Mandatory Individual Amendments: a Change in the Pattern of Executive Dominance in the Brazilian Budgetary and Financial Cycle^*

Abstract

Until 2013, the Brazilian executive branch had control over the execution of all discretionary public spending. In that context, all Brazilian legislators could do was amend the budget proposal prepared by the government. This article analyzes whether EC 86/2015, known as the mandatory budget amendment EC, has reduced executive dominance over the financial cycle of individual budget amendments. To this end, we examine descriptively and inferentially the main changes proposed by EC 86/2015, as well as data about the individual amendments executed before and after the constitutional change. Our results indicate that executive dominance was reduced after the change in the budget execution rules: the amount paid and the number of executed individual amendments have increased, while the profile of legislators with executed amendments also changed: Being part of the government coalition became less important. This article contributes to legislative studies by exploring the empirical implications of this important constitutional change. From a theoretical point of view, our findings challenge the expectation that the executive uses amendment execution as a bargaining tool to secure legislative support in roll-call votes in the National Congress, especially in the post-EC 86/2015 period.

Federal budget; mandatory individual budget amendments; executive branch; legislative branch; coalitional presidentialism

On March 17, 2015, the National Congress in Brazil promulgated the Constitutional Amendment Bill (PEC, in its Portuguese acronym) of the ‘Orçamento Impositivo’ (mandatory budget), resulting in Constitutional Amendment (EC, in its Portuguese acronym) Nº 86/2015. The main change brought by this EC is that individual budget amendments no longer need to be authorized by the executive to be executed. Before this constitutional amendment, the executive had the power to decide which amendments would be executed or not during each fiscal year; after EC 86/2015, the execution of all individual budget amendments became mandatory in the letter of the law.

The legislators who proposed and approved this change in the rules of the game were partly motivated by what could be summarized by an excerpt taken from the speech of the former president of the Chamber of Deputies, Henrique Alves (affiliated to the MDB):

Since its birth [of the PEC], people do not understand, but I know, and if today you know the importance of an individual amendment, which we will seek in order to meet the need of the poorest and the smallest. If it wasn't for us, they wouldn't get here. And then, the most humiliating way, the dropper, which forces this House to humiliate itself, the legislators to humiliate themselves (BRASIL, 2013, literal translation).

The fact that individual budget amendments are an important revenue source that is used to meet local needs was highlighted among federal deputies and senators during the processing of EC 86/2015¹. Budgetary amendments is seen as a means to ensure that schools, hospitals, highways, sports centers, and bridges, for example, are built in locations in which the federal executive is not always able to operate. What is more, mandatory execution of these amendments means that Congress has the freedom and autonomy to participate in the federal budget.

Many legislators contended that executive dominance and control over the budget process had greatly favored coalition members, as highlighted by Henrique Alves (MDB) – a claim also supported by part of the literature on individual budget amendments (FIGUEIREDO and LIMONGI, 2008; LUZ, 2017; PEREIRA and MUELLER, 2002; RAILE, PEREIRA, and POWER, 2011; VASSELAI and MIGNOZZETTI, 2014). Holder of the final decision on resources to be executed, the president, as well as his ministers, mostly executed individual amendments proposed by coalition legislators. EC 86/2015 aimed at reducing this bias by mandating that all approved amendments be automatically executed, regardless of authorship.

Legislators' aspiration to participate more actively in the budget process and to be less dependent on executive discretionary decisions is not an isolated case. The approval of the PEC of the mandatory budget amendment is part of a growing trend toward the legislature taking a larger role in the Brazilian decision-making process. In this sense, Almeida (2018; 2015) points out that, since the end of the 2010s, the federal executive's dominance over the process of drafting and approving laws has been reduced, while the legislature's control over this process has increased.

This article aims to analyze the changes in the rules governing federal budget execution provided for in EC 86/2015, as well as their implications for the executive's capacity to dominate and control the financial cycle of individual budget amendments. We thus seek to fill the knowledge gap created by this important institutional change. If EC 86/2015 produces the expected effects, the executive will lose one of its constitutional prerogatives, whereas both opposition and coalition legislators will rely on knowing that their individual amendments will surely be executed.

This article seeks to answer the following questions: Which changes did EC 86/2015 propose to reduce executive dominance and control over the federal budget? By reducing executive dominance in Brazil, was EC 86/2015 able to alter the pattern of executive-legislative relations in the financial and budgetary structure of individual amendments?

To answer these questions, we take two paths. First, we descriptively analyze whether the pattern of individual amendment execution has changed after EC 86/2015. To do so, we examined the total amount approved in the Annual Budget Law (LOA, in the Portuguese acronym) and subsequently paid by the executive, the variation in the number of total amendments executed, and the coalition's participation in this process, before and after EC 86/2015. We demonstrate that both the amounts being paid and the total number of executed amendments increased after the legal change. This suggests that the executive has lost some of its autonomy vis-à-vis the legislature when it comes to individual amendments. We also found that the disparity between the number of amendments executed for the coalition and the opposition was reduced.

After the descriptive analyses, we identified which were the defining features of the federal deputies with the highest number of executed amendments, comparing the contexts before and after EC 86/2015. The purpose was to test the hypothesis that this institutional change reduced the executive's control over individual amendment execution. To this end, we used the difference-in-difference method to compare the patterns of amendment execution between the treatment groups we established: coalition and non-coalition. The tests indicate that the likelihood that an individual amendment will be executed has increased in the post-EC 86/2015 period; moreover, after the new rule was implemented, the prevalence of amendment execution among coalition deputies was no longer observed.

The data used in the analysis were extracted from the Chamber of Deputies' website, the Integrated System of Financial Administration of the Federal Government (SIAFI, in its Portuguese acronym), the Federal Senate's information system (Siga Brasil), the Integrated System of Planning and Budget (SIOP, in its Portuguese acronym), the Legislative Database of CEBRAP, and from Volpe e Cambraia (2015). We offer the sources and time frame for each result we present.

The remainder of the article has four sections. In the first section, we show how the literature understands executive dominance over individual budget amendments before the constitutional amendment. The second section explores the most important changes proposed by EC 86/2015 regarding the legislature's larger role in the decision-making process and the potential impacts of this constitutional amendment in case the letter of the law is observed in practice. In the third section, we present and analyze the data on executive dominance before and after EC 86/2015. Finally, the last section concludes.

Executive dominance in Brazilian legislative production and its impact on individual budgetary amendments

The Federal Constitution of 1988 has a set of rules and procedures instructing the regulation of the Union's federal budget. The 1988 Constitution has granted the executive branch legal dominance over the initiative of budget bills (PLNs, in its Portuguese acronym); at the same time, it restricted the participation of the National Congress in this process, particularly its involvement in new public contracts, programs, and budget execution. And even though Congress is allowed to propose budget amendments to the government's original budget proposal to allocate and reallocate budgetary resources, the decision of which amendments will be executed remained in the hands of the executive until 2013. Therefore, for many years, the executive dominated and controlled the formulation and execution of the budget agenda – particularly discretionary spending – and, in parallel, restricted the individual participation of legislators in the federal budget process.

The fact that the executive had to authorize the execution of federal budgetary resources has put the executive in a privileged position vis-à-vis members of Congress. Although Congress is responsible for approving the LOA, it is not responsible for executing it. According to the 1988 Constitution, the executive has exclusive control over discretionary spending of the Union linked to the ministries. This mechanism had secured the executive a privileged position to negotiate the transfer of these resources with members of Congress.

The effects of this concentration of power on the part of the executive have been extensively analyzed for years. Individual budget amendments have been interpreted from two different perspectives. The first is based on the distributivity theory, which explains the workings of the United States Congress through individualistic exchanges that result in localist policies for which legislators claim credit in their electoral districts (CAIN, FEREJOHN and FIORINA, 1990; FIORINA, 1989; MAYHEW, 1974;). In the case of Brazil, it is argued that having the power to execute the budget without necessarily committing to the approved proposal gives the president and his ministers an efficient instrument to control and buy legislative support to approve their political agenda (MUELLER and ALSTON, 2001; PEREIRA and MUELLER, 2003, 2002; PEREIRA and RENNÓ, 2007; RAILE, PEREIRA, and POWER, 2011). According to this argument, federal deputies vote in support of the government's agenda, and, in return, local-level policies are brought to their constituencies through individual amendments.

The fact that Brazil adopted a presidential system and has an open-list proportional representation electoral system would serve as the basis for understanding how legislative support is purchased. According to Shugart and Carey (1992), Brazilian presidents are strong participants in the legislative process because they have legislative powers such as the exclusive authority to introduce and execute the budget law and to issue provisional measures (executive decrees that have the force of law). Such prerogatives, coupled with a legislature concerned with particularist and localist interests (due to the proportional distribution of votes), would help Brazilian presidents bypass the legislature.

In this context, the power to approve individual amendments is the executive's most important instrument for building legislative support. The strategy is as follows: Since the executive is not obligated to execute discretionary spending, it approves individual budget amendments to attract legislative support during voting sessions. Individual budget amendments thus are approved contingent on the legislator's vote in favor of government bills. Legislators are satisfied with this negotiation because they get resources for their electoral strongholds (AMES, 2003; PEREIRA and MUELLER, 2002).

The second perspective argues that the approval of individual amendments does not fully explain how legislative support for the government's agenda is built, given how central political parties are in the legislature (FIGUEIREDO and LIMONGI, 1999). According to this view, the executive does not negotiate legislative support directly with legislators. Agreements are established with party leaders to minimize the costs of political negotiation (FIGUEIREDO and LIMONGI, 2008; MESQUITA, 2008). In this case, in what concerns legislative support, the explanatory variable with the greatest predictive power is being a member of the government coalition. Individual bargains do not fully explain the executive's success rate in getting its projects approved. Rather, ministerial appointments and the distribution of positions and influence on the decision-making process are the key explanatory factors (FREITAS, 2016).

Moreover, since the executive dominated and controlled the budget and the stages in which budgetary resources are released, legislative demands on this issue take a back seat – especially the approval of individual budget amendments. These resources would only be released at the end of each year, in values lower than what was requested and approved at the beginning of the budget cycle, contingent, above all, on the government's ability to meet the fiscal results established by the Budgetary Guidelines Act (LDO, in the Portuguese acronym). According to Figueiredo and Limongi (2005) and Luz (2017), the rate at which individual amendments were approved did not exceed the average of 20% over the years, and most of the benefited legislators were members of the government coalition. Vasselai and Mignozzetti's (2014) analysis of data on the execution of these amendments and on the votes on executive bills corroborates this argument as it demonstrates that the transfer of budgetary resources is not linked to legislators' support for the government's agenda.

Finally, other studies argue that the execution of individual amendments does not guarantee the electoral gains expected by legislators (FIRPO et al., 2015; MESQUITA, 2008; MESQUITA et al., 2014). The argument is that these amendments did not increase legislators' likelihood of winning the next elections and that, contrary to what was expected, legislators do not allocate the largest share of their resources to the municipalities where they received the most votes in previous elections. This trend became more evident after 2004, when budgetary resources from individual amendments could no longer be allocated to municipalities but instead had to be directed to states. Therefore, legislators have no guarantee that the amendments they fought for will turn into public policies exactly in their strongholds. Consequently, it is uncertain whether they will be able to convert the credit claimed through the amendments into votes (MESQUITA et al., 2014).

Although they reach different conclusions, both interpretations are based on the same premise: the executive dominates the execution of individual budget amendments. It is only through this form of dominance that budget amendments may or may not be used in exchange for support or be relegated to the background by the government. This premise, however, only holds until 2013, when individual budget amendments approved by the legislators were not automatically executed but instead were contingent on the executive defining and coordinating the release of these resources. A change in the rules – with the execution of budget amendments becoming mandatory after EC 86/2015 – may have affected this dynamic.

Changes in the dynamics of executive dominance over the legislature and the approval of the mandatory nature of individual budget amendments

As highlighted, the Brazilian executive played a central role in the decision-making process for a long time. The 1988 Constitution conferred broad legislative powers to the executive, which helps explain the executive’s dominance in passing bills compared to the legislature (FIGUEIREDO and LIMONGI, 1999, 1996; PEREIRA and MUELLER, 2000; SANTOS, 2002). However, when the majority in the legislature is dissatisfied with a given rule, Congress can change it. According to Congleton (2011), a branch of government with less influence in the decision-making process can still use its resources – although scarce – to obtain additional resources in favorable circumstances. That is, in cases where the majority principle applies, the majority may influence not only policy outcomes but also procedural choices (KREHBIEL, 1992).

In this sense, Almeida (2018, 2015) points out that the participation of the legislative branch in the decision-making process has changed over the years, which is attested by the fact that executive dominance in approving laws in Brazil was reduced. Almeida (2018, 2015) shows that the number of executive bills that were passed in Congress has reduced since 2008, whereas the legislature's role in approving their own bills has increased. Therefore, legislators have demanded more and more influence and participation in the legislative process in exchange for their support for the government's agenda.

Almeida (2018) also demonstrates that, since 2007, the executive has been using less of its legislative prerogatives to accelerate the examination phase of bills in Congress. In this dynamic, congressional thematic committees – in the Chamber of Deputies and Federal Senate – have been increasingly active and prominent. These findings, along with the decreased executive dominance, have led Almeida to conclude that, over the last years, the legislature has gained more and more ground in the Brazilian legislative production.

EC 86/2015 represents thus another chapter in this more active role of legislators in the legislative process. Not surprisingly, the need to get the green light from the executive on budget execution has long been a cause of dissatisfaction and discussions among legislators – even among coalition members, who have always obtained more budgetary resources compared to opposition legislators. An example is PEC 22A/2000, which intended to make the execution of all amendments² to the federal budget³ mandatory, including collective amendments. This proposal was prepared by the then president of the Senate, Antônio Carlos Magalhães, who was affiliated to the PFL in the State of Bahia. Magalhães was one of the top PFL leaders, a party that was not only part of the government coalition but also had the vice presidency of the Republic, then occupied by Marco Maciel (affiliated to the PFL in the State of Pernambuco). Although Magalhães was careful to propose that, if approved, the PEC would only go into effect by 2003 so the government at the time would not be politically harmed (BRASIL 2000), it is still significant that such a proposal was introduced by a prominent member of the government coalition.

As Bonfim (2016) has shown, the speeches of party leaders in the Chamber of Deputies during the first round of voting for the PEC of the mandatory budget amendment were unanimously in favor of passing EC 86/2015. The party leaders' main argument was that approving this proposal would lead to more equality in the execution of individual budget amendments, so the legislature would no longer have to bargain with the executive to have these resources released. In this way, drawing on a dissatisfaction shared by opposition and coalition members in both legislative houses, Brazilian legislators eliminated the need of authorization from the executive to get individual budget amendments executed. The proposal was approved by a large majority: 452 votes in favor, 18 against, and one abstention in the second round of voting in the Chamber of Deputies. Since 2014, thanks to EC 86/2015, individual amendments have gained the status of amendments of mandatory execution.

Despite that, Santos and Gasparini (2020) contend that the resulting constitutional changes did not change the nature and the ‘modus operandi’ of the federal budget. That is because, according to these authors, the legislation approved included provisions that allow for the executive not to execute individual budget amendments – for example, in cases of technical constraints or spending restriction –, thus hindering the achievement of the legislators’ original goals. However, the data we will present next shows the opposite. Our findings indicate that, although individual amendments did not reach the 1.2% threshold of the Net Current Revenue (NCR) in relation to their payment, the amounts executed after EC 86/2015 went into effect are the highest ever registered. In addition, opposition legislators started to get more amendments executed compared to the period before this constitutional change.

Mandatory individual budget amendments and their effects on executive dominance

The Brazilian executive branch prepares the budget based on three instruments: the Multiannual Plan (PPA, in its Portuguese acronym), the LDO, and the LOA. As for the legislature, although it does not have the prerogative to initiate matters related to the budget, it can participate in the budget process by introducing amendments to executive bills.

In Congress, budget bills are analyzed by the Joint Committee on Planning, Public Budgets, and Oversight (CMO, in its Portuguese acronym). This committee is composed of 80 legislators, of which 40 are full members: 30 deputies and 10 senators, and an equal number of committee substitutes appointed according to the principle of party proportionality. This committee is charged with evaluating and issuing an opinion on all laws related to the PPA, the budgetary guidelines, the annual budget, additional credits, the accounts presented annually by the President of the Republic, and the national, regional, and sectoral plans and programs provided for in the Constitution.

During the processing of the budget proposal in the CMO, legislators can propose modifications to reallocate part of the budget to states, municipalities, and civil society organizations. Budget amendments could thus be defined as an institutional tool used by the legislature to propose changes to the PLNs. In the case of the Annual Budget Bill (PLOA, in its Portuguese acronym), these changes include the addition, cancellation, and modification of elements in the original proposal⁴. Since article 166 of the 1988 Constitution defines that individual amendments cannot be used to reallocate spending on personnel, debt service, and constitutional tax transfers to states, municipalities, and the Federal District, legislators concentrate on investment-related discretionary resources (FIGUEIREDO and LIMONGI, 2008).

In addition, the amendment process must follow what is already provided for in the LDO and PPA, that is, the amendments must cover programs that were previously defined by the ministries, and they should meet the objectives established in the PPA. Therefore, when proposing individual amendments, legislators must choose to which of the executive's projects they will allocate resources and to what locality the financial resources will be sent.

It is important to stress that only a small share of the federal budget is subject to amendment by individual legislators. The average percentage of investments in the annual budgets between 2001 and 2018 was only 2.5% of the overall budget approved in the period – investments are the type of expenditure over which legislators have the most influence. In other words, most of the federal budget resources are earmarked by the Constitution and are thus outside the reach of both the executive and the legislature.

The change in the rules and the adoption of the mandatory budget amendment, however, could limit the executive's control over the Union's discretionary spending, at least over the payment of individual amendments approved in the LOA. In short, EC 86/2015 amended articles 165, 166, and 198 of the 1988 Constitution and turned the execution of individual budget amendments mandatory for up to 1.2% of the NCR of the previous year, with 50% of these resources necessarily allocated to cover health expenditures. Therefore, changes in budget execution rules may potentially affect executive-legislative relations.

Before this constitutional change, the National Congress had already included the mandatory execution of amendments in the 2014 budget proposal via article 52 of the LDO. The year 2014 was thus the first year susceptible to the effects of this change in budget rules. As of 2015, with the approval of EC 86/2015, the new rules became part of the Constitution. The first real effect of these changes was a significant increase in the number of individual amendments executed (including ‘unpaid commitments’⁵). The values were adjusted with the National Consumer Price Index (IPCA, in its Portuguese acronym) of December 2018 (see Graph 01). The series begins in 2004 since no data on ‘unpaid commitments’ paid in previous years was available. The same criterion applies to Graphs 02 and 03. For other empirical data, the series begins in 1999.

Graph 01
Individual amendments approved and paid (including ‘unpaid commitments’) for Brazilian federal deputies and senators (2004-2018)

The year 2018 is the endpoint of all analyses in this study. We acknowledge that this strategy reduces the explanatory power of our results, as our post-constitutional change data is limited to only five years. However, this is a justified strategy because, otherwise, we would have to adopt another criterion to define who is part of the government coalition and who is in the opposition during the administration of Jair Bolsonaro (affiliated to the PL) – which began in 2019 –, which in turn would undermine our ability to draw comparisons with data on previous years. According to Figueiredo (2007), coalition members are those who belong to parties that control a cabinet ministry and formally declare support for the government. Bolsonaro's campaign strategy was to defend an anti-party platform, associating coalition presidentialism with corruption. He appointed military ministers or outsiders, people disconnected from the political community or experts in their fields (LIMONGI et al., 2022). The president himself governed without a political party for two years, from November 2019 to November 2021; it is thus impossible to use party criteria to define who was part of his coalition.

Graph 01 shows that, since 2012, the amounts approved in the LOA fall continuously, reaching their lowest level in 2018. However, while the amounts approved have decreased, the amounts paid (including ‘unpaid commitments’) in all years after EC 86/2015 were the highest ever registered. This suggests that the executive's discretion has decreased, affecting its ability to relegate legislators' individual amendments to the background. For the sake of comparison, we calculated these amounts as a percentage of the NCR of the previous year. The results in Graph 02 demonstrate that, although the amounts paid after the constitutional change did not reach the threshold of 1.2% of the NCR established by EC 86/2015, they are the highest observed in the period under study.

Graph 02
Individual amendments of Brazilian federal deputies and senators approved and paid (including ‘unpaid commitments’) as a percentage of the NCR (2004-2018).

It is worth mentioning three other aspects related to the changes in the budget rules brought by EC 86/2015. Firstly, the new rule established the threshold of 0.6% of the RCL of the previous year for unpaid commitments to comply with the financial execution. Second, it is only in cases of legal and technical impediments that individual amendments are not executed⁶. Third, mandatory budget amendments must be executed in an egalitarian and impersonal manner, regardless of authorship.

It should be noted that the amounts approved as a percentage of the NCR after the change in the budget implementation rules (2014-2018) remained below what was observed in previous years (2010, 2011, and 2012). However, when taking into account the amount paid (including ‘unpaid commitments’), we see a pattern of growth following the legal change: There is a quantum leap between 2016 and 2018, and never, in the entire series, have these values been so close to the 1.2% threshold established by EC 86/2015. In short, even if we take the amounts as a percentage of the NCR as a reference, we see a similar pattern in Chart 02 and Chart 01. In other words, there is evidence that the executive has lost a share of its control after EC 86/2015 both in absolute and proportional terms.

Regarding the first aspect, we stress that the execution of ‘unpaid commitments’ is limited to 0.6% of the NCR during the corresponding financial year to prevent individual amendment resources from being released gradually and slowly over several years. Graph 03 shows how this pattern changed after EC 86/2015 was implemented.

Graph 03
Amounts paid and ‘unpaid commitments’ of individual amendments of federal deputies and senators (2004-2018) as a percentage of the NCR.

As we can see, not only did the ‘unpaid commitments’ as a percentage of the NCR of the previous year increase continuously but the amounts paid also rose significantly. Although the amounts paid only started to grow in 2016 (two years after the institutional change), it remained stable in 2017, and, in 2018, it exceeded the total ‘unpaid commitments’ that were paid⁷. This suggests that the new rule had become institutionalized, since these amounts have increased without an increase in the requested amounts.

As for the legal and technical impediments, EC 86/2015 did not initially define which technical criteria should be used to guarantee amendment execution. In this case, the body responsible for the execution would have the authority to define the criteria and parameters for implementing individual amendments. Under the terms of the legislation, it is the responsibility of the branches, the Federal Public Prosecutor’s Office (MPU, in the Portuguese acronym), and the Public Defender's Office of the Union (DPU, in the Portuguese acronym), within 120 days of the publication of the budget law, to examine the impediments to the execution of the scheduled individual amendments, as well as to inform the National Congress of the justifications (BRASIL, 1988, Art. 166, § 14, of the Constitution). Diniz (2016) maintains that the executive did not allow the National Congress to conceptually determine what a technical impediment is, as the provisions related to the subject in the 2015 LDO were vetoed by then President Dilma Rousseff (affiliated to the PT).

However, in June 2015, the government issued Interministerial Directive Nº 221/2015 determining that the following situations would be classified as technical impediments to the conclusion of an agreement or transfer contract through individual amendments: 01. if the beneficiary and the amount involved are not indicated by the author of the individual amendment by the deadlines established in the directive; 02. if the proposal and the work plan are not presented within the period stipulated in item III of Art. 04 or if the complementation and adjustments requested in the work plan are not made within the period stipulated in item V of Art. 04; 03. if the proponent withdraws the proposal; 04. if there is an incompatibility between what is being proposed and the purpose of the budget action; 05. if there is an incompatibility between what is being proposed and the program of the governmental body or entity in charge of the execution; 06. if the proposed value lacks reasonableness, if it is incompatible with the project's execution schedule, or if it prevents the completion of an essential stage of the project; 07. if the work plan is not approved; and 08. other duly justified technical reasons⁸.

Thus, although the expectation regarding EC 86/2015 was to reduce executive dominance over individual amendments, some loopholes may have been left open so that control was not completely relinquished. Between 2014 and 2018, 15,381 amendments had technical problems, completely or partially, covering a total sum of R$10,897 billion, which corresponds to 24% of all approved amendments in the period. However, this is not to say that this amount was not totally executed. CMO Normative Ruling Nº 01 of 2014 guaranteed legislators the opportunity to indicate corrective measures to ensure that their amendments are executed. In other words, legislators can introduce administrative measures to correct technical problems in their amendments.

With respect to the third aspect, the equitable execution of individual amendments, the PEC of the mandatory budget amendment established that approved amendments must be executed in an impersonal and equal manner. However, this provision has not defined how this would be achieved. In addition, in the context of spending restriction, each cabinet ministry had the authority to decide what will be settled and paid, within the financial limitations of each ministry. In this sense, it is worth asking whether the total number of individual amendments approved increased after EC 86/2015 was passed. More importantly, given that the new constitutional text addresses equitable execution, we should ask whether the number of individual amendments executed has increased. Graph 04 shows federal deputies' amendments that were introduced and executed before and after the institutional change (1999-2018).

Graph 04
Total number of individual amendments of federal deputies approved and executed (1999-2018)

Graph 04 shows that the change in the constitutional text did not lead to an increase in the total number of individual amendments approved. Still, the number of amendments approved for the years 2016, 2017, and 2018 was lower than that in the years prior to the approval of EC 86/2015. The modifications promoted by this constitutional amendment, however, refer more to the execution of individual amendments than their approval. In other words, to evaluate whether executive dominance was reduced, it is essential to investigate how much the dynamics of amendment execution have changed.

Graph 04 shows that the pattern holds until 2017. However, the total individual amendments executed rose sharply in 2018, clearly the highest value in the period under analysis. It might be that the same phenomenon observed in Graph 03 – in which the effect of the change only became evident as of 2016 – is present in this case, here with consequences arising only as of 2018. This suggests that some of the effects of the new rule have been gradually building. However, despite the discrepancy in the 2018 figures, we need more years to know if this is in fact a new trend, or if it was simply an atypical year.

Graph 04 also shows that the execution rate of individual amendments is higher in electoral years (even-numbered years) in the period under analysis. This information suggests that the execution of individual amendments is considered strategic in municipal, state, and federal elections. It should be noted, however, that most of these resources have been allocated to states (MESQUITA et al., 2014). As Mesquita et al. (2014) point out, legislators do not allocate most of the resources they get from individual amendments to the municipalities where they obtained the most votes in past elections.

In addition, the federal budget execution rules that were established by Supplementary Law Nº 101 of 2000, also known as the Fiscal Responsibility Law, make it more difficult for Brazilian legislators to use individual amendments strategically. This law and the LDO established that municipalities in default with the Union cannot receive voluntary transfers. According to Baião et al. (2018), only 17.7% of Brazilian municipalities (990 out of a total of 5570) had complied with all tax and transparency obligations listed in the Auxiliary Information Service for Voluntary Transfers (CAUC, in its Portuguese acronym).

Therefore, budget norms establish a set of barriers that inhibit the use of individual amendments for electoral purposes, especially since the 2000s. The interpretation that budget amendments are executed mostly in electoral years should thus be further examined, as there is a discrepancy between what is being immediately interpreted from the data and the current rules governing the distribution of these resources. This endeavor, however, is beyond the scope of this article.

Despite the variations in execution between even- and odd-numbered years, the data we present indicate that the executive has lost relevance in what concerns individual amendments. After all, the amount being paid after EC 86/2015 increased, and the total number of amendments executed also increased significantly, at least in 2018. However, this does not confirm that the executive's bargaining power in using these amendments was reduced, as the executive might have increased the execution rate but kept favoring coalition legislators. For that reason, to further explore the bargaining issue, we examine whether the profile of executed amendments changed after EC 86/2015, more specifically, we analyze if the benefited legislators were part of the government coalition or opposition. To that end, we adopted Figueiredo's (2007) definition of what constitutes a member of the government coalition: to be affiliated to a party that holds a cabinet ministry and formally declares support for the government. All other deputies were considered members of the opposition. The data are shown in Graph 05.

Graph 05
Executed Individual amendments of federal deputies, separated into coalition and opposition (1999-2018)

Graph 05 shows that the number of executed amendments among opposition legislators gradually increased. But this pattern began years before EC 86/2015, and the opposition had a higher execution rate in 2013 than in 2015 and 2016. In other words, opposition legislators were already increasing their amendment execution rates before EC 86/2015 came into force. This trend, however, did not discourage the congressional majority from transforming the proposal for equal and impersonal execution of individual amendments into law: The PEC of the mandatory budget amendment was voted and approved. After that, the difference between the execution rates of coalition and opposition legislators remained stable, which suggests that EC 86/2015 contributed to maintaining this trend of greater equality. Thus, apart from the years 2002, 2003, and 2013, the smallest differences in the individual amendment execution rates between government and opposition were found in those years in which the constitutional amendment was in effect, which suggests that executive dominance has been firmly declining.

To confirm these indications, we tested whether ‘the institutional change brought by EC 86/2015 reduced the executive's discretion regarding individual amendments’. We took EC 86/2015 as a treatment that affects coalition deputies differently compared to those in the opposition. Likewise, we assume that EC 86/2015 affects the reelected and the non-reelected differently. After all, since the execution of individual amendments became mandatory, the preferences of the executive and the ‘expertise’ of the legislator should matter less for the execution of these amendments. In other words, we will compare the execution of individual amendments before and after the treatment, between the two pairs of groups formed: coalition and non-coalition deputies; reelected and non-reelected deputies.

This arrangement is appropriate for the difference-in-difference method, which consists in comparing changes in the pattern of the dependent variable between the treatment groups – coalition and reelected deputies after EC 86/2015 – and the control group. However, to validate the method, we must verify the parallel trends assumption. To that end, we analyzed whether there was a stable pattern of individual amendment execution, before the treatment, among the groups under analysis, and we examined the effects of the coalition and reelected groups on the execution of individual amendments in periods before 2014: 1999-2002, 2003-2006, 2007-2010, and 2011-2013. These periods refer to the four-year mandate of different presidencies (FHC2, Lula01, and Lula02) and the first three years (2011-2013) of Dilma Rousseff's presidency (Dilma 01) – the year 2014 was excluded because amendments approved in 2014 were already affected by EC 86/2015.

The parallel trends assumption was tested with a multivariate logistic regression. The variables are listed below, followed by a graph with the odds ratio for each of the model’s independent variables, in each time interval. We used data from all individual amendments of federal deputies approved between 1999 and 2013 – only the amendments paid in the financial year for which they were approved were considered executed. The same criterion was used in the next test, the difference-in-difference model.

Model variables to test the parallel trends assumption

Dependent variable – ‘Execution’: takes the value 01 if the deputy's amendment was executed and 0 if it was not.

Independent variable 01 – ‘Coalition’: takes the value 01 if the deputy's party was in the government coalition in the year the amendment was introduced and 0 if it was not. The assumption is that if the executive uses amendments as a bargaining tool, being part of the coalition is a prevailing feature of legislators who had amendments executed before EC 86/2015. Following Figueiredo and Limongi (2008), we took the 31st of October of the year in which the amendment was executed as a reference to define whether the deputy belonged to a coalition party.

Independent variable 02 – ‘Reelected’: takes the value 01 if the deputy had a seat in the legislature prior to that in which he or she presented the amendment, and 0 if he or she did not. The assumption is that if legislators use individual amendments as an electoral connection (AMES, 2003; PEREIRA and MUELLER, 2003; PEREIRA and RENNÓ, 2001), this should be a distinguishing feature of the reelected until EC 86/2015 was implemented. After EC 86/2015, the effect of this variable on ‘Execution’ would decrease because, given that execution is mandatory, deputies' expertise would have a lower impact on the likelihood of individual amendments being executed.

Graph 06 shows that there is a pattern in how the variable ‘Coalition’ affected the likelihood of having an amendment executed before EC 86/2015, which is in line with the studies we discussed previously (FIGUEIREDO and LIMONGI, 2005; LUZ, 2017). Although the impact varied in intensity, the effect is positive and statistically significant for a 95% confidence interval in all periods considered. It is important to stress that our time series includes four governments led by different parties and presidents: the second government of Fernando Henrique Cardoso (FHC, affiliated to the PSDB); the two Lula governments (affiliated to the PT), and the first government of Dilma Rousseff (affiliated to the PT). This variation reduces the chances that the use of individual amendment as a bargaining tool by the executive is a strategy restricted to a few actors and contexts.

Graph 06
Odds ratio of each variable in the model of the parallel trends assumption

On the other hand, the same is not valid for the variable ‘Reelection’. Of the four periods analyzed, one has a negative effect (FHC2), two have positive effects and statistical significance for the established confidence interval (Lula01 and Dilma01), and one has a very low positive effect and no statistical significance (Lula02). In the latter case, the odds ratio fluctuates between 0.964 and 1.074, with a 95% confidence interval. Thus, Pereira and Rennó's (2001) and Ames' (2003) argument that amendments are important resources to secure legislators' reelection does not find solid ground in this article. In short, our results confirm what Mesquita (2008), Mesquita et al. (2014), and Firpo et al. (2015) contend: Individual amendment execution does not directly affect legislators' electoral success. Of all the variables we used, only ‘Coalition’ meets the parallel trends assumption.

Based on this information, we created our difference-in-difference model: a multivariate logistic regression to verify how EC 86/2015 has affected the execution of individual amendments for the control group (coalition). We have fixed the effects for each federal deputy, which gives us two advantages. First, it ensures that the effect observed is linked to whether or not the deputy belongs to the coalition, not to the legislator's individual behavior. Second, it makes the logistic regression formally consider the characteristics of panel data, a condition intrinsic to the difference-in-difference model. Next, we detail the variables and present the odds ratio of each independent variable (see Table 01 and Graph 07). This time, we used data from all individual amendments of federal deputies passed between 1999 and 2018.

Table 01
Odds ratio of the difference-in-difference model – effect of EC 86/2015 on the execution of individual amendments in the treatment group (coalition)

Execution		Odds Ratio	Standard Deviation	95 % Confidence Interval
EC 86/2015		1,369***	0,041	1.352 – 1.473
Coalition		1,411***	0,031	1.290 – 1.452
EC x Coalition		0,677***	0,026	0.627 – 0.730
N	139.423
N Groups	1675

Source: Created by the authors based on the Stata Software. Notes: ***p<0,001.

Graph 07
Odds ratio of the difference-in-difference model

Variables of the difference-in-difference model – effect of EC 86/2015 on the execution of individual amendments between treatment groups (coalition and reelected)

Dependent variable – ‘Execution’: takes the value 1 if the deputy's amendment was executed and 0 if it was not.

Independent variable 01 – EC 86/2015: takes the value 01 if the individual amendment was executed in a post-EC 86/2015 year and 0 if it was executed before.

Independent variable 02 – ‘Coalition’: takes the value 01 if the deputy's party was in the coalition in the year the amendment was introduced and 0 if it was not.

Independent variable 3 – EC x ‘Coalition’: takes the value 01 if the deputy's party was in the coalition in the year in which the amendment was introduced and if this year was after EC 86/2015 had come into effect. Otherwise, it takes the value 0. We used this interactive variable as a reference to test our coalition hypothesis: If the executive used individual amendments as a bargaining tool among coalition members in pre-EC 86/2015 years, and assuming that the treatment had a negative effect in terms of amendment execution for this control group, its coefficient is expected to be lower than that generated by the variable ‘Coalition’, which indicates that executive dominance was reduced.

We first highlight the effect of the variable ‘EC 86/2015’ on individual amendment execution. Because of mandatory execution, an approved amendment is 37% more likely to be executed after EC 86/2015 than before, all other variables fixed in their average values. One of the objectives of making amendment execution mandatory was precisely to minimize the government's discretion regarding execution and make sure all amendments – except in cases involving technical problems or spending restrictions – are executed.

In what concerns the impact of EC 86/2015 on the likelihood that coalition deputies will have their amendments executed, it is striking that the odds ratio of the variable ‘Coalition’ is higher than that of the interactive variable ‘EC x Coalition’; also, while the former is greater than 01, the latter is lower than 01(negative effect). Therefore, the difference in the likelihood of having an individual amendment executed between coalition and opposition legislators drastically reduced after EC 86/2015 was implemented – at least in the first years since it came into effect.

The results involving the variable ‘EC 86/2015’ point to a reduction of the executive's discretion over individual amendment execution. As the executive lost autonomy, inequality in amendment execution between coalition and opposition legislators decreased during the period we examined, as attested by the difference in the odds ratios of the variables ‘Coalition’ and ‘EC x Coalition’.

Although our results seem to confirm our hypothesis, we must consider some empirical limitations. Given the criteria we used to define who is and who is not part of the government, we only have data to compare the effects of EC 86/2015 between 2014 and 2018, which leads us to two considerations. The first refers to an operational problem. Since our time series includes only a few years, there is a greater risk that our results are associated with random factors not considered in this study. The second consideration involves a contextual issue, as the period we analyzed was a critical moment for the Brazilian political system. In 2016, President Rousseff was removed from office after an impeachment process had started at the end of 2015, the first year of her second term. Vice president Michel Temer took office, which led to a reshaping of the government coalition. Therefore, we cannot disregard that the effects we found might be linked to the methodological limitations of our data or to an atypical political situation.

Conclusions

This article provides contributions of two natures: theoretical and empirical. From a theoretical perspective, our findings suggest that the changes caused by EC 86/2015 should be taken as a turning point by scholars who address individual budget amendments as a bargaining tool in executive-legislative relations. The fact that the executive has lost autonomy in amendment execution raises the question of whether individual amendments are still one of the executive's main tools for buying legislative support.

From an empirical perspective – and in line with our theoretical contribution –, we have demonstrated that, since the constitutional amendment, opposition and coalition legislators have similar chances of receiving budgetary resources. Therefore, our results show, for the first time, how a change to a recent rule was able to quickly modify the patterns of individual amendment execution and grant the legislative branch more control over the legislative agenda, as already indicated by Almeida (2018; 2015).

The constitutional amendment we analyzed here was not the last attempt to reduce the executive's control over the federal budget. After EC 86/2015, the National Congress approved two additional constitutional amendments to expand legislators' participation in the budget cycle and make them less dependent on resources controlled by the executive. They enacted EC 100/2019, which made the execution of amendments from state parliamentary groups mandatory, and EC 105/2019, which created the ‘special transfers’ modality for mandatory individual budget amendments: It allows states and municipalities to receive federal resources without the need for any type of compensation or even any connection to the government's policy priorities. In addition, the legislature included the ‘primary fiscal surplus equal to 09’ for that financial year through the 2019 LDO (which was the basis for the 2020 LOA), allowing the general rapporteur of the proposal to control a portion of the discretionary resources without necessarily informing the locality and the legislator benefiting from these resources. This mechanism thus allows some members of Congress to benefit from federal resources at the expense of others.

The changes in budget rules promoted by the National Congress in recent years resulted in legislators' greater access to discretionary federal resources. While in 2017 R$10,738 billion were committed to budget amendments, in 2021 the amount rose sharply to R$33,398 billion, including individual and collective amendments, as well as those of the general rapporteur. In short, in recent years, the legislature has launched an important campaign to reduce executive dominance in the budgetary and financial cycle in Brazil. In this first examination of the effects of one of these constitutional amendments, we present evidence that EC 86/2015 reduced executive dominance in executing individual amendments, which should impact executive-legislative relations and the coordination of this process. For if the executive has lost some prerogatives, some other actor must have taken them. In this sense, in addition to identifying the loss of dominance, we should determine what pattern this process has taken. After all, one must know who ultimately wins in the process of individual budget amendment execution.

References

• ALMEIDA, Acir dos Santos (2018), Governo presidencial condicionado: delegação e participação legislativa na Câmara dos Deputados. Doctoral thesis. Programa de Pós-Graduação em Ciências Política. Instituto de Estudos Sociais e Políticos. Universidade do Estado do Rio de Janeiro.
• ALMEIDA, Acir dos Santos (2015), Processo legislativo: mudanças recentes e desafio. Boletim de Análise Político-Institucional. Vol. 07, pp. 43-50.
• AMES, Barry (2003), Os entraves da democracia no Brasil. Rio de Janeiro: Fundação Getúlio Vargas. 410 pp..
• BAIÃO, Alexandre Lima; COUTO, Cláudio Gonçalves, JUCÁ, Ivan Chaves (2018), A execução das emendas orçamentárias individuais: papel de ministros, cargos de liderança e normas fiscais. Revista Brasileira de Ciência Política. Vol. 25, pp. 47–86.
• BRASIL (2013), Henrique Alves defende orçamento impositivo para emendas parlamentares. Câmara dos Deputados. 04 de Fevereiro de 2013. Available at ˂https://www.camara.leg.br/radio/radioagencia/395079-henrique-alves-defende-orcamento-impositivo-para-emendas-parlamentares/˃ Accessed on March, 02, 2021.
  » https://www.camara.leg.br/radio/radioagencia/395079-henrique-alves-defende-orcamento-impositivo-para-emendas-parlamentares/˃
• BRASIL (2000), ACM apresentará PEC para tornar orçamento impositivo. Senado Federal. Da redação. 05 de Junho de 2000. Available at ˂https://www12.senado.leg.br/noticias/materias/2000/06/05/acm-apresentara-pec-para-tornar-orcamento-impositivo˃ Accessed on March, 02, 2021.
  » https://www12.senado.leg.br/noticias/materias/2000/06/05/acm-apresentara-pec-para-tornar-orcamento-impositivo˃
• BRASIL (1988), Constituição da República Federativa do Brasil. Brasília. Available at ˂http://www.planalto.gov.br/ccivil_03/constituicao/constituicaocompilado. htm˃. Accessed on June, 14, 2018.
  » http://www.planalto.gov.br/ccivil_03/constituicao/constituicaocompilado. htm
• BONFIM, Raul Wesley Leal (2016), A relação entre os Poderes Executivo e Legislativo no processo orçamentário brasileiro: o caso da Proposta de Emenda à Constituição (PEC) do orçamento impositivo. Cadernos de Pesquisa em Ciência Política. Vol. 05, Nº 02, pp. 01–27.
• CAIN, Bruce; FEREJOHN, John, and FIORINA, Morris (1990), The personal vote: constituency service and electoral independence. Cambridge: Harvard University Press. 268 pp..
• CONGLETON, Roger D. (2011), Perfecting Parliament: constitutional reform, liberalism, and the rise of Western democracy. Cambridge: Cambridge University Press. 658 pp..
• DINIZ, Vítor de Almeida (2016), PEC do orçamento impositivo: um sonho que virou realidade? Master’s dissertation. Faculdade de Filosofia, Letras e Ciências Humanas. Universidade de São Paulo.
• FIGUEIREDO, Argelina Cheibub (2007), Government coalitions in Brazilian democracy. Brazilian Political Science Review. Vol. 01, Nº 02, pp. 182-216.
• FIGUEIREDO, Argelina Cheibub and LIMONGI, Fernando de M. Papaterra (2008), Política orçamentária no presidencialismo de coalizão. Rio de Janeiro: Editora Fundação Getúlio Vargas. 184 pp..
• FIGUEIREDO, Argelina Cheibub and LIMONGI, Fernando de M. Papaterra (2005), Processo orçamentário e comportamento Legislativo: emendas individuais, apoio ao Executivo e programas de governo. Dados. Vol. 48, Nº 04, pp. 737–776.
• FIGUEIREDO, Argelina Cheibub and LIMONGI, Fernando de M. Papaterra (1999), Executivo e legislativo na nova ordem constitucional. Rio de Janeiro: Editora Fundação Getúlio Vargas. 108 pp..
• FIGUEIREDO, Argelina Cheibub and LIMONGI, Fernando de M. Papaterra (1996), Congresso nacional: organização, processo legislativo e produção legal. Cadernos de Pesquisa CEBRAP. Nº 05, 91 pp..
• FIORINA, Morris P. (1989), Congress: keystone of the Washington establishment. Yale University Press. 192 pp..
• FIRPO, Sergio; PONCZEK, Vladimir, and SANFELICE, Viviane (2015), The relationship between federal budget amendments and local electoral power. Journal of Development Economics. Vol. 116, pp, 186–198.
• FREITAS, Andréa Marcondes de (2016), O presidencialismo da coalizão. Rio de Janeiro: Konrad Adenauer Stiftung. 150 pp..
• GIACOMONI, James (2012), Orçamento público. São Paulo: Editora Atlas. 338 pp.
• KREHBIEL, Keith (1992), Information and Legislative organization. Ann Arbor: University of Michigan Press. 314 pp..
• LIMONGI, Fernando de M. Papaterra; FREITAS, Andréa; MEDEIROS, Danilo, and LUZ, Joyce Hellen (2022), Government and Congress. In: Brazil under Bolsonaro. How endangered is democracy? Edited by BIRLE, Peter and SPECK, Bruno. Berlim: Ibero-Amerikanisches Institut Preußischer Kulturbesitz. pp. 30-43.
• LUZ, Joyce Hellen (2017), Política ministerial: as emendas individuais orçamentárias no presidencialismo de coalizão brasileiro. Master’s dissertation. Faculdade de Filosofia, Letras e Ciências Humanas. Universidade de São Paulo.
• MAYHEW, David R. (1974), Congress: the electoral connection. New Haven: Yale University Press. 216 pp..
• MESQUITA, Lara (2008), Emendas ao orçamento e conexão eleitoral na Câmara dos Deputados brasileira. Master’s dissertation. Faculdade de Filosofia, Letras e Ciências Humanas. Universidade de São Paulo.
• MESQUITA, Lara; SILOTTO, Graziele; LUZ, Joyce Hellen, and HUBERT, Paulo (2014), Emendas individuais e concentração de votos: uma análise exploratória. Teoria & Pesquisa: Revista de Ciência Política. Vol. 23, Nº 02, pp. 82-106.
• MUELLER, Bernardo and ALSTON, Lee (2001), Coalitional stability and the gains from trade between the executive and the legislature in Brazil. Anais do XXIX Encontro Nacional de Economia (ANPEC). Salvador.
• PEREIRA, Carlos and MUELLER, Bernardo (2003), Partidos fracos na arena eleitoral e partidos fortes na arena legislativa: a conexão eleitoral no Brasil. Dados. Vol. 46, Nº 04, pp. 735–771.
• PEREIRA, Carlos and MUELLER, Bernardo (2002), Comportamento estratégico em presidencialismo de coalizão: as relações entre Executivo e Legislativo na elaboração do orçamento brasileiro. Dados. Vol. 45, Nº 02, pp. 265-301.
• PEREIRA, Carlos and MUELLER, Bernardo (2000), Uma teoria da preponderância do Poder Executivo: o sistema de comissões no Legislativo Brasileiro. Revista Brasileira de Ciências Sociais. Vol. 15, Nº 43, pp. 45–67.
• PEREIRA, Carlos and RENNÓ, Lucio (2007), O que é que o reeleito tem? O retorno: o esboço de uma teoria da reeleição no Brasil. Brazilian Journal of Political Economy. Vol. 27, Nº 04, pp. 664-683.
• PEREIRA, Carlos and RENNÓ, Lucio (2001), O que é que o reeleito tem? Dinâmicas político-institucionais locais e nacionais nas eleições de 1998 para a Câmara dos Deputados. Dados. Vol. 44, Nº 02, pp. 133–172.
• RAILE, Eric; PEREIRA, Carlos, and POWER, Timothy (2011), The executive toolbox: building legislative support in a multiparty presidential regime. Political Research Quarterly. Vol. 64, Nº 02, pp. 323-334.
• SANTOS, Fabiano (2002), Partidos e comissões no Presidencialismo de Coalizão. Dados. Vol. 45, Nº 02, pp. 237–264.
• SANTOS, Núbia Cristina Barbosa and GASPARINI, Carlos Eduardo (2020), Orçamento impositivo e relação entre poderes no Brasil. Revista Brasileira de Ciência Política. Vol. 31, pp. 339–396.
• SHUGART, Matthew Soberg and CAREY, John M. Carey (1992), Presidents and Assemblies: constitutional design and electoral dynamics. Cambridge: Cambridge University Press. 332 pp..
• VASSELAI, Fabricio and MIGNOZZETTI, Umberto (2014), O efeito das emendas ao Orçamento no comportamento parlamentar e a dimensão temporal: velhas teses, novos testes. Dados. Vol. 57, Nº 03, pp. 817–853.
• VOLPE, Ricardo Alberto and CAMBRAIA, Túlio (2015), A experiência do orçamento impositivo na lei de diretrizes orçamentárias para 2014. RBPO – Revista Brasileira de Planejamento e Orçamento. Vol. 05, Nº 02, pp. 100–131.

• 1
  Available at ˂http://www.camara.gov.br/proposicoesWeb/fichadetramitacao?idProposicao= 602633˃
• 2
  There are individual and collective budget amendments. The first refers to amendments introduced individually by federal deputies and senators. The second refers to amendments introduced by state parliamentary groups and standing committees in the National Congress. For more details on how amendments are submitted, see Resolution Nº 01 of 2006 of the National Congress.
• 3
  Available at ˂https://www25.senado.leg.br/web/atividade/materias/-/materia/ 114143˃. Accessed on March, 25, 2019.
• 4
  There are three types of amendments: amendment to expenditures – indicates the inclusion, relocation, or cancellation of budget allocations; amendment to revenue – proposes the inclusion, exclusion, or alteration of revenues in cases of error; and amendment to the text – changes in the writing of the bill.
• 5
  ‘Unpaid commitments’ are resources the executive cannot effectively pay within a given budget year and are thus necessarily included as a payment order in the LOA in the following year (GIACOMONI, 2012).
• 6
  Amendments also might not be executed if the government is expected to not meet the fiscal target established in the LDO; in this case, budget resources allocated to amendments may be frozen at the same rate as all other discretionary expenditures (BRASIL, 1988, art. 166, § 17).
• 7
  According to information from the Budget and Financial Oversight Consultancy (CONOF, in the Portuguese acronym) of the Chamber of Deputies, it was taking several years to pay individual amendments. Between 2005 and 2015, the largest amount paid in the same financial year in which the amendment was approved was never higher than 15% of the authorized amount. Even in the first two fiscal years in which the mandatory budget was in effect (2014 and 2015), the amount paid in amendments was not financially representative.
• 8
  In addition to Interministerial Directive Nº 221/2015, the government also adopted Interministerial Directive Nº 222/2015 to set the proceedings and deadlines for submitting and registering individual amendments.
• Data replication: https://doi.org/10.7910/DVN/O08RNG
• Translated by Karin Blikstad

• *
  This research was financed by the National Council for Scientific and Technological Development (CNPq, in Portugues) – Process No. 140566/2018-0 and São Paulo Research Foundation (Fapesp, in Portuguese) – Process No. 2018/00013-9 and No. 2018/01513-5.

Publication Dates

• Publication in this collection
  02 June 2023
• Date of issue
  2023

History

• Received
  16 Mar 2021
• Accepted
  28 Oct 2022