most voluntary transfers are more likely to meet internal transparency requirements. However, the costs of not fulfilling such requirements are low and easily reversible in the short-term. In the long-term, municipal officials who fail to comply with budget regulation, including internal transparency statutes, can face more severe consequences, such as the possibility of being barred from office. However, our results show relatively few instances of punishment, and, on average, punishments are only levied after ten years of legal processes. Overall, our results show how, as the lynchpin agents of budgetary accountability and enforcement for Brazil’s 5570 municipalities, the courts of audit (Tribunais de Contas) are important but ineffective.

Keywords:
public transparency; intergovernamental transfers; educational budget; Audit Courts

Resumo

Muito tem sido escrito a respeito da transparência externa no setor público. No entanto, a transparência interna - a divulgação inter- e intra-governamental de informações - é comumente negligenciada como objeto de estudo. Subestimar a transparência interna é surpreendente, pois ela é um fator determinante para a coordenação de políticas públicas e para uma governança democrática de qualidade. O presente artigo analisa os custos de curto e longo prazos do não cumprimento de requisitos de transparência interna vinculados aos orçamentos municipais em geral e aos gastos com educação em particular. Neste estudo, utilizamos métodos mistos, analisando quantitativamente os incentivos de curto prazo para cumprir os requisitos de transparência do financiamento da educação e, qualitativamente, as consequências futuras decorrentes do seu descumprimento. Os resultados revelam que municípios que recebem maior volume de transferências voluntárias têm maior probabilidade de atender aos requisitos de transparência interna; contudo, os custos de não cumprir tais requisitos são baixos e facilmente reversíveis a curto prazo. No longo prazo, os prefeitos que não cumprirem a regulação orçamentária, incluindo a legislação para a transparência interna, podem enfrentar consequências mais severas, como a possibilidade de serem futuramente impedidos de exercer seus cargos. No entanto, nossos resultados mostram relativamente poucos casos de punição e que, em média, essas punições só são aplicadas após dez anos de processos judiciais. De modo geral, demonstramos que os Tribunais de Contas - principais agentes para a accountability orçamentária dos 5.570 municípios brasileiros - são tão importantes quanto ineficazes para a transparência interna.

Palavras-chave:
transparência pública; transferências intergovernamentais; orçamento educacional; Tribunais de Contas

Resumen

Mucho se ha escrito sobre la transparencia externa en el sector público. Sin embargo, la transparencia interna -divulgación intra e intergubernamental de información- se suele pasar por alto como objeto de estudio. Subestimar la transparencia interna es sorprendente, ya que es un factor determinante para la coordinación de las políticas públicas y para una gobernabilidad democrática de calidad. Este artículo analiza los costos a corto y largo plazo del incumplimiento de los requisitos de transparencia interna, vinculados a los presupuestos municipales en general y al gasto educativo en particular. En este estudio, utilizamos métodos mixtos, analizando cuantitativamente los incentivos a corto plazo para cumplir con los requisitos de transparencia del financiamiento de la educación y, cualitativamente, las consecuencias futuras del incumplimiento. Los resultados muestran que los municipios que reciben mayores transferencias voluntarias tienen más probabilidades de cumplir con los requisitos de transparencia interna; sin embargo, los costos de incumplir dichos requisitos son bajos y fácilmente reversibles a corto plazo. A largo plazo, los alcaldes que no cumplan con la regulación presupuestaria, incluyendo la legislación de transparencia interna, podrían enfrentar consecuencias más severas, como la posibilidad de verse impedidos de ejercer sus cargos en el futuro. Sin embargo, nuestros resultados muestran relativamente pocos casos de sanción y, en promedio, estas penalidades solo se aplican después de diez años de procesos judiciales. En general, demostramos que los tribunales de cuentas, principales agentes del accountability presupuestario de los 5.570 municipios brasileños, son tan importantes como ineficaces para la transparencia interna.

Palabras clave:
transparencia pública; transferencias intergubernamentales; presupuesto de educación; tribunales de cuentas

1. INTRODUCTION

Transparency and information-sharing are fundamental to the effective operation of organizations (Yang & Maxwell, 2011). Yet nearly all analyses of government have focused on external transparency, with most trained on processes related to accountability and participation (Michener, 2019). This dominant focus has distracted us from the importance of internal transparency.¹ The sharing of information within government is a prerequisite of horizontal accountability among governmental actors and essential for effective policy coordination and collaboration (O’Donnell, 1998; Pinho & Sacramento, 2009; Rocha, 2013). The growing realization of its importance has given rise to new initiatives in Brazil, such as Decree n. 10,046 (2019), which creates obligations and incentives for federal agencies to share databases with each other. Such initiatives respond to the notion that deficits of internal transparency and information sharing within one or more governmental units can have perverse effects on the operational soundness of the state as a whole (Santos & Rover, 2019).

This article looks at the extent to which Brazilian municipalities abide by intragovernmental transparency obligations and to what degree they are brought to account for compliance failures. In particular, we analyze the short-and-long-term incentives of complying with federal laws on the disclosure of information regarding budgetary expenditures, especially in the realm of education budgeting.

Brazil is an unusual case of federalism, as it combines high levels of political autonomy with strict centralized control of public finances (Soares & Melo, 2016). We use education as an object of analysis because, on the one hand, states and municipalities exert significant autonomy in administrating basic education and, on the other, educational finance contains a number of internally linked obligations, including redistribution policies and the intragovernmental pooling of resources. In this sense, basic education in Brazil occupies the crossroads of federalism, furnishing important lessons on the challenges of internal transparency policies and, therein, incentives for compliance.

As stipulated by law - the Fiscal Responsibility Law (Lei Complementar no 101, 2000) and the Access to Information Law (Lei no 12.527, 2011) - all local governments must send budgetary expenditures to the federal government and publish this information online. On the books, receiving transfers from the Federal Ministry of Education and associated institutions is conditional upon submitting and publishing this information. Furthermore, every municipality and state government must have its budgetary expenditures approved by state or municipal Audit Courts (Tribunais de Contas). Notwithstanding these requirements, many municipalities disclose information late or fail to do so at all, which should raise concerns surrounding intergovernmental transparency in Brazil.

In 2012, the central government launched an electronic system referred to as CAUC (Auxiliary System for Information on Voluntary Transfers) to consolidate municipal data pertaining to transfers from the central government. Based on data from the CAUC, we analyze incentives in the ‘short term’ to comply with statutory transparency requirements - a condition to receive funds from the central government. Using a linear regression with fixed effects, we estimate the relationship between the volume of voluntary transfers and compliance with the transparency of budgetary expenditure conditionalities. We find that municipalities receiving larger voluntary transfers tend to exhibit better compliance with obligations.

We also find that, despite the incentives generated by conditionalities, the ‘costs’² of not complying are low in the short term. The federal government will halt transfers in response to noncompliance, but municipalities have several means of reverting this situation and only tend to comply with transparency conditionalities when they seek federal resources.

In order to confirm the long-term costs of noncompliance with internal transparency norms, we analyze data from the 2016 municipal elections and the supplementary elections of 2017 and 2018, provided by the Superior Electoral Tribunal (Tribunal Superior Eleitoral - TSE). Given the possibility of being barred from holding office and corresponding reputational damage, the long-term costs of not complying with transparency obligations are more severe. The Audit Courts³ assume a significant role in enforcing transparency obligations in the long term. The work of the Audit Courts is sent to the State Legislative Assembly, where approval of a municipality’s accounts is decided. The Audit Courts also exercise a corrective role when they encounter irregularities in the use of public funds. In such an eventuality, the Audit Courts may dispatch cases to the Public Prosecutor (Ministério Público) for analysis or prosecution.

This study differs from a typical transparency audit in that we do not analyze the disclosure of information to the public. In sending data on budgetary expenditures to the federal government, municipalities are engaging in indirect disclosure, as it is the federal government that will render these data publicly transparent. In this sense, internal transparency implies accountability concerns that emanate not only from civil society but also from within government (Pinho & Sacramento, 2009). Although the comparison of internal versus external transparency is not the object of this article, it is important to note that the costs of not complying with internal transparency obligations should be greater than simply failing to post information intended for the benefit of citizens alone.

Whereas for short-term costs we quantitatively estimate whether the value of transfers is associated with greater internal transparency, for long-term costs we utilize process tracing methods (Lima, 2017) to cast light on what ultimately happens to infractors. Here, we include interview data from discussions undertaken with two former administrators of the Ministry of Education⁴ in order to untangle the complex accountability processes involved in the disclosure of budgetary expenditure information. It is important to note that the interviews do not represent the source of our conclusions, but merely help to support and illustrate the facts presented.

This article is divided into five sections, including this introduction. In the second section, we present the context of voluntary transfers in Brazil and the process of approving municipal expenditures; in the third and fourth section we analyze the long-term costs of not complying with internal transparency. The fifth section concludes.

2. CONTEXT

2.1. Intergovernmental Transfers and the Importance of Voluntary Transfers

The Brazilian federation is distinguished by decentralized tax attributions among the states and municipalities and its complex system of intergovernmental transfers (Prado, 2001). As Silva (2005) contends, one of the central elements of coordinating a decentralized federal system is the implementation of a stable, transparent system of transfers.

Intergovernmental transfers are divided into four types. First, constitutional transfers represent a compulsory responsibility of the federal government and include the Municipal Participation Fund (Fundo de Participação dos Municípios - FPM) and the Fundeb, a pool of funds from all three levels of government. Second, legal transfers are regulated by specific laws and can include conditional transfers (when transferred from fund to fund, for example) or unconditional transfers (originating in petroleum royalties, for example). Third are transfers from the Unified Health System (Sistema Único de Saúde - SUS), which finance the operation of the health system. Fourth, voluntary transfers by the central government represent transfers that are discretionary or negotiated (Soares & Melo, 2016).

In the current article we analyze this fourth category, voluntary transfers by the central government to municipalities. These are noncompulsory transfer agreements between the central government and local governments. Voluntary transfers enable local (state and municipal) governments to invest more heavily in sectoral policies and the quality of public policies (Soares & Melo, 2016). Unsurprisingly, these funds are highly prized by local governments for their potential electoral and civic benefits.

According to article 25 of the Fiscal Responsibility Law (LRF):

[...] voluntary transfers are understood as the delivery of current resources or capital to another federal entity, by reason of cooperation, help or financial assistance, and are not determined by obligations germane to constitutional, legal, or the Unified Health System [transfers] (Lei Complementar no 101, 2000, authors’ translation).

According to Baião, Cunha, and Souza (2017), in 2010 voluntary transfers represented 6% of total municipal receipts. According to 2017 data from the Accounting and Fiscal System of the Brazilian Public Service (Sistema de Informações Contábeis e Fiscais do Setor Público Brasileiro - SICONFI), central government transfers to municipalities represented, on average, 2.5% of total municipal receipts. This proportion is slightly above the average amount of transfers received by municipalities via transfer agreements with the states, which correspond to 2.1% of total municipal receipts.⁵

Despite the apparently low value municipalities receive via voluntary transfers, it is important to recognize the degree of discretion implied in the management of these funds compared to other types of transfers. The latitude that local governments have in managing these funds provides them with a significant source of resources for strategic spending (Bijos, 2013). These funds contribute importantly to augmenting the acquisitive capacity of municipalities, especially during periods of recession, fiscal rigidity, and low tax collection (Soares & Melo, 2016). Voluntary transfers have been extensively studied; diverse sources of empirical evidence show that municipalities distribute resources in heterogeneous manners, (Brollo & Nannicini, 2012; Meireles, 2019; Moutinho, 2016), with diverse impacts (Bertholini, Pereira & Renno, 2018; Sakurai, 2013).

2.2. Legal Requirements for Voluntary Transfers

The operation of internal transparency in Brazil places an obligation on municipal governments to send their budgetary information to both the federal government and the Audit Courts (state or municipal). Confirmation that a municipality has complied with these transparency requirements precedes the release of voluntary transfers by the central government. Rules surrounding these requirements are found in several laws and, on the whole, the system is surprisingly complex. Due to this complexity, a system to track voluntary transfers, called the CAUC, was launched in 2012 to keep track of transfers and how they are spent. The CAUC contains municipal disclosures and corresponding balance of budgetary expenditures online, as obligated by the federal government’s rules on voluntary disclosures (Fraga, 2016). Local governments update the system daily (called ‘overnight’), which is compiled from diverse databases, such as the SIOPE database on education spending (Sistema de Informações sobre Orçamentos Públicos em Educação). Currently, the CAUC compiles information on and monitors fourteen municipal spending requirements and fifteen from the states and the Federal District.⁶

One of the CAUC’s principal requirements refers to expenditures on education. As established by article 212 of the Brazilian Constitution, all states must apply - at a minimum - 25% of their budget (from taxes and transfers) to public education. The National Fund for the Development of Education (Fundo Nacional de Desenvolvimento em Educação - FNDE), an agency of the Ministry of Education, verifies conformity with this requirement, and education spending data is entered into the SIOPE database (SIOPE, 2019). Municipalities have until the 30^th of April to enter this information, and states until the 31^st of May, proving their compliance with minimal constitutional spending requirements on education. The day after inputting this information, data on educational spending is rendered transparent on the SIOPE database.

The Fiscal Responsibility Law (LRF) furnishes one means of generating compliance with internal transparency provisions. The LRF, which contains numerous transparency provisions, stipulates that inter-governmental transfers of resources will be blocked if requirements are not met. However, the LRF makes an exception for resources sent to the areas of health, education, and social assistance; these resources cannot be blocked (Bijos, 2013; Lei Complementar n^o 101, 2000). States and municipalities that fail to comply with the CAUC are legally barred from celebrating new transfer agreements and receiving additional voluntary transfers. In this way, the central government’s ability to block voluntary transfers based on the LRF is central to an analysis of the costs and incentives of internal transparency.

However, the LRF does not automatically block voluntary transfers if requirements are not met. As illustrated by Figure 1, in addition to sending information to the federal government through the SIOPE, municipalities must have their accounts approved by the State Audit Courts. In practice, it is common that municipal governments send data on the execution of their budgets to the Audit Courts without reporting them to the federal government. Despite the CAUC being updated on a daily basis, in many cases noncompliance does not translate into real consequences if the offending municipality’s accounts are approved by the Audit Courts. The blocking mechanism of the LRF can therefore have no effect if the Audit Courts do give their assent.

We observe that the Audit Courts effectively decide whether to block voluntary transfers or not. The states of São Paulo and Minas Gerais illustrate Audit Courts that interpret internal transparency obligations differently from the Federal Treasury. São Paulo, for example, currently considers the pension of retired education workers to be part of the 25% constitutional minimum spending threshold dedicated to public education.⁷ Minas Gerais, on the other hand, created a ‘Management adjustment agreement’ (Termo de ajuste de gestão), with which the Audit Courts accept that a municipality does not comply with the 25% minimum as long as it eventually adjusts the irregularity. According to Official 2 interviewed for this study, “there are 27 Audit Courts, each one with its own interpretation”. The interviewee also described the role of the Audit Courts as, “the only institution that has a say over voluntary transfers”.

In relation to complying with the 25% minimum that must be spent on public education, the municipalities that do not fill in the SIOPE can prove their compliance with the constitutional minimum by means of a certificate from the Audit Courts. In our interview with managers, it was summarized in the following terms:

[...] let’s suppose that one of these items on the CAUC is a question of complying with the 25% constitutional minimum in public education and the municipality did not comply. In the records of the SIOPE, the municipality only spent 24.99%. In this case, the municipality can be noncompliant with the CAUC and still receive the transfer, because what is more important - above the SIOPE and above the analysis of the federal government - are the Audit Courts, which constitutionally analyze the municipality’s accounts and have the power to say that the 25% was applied. So that municipality can be ostensibly inadequate by the CAUC because of noncompliance with the 25%, but it wields an approval from the Audit Courts (Official 1, authors’ translation).

If the proof of complying with the Audit Courts is sufficient to unblock transfers, that means that the decisions and analyses of the Audit Courts trump all other statutory obligations (in particular - the obligation of internal transparency, as determined by the Fiscal Responsibility Law). Therefore, in order to understand the short-term costs of not complying with the exigencies of internal transparency, it is imperative to analyze the processes that lead to the approval of a municipality’s accounts by the Audit Courts.

2.3. The approval process of the State Audit Courts (TCEs)

Article 31 of the Federal Constitution establishes that the approval of accounts is the responsibility of the legislative power. The legislature acts based on a technical report from the municipal, state, or federal Audit Courts. Both the Fiscal Responsibility Law and the Access to Information Law point to the administrative autonomy of the Audit Courts as the means of maintaining sustainable and transparent accounts. The primary role of the Audit Courts is to assist the legislatures with an evaluation and decision on the legality of accounts, which the legislature is free to follow or ignore (Santana, Libonati, Vasconcelos & Slomski, 2009).

In this sense, the Audit Courts are entities encharged with the legal and institutional responsibility to promote transparency and accountability (O’Donnell, 1998; Rocha, 2013). Although Brazilian legislation does not actually use the term accountability to describe the role of the Audit Courts, the Courts serve as the lynchpin institution in terms of horizontal accountability, enforcing budgetary and financial governance within their jurisdictions. The Audit Courts are also responsible for evaluating how budgetary and financial governance affect the economic and social development of those jurisdictions under their authority. In short, they represent the guarantee that governments exercise control over themselves and provide the information that civic actors need to demand vertical accountability.

Each Audit Court is responsible for supervising municipal budgets, which are then ‘judged’ by legislative assemblies. Audit courts share some rules and procedures, but each one tends to have its own particularities. The timeline for ‘judging’ a municipality’s accounts is emblematic of these particularities. The process begins with municipalities presenting their balances and annual accounts, but some states - such as São Paulo, Paraná e Paraíba - impose a March 31^st deadline for the presentation of these accounts, whereas others give up to 180 days after the end of a fiscal year. Such is the case of Rio de Janeiro.

The evaluation of accounts by the Audit Courts is then presented to the local or state assembly for approval (Lopes, Pederneiras & Silva, 2009). The evaluation of accounts also varies in terms of its timeline. The majority can emit a recommendation to approve or reject accounts until the last day of the next year that those accounts were submitted. Such is the case of São Paulo e Rio de Janeiro. Others, such as Paraná and Minas Gerais, however, have until one year after accounts have been submitted to issue their evaluation and recommendation.

It is important to observe, however, that the legislative branch is ultimately responsible for approving or rejecting accounts. The recommendation emitted by the Audit Courts will prevail unless 2/3 of legislators vote against it, and any rejection must be accompanied by a clear justification (Andrada & Barros, 2010). Technical committees in the assemblies are also responsible for preparing briefs on how to vote. Yet despite a few exceptions - such as Rio de Janeiro’s legislature, which approved the accounts in 2016 against the recommendations of the Audit Court⁸ -- the evaluations and decisions of the Audit Courts are routinely followed by legislatures. What corroborates the generally amenable disposition of legislatures towards the Audit Courts is the fact that most Audit Court judges are ex-legislators (Speck, 2011). Following a decision by the legislature, a legislative decree approving or rejecting the accounts is published.

The period for ‘judging’ accounts also varies among the Brazilian states. In Minas Gerais, for example, if an assembly does not act on the Audit Courts’ evaluation within 120 days, the process is sent to the state Public Prosecutor’s office (Ministério Público).

If the accounts are approved by the assembly, the municipality has the legal backing to continue to receive voluntary transfers and new transfer agreements, even when it is not complying with internal transparency requirements. In sum, it is the interpretation of the Audit Courts that really matters and that may be discouraging municipal governments from respecting the rules governing budgets and transparency.

Finally, there is also the case of judicializing federal decisions based on irregularities in the CAUC database. The governments of the states of Minas Gerais and Acre, for example, were blocked by reason of the CAUC for failing to meet education conditionalities. However, these states obtained an injunction from the Supreme Court.⁹ With the judicialization of decisions, many times the judiciary does recognize the irregularities, but weighs these technicalities as less important than what is at stake when resources are frozen.

Figure 1
Internal transparency mechanisms (for receiving voluntary transfers)

3. SHORT TERM COSTS AND INCENTIVES

Initially, we examine the short-term costs of disobeying statutory budgetary exigencies of public education. If municipalities have their voluntary transfers blocked for failing to comply with the Fiscal Responsibility Law, we verify if in fact a penalty has been applied. We therefore consider a penalty - the blocking of further voluntary transfers - a short term financial cost for the municipalities.

Other mechanisms could generate different short-term costs for noncompliance. The most salient example is Directive 844 of 2008 (Portaria/MEC nº 844, 2008), issued by the Ministry of Education. This Directive determined that a failure to disclose information on the SIOPE system can impede new transfer agreements or other types of contracts with the Ministry of Education. However, given the fact this ‘directive’ is of inferior legal standing (a directive is hierarchically inferior to a law, for example), municipalities have the prerogative to question it in court. According to the education managers we interviewed, the Directive serves merely to “scare” administrators and has little practical effect on voluntary transfers. For this reason, we do not count the Directive as a legitimate short-term cost of noncompliance.

3.1. Data

In order to analyze short term costs, we relied on data from the following databases: (I) CAUC and (II) SICONV. The data from the CAUC are curated by the Secretary of the National Treasury and provide information on compliance with requirements germane to receiving voluntary transfers from the federal government. We obtained information on the CAUC reports for 5 April 2018, via a freedom of information request.¹⁰ Such reports compile data on all Brazilian governmental entities, informing the date and information relating to the compliance of each requirement. Of the 5570 Brazilian municipalities, 4816 municipalities (86.46% of the total) had not completed at least one of the 14 requirements constituting the CAUC.

Two of the CAUC’s requirements are critical for our analysis: item 3.2 (Submission of a summary Budgetary Execution Report - [known by its acronym, RREO in Portuguese]) and 4.2 (Application of resources set out as the constitutional minimum in education). These two items encompass the principal transparency obligations on education spending. Information on the Budgetary Execution Report originates in the STN/SICONFI database, whereas data on the constitutional minimum spending is found in the SIOPE/FNDE database. Although data from the SIOPE serve to feed the executive Budgetary Execution Report, it is possible that a municipality sends the Budgetary Execution Report to the federal government without having submitted it to the SIOPE database, and vice-versa.

According to article 52 of the Fiscal Responsibility Law, the federal administration must publish data on the Budgetary Execution Report every two months, with the deadline being 30 days following each bimonthly period. We analyze all Reports submitted until 5 of April 2018. On this date, municipalities across Brazil should have submitted the Budgetary Execution Report for the first two months of 2018, the end of March being the deadline. Of the 5570 municipalities, 63.78% had not complied with the requirement to disclose the Budgetary Execution Report by the deadline. Only 2017 municipalities (36.22% of the total) had sent the Budgetary Execution Report by the end of the second two-month period in 2018, namely, 30 May 2018. It is important to note that the system remains open for receiving reports after the deadline.

CAUC items that involve constitutional liabilities - particularly those involving constitutional minimum spending obligations in health and education - should be more difficult to flout. We would therefore expect a higher level of compliance with these obligations. As the CAUC is updated daily, we would expect that items that are easily submitted late to the system - for example, the submission of the Budgetary Execution Report - motivate the municipalities (as obligations) less than those that cannot be submitted late or reverted, such as disclosures relating to the constitutional minimum spending requirements.

In effect, we can confirm that compliance with the second of our analyzed items, constitutional minimum spending on education - is significantly higher. On 5 April 2018, 82.96% of municipalities complied with constitutional minimum requirement; the other 17.04% had not submitted this item at the time of our analysis. The majority of the 949 municipalities that did not disclose the constitutional minimum had also not submitted the Budgetary Execution Report. In other words, they had flouted both requirements. The proportion of municipal governments complying with the constitutional minimum must increase over the course of the year, as Budgetary Execution Reports are submitted.

The second database used to analyze the short-term costs of complying with internal transparency obligations is the SICONV.¹¹ Published by the Ministry of the Economy, the information contained in SICONV pertains to transfer agreements, accords, and the value of voluntary transfers.

3.2. Internal Transparency and Voluntary Transfers

In order to understand if the blocking of voluntary transfers by the federal government is, in effect, a short-term cost of failing to comply with internal transparency obligations, we cross-referenced the information on the fulfillment of requirements from the CAUC and the SICONV databases.

Before assessing the relationship between internal transparency compliance and the amount of funds received via voluntary transfers, it is important to emphasize how difficult it is to block a transfer already in course. After all, the principle of respecting the continuity and smooth operation of the public administration looms large among governmental priorities. Blocking a transfer agreement is different than blocking funds from a previously signed transfer agreement. In other words, a municipality that flouts the requirement of the CAUC can continue receiving resources from transfer agreements made in the past. In the words of one of the public officials interviewed:

[...] We cannot impede the continuity of an action. So, if the municipality received 40% of the funds due, the project is already in course and [the government] will stop the funds flowing just because the guy [municipality] fell out of line in the CAUC? It does not make sense (Official 1, authors’ translation).

In effect, despite 63.78% of municipalities having failed to submit their Budgetary Execution Report in April 2018, only a miniscule number of transfer agreements were cancelled that year. Examining the history of transfer agreements signed since September 2008, the data show that 3.3% were cancelled and 8.87% were annulled. Of those that were cancelled or annulled over the last decade, less than 1% were eventually executed (99.87% maintained their annulment).

Figure 2 illustrates the 8.87% of transfer agreements that were annulled, distributed over the length of a decade. Of all those annulled since 2008, only 11.7% occurred in 2018, which confirms that among the 63.78% of municipalities that did not disclose their Budgetary Execution Report, few had their transfer agreements annulled. It is therefore clear that the cost of noncompliance with transparency obligations does not imply the annulment of an existing transfer agreements. In other words, once a transfer agreement is signed, disobeying the transparency requirements of the Fiscal Responsibility Law does not come at the cost of blocked transfers from the federal government.

Figure 2
History of Transfer Agreements

Notwithstanding these results, it is likely that what matters most to subnational governments is the amount of resources being transferred by the central government and not the quantity of transfer agreements signed. This is to say that, even if the annulment of current transfer agreements does not represent a real potential cost for municipalities, the amount of money at stake in transfer agreements can itself motivate compliance with internal transparency requirements. The logic is that municipalities seek to secure future transfers through compliant behavior. In order to test whether the value of transfers is related to compliance with the exigencies of the Fiscal Responsibility Law, we estimate the relationship between the amount of voluntary transfers municipalities receive and the extent to which municipalities comply with the requirements of the CAUC. Formally, we are estimating a linear regression of the log (ln) value of voluntary transfers received from previously established transfer agreements (until 2017) and compliance with requirement in 2018. The following equation (1) illustrates:¹²

${\mathrm{C}}_{jk}=\mathrm{\beta }0+\mathrm{\gamma }\mathrm{l}\mathrm{n}{\left(\mathrm{R}\mathrm{\$}\mathrm{}\mathrm{T}\mathrm{r}\mathrm{a}\mathrm{n}\mathrm{s}\mathrm{f}\right)}_{jk}+\mathrm{l}\mathrm{n}{\left(\mathrm{P}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\right)}_{jk}+\mathrm{l}\mathrm{n}{\left(\mathrm{T}\mathrm{a}\mathrm{x}\mathrm{}\mathrm{R}\mathrm{e}\mathrm{v}\mathrm{e}\mathrm{n}\mathrm{u}\mathrm{e}\right)}_{jk}+{\mathrm{P}\mathrm{a}\mathrm{r}\mathrm{t}\mathrm{y}}_{jk}+{\mathrm{D}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{}\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{}\mathrm{S}\mathrm{i}\mathrm{g}\mathrm{n}\mathrm{e}\mathrm{d}}_{jk}+{\mathrm{I}\mathrm{n}\mathrm{s}\mathrm{t}\mathrm{i}\mathrm{t}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{}\mathrm{i}\mathrm{n}\mathrm{}\mathrm{P}\mathrm{u}\mathrm{b}\mathrm{l}\mathrm{i}\mathrm{c}\mathrm{}\mathrm{A}\mathrm{d}\mathrm{m}\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}_{jk}+{\mathrm{L}\mathrm{e}\mathrm{g}\mathrm{a}\mathrm{l}\mathrm{}\mathrm{S}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{i}\mathrm{n}\mathrm{g}}_{jk}+{\mathrm{S}\mathrm{t}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{}\mathrm{F}\mathrm{E}}_k+{\mathrm{\epsilon }}_j$ (1)

In equation (1), C_jk is a binary result that assumes the value 1 if the municipality fails to comply and zero if it complies. We test three possible scenarios of noncompliance: (i) “The municipality does not comply with at least one of the CAUC’s requirements, (ii) “The municipality does not corroborate having submitted the Budgetary Execution Report” and (iii) “The municipality does not comply with the constitutional minimum spending in education (25% of revenue)”. The independent variable (ln Transfers, in Real$) is the natural logarithm of voluntary transfers from the federal government to each municipality j in state k. We restrict our analysis to transfer agreements that were previously established in all areas of the public administration except for education, health, and social assistance between 2008 and 2017. As discussed, these ongoing transfer agreements are not cancelled if municipalities flout internal transparency obligations.

Given that the aim of estimating equation (1) is to test whether the amount of funds involved in voluntary transfers is related to greater compliance with the requirements of the CAUC, it is necessary to apply statistical controls to variables that can endogenously impact both the value of transfers as well as the compliance of requirements. We include in our estimation control variables for population, municipal tax revenue, partisan affiliation of the mayor, the number of days since a transfer agreement is signed and the period of analysis, the granting agency or entity (the ministry or area of the public administration that grants the transfer agreement),¹³ and the legal standing of the recipient (if municipal public administration or others¹⁴). Additionally, we include fixed effects by state (k) in order to account for differences arising from regional characteristics.

If municipalities with larger populations receive larger voluntary transfers and, at the same time, evince greater compliance with CAUC requirements - perhaps because they receive greater attention from the federal government - population could be biasing results. Similarly, a mayor’s party might be related to the likelihood of compliance and also the amount of resources received via voluntary transfers. Diverse authors argue that political alignment is a crucial factor in the volume of resources received via intergovernmental transfers (Bracco, Lockwood, Porcelli & Redoano, 2015; Brollo & Nannicini, 2012; Feld & Schaltegger, 2005; Veiga & Pinho, 2007). On the other hand, Cruz & Silva (2020) show that we do not find any statistical relationship between a mayor’s political alignment and the amount of funds received from the Ministry of Education. In any case, the inclusion of the mayor’s party in our estimation helps us avoid some potential biases arising from the effects of the party’s preferences regarding compliance with the CAUC requirements and the amounts received by transfer agreements.

Table 1
CAUC & Transfer Agreements

	Does Not Comply with at Least 1 Requirement	No Confirmation of Having Sent Budgetary Execution Report	Does Not Meet Constitutional Minimum Spending Requirements on Education
	[1]	[2]	[3]
ln(R$ Transfers)jk	-0.030	-0.025	-0.030
	[0.007]***	[0.007]***	[0.006]***
ln(Population)jk	0.012	-0.012	0.022
	[0.007]*	[0.007]*	[0.007]***
ln(R$ Tax Revenue)jk	-0.031	-0.029	0.028
	[0.008]***	[0.009]***	[0.009]***
Partyjk	Yes	Yes	Yes
Date contract signedjk	Yes	Yes	Yes
Institutionjk	Yes	Yes	Yes
Legal Standingjk	Yes	Yes	Yes
State Fixed Effectk	Yes	Yes	Yes
N	5,114	5,114	5,114
R2	0.13	0.26	0.07

• * p<0.1; ** p<0.05; *** p<0.01 Robust standard errors. Sources: CAUC/STN (April/2018), Siconv (May/2018), SIOPE/FNDE and TSE. R$ Transfers refers to the value of voluntary transfers executed via current transfer agreements (except for education, health and social assistance) established until 2017. R$ Tax Revenue refers to tax revenue per capita.

As illustrated by Table 1, there is a negative correlation between the value of voluntary transfers received from transfer agreements in previous years and compliance of at least one requirement with the CAUC in 2018 (column [1]). More specifically, for every 1% increase in the value of transfers, the probability of noncompliance with one of the requirements of the CAUC diminishes by three percentage points, holding all other variables constant. The results of column [2] also show that as transfers increase in value, the probability that a municipality has not sent the Budgetary Execution Report decreases. According to column [2], the probability of failing to submit the Budgetary Execution Report is -2.5% per unit change in the log of transfers. Column [3] demonstrates that increases in the value of voluntary transfers also lead to less noncompliance with the constitutional minimum (-3%).

It is important to note that in restricting ourselves to transfer agreements executed in previous years we are not analyzing an immediate relationship between transfers and compliance. The largest volume of transfers is not explained by compliance (which would imply a negative relationship between the dependent and independent variables) precisely because the transfers correspond to already established transfer agreements. As previously discussed, these are not cancelled because of noncompliance with internal transparency obligations. In other words, compliance with the transparency obligations of the Fiscal Responsibility Law can improve the relationship between the subnational unit in question and the federal government, ensuring increases in future transfers.

What we observe is, in effect, the larger the transfers in the past, the greater the incentives to comply in the future. Two possible rationales might explain these findings. First, the value of transfers can increase the cost of noncompliance. If the value of the voluntary transfers is large, the cost of having these resources delayed or not receiving them becomes relatively higher. A second rationale to explain the relationship between the value of voluntary transfers and noncompliance with transparency obligations has to with an “endogenous” relationship, namely, the characteristics of public officials. Similar to other studies in the area (particularly Soares & Melo (2016)) we use municipal tax revenue per capita as a proxy for a public official’s technical capacity. However, it is still possible that public officials who tend to be more proactive in securing federal funds for their municipalities are also preoccupied with compliance. This possible endogeneity speaks to the importance of public officials in ensuring compliance with internal transparency obligations. Put differently, endogeneity would show that “more effective public officials” value internal transparency, which is a result that has intrinsic value.

4. LONG TERM COSTS AND INCENTIVES

In order to investigate the long-term costs of noncompliance with internal transparency or, similarly, incentives for compliance, we use process tracing as a method to understand the consequences of flouting internal transparency requirements. Process tracing refers to the systematic mapping and analysis of evidence in light of predefined hypotheses (Bennett & Checkel, 2014; Lima, 2017).

Our working hypothesis is that the long-term costs of not complying with internal transparency exigencies are more significant than those in the short term due to the potential consequences faced by public officials, including the possibility of being barred from office. To test this hypothesis, it is necessary to effectively identify these consequences, most of which constitute “political costs”. We therefore analyze both the degree to which consequences affect noncompliant managers and the time it takes for these consequences to occur.

4.1. Data from the Supreme Electoral Court

Crossing two public databases available from the Superior Electoral Tribunal (TSE), we analyze municipal elections in 2016 and supplementary elections in 2017 and 2018. The first database was constructed from all the mayoral candidates in 2016 and the reasons for their disqualification. During 2016 municipal elections, 458 mayoral candidates (of a total of 5570 municipalities) were disqualified from running from office, while eight had their candidacies revoked.

We specifically focused on candidates who were disqualified because of their management of public accounts during previous terms in office. According to a reform from 1990 (Lei Complementar n^o 64/1990, 1990), those who are considered ineligible include the following:

[...] Those associated with public positions or functions who have had their accounts rejected because of irregularities that constitute an act of administrative improbity, and for the irrevocable decision of the competent institution - unless that decision has been suspended or annulled by the Judiciary, are barred from competing in the elections] to be realized during the eight subsequent years […]

Text from the Complementary Law nº 135 of 2010 (Lei Complementar n^o 135, 2010, popularly known as the “Clean Slate Law”, Authors’ translation).

The second database consists of information from municipal supplementary elections during 2017 and 2018. The goal is to understand the reason behind supplementary elections, and of particular interest are elections that occurred because of mayors who were removed or disqualified. We sourced details from the corresponding Regional Electoral Tribunals as well as the Supreme Electoral Tribunal. The search was conducted in the portal, “Simultaneous search of jurisprudence on the Electoral Tribunals” in which the universally used word indeferimento (overrule) was used, along with the name of the municipality. In the eventuality that this search produced no results, we looked up the number of the legal case for each municipality, which gave each resolution calling for new elections and the reasons therein.

Using these search strategies, we discovered the reasons behind each of the 106 supplementary elections during the period. It is important to note that five of the denied candidacies were reversed when mayors secured injunctions. The total sample therefore consisted of 101 cases of supplementary elections. Although this number is small in relation to the total number of Brazilian municipalities, it shows a potentially high political cost in not complying with budgetary requirements. The punishment of a few officials can have a learning effect on others.

4.2. Analysis of Data from the Superior Electoral Tribunal (TSE)

Based on data from the 2016 municipal elections we identified 13 cases where the motive for disqualifying an official was related to the rejection of accounts or for irregular accounts (Complementary Law 64). Furthermore, 203 cases of disqualified mayoral candidates originated by reason of the “Clean Slate Law”. In these cases, it was impossible to specify the exact reason for a candidate’s disqualification.

The second database, consisting of the supplementary elections of 2017 and 2018, allowed us to obtain more precise results. Of the 101 supplementary elections, 24 occurred because the mayors elected in 2016 had had their accounts rejected (refer to Table 2). Table 2 illustrates those who suffered the long-term political costs of budgetary noncompliance. It is important to keep in mind that there is a gap between the time when accounts are sent and the approval and effective analysis of accounts, especially because accounts are analyzed by the Electoral Courts and are voted on by the legislative assembly, as described in section 2.2. Surprisingly, we found that the disapproval of accounts for the average disqualified candidate (for the 24 cases) occurred in the year 2007 - almost a decade before the municipal elections of 2016! The delay between the rejection of accounts and a judgment against the corresponding public official took about a decade to produce material consequences.

Table 2
Mayors elected in 2016 municipal elections disqualified for rejection of public accounts

State	Municipality	Year of Accounts Rejected
AM	Novo Aripuanã	2011
ES	Fundão	2011
ES	Muqui	2008
MG	Ibiracatu	2003 and 2004
MG	Santana da Vargem	2005
MG	Campo Florido	2003 and 2004
MG	Santa Rita de Minas	2007 and 2008
MG	Cristiano Otoni	2012
MG	Guaraciama	2001
MG	Ervália	2011
MG	Santa Cruz de Salinas	2015
PR	Nova Laranjeiras	2011
RS	Butiá	2009, 2010, 2011
RS	Salto do Jacaí	2007
SP	Sebastianópolis do Sul	2011
SP	Mombuca	2012
SP	Mairinque	1992
SP	Itatinga	2011 and 2012
SP	Rincão	2012
TO	Taguatinga	2009 and 2010
MA	Bacabal	1997 and 2004*
MG	São Bento Abade	2001 to 2004*
MG	Campo Azul	2009 to 2012*/2009 to 2012*
RS	São Vicente do Sul	2009 to 2012*/2009 to 2012*

• *Cases in which the TRE/TSE did not specify the year accounts were rejected. In these cases, we put the years that candidates held office or the equivalent. Source: Elaborated by the authors based on Superior Electoral Tribunal (TSE).

Furthermore, the results seem to show regional asymmetries in the process of approving accounts. Of the 24 cases analyzed, 15 (62.5%) occurred in the states of Minas Gerais and São Paulo, suggesting that these tribunals are more efficient in processing their accounts than others. At the same time, the number of cases coming from these states is clearly disproportional. In Minas Gerais there are 853 cities and 654 in São Paulo, which represent approximately 27% of all Brazilian cities.

Given our objective of understanding long term costs, the data allow us to conclude that legal repercussions do occur at the individual level (mayors) for reason of rejected public accounts. However, legal repercussions are a long time coming and extreme differences in the rejection of accounts exist among states.

5. CONCLUSION

Transparent budgetary expenditures can contribute to the quality of public policies in diverse ways, such as the reduction of corruption and the generation of greater efficiencies and effectiveness in spending and policy outcomes (De Renzio, Wehner & Wehner, 2017). However, governments can decide to hide or manipulate budgetary expenditure information when the costs of doing so are low (Michener, Contreras & Niskier, 2018). Such appears to be the case in Brazil, much to the detriment of critical policy expenditures, such as public education.

The current study sought to analyze the costs of failing to comply with internal transparency obligations on budgetary expenditures at the municipal level in Brazil, focusing specifically on requirements to render expenditures on education transparent. Although there are financial and political costs (including reputational costs) in the short and long terms for flouting statutory budget regulations, they are extremely weak; moreover, they can be reverted by the Audit Courts as well as by the judiciary.

It is the federal government that imposes short-term costs on municipalities for compliance failures, by blocking ‘voluntary transfers’ of funds. Nevertheless, an analysis of data in both the CAUC and the SICONV databases show that many municipalities failed to disclose information on at least one of the items associated with the transparency of voluntary transfers. Effectively, 86.48% of municipalities did not comply with at least one of the 14 reporting requirements of the CAUC, 63.78% had not submitted their Budgetary Execution Report (RREO), and 17.04% had not proven that they spent the constitutionally mandated minimum on public education.

As for the long-term costs of not complying with budgetary regulations, even though punishments can be more severe, they take extremely long periods of time to materialize and appear to occur infrequently and in regionally unequal ways. To this last point, 62.5% of those punished between 2007-2018 came from states with only 27% of Brazil’s municipalities (São Paulo and Minas Gerais). On average, those who held public office were punished only after 10 years. In other words, an offending mayor can enjoy two consecutive terms (8 years) of office even if offenses in the first term should have prevented him or her from being reelected. Given these conditions, it is not clear whether the long-term costs of complying with budgetary regulation - much less reporting obligations germane to internal transparency - are capable of positively influencing behavior.

Finally, the current study contributes to the study of accountability processes in Brazil. Our analysis of enforcement processes highlights the lynchpin importance of the Audit Courts (Tribunais de Contas) in determining levels of municipal accountability in Brazil. Our research confirmed that the release of voluntary transfers is ultimately and ironically not in the hands of the funder, namely the federal government, but instead in the Audit Courts. Decisions by the Audit Courts can and do reverse hardline positions of the federal government. The apparent laxity of the Audit Courts in relation to municipal budgetary compliance appears to contribute to the prevailing incentive system, one in which municipalities have few motivations to comply with statutory budget requirements. In sum the ultimate authority of the Audit Courts effectively weakens the enforcement opportunities of the federal government and thus weakens municipal commitments to transparency. The essential implication is that the Audit Courts are critical in understanding the dysfunctional accountability systems of subnational governance in Brazil and merit further research and debate.

ACKNOWLEDGEMENTS

This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brazil (CAPES) - Finance Code 001.

REFERÊNCIAS

• Andrada, A. C., & Barros, L. C. (2010). O parecer prévio como instrumento de transparência, controle social e fortalecimento da cidadania. Revista do tribunal de contas do estado de Minas Gerais, 77(4), 53-75.
• Baião, A. L., Cunha, A. S. M., & Souza, F. S. R. N. (2017). Papel das transferências intergovernamentais na equalização fiscal dos municípios brasileiros. Revista do Serviço Público, 68(3), 583-610. Recuperado dehttps://doi.org/10.21874/rsp.v68i3.1406
  » https://doi.org/10.21874/rsp.v68i3.1406
• Bennett, A., & Checkel, J. (Orgs.). (2014). Process Tracing: From Metaphor to Analytic Tool Cambridge, UK: Cambridge University Press. Recuperado de https://doi.org/10.1017/CBO9781139858472
  » https://doi.org/10.1017/CBO9781139858472
• Bertholini, F., Pereira, C., & Renno, L. (2018). Pork is policy: Dissipative inclusion at the local level. Governance, 31(4), 701-720. Recuperado dehttps://doi.org/10.1111/gove.12331
  » https://doi.org/10.1111/gove.12331
• Bijos, D. (2013). Federalismo, instituições políticas e relações intergovernamentais: um estudo sobre os elementos determinantes das transferências voluntárias da União para os municípios do estado de Minas Gerais (Dissertação de Mestrado). Universidade de Brasília, Brasília, DF. Recuperado de http://repositorio.unb.br/handle/10482/13434
  » http://repositorio.unb.br/handle/10482/13434
• Bracco, E., Lockwood, B., Porcelli, F., & Redoano, M. (2015). Intergovernmental grants as signals and the alignment effect: Theory and evidence. Journal of Public Economics, 123, 78-91. Recuperado de https://doi.org/10.1016/j.jpubeco.2014.11.007
  » https://doi.org/10.1016/j.jpubeco.2014.11.007
• Brollo, F., & Nannicini, T. (2012). Tying Your Enemy’s Hands in Close Races: The Politics of Federal Transfers in Brazil. American Political Science Review, 106(4), 742-761. Recuperado de https://doi.org/10.1017/S0003055412000433
  » https://doi.org/10.1017/S0003055412000433
• Cruz, T., & Silva, T. (2020). Minimum Spending in Education and the Flypaper Effect. Economics of Education Review, 77, 102012. Recuperado dehttps://doi.org/10.1016/j.econedurev.2020.102012
  » https://doi.org/10.1016/j.econedurev.2020.102012
• De Renzio, P., Wehner, J., & Wehner, J. (2017). The Impacts of Fiscal Openness (SSRN Scholarly Paper ID 2602439). Rochester, NY: Social Science Research Network. Recuperado de https://papers.ssrn.com/abstract=2602439
  » https://papers.ssrn.com/abstract=2602439
• Decreto Decreto n. 10.046, de 9 de outubro de 2019 Recuperado de http://www.planalto.gov.br/ccivil_03/_Ato2019-2022/2019/Decreto/D10046.htm
  » http://www.planalto.gov.br/ccivil_03/_Ato2019-2022/2019/Decreto/D10046.htm
• Feld, L. P., & Schaltegger, C. A. (2005). Voters AS A Hard Budget Constraint: On the Determination of Intergovernmental Grants. Public Choice, 123(1-2), 147-169. Recuperado dehttps://doi.org/10.1007/s11127-005-0265-1
  » https://doi.org/10.1007/s11127-005-0265-1
• Fraga, F. V. (2016). O Descumprimento dos Requisitos Fiscais do CAUC e o Financiamento Municipal: Uma Análise sobre a Burocratização do Serviço e o Princípio da Eficiência (Dissertação de Mestrado). Universidade Católica do Salvador, Salvador, BA.
• Graça, L. F. G. (2016). O governo do estado e a assembleia legislativa: entre a submissão e os limites da independência Curitiba, PR; Belo Horizonte, MG: Appris Editora; Livraria Eireli.
• Hellevik, O. (2009). Linear versus logistic regression when the dependent variable is a dichotomy. Quality & Quantity, 43(1), 59-74. Recuperado de https://doi.org/10.1007/s11135-007-9077-3
  » https://doi.org/10.1007/s11135-007-9077-3
• Lei Lei Complementar nº 64, de 18 maio 1990 (1990). Estabelece, de acordo com o art. 14, § 9º da Constituição Federal, casos de inelegibilidade, prazos de cessação, e determina outras providências. Brasília, DF. Recuperado de http://www.planalto.gov.br/ccivil_03/leis/lcp/Lcp64.htm
  » http://www.planalto.gov.br/ccivil_03/leis/lcp/Lcp64.htm
• Lei Lei Complementar nº 101, de 4 de maio de 2000 (2000). Estabelece normas de finanças públicas voltadas para a responsabilidade na gestão fiscal e dá outras providências. Brasília, DF. Recuperado dehttp://www.planalto.gov.br/ccivil_03/leis/lcp/lcp101.htm
  » http://www.planalto.gov.br/ccivil_03/leis/lcp/lcp101.htm
• Lei Lei Complementar nº 135, de 4 de junho de 2010 (2010). Altera a Lei Complementar no 64, de 18 de maio de 1990, que estabelece, de acordo com o § 9o do art. 14 da Constituição Federal, casos de inelegibilidade, prazos de cessação e determina outras providências, para incluir hipóteses de inelegibilidade que visam a proteger a probidade administrativa e a moralidade no exercício do mandato. Brasília, DF. Recuperado de http://www.planalto.gov.br/ccivil_03/leis/lcp/Lcp135.htm
  » http://www.planalto.gov.br/ccivil_03/leis/lcp/Lcp135.htm
• Lei Lei nº 12.527, de 18 de novembro de 2011 (2011). Regula o acesso a informações previsto no inciso XXXIII do art. 5º , no inciso II do § 3º do art. 37 e no § 2º do art. 216 da Constituição Federal; altera a Lei nº 8.112, de 11 de dezembro de 1990; revoga a Lei nº 11.111, de 5 de maio de 2005, e dispositivos da Lei nº 8.159, de 8 de janeiro de 1991; e dá outras providências. Brasília, DF. Recuperado de http://www.planalto.gov.br/ccivil_03/_ato2011-2014/2011/lei/l12527.htm
  » http://www.planalto.gov.br/ccivil_03/_ato2011-2014/2011/lei/l12527.htm
• Lima, I. A. (2017). Técnicas Qualitativas em análises de Causalidade: Aplicações do Process Tracing. Revista Eletrônica de Ciência Política, 8(1), 128-143. Recuperado de https://doi.org/10.5380/recp.v8i1.50210
  » https://doi.org/10.5380/recp.v8i1.50210
• Lopes, J. E. G., R.Filho, J. F., Pederneiras, M. M. M., & Silva, F. D. C. (2009). Requisitos para Aprovação de Prestações de Contas de Municípios: Aplicação de Análise Discriminante (AD) a partir de Julgamentos do Controle Externo. Contabilidade Vista & Revista, 19(4), 59-83. Recuperado de http://revistas.face.ufmg.br/index.php/contabilidadevistaerevista/article/view/368
  » http://revistas.face.ufmg.br/index.php/contabilidadevistaerevista/article/view/368
• Massiani, J., & Picco, G. (2013). The Opportunity Cost of Public Funds: Concepts and Issues. Public Budgeting & Finance, 33(3), 96-114. Recuperado de https://doi.org/10.1111/j.1540-5850.2013.12016.x
  » https://doi.org/10.1111/j.1540-5850.2013.12016.x
• Meireles, F. (2019). Alinhamento partidário e demanda por transferências federais no Brasil. Revista de Administração Pública, 53(1), 173-194. Recuperado de https://doi.org/10.1590/0034-761220170282
  » https://doi.org/10.1590/0034-761220170282
• Michener, G. (2019). Gauging the Impact of Transparency Policies. Public Administration Review, 79(1), 136-139. Recuperado de https://doi.org/10.1111/puar.13011
  » https://doi.org/10.1111/puar.13011
• Michener, G., Contreras, E., & Niskier, I. (2018). Da opacidade à transparência? Avaliando a Lei de Acesso à Informação no Brasil cinco anos depois. Revista de Administração Pública, 52(4), 610-629.
• Moutinho, J. A. (2016). Transferências voluntárias da União para municípios brasileiros: Mapeamento do cenário nacional. Revista de Administração Pública, 50(1), 151-166. Recuperado dehttps://doi.org/10.1590/0034-7612139003
  » https://doi.org/10.1590/0034-7612139003
• O’Donnell, G. (1998). Accountability horizontal e novas poliarquias. Lua Nova: Revista de Cultura e Política, 44, 27-54. Recuperado de https://doi.org/10.1590/S0102-64451998000200003
  » https://doi.org/10.1590/S0102-64451998000200003
• Pinho, J. A. G., & Sacramento, A. R. S. (2009). Accountability: Já podemos traduzi-la para o português?. Revista de Administração Pública, 43(6), 1343-1368. Recuperado de https://doi.org/10.1590/S0034-76122009000600006
  » https://doi.org/10.1590/S0034-76122009000600006
• Portaria/MEC nº 844, de 8 de julho de 2008 (2008). Define os objetivos do Sistema de Informações sobre Orçamentos Públicos em Educação (SIOPE) e aprova o manual de Instruções para o Usuário do SIOPE. Brasília, DF. Recuperado de https://www.fnde.gov.br/index.php/acesso-a-informacao/institucional/legislacao/item/4024-portaria-mec-n%C2%BA-844,-de-8-de-julho-de-2008
  » https://www.fnde.gov.br/index.php/acesso-a-informacao/institucional/legislacao/item/4024-portaria-mec-n%C2%BA-844,-de-8-de-julho-de-2008
• Prado, S. (2001). Transferências Fiscais e Financiamento Municipal no Brasil: Projeto Descentralização Fiscal e Cooperação Financeira Intergovernamental (Relatório de Pesquisa). Rio de Janeiro, RJ: Fundação Konrad Adenauer.
• Rocha, A. C. (2013). A realização da accountability em pareceres prévios do Tribunal de Contas de Santa Catarina. Revista de Administração Pública, 47(4), 901-926. Recuperado de https://doi.org/10.1590/S0034-76122013000400005
  » https://doi.org/10.1590/S0034-76122013000400005
• Sakurai, S. N. (2013). Efeitos Assimétricos das Transferências Governamentais sobre os Gastos Públicos Locais: Evidências em Painel para os Municípios Brasileiros. Pesquisa e Planejamento Econômico, 43(2), 309-332.
• Santana, J. J. B Jr., Libonati, J. J., Vasconcelos, M. T. C., & Slomski, V. (2009, dezembro). Transparência fiscal eletrônica: uma análise dos níveis de transparência apresentados nos sites dos poderes e órgãos dos Estados e do Distrito Federal do Brasil. Revista de Educação e Pesquisa em Contabilidade, 3(3), 62-84. Recuperado dehttps://doi.org/10.17524/repec.v3i3.101
  » https://doi.org/10.17524/repec.v3i3.101
• Santos, R. R. dos, & Rover, S. (2019). Influência da governança pública na eficiência da alocação dos recursos públicos. Revista de Administração Pública, 53(4), 732-752. Recuperado de http://bibliotecadigital.fgv.br/ojs/index.php/rap/article/view/79919
  » http://bibliotecadigital.fgv.br/ojs/index.php/rap/article/view/79919
• Silva, M. S. (2015). Teoria do federalismo fiscal: notas sobre as contribuições de Oates, Musgrave, Shah e Ter-Minassian. Nova Economia, 15(1), 117-137. Recuperado dehttp://www.redalyc.org/resumen.oa?id=400437538005
  » http://www.redalyc.org/resumen.oa?id=400437538005
• Sistema de Informações sobre Orçamentos Públicos em Educação. (2019). Portal do FNDE Recuperado de http://www.fnde.gov.br/fnde_sistemas/siope
  » http://www.fnde.gov.br/fnde_sistemas/siope
• Soares, M. M., & Melo, B. G. (2016). Condicionantes políticos e técnicos das transferências voluntárias da União aos municípios brasileiros. Revista de Administração Pública, 50(4), 539-561. Recuperado dehttp://bibliotecadigital.fgv.br/ojs/index.php/rap/article/view/63583
  » http://bibliotecadigital.fgv.br/ojs/index.php/rap/article/view/63583
• Speck, B. (2011). Auditing Institutions. In T. J. Power, & M. M. Taylor (Eds.), Corruption and Democracy in Brazil: The Struggle for Accountability South Bend, Indiana: University of Notre Dame Press. Recuperado de https://www.bookdepository.com/Corruption-Democracy-Brazil-Timothy-J-Power/9780268038946
  » https://www.bookdepository.com/Corruption-Democracy-Brazil-Timothy-J-Power/9780268038946
• Supremo Tribunal Federal. (2020, agosto 24). STF declara inconstitucional norma de SP sobre inclusão de gastos previdenciários nas despesas com educação Recuperado dehttp://portal.stf.jus.br/noticias/verNoticiaDetalhe.asp?idConteudo=450203
  » http://portal.stf.jus.br/noticias/verNoticiaDetalhe.asp?idConteudo=450203
• Veiga, L. G., & Pinho, M. M. (2007). The political economy of intergovernmental grants: Evidence from a maturing democracy. Public Choice, 133(3-4), 457-477. Recuperado de https://doi.org/10.1007/s11127-007-9208-3
  » https://doi.org/10.1007/s11127-007-9208-3
• Yang, T.-M., & Maxwell, T. A. (2011). Information-sharing in public organizations: A literature review of interpersonal, intra-organizational and inter-organizational success factors. Government Information Quarterly, 28(2), 164-175. Recuperado dehttps://doi.org/10.1016/j.giq.2010.06.008
  » https://doi.org/10.1016/j.giq.2010.06.008

• 1
  We use the following terms interchangeably: internal transparency and intergovernmental transparency.
• 2
  We refer to ‘costs’ in the economic sense of the word: the cost of not complying with a legal exigency is not only financial but also one of “opportunity” (Massiani & Picco, 2013). In this sense, we aim to analyze the implication of choices made by officials in deciding to comply - or not to comply - with statutory requirements.
• 3
  Throughout the article, we refer to municipal (TCM) and state Audit Courts (TCE), simply as ‘Audit Courts’, as they perform analogous functions.
• 4
  One former minister of education (official 1) and one former president of the National Fund for the Development of Education (FNDE) (official 2). The interviews took place as semi-structured interviews with both officials at the same time.
• 5
  These proportions only take into account municipalities that received some sort of transfer via an agreement with the central government or a state.
• 6
  The requirements: (1) Tax regularity, contributions to the federal pension and national debt; (2) Regularity in terms of contributions to the FGTS (Severance Regularity Fund); (3) Regularity in terms of debt, loans, and financing provided by the federal government; (4) Legal and criminal liability; (5) Regularity in terms of previous reporting obligations; (6) Publication of the Fiscal Management Report - RGF; (7) Publication of the Budgetary Execution Report - RREO; (8) Submission of annual accounts; (9) Submission of the accounting balance matrix; (10) Submission of information for the Public Debt Registry - CDP; (11) Regiment of full tax competence;(12) Application of constitutional minimum spending in public education; (13) Application of constitutional minimal spending in health; (14) Pension regularity; (15) Regularity in terms of the provision of fiscal incentives (item 15 only for states).
• 7
  In a recent decision, the Supreme Court (STF) issued a writ of unconstitutionality on a São Paulo-issued legal norm relating to the inclusion of pension expenditures in public education reporting (Supremo Tribunal Federal, 2020).
• 8
  See Graça (2016) for an analysis of the relationship between the Executive and Legislative branches in the state of Rio de Janeiro.
• 9
  Tutela Provisória na Ação Cível Originária 3.111, Minas Gerais; and Medida Cautelar na Ação Cível Originária 3.104, Acre.
• 10
  We asked for the report “Relatório de Situação de Vários Entes” (Report on the Situation of Various Units), and in particular for the “CNPJ principal do Ente Federado” (Principal Organization of the Federal Unit).
• 11
  Retrieved from http://portal.convenios.gov.br/download-de-dados
• 12
  Since the proportions of the outcome variables are not in the tails of the distribution, a linear regression is preferable to a logistic model for this analysis (Hellevik, 2009).
• 13
  We included three variables as controls of the granting institution, which refer to the proportion of the municipality that are: from the Ministry of Sport, the Ministry of Tourism, and the Ministry of Agriculture, Livestock, and Inventory. These three institutions possess the highest proportion of transfer agreements in the federal government.
• 14
  ‘Others’ include state public administration, public consortium, public company / mixed capital company, civil society organization.

• [Translated version] Note: All quotes in English translated by this article’s translator

Publication Dates

• Publication in this collection
  05 May 2021
• Date of issue
  Mar-Apr 2021

History

• Received
  14 Oct 2019
• Accepted
  12 Sept 2020