Abstract

Latin American political systems are recognized as having presidents with broad legislative powers (SANTOS, PÉREZ-LIÑÁN and GARCIA MONTEIRO, 2014SANTOS, M. L., PÉREZ-LIÑÁN, A. and GARCIA MONTEIRO, M. (2014), El control presidencial de la agenda legislativa en América Latina. Revista de Ciencia Política. Vol. 34, Nº 03, pp. 511-536.). This feature has alerted political scientists to possible deficiencies regarding the legislators' ability to fully exercise their role. Legislatures in these countries are usually defined as weak, or as mere "rubber stamps" of the presidential agenda (COX and MORGENSTERN, 2001COX, G. W. and MORGENSTERN, S. (2001), Latin America's reactive assemblies and proactive presidents. Comparative Politics. Vol. 33, Nº 02, pp. 171–189.; O'DONNELL, 1994O'DONNELL, G. (1994), Delegative democracy. Journal of Democracy. Vol. 05, Nº 01, pp. 55-69.).

Brazil is a notable example of a presidency with strong legislative powers. The president approves most of the proposals sent to the legislature – with high success rates – and most of the laws come from bills initiated by the executive – with high rates of dominance (FIGUEIREDO and LIMONGI, 1999FIGUEIREDO, A. and LIMONGI, F. (1999), Executivo e Legislativo na Nova Ordem Constitucional. Rio de Janeiro: Editora FGV. 232 pp.). The explanations for this success and dominance are based on presidential powers – powers over agenda and discretionary control of appointments and budget (AMORIM NETO, 2000AMORIM NETO, O. (2000), Gabinetes presidenciais, ciclos eleitorais e disciplina legislativa no Brasil. Dados – Revista Brasileira de Ciências Sociais. Vol. 43, pp. 479-519.; AMORIM NETO, COX, and MCCUBBINS, 2003AMORIM NETO, O.; COX, G. W. and McCUBBINS, M. D. (2003), Agenda power in Brazil’s Câmara Dos Deputados, 1989-98. World Politics. Vol. 55, Nº 04, pp. 550-578.; FIGUEIREDO and LIMONGI, 1999FIGUEIREDO, A. and LIMONGI, F. (1999), Executivo e Legislativo na Nova Ordem Constitucional. Rio de Janeiro: Editora FGV. 232 pp., 2008FIGUEIREDO, A. and LIMONGI, F. (2008), Política orçamentária no presidencialismo de coalizão. Rio de Janeiro: Editora FGV. 184 pp.; MELO and PEREIRA, 2013MELO, M. A. and PEREIRA, C. (2013), Making Brazil work: checking the president in a multiparty system. London: Palgrave MacMillan. 224 pp.; PEREIRA and MUELLER, 2002PEREIRA, C. and MUELLER, B. (2002), Comportamento estratégico em presidencialismo de coalizão: as relações entre Executivo e Legislativo na elaboração do orçamento brasileiro. Dados – Revista Brasileira de Ciências Sociais. Vol. 45, Nº 02, pp. 265-301.). In this context, in which the explanations for governability focus exclusively on the ability of the executive to promote it, doubts emerge as to the ability of the legislative branch to fully exercise its functions. This is particularly true when dealing with its main function, which is to assess, debate and produce laws.

However, what is the real ability of the legislature to affect the content of policies in an unbalanced context between the branches? Is the ability of the legislative branch to deliberate and modify executive proposals affected by the legislative powers of the president? These questions guide this article. It will be shown that the extensive executive authority is not an impediment to the legislature action, to the extent that none of the resources available to the executive are capable of inhibiting the active participation of the legislative branch in the process of making laws. To demonstrate this point, I will analyze amendments and substitute bills made by Congress to ordinary bills (projetos de lei ordinária – PL), provisional decrees (medidas provisórias – MPV) and complementary bills (projetos de lei complementar – PLP) sent by the executive during the governments of Fernando Henrique Cardoso and Luiz Inácio Lula da Silva, which were partially or totally vetoed¹ 1 Matters that had been vetoed because they make explicit a possible conflict between the executive and legislative branches were chosen. It is important to highlight that, among these matters, there are central themes to Brazilian politics in recent years, such as: bills addressing minimum wage, the Law of Fiscal Responsibility, income tax and other taxes, bills that reorganize the careers of public servants, the ProUNI bill, and Minha Casa, Minha Vida, (My House, My Life), among others. For a complete list, see Apendix A published online (see http://bpsr.org.br/files/archives/Dataset_Freitas). . Therefore, what is intended to show is that the high rates of success and executive predominance do not explain the high degree of participation of the legislative branch in the legal output of the country.

This article is divided into five parts. In the first part, brief considerations are made on the balance of power between the executive and legislative branches, attempting to show that a strong executive does not necessarily imply a weak legislature. The second part presents the methodology used to evaluate the modifications made by the legislature. The third part presents the impact of the modifications made by legislative amendments and substitute bills in the laws. The fourth part shows that the powers of agenda of the executive and party leaders do not affect legislature ability to change executive proposals. Lastly, final considerations are given.

Strong executive, weak legislature?

The presidency in Brazil has broad legislative powers, it is able to issue provisional decrees (medidas provisórias) which have the force of law from the moment of publication and can request urgency for the bills that it proposed, significantly limiting the time for evaluating its proposals by the legislature (45 days in each legislative house). The presidency also has exclusive initiative over budgetary and administrative matters, that is, it has agenda powers. The president also has negative powers and is able to partially or totally veto bills ratified by the legislative power. Furthermore, Brazilian presidents have a wide range of resources at their disposal, discretionary control over the budget² 2 This power was reduced since the introduction of the imposed budget, which requires the president to implement the parliamentary amendments to the budget. The recent change, introduced for the 2014 budget through the budgetary law (Law nº 12,919), confirmed by Constitutional Amendment nº 86, still does not allow us to veriy its effects. and over appointments.

This broad set of presidential powers is associated with high rates of success and the power this branch has in the approval of its agenda. However, it raises important considerations about the ability of the legislature to fully exercise its role. The discussion about the medidas provisórias (provisional decrees – MPVs) exemplifies this issue. Considered by Amorim Neto and Tafner (2002AMORIM NETO, O. and TAFNER, P. (2002), Governos de coalizão e mecanismos de alarme de incêndio no controle legislativo das medidas provisórias. Dados – Revista Brasileira de Ciências Sociais. Vol. 45, Nº 01, p. 05-38., p. 28) as an anti-majority instrument, the MPVs are always mentioned when it comes to the low capacity of the legislative branch facing the powers of the president. It is not without reason that a large part of the debate around this topic is dedicated to understanding whether or not the legislative branch abdicates its authority to legislate in favor of the executive (AMORIM NETO and TAFNER, 2002AMORIM NETO, O. and TAFNER, P. (2002), Governos de coalizão e mecanismos de alarme de incêndio no controle legislativo das medidas provisórias. Dados – Revista Brasileira de Ciências Sociais. Vol. 45, Nº 01, p. 05-38.; FIGUEIREDO and LIMONGI, 1997FIGUEIREDO, A. and LIMONGI, F. (1997), O Congresso e as medidas provisórias: abdicação ou delegação? Novos Estudos Cebrap. Vol. 47, pp. 127-154.). The conclusions that the authors reach are diverse, and substantives as usurpation, abdication and delegation are used to refer to the modus operandi in the relationship between the executive and legislative branches.

It is common that, referring to governability, the focus falls on the means used by the executive to annul the private interests that would guide the action of the legislature. If taken to the extreme, the result would be that the powers of the president would take away any possibility of legislative action. Even when the explanation has partisan grounds, it tends to be concentrated on how the centralization of the legislative process, in the hands of the party leaders would reduce the fragmented and personal incentives coming from the electoral system. Thus, even though governability is recognized as a fact today, it seems to be assured in spite of the legislative branch by means of its neutralization.

Nevertheless, are strong presidents necessarily accompanied by weak legislatures? How, and to what extent, do presidential powers reduce the ability of the legislative branch? By centering on the success and dominance of the executive, the discussions around the issue of governability leave aside an important aspect of the functioning of the system, which is the ability of the legislature to fully evaluate the proposals of the executive. It has been deducted that the strengths of the powers are correlated. Thus, if the executive is powerful, the legislature is weak, and vice-versa. The same may be said of success: if the executive is successful, this means that the legislature is not³ 3 Obviously, it must be considered that the resources of a State are limited, and a large part of the work of the representatives is precisely to decide how these resources will be distributed. Furthermore, if the total amount of resources is limited, then in order to make new policies it must be determined which policies will not be implemented. The argument that I seek to develop does not harm this logic. The point is that, given that none of the powers can make policy unilaterally, the legislative and the executive negotiate to arrive at a minimum consensus that permits some policy instead of no policy. .

However, this is done without proper consideration of how the presidential powers in fact minimize the ability of the legislative branch. Do the three types of presidential power – i.e., power of veto, control of resources and powers of agenda – affect the decision-making ability of the legislative branch?

The resources at the disposal of the executive, in the Brazilian case, discretionary control of appointments and budget, change the incentives for the legislators, facilitating the process of coordination in order to make it easier for the executive to build a systematic base of support in the legislative. This encourages, for example, the formation of coalitions that support the executive. However, the presence of incentives does not imply that some legislators cannot choose the opposition. For certain parties, the incentives to remain in the opposition will be greater than those provided by the resources of the executive. On the other hand, even though this is an important instrument for the process of negotiation, these resources do not hinder the evaluation and modification of bills coming from the executive. At most, they limit with whom this branch will negotiate in forming its agenda (FREITAS, 2013FREITAS, A. (2013), O presidencialismo da coalizão. PhD thesis. Faculdade de Filosofia, Letras e ciências Humanas. Programa de pós-graduação em Ciência Política. Universidade de São Paulo.).

Veto power, insofar as it leads the legislature to anticipate the preferences of the executive, fosters negotiation between the powers, particularly where the partial veto is permitted. However, here it should also be considered that unilateral action is not possible; or, if it is, only the legislature can pass legislation without the consent of the executive, but this would require the formation of large majorities⁴ 4 Also, the legislative bills and the proposed changes within this branch are subject to presidential veto. Unless they have raised a sufficient majority to overturn the presidential veto, the legislative branch has no way to act unilaterally. . The legislators can anticipate that a given modification will be vetoed, but they cannot assume that they will simply ratify a proposal much different from their preferences. It is more likely that, in the process of anticipation, both powers will moderate their proposals in order to make them viable.

The legislative powers of the president, on the other hand, interfere in the legislative agenda. Agenda, in political science, is always considered to have a double meaning – timing and content (FIGUEIREDO and LIMONGI, 1999FIGUEIREDO, A. and LIMONGI, F. (1999), Executivo e Legislativo na Nova Ordem Constitucional. Rio de Janeiro: Editora FGV. 232 pp.). However, the powers of the president are not able to define the content itself, but they can define the topic. Powers of agenda, as provisional decrees or of constitutional urgency, can determine when a bill will be voted on, and they are able to determine what topic the legislature will debate, even if the legislature does not want the topic to be debated. Nevertheless, different from what is normally assumed, powers of agenda do not imply that the executive has control of the content to the extent that the executive is unable either to prevent changes thereto or to avoid rejection of the bills. Powers of agenda, like the other powers of the Executive, change the incentives. Provisional decrees, for example, can change the incentives for the legislators to the extent that they may provide cover for them to support the executive (FIGUEIREDO and LIMONGI, 1999FIGUEIREDO, A. and LIMONGI, F. (1999), Executivo e Legislativo na Nova Ordem Constitucional. Rio de Janeiro: Editora FGV. 232 pp.; HUBER, 1996HUBER, J. D. (1996), Rationalizing parliament: legislative institutions and party politics in France. Cambridge: Cambridge University Press. 232 pp.). Provisional decrees also create a legal problem when they are altered or rejected. The legislature is responsible to solve this kind of problem, which reduces the incentives for legislative action. However, these decrees are still neither rejection- nor change-proof.

In theory, none of these mechanisms available to the executive reduce the ability of the legislative branch to change or reject what is being proposed. As Immergut (1996)IMMERGUT, E. M. (1996), As regras do jogo: a lógica da política de saúde na França, na Suíça e na Suécia. Revista Brasileira de Ciências Sociais. Vol. 30, Nº 11, pp. 139-163. reminds us, institutions provide the context in which politicians will act. They define incentives and shape strategies, but do not determine results.

None of the powers of the president permits this branch unilateral action because the legislature approval is always necessary. The opposite is also true: the legislature is not able to approve bills too far from the preference of the executive, unless they are able to amass a vast majority.

The presidential powers do not imply any weakening of the legislative branch. Strong presidents do not imply weak legislatures because the powers given to one or another branch are not complementary. The same may be said about success: the president's success does not imply failure of the legislature. Even though presidential systems are characterized by independence between branches, what guarantees the mutual oversight capacity of the branches is just a high degree of overlap of their functions, which demands a high degree of coordination. The powers of the president, together with the centralization of the legislature around the party leaders, promote the capacity for coordination between the branches, as Figueiredo and Limongi (1999)FIGUEIREDO, A. and LIMONGI, F. (1999), Executivo e Legislativo na Nova Ordem Constitucional. Rio de Janeiro: Editora FGV. 232 pp. point out, diminishing the fragmenting effects of the electoral arena. Yet, this does not imply weakening of the legislative branch to the extent that president powers are not able to hinder the action of the legislators. Institutions provide incentives and regulate the strategies available to the actors, but do not define the results.

However, what does alteration of the executive's bill mean, from the point of view of presidential success? Bonvecchi and Zelaznik (2012)BONVECCHI, A. and ZELAZNIK, J. (2012), Measuring legislative input on presidential agendas (Argentina, 1999-2007). Journal of Politics in Latin America. Vol. 03, Nº 03, pp. 127-150., dealing with presidential success, claim that the analysis in search of evaluating success by counting both the bills introduced and signed into law are insufficient because they do not take into account to what extent the bills are changed. Barrett and Eshbaugh-Soha (2007)BARRETT, A. W. and ESHBAUGH-SOHA, M. (2007), Presidential success on the substance of legislation. Political Research Quarterly. Vol. 60, Nº 01, pp. 100-112. and Eshbaugh-Soha (2010)ESHBAUGH-SOHA, M. (2010), The importance of policy scope to presidential success in Congress. Presidential Studies Quarterly. Vol. 40, Nº 04, pp. 708-724. take a similar path and show that success may be better understood when the content proposed by the executive that remains after passing through the legislature is examined. Martin and Vanberg (2011)MARTIN, L. W. and VANBERG, G. (2011). Parliaments and coalitions: the role of legislative institutions in multiparty governance. Oxford: Oxford University Press. 208 pp., looking at parliamentary countries, point out that modifications to executive bills are consequences of the negotiations inside the coalition. For the latter authors, the legislature would be a privileged arena for the members of the coalition to control each other, avoiding the pitfalls imposed by the delegation.

Stepan (1999)STEPAN, A. (1999), Para uma nova análise comparativa do federalismo e da democracia: federações que restringem ou ampliam o poder do demos. Dados – Revista Brasileira de Ciências Sociais. Vol. 42, Nº 02. and Ames (2003)AMES, B. (2003), Os Entraves da democracia no Brasil. Rio de Janeiro: Editora FGV. 412 pp., looking at the Brazilian case and taking the law of anticipation of reactions as their premise, claim that the success of the president in Brazil is overestimated primarily because the executive branch would not send bills that would probably be rejected. Second, because bills sent are so heavily altered that they would lose their original features. From this point of view, the changes would be unsuccessful for the executive. However, if this were the case, why would the president not veto such changes?

The law of anticipation of reactions states that the president will anticipate negative reactions to the bills, possibly not sending a bill to the legislature in anticipation of its rejection. The same may be said of legislators about changing executive bills; anticipating that the changes will be vetoed, the legislators may prefer not to make them. What the authors do not consider is that the legislators and/or the president may insist on a bill that will be rejected in order to place the blame for inaction on the other branch. They also do not consider that policies may be used to force a negotiation. Thus, the president may send a bill to the legislature, radicalizing their preferences, in order to leave room for negotiation. Finally, it must be considered that the anticipation may fail, or that the context may change and the preferences that seemed evident may change during the process (DINIZ, 2005DINIZ, S. (2005), Interações entre os poderes executivo e legislativo no processo decisório: avaliando sucesso e fracasso presidencial. Dados – Revista Brasileira de Ciências Sociais. Vol. 48, Nº 02, pp. 333-369.; LIMONGI, 2006LIMONGI, F. (2006), A democracia no Brasil: presidencialismo, coalizão partidária e processo decisório. Novos Estudos – CEBRAP. Vol. 76, pp. 17-41.).

The fact is that if the anticipation were perfect, the executive would never have a bill rejected or altered. The same goes for the legislature, there would be vetoes neither to its bills nor to changes made to executive bills. Given that the anticipation is not perfect, a negotiation stage is expected. The legislature is then a privileged arena in which the two branches, through the alteration of bills, form the consensus necessary for the adoption of policies.

I believe that changes made by the legislature to executive bills show, as pointed out by Barrett and Eshbaugh-Soha (2007)BARRETT, A. W. and ESHBAUGH-SOHA, M. (2007), Presidential success on the substance of legislation. Political Research Quarterly. Vol. 60, Nº 01, pp. 100-112., that the executive must negotiate and this is not only obvious; it is expected in a democracy. Thus, changes to bills are not necessarily failures but a natural process of negotiation. If the bills are modified without considering the preferences of the executive, then they will be vetoed. In addition, the legislators will be able to anticipate this action and moderate their amendments. If this were true, the failure of the executive regarding a certain bill would be a veto that is then overturned. Otherwise, we would only have moderation to the point where the preferences of the executive and the majority of the legislature are considered.

With this in mind, the aim of this study is not to discuss how successful the executive is, but to show that the legislative branch has an important role in the shaping of laws. The issue to be considered is: when the executive sends a bill to the legislature, an intense process of negotiation about that bill begins, which may involve modification of the initial proposal.

If this were true, then despite the powers of the executive, the legislature neither abdicates nor delegates its legislative capacity to the executive. It fulfills its role with regard to the production of laws. Therefore, the direct relationship between strong legislative powers of the president and weakness of the legislative branch is not necessarily true.

Methodology

When a bill begins its proceedings, the legislators, deputies and senators, each in their own part of the legislative process, may present amendments to the bill: insertions, withdrawals, modifications, substitutions or combinations. Still, the rapporteur may present a substitute bill, that is, a new bill to substitute the prior text. In the substitute bill, the rapporteur may withdraw part of the original bill, or may rewrite it entirely, incorporating amendments of his own or from other legislators.

Legislators have few restrictions with regard to the presentation of amendments. The restrictions that do exist relate to the moment at which the amendments may be presented. Even those bills with extraordinary processing rituals, bills with urgency or provisional decrees are subject to amendments from all legislators.

For bills with ordinary processing, any member of the committee that is going to debate them may present amendments. After offering the report of the committee within a limit of five working days – for the Senate – or five sessions – for the Chamber – any legislator may present amendments. These amendments are sent back to the committees for the rapporteur to offer their report on them. For bills with extraordinary procedures⁵ 5 Code bills, bills with urgency, MPVs, or bills that pass only in the committees (poder terminativo ou conclusivo). , any legislator may present amendments in the committees before they have deliberated and offered report on the matters.

Given the institutional rules, the costs of presenting amendments are very low. Therefore, any deputy or senator unhappy with a bill sent by the executive may propose modifications thereto. They may propose, but not necessarily get it approved. Approval is a second step, which demands coordination and the forming of majorities. However, how are legislative changes measured?

Analysis used to understand the process of alteration in the legislature, or even the constitutional text, usually use the count of articles (MARTIN and VANBERG, 2011MARTIN, L. W. and VANBERG, G. (2011). Parliaments and coalitions: the role of legislative institutions in multiparty governance. Oxford: Oxford University Press. 208 pp.), words, or amendments (FREITAS, 2010FREITAS, R. (2010), Poder de agenda e participação legislativa no presidencialismo de coalizão brasileiro. Master's thesis. Faculdade de Filosofia, Letras e ciências Humanas. Programa de pós-graduação em Ciência Política. Universidade de São Paulo.; SILVA and ARAÚJO, 2012SILVA, R. S. and ARAÚJO, S. M. V. G. (2012), Os agenda holders no Congresso Nacional. Paper presented at the I Seminário Internacional Instituições, Comportamento Político e Geografia do Voto, Brasília.). Martin and Vanberg (2011)MARTIN, L. W. and VANBERG, G. (2011). Parliaments and coalitions: the role of legislative institutions in multiparty governance. Oxford: Oxford University Press. 208 pp., for example, try to measure the proposed alterations to a bill by comparing the number of original articles with the number of final articles. R. Freitas (2010)FREITAS, R. (2010), Poder de agenda e participação legislativa no presidencialismo de coalizão brasileiro. Master's thesis. Faculdade de Filosofia, Letras e ciências Humanas. Programa de pós-graduação em Ciência Política. Universidade de São Paulo. verifies which amendments were approved and quantifies the modified bills. Silva and Araújo (2012)SILVA, R. S. and ARAÚJO, S. M. V. G. (2012), Os agenda holders no Congresso Nacional. Paper presented at the I Seminário Internacional Instituições, Comportamento Político e Geografia do Voto, Brasília. analyze the attractiveness of a bill from the ratio between the number of amendments presented and the number of articles in the original bill.

The authors cited above show the limitations in their methods, which lose many of the nuances of the legislative process. Comparing the number of original and final articles of the bills does not allow evaluation of the real extent of alterations. As an example, one can imagine a bill to which two articles were added, and two others were removed. The comparison between the original and final bills would have, as a result, zero modifications. That is, it would seem that the legislature made no alterations when, in fact, it did. This methodology also does not allow the identification of who the authors of each of the modifications, or their parties, are⁶. This, in turn, makes it impossible to see the position of the actors in forming the necessary consensus.

If, on the one hand, the analysis of the number of amendments approved – that is, inserted into the final text of the law – allows the identification of the author of the changes, on the other hand it does not allow an evaluation of the changes the bill went through. An amendment may change a bill a little bit or it can make significant changes. A modifying amendment, for example, may make only small changes in the text without, however, changing the meaning of what is being proposed, that is, without changing it in fact. An amendment is a unit that does not allow comparison and does not measure what one wants to measure; that is, how much the original bill is changed. What is needed, then, is to deconstruct the original bill into basic units, comparing them to the bill that was approved.

Thus, the methodology developed by Arantes and Couto (2009ARANTES, R. and COUTO, C. (2009), Uma constituição incomum. In: A Constituição de 1988: passado e futuro. Edited by CARVALHO, M. A. R. de. São Paulo: Hucitec, Anpocs, pp. 17–51., 2010ARANTES, R. and COUTO, C. (2010), Construção democrática e modelos de Constituição. Dados – Revista Brasileira de Ciências Sociais. Vol. 53, Nº 03, pp. 545-585.) for the constitutional analysis was used. The authors separated the constitutional text into provisions, which are:

The same method was used in the present study. Each matter submitted by the executive was deconstructed into provisions, in an attempt to identify the smallest possible legal unit containing legal elements.

Law nº 9,252 from 1995 is used in Figure 01 above to illustrate it. This law has 04 provisions. The first provision is the first article of the law; the second provision is the second article. The third provision is formed from the combination of the third article and subsection I; in this case, only when the third article is combined with its respective subsections is the complete legal meaning of this text apparent. The same holds for the fourth provision, formed from the combination of the third article with subsection II. The fourth article has no legal content; as such, it is not counted as a provision. In this way, the law is deconstructed into the smallest unit that still has legal content.

The bill sent to the executive, thus unraveled, was compared to the bill that came from the legislature, deconstructed by the same method, for presidential sanction. Hence, it was possible to identify, one by one, the changes made by the legislature.

As the modifications were identified, each of them was compared to the amendments and substitute bills introduced in the House of Representatives and the process was repeated in the senate. Finally, the final bill that left the congress was compared with the presidential vetoes⁷ 7 Thanks to Andréa Junqueira Machado for her help with data collection. . Thus, it was possible to identify the exact points in the bills where there were divergences between the executive and the legislative branches.

Take, for example, Law nº 11,648 from 2008⁸ 8 The law that recognizes Trade Unions in Brazil as nationwide established private entities representing the workers. . In its final version, the law had seven articles. The original bill, sent to the Congress, had five articles. When submitted to the president to sanction, it had eight articles, with one of them having been vetoed, which left the final count at seven articles.

If we only take the number of articles into account, we would affirm that the legislature, by inserting three articles, significantly changed the law. However, this would be a hasty analysis.

The cited law has numerous subsections, paragraphs and subparagraphs. If we classify the law by the number of provisions, we can see that the original law has 36 provisions. The congress changed, added and/or removed seven provisions of the matters. The president vetoed one provision. The separation of the smallest unit of the matter permits the correct sizing of the contribution of each of the branches to the law. It also permits a more detailed and precise analysis of the modifications, avoiding the difficulties noted in studies that rely on the count of articles, words and even amendments.

I suggest that the analysis based on provisions is the most appropriate way to assess not only the alteration itself, but also its impact on the text without losing the authorship of each change. It thus permits not only the reconstruction of the path of each bill within the legislature, but also understanding how the legislators and the executive construct the necessary consensus that leads to the high levels of discipline and approval on the floor.

The degree of detail in the methodology makes it impossible to analyze all proposals sent by the executive. Thus, the universe of analysis herein was the executive bills that underwent some type of veto⁹ 9 Thanks to Patrick Silva for the idea. , either partial or total. These are certainly the most conflicting matters, or the only ones in which the conflict between executive and legislative branches is explicit and measurable. As such, it is possible to evaluate not only whether the legislature modifies the bills, but also to understand how the executive reacts to these modifications. Matters originating in the executive branch, presented and approved between 01/01/1995 – the beginning of Fernando Henrique Cardoso's government – and 31/12/2010 – the end of Luís Inácio da Silva's government – were analyzed. The normative species that compose the sample are: MPVs issued after 2001 (provisional decrees), PLs (ordinary bills) and PLPs (complementary bills)¹⁰ 10 PLPs are projects that regulate constitutional amendments. MPVs prior to the modification of the processing procedure, regulated by EC 32, were not included due to the difficulty imposed by the very processing procedure of the materials, since the possibility of reissuing the measure makes it impossible to identify if the author of the modifications is the President or the Legislature, given that the executive can modify the MPV when he reissues it. PLNs dealing with the federal budget were also not included because their effect ends after their application. PECs (proposed amendments to the Constitution) do not enter because they cannot be vetoed. .

Before analyzing the data, it is necessary to call attention to the terms used in the analysis from this point forward. The primary unit of analysis comprises the provisions; as stated above, I defined provision as the smallest unit of a bill (or of a law) which still has legal content.

The provisions were recorded in each of the stages of the legislative process and, therefore, it is necessary to differentiate proposal from result. When the executive sends a bill to the legislative branch, it has a specific number of provisions that reflect the executive's preference. This bill may be altered within the legislature, as mentioned above, through the introduction of amendments or substitute bills. The proposed changes, in turn, are not limited to the inclusion of new content. Thus, the provisions were divided into:

Additions – proposing new content to a specific bill, that is, proposing new articles, subsections, paragraphs, sections;

Modifications – modifying existing content in the original bill, that is, reformulating a specific provision in such a way as not to change its character;

Suppressions – suppress provisions, subsections, paragraphs or sections of the original bill.

In other words, when the provisions proposed by the executive arrive in the legislature, this latter may not only add other provisions, it may also remove or modify some of the provisions proposed by the executive. Having passed through legislative treatment, if approved, the bill is sent for presidential sanction. The president may veto the content proposed and/or modified within the legislature, as well as veto articles proposed by himself. Only after presidential veto there is a final result. With the law enacted, it is possible to trace the inverse path and verify how many provisions proposed by the executive and how many proposed by the legislature are contained in the legal text.

Therefore, separating preferences from results, throughout the present study and in the tables, reference is made to the articles in terms of five categories: executive provisions, legislative provisions and provisions in the law. This last category, however, involves two others: executive and legislative provisions in the legal text.

There are, then:

Executive provisions – found in the bill sent by the executive to the legislature, therefore constituting the preferences of the executive, counted before any legislative action;

Legislative provisions – added, removed or modified within the legislature. They constitute a means of taking into account the different types of changes that the legislature may make, counted before a veto. That is, they indicate the preferences of the legislature;

Provisions in the Law – final articles, with legislative actions and presidential vetoes removed. They constitute the result, that is, those articles that survived the entire process and passed into law. To be clear, the provisions in the law are those that came from the executive and were not altered in the legislature, counting the additions made by the legislature minus the presidential vetoes. These may be divided into two categories:

Executive provisions in the law – found in the original bill of the president, suffered no changes within the legislature and were not vetoed;

Legislative provisions in the law – underwent changes, were added to or modified within the legislature¹¹ 11 Note that, in this category, obviously those provisions removed are not recorded, as they are not part of the legal text. but were not vetoed by the executive.

In Table 01, we can observe the universe of analysis. This makes it possible to visualize the number of bills analyzed, separated by types. As can be seen, 179 bills were analyzed, of which 178 became law and one was totally vetoed. This set of bills represents about 15 percent of all bills sent by the executive to the legislature during this period. Although the number of bills in each group varies, the samples are reasonably balanced when separated by normative types and represent 15 percent of all MPVs, 12 percent of the PLs and 21 percent of the PLPs sent.

In Table 01 it is also possible to observe the number of executive provisions, that is, those that appeared in the original 179 bills sent by the executive; the total number of legislative provisions, which are those modified by the Congress; and the total number of provisions of the law, which are those that remained in the text transformed into a legal standard. However, as seen above, the provisions presented in this table are an indication of activity but not of the contribution to the legal text. In other words, it is not possible to add the provisions of the legislature to those of the executive and to ascertain the total number of provisions in the law. This is because the provisions of the legislature are the sum of the various types of action, including the removal of provisions from the bill from which the law originates.

The number of legislative changes itself shows the extent of participation of this branch in defining policies. It is also possible to see that the number of altered provisions is high, regardless of the type of proposal. Even the MPVs, in which legislators have little time to evaluate the bill, do not pass unscathed through the legislative process. This point will be discussed below, since what matters here is to recall what the universe of analysis is and, as shown in Table 01, that legislative alteration does not always add articles to the executive bill.

Thus, the legislature proposes new content (additive provisions), vetoes the content (suppressive provision), or modifies what the executive proposed (modified provisions). A large part of the legislative provisions (7,638) are additive. Those suppressed are also not few, they represent 1,901. Finally, there are 1,191 modified articles.

It must also be emphasized that the distribution of provisions is not uniform. There are outliers in all types of bills, whether in the number of articles sent from the executive to the legislature or in the number of changes made in the legislature. Chart 01 shows the distribution of provisions in the MPVs, from the executive and legislative branches. As noted, most MPVs arrive in the legislature with between 02 and 200 provisions, which represent 80 percent of the MPVs analyzed. There are, however, two that arrived with about 800 provisions and, in one very extreme case, with 2,147 provisions. The provisions of the legislature in the MPVs range from 01 to 584, but in nearly 90 percent of the bills the changes go from 01 to 100 provisions.

Chart 02 shows the distribution of the number of executive and legislative provisions per bill in the PLs. The greatest concentration of executive provisions is between 03 and 100 provisions. Those from the legislature are between 0 and 25. There are few outliers, where the provisions number more than 400. It is worth noting here that only 04 bills in the sample received no change within the legislature, and they were all PLs.

Finally, Chart 03 shows the number of provisions in PLPs. Among the PLPs, the concentration of executive provisions ranges between 24 and 150; there is also an exceptional case starting its processing in the executive branch with 694 provisions. Those from the legislature are concentrated between 22 and 250, with one extreme case, a bill that received more than 1,114 changes.

Impact of the legislative branch on legal output

As shown, the number of changes made in the legislature is not small. However, what is the impact of these changes on the legal output? After all, it makes no sense to concentrate on the legislative changes if they are insignificant. Therefore, the first question is how much do the legislative modifications actually affect the legal output? To answer this question we need to understand how much each branch contributes to the legal text, and for that I use what I call rate of contribution.

The first step in understanding the calculation of this rate is to separate the actions of each branch in the process of shaping a bill into law. Beginning with the executive, this branch sends a bill containing a certain number of legislative provisions. At the end of the process, it may veto as many provisions as the bill sent for sanction contains. Thus, the actions of the executive on the legal text may be divided in two:

Executive provisions (P_E) – constitute the initial input, that is, the provisions present in the bill sent to the legislature;

Vetoed provisions (P_V).

The vetoes, in turn, may influence legislative provisions and/or the executive's own provisions. The adopted methodology allows identification of whether the executive vetoes provisions altered within the legislature, or if it vetoes provisions found in the original bill that were not altered within the legislature. Thus, vetoed provisions may be deconstructed according to whomever proposed them, that is:

P_V = V_E + V_L

Where:

V_E = executive vetoes of provisions found in the original bill;

V_L = executive vetoes of legislative provisions.

The legislative branch, in turn, may add new content to the bill – amending provisions: it may suppress content from the bills – suppressive provisions; and/or it may modify the content of the bills – modified provisions. Thus, the actions of the legislature on the legal text may be made through:

Added provisions – P_La

Suppressed provisions – P_Ls

Modified provisions – P_Lm

Therefore, the total number of provisions in the legal text or the provisions transforming in law (TP_L)¹² 12 Note that, in the calculation of TPL, PLm is contained in PE below. It is taken from the contribution of the executive but, from the point of view of formalization, including it in this step would imply taking it from the executive in order to add it to the legislative. It would be, thus: TPL = (PE - PLm) + PLa + PLm - PLs - PV. In this case, PLm it is not needed since it is added and subtracted in the representation. is represented as:

TP_L = P_E + P_La - P_Ls - P_V

The percentage rate of contribution of the executive is represented as:

C_E = ((P_E - P_Ls - P_Lm - V_E)/ TP_L)*100

Finally, the percentage rate of contribution of the legislature is:

C_L = ((P_La + P_Lm - V_L)/ TP_L) *100

For example, one can imagine that a bill with 24 provisions is sent by the executive. When it arrives in the legislature, 08 provisions are added, 01 provision is removed and 03 others are modified. The bill is sent for presidential sanction and the president vetoes 02 executive provisions and 04 from the legislature. In this case, the total number of vetoes would be:

P_V = 02 +04 = 06

The total number of provisions in the legal standard would be:

TP_L = 24 + 08 - 01 - 06 = 25

The rate of contribution of the executive would be:

C_E = ((24 - 01 - 03 - 02)/25)*100 = 72

The rate of contribution of the legislature would be:

C_L = ((08 + 03 - 04)/25)*100 = 28

Table 02 shows the average contribution of the executive and of the legislative branches to the promulgation of laws. We can see that the contribution of the legislature is significant: on average, 36 percent of the text of the laws is produced within the legislature. It is also noted that this contribution varies significantly among the types of bills. In the PLPs, the average contributions are inverted, with the executive contributing 33.5 percent of the legal content and the legislature 66.5 percent.

Again, it is worth stressing that even the provisional decrees (MPVs) – seen as the sole expression of the will of the president, as a weapon that he would use to impose his will on the legislature – are objects of significant contribution of the legislative branch. The legislature is responsible for 40 percent of the text of the laws produced by this type of bill. That is, the president does not govern alone, nor is he able to impose his will on the legislature, not even in editing provisional decrees. The legislative branch, even under these extreme conditions, in no way is led to inaction or submission.

There is an enormous variation in the rate of contribution of each branch when the bills are deconstructed and taken separately. In Chart 04, it is possible to see this variation. What can be seen is that the contribution of the two branches ranges from 0 to 100 percent. There is no doubt that, in quantitative terms, the executive contributes more. However, the contribution of the legislature is far from what is expected in a context where the president possesses an immense range of powers.

These data allow us to realize, in a clear way, that although most of the laws come from executive bills, the legislature does not exempt itself from its role. It deliberates on the bills and alters them when it considers that it is necessary, and the alterations have considerable impact on the legal system of the country.

Impact of the powers of agenda on the changes

In this section, the impact of the characteristics of the types of bills and the powers of the agenda – of the executive and the partisan leaders – will be analyzed regarding the capacity of the legislature to make changes. The three types of bill that are being analyzed have differences regarding their content and their procedures. PLPs regulate the constitution and, because of this, require a qualified quorum in order to be approved; that is, the support of an absolute majority of legislators. Even the changes or the amendments presented to this type of bill require a qualified majority in order to be approved, which means that the cost of alteration is high. Bills of this type, at least at first sight, seem to be more complex, longer and, for this reason, they take more time in the legislature.

MPVs are urgent measures and, because of that, their processing is extremely fast: they only go through one special committee formed ad hoc, their vote in the plenary may be done by the symbolic or nominal method and their approval requires only that a majority of the ones present vote "Yes". PLs, in turn, take much more time than MPVs, but much less than PLPs. They pass through at least two committees and, like MPVs, do not require a special quorum in order to be approved. PLs and PLPs can be processed on an urgency basis, whether by request of the party leaders or of the executive branch. In the latter case, by a constitutional urgency request.

It would not be wrong to assume that the time, the complexity and the type of processing to which the matters are submitted affect the opportunity of the legislators to modify their proposals. It would also be reasonable to imagine that the initial input of the executive affects the number of changes: very large bills would offer more material to be modified.

In fact, based on the averages of the executive provisions in Table 03, there is a difference between the number of provisions sent according to the type of bill. PLPs arrive in the legislature with more than double the provisions of PLs, and about 50 percent more articles than MPVs. There are, respectively, 163.4, 73.1 and 103.5 provisions for each type of bill. However, as seen in the previous section, the distribution is not uniform and there are some outliers.

Among the MPVs, there are 12 outliers. The MPV¹³ 13 MPV 441 from 2008, transformed into Law 11907 from 2009. that draws more attention deals with the reformulation of the career plan and the bonuses of several public service agencies. This MPV was sent to the legislature with 2,147 provisions. Removing it from the sample, the average number of provisions in the original bill falls to 83, very much like the PLs. However, there are also 08 outliers among the PLs. The PL¹⁴ 14 PL 4,776 of 2005, transformed into Law 11284/2006. that provides for the management of public forests for sustainable production arrived at the legislature with 429 provisions, for example. Among the PLPs, there is only one outlier; the PLP was transformed into the Law of Fiscal Responsibility¹⁵ 15 PLP 18 of 1999 transformed into Complementary Law 101/00, known as the Law of Fiscal Responsability imposes limits on state and municipal expenditures. which was sent by the executive with 694 provisions. Due to these deviating cases, the medians are the best way to evaluate the data. There are, then, the medians of provisions from the executive being 29, 39 and 77 in the MPVs, PLs and PLPs, respectively. Nevertheless, PLPs arrive in the legislature considerably larger than the other two types of bills. However, the MPVs and PLs, while still close, invert their positions.

The bills analyzed received, on average, 60 alterations within the legislature. However, when we examine the provisions of the legislature, there are also cases deviating from the average. There are 05 cases among the MPVs. For example, the MPV¹⁶ 16 MPV 255 from 2005, transformed into Ordinary Law 11196/2005. that deals with personal Income Tax withholding received 584 alterations in the legislature. There are 07 outliers among the PLs, with the outstanding case being the one that deals with the restructuring of land and waterway transportation. Only this PL¹⁷ 17 PL1615 from 1999, transformed into Ordinary Law 10233/2001. received 641 alterations in the legislature. There is also only one case, identified above, among the PLPs: the bill that deals with the Law of Fiscal Responsibility, to which 1,114 alterations were made.

Therefore, when the alterations are examined, it is also appropriate to look at the medians. Upon separating the legislative provisions by type of bill, we see that the medians are 12.5, 24 and 115 in the PLs, MPVs and PLPs, respectively. As the PLPs stand out regarding the alterations, the cost of gathering majorities must be minimized then. The little time that legislators have to evaluate MPVs also does not seem to be a problem, since they are more altered than the PLs. The curiosity here is that PLs, which take longer than the MPVs as well as not needing nominal and qualified majorities to be approved, as do the PLPs, receive the fewest alterations.

There is a correlation¹⁸ 18 The correlation of the number of legislative provisions with executive provisions is significant at the 01 percent level, at the rate of 0.29; but, when separated by type of bill, the result is different. PLPs are significant at the1 percent level, at the rate of 0.93. In MPVs, the relationship is not significant. PLs are significant at the 01 percent level, at the rate of 0.53. , albeit small, between the size of the bill that is sent and the number of alterations; that is, between the number of provisions from the executive and the legislature. This can be seen in Chart 05. To allow visualization of the variables, their logs and not their values were used. What the chart shows is that the initial input from the executive impacts the legislative alterations in a moderate way. Although there were bills that were initially quite large, which received almost no legislative alterations, the inverse is also true; that is, there are bills sent by the executive with a small number of provisions that are greatly altered.

The processing times for each type of bill are also very different. PLPs are evaluated, on average, in 882 days; while PLs are processed, on average, in half that time, in 479 days. The average number of days for processing the MPVs, obviously, is the least of all, at 129 days¹⁹ 19 The average is slightly greater than what should be the maximum term of processing, which is 120 days. This happens because, when the National Congress goes on recess, the time of the MPVs stops counting. . However, will there be a relationship between the number of days of bills being processed with the number of alterations?

Chart 06 shows the ratio between the number of altered provisions and the days of processing in the legislature. What is observed is that, relating the legislative alterations to the days that each proposal spends in the legislature, the level of the alterations of the different projects is close and the averages of the ratios are approximately 0.42 for MPVs, 0.14 for PLs and 0.46 for PLPs. Therefore, seen in this way, the alterations of MPVs and PLPs are approximately of the same magnitude, and PLs remain the least altered²⁰ 20 The correlation between the two variables – legislative articles and number of days of processing – is neither significant with the grouped cases, nor with the separated by type of bill. .

Chart 06 clearly shows that the number of days is not an important variable for explaining alterations of the bills, not only by the lack of any correlation between the days of processing but also by the legislative changes to provisions as well. Again, although the legislature has very little time to evaluate the MPVs, this factor is not enough to hinder the deliberation and the evaluation thereof. Therefore, the debate about provisional decrees as an abdication or delegation of the legislative branch should be mitigated. It is worth pointing out that, although the legislature has handed the executive an extremely powerful instrument, the MPVs do not constitute a "blank check". Each MPV within the legislature is duly considered, there being no impediment to the action of the legislators regarding rejection, alteration or approval.

However, it cannot be inferred that this power of agenda, the provisional decrees, has no effect on the alterations. Table 04 shows the distribution of the provisions per type of bill and per type of alteration. It is evident that the type of bill matters to the understanding of the type of alteration. Even if, in general, most of the alterations proposed by the legislature are in the sense of including new content in the bills, there is significant variation in the types of alterations among the types of bills. While an important part of the changes made in the PLPs is suppressive, this type of alteration hardly ever occurs in the MPVs.

It should be remembered that MPVs change the status quo as they are being edited, that is, before the Congress becomes involved with them. As Figueiredo and Limongi (1999)FIGUEIREDO, A. and LIMONGI, F. (1999), Executivo e Legislativo na Nova Ordem Constitucional. Rio de Janeiro: Editora FGV. 232 pp. argued so well, by changing the status quo, the MPVs interfere directly with the preferences of the legislators. The same may be said about alterations to the MPVs. Since legislators are not dealing with a possibility, but with a policy already in practice, removing a provision from a MPV is more costly than from a PL or a PLP. Another factor that also should be considered is that when legislators alter a MPV, they must also deal with the consequences of the legal relationships resulting from the alteration thereof. The same thing happens when the congress rejects a MPV since, in this case, it is the responsibility of the congress to deal with the consequences of the legal modification caused by the rejection of the MPV.

In addition to the power to issue provisional decrees, Brazilian legislators have provided the executive with another type of agenda power. As stated above, the executive may unilaterally request that its bills receive urgency treatment; so, it does not depend on the consent of the legislature and may be made at any time, even after the bill has been sent to the legislature. The urgency request may be made for PLs and PLPs but, in practice, it has never been made for this latter type of bill. When the urgency request is made, there is no formal impediment to making the amendment, although it significantly reduces the processing time of the bills. If the urgency request is made, each chamber of the congress has 45 days to consider the bill; otherwise, as happens with the MPVs, it "locks up" the agenda. Thus, there is a reduction of the time available to evaluate bills, which may limit the opportunity to amend them²¹ 21 It is important to stress that constitutional urgency (urgency of the President) is very rarely requested. Between 1988 and 2011, it was requested for only 3.6 percent of the bills sent by the president to the legislature. Among the bills being evaluated here, of the vetoed bills, constitutional urgency was requested for 13 percent thereof, which is an indication of the relevance of that set. . Although there was no correlation between processing time and alterations, one can be more rigorous and check the effect of this power of agenda on the alterations.

Another power of agenda that could affect the opportunity for legislative alteration is the urgency of the leaders – this time, not a power of the President but of the party leaders within the legislature²² 22 For the Senate, see art 336 of the RISF (Regimento Interno do Senado Federal, Federal Senate Internal Rules). For the House, see Articles 154 and 155 of the RICD (Regimento Interno da Câmara dos Deputados, House of Representatives Internal Rules). .

Both Chamber and Senate approval of legislative urgency allows the withdrawal of the bill from the committees, without them having deliberated on the matter. Therefore, it could be assumed that bills following this process may, in theory, be withdrawn from the committees and have their processing time reduced, limit the process of deliberation and, consequently, the ability of the legislators to amend them.

In Chart 7 ²³ 23 All the PLPs were processed with legislative urgency and none was treated with presidential urgency, which makes it impossible to compare the effect of urgency on this type of bill. , the distribution of the provisions in the PLs can be seen, separated by the incidence or not of urgency and, in the cases where it occurs, by constitutional or legislative urgency. The medians of the provision in the bills without urgency, with executive and with legislative urgency are 04, 10 and 18, respectively. At first glance, bills with legislative urgency would be more altered than those with executive urgency, which, in turn, are more altered than bills without urgency. A more accurate test, the test of averages, revealed that there are no differences between the averages at the 95 percent Confidence Interval. Thus, what is verified is that the occurrence or not of urgency does not affect the number of alterations made in the legislature. In such aspect, the type of urgency notwithstanding, the legislators continue to exercise their power to deliberate and modify the proposals.

A consideration should be made about the presidential veto. The president vetoes 12.6 percent of the legislative alterations, 1,033 alterations of the 8,219 that could be vetoed²⁴ 24 Suppressions made by the Legislature are not subject to presidential veto. For more about the veto, see A. Freitas (2013). . None of the vetoed alterations were overturned in the legislature. I believe that this demonstrates that the alterations could not be seen as an executive failure. They are part of the process of negotiation that makes the laws the result of the majority decision.

Conclusion

The legislative branch has an important role in the production of laws. It is through deliberation within the legislature that bills are publicized, participation of organized civil society is guaranteed through public hearings, numerous actors gain their voice in the decision-making process. The number of preferences considered in each bill is increasingly necessary.

The question raised in this article is: are strong presidents necessarily accompanied by weak legislatures? Do the legislative powers of the president influence the ability of the legislative branch to evaluate proposals from the executive?

What we observe is that despite the role of the legislative branch in formulating policy being constantly questioned, we found that legislators deliberate and alter the proposals coming from the executive and their alterations have a significant impact on the Brazilian legal system. The legislature, through its alterations, is responsible for nearly 40 percent of the content of the laws examined in this study. The sample used, that is, the set of laws proposed by the executive that were vetoed, represents 15 percent of the total laws initiated by the executive branch in the governments of FHC and Lula. There are good reasons to assume that the participation of the legislature remains at similar levels in other proposals. However, this statement should be verified on an empirical basis.

Going beyond the alterations to bills from the executive, the legislature is responsible for 20 percent of the legal production. Taking the budgetary bills, the exclusive initiative of the executive, out of this account, this value increases to 37 percent. In summary: the participation of the legislature in altering the status quo is far from being small or insignificant. Furthermore, so far, the specialized literature has given little importance to this action.

The powers of agenda of Brazilian Presidents and party leaders should not be seen as obstacles to the ability of the legislature to alter proposals coming from the executive branch. The legislature makes more alterations to MPVs and bills being processed with urgency, where issues related to time and the procedures of processing would supposedly impose limits on the legislative action, than they make to PLs processed under normal conditions.

More important than the amount of legislative participation is the understanding that the powers of agenda of the executive and of the party leaders cannot hinder the active participation of the legislative branch. What this study shows is that there is no hindrance to legislative action. Thus, the dominance of the executive in the legal production should be relativized, understanding that such matters, although they originate in the executive, are not approved by circumventing the legislative branch; on the contrary, they have the agreement of that branch.

The negative characterization of Brazilian democracy, derived from the observation of the dominance that the executive has on the legal production, is not borne out in reality. Governability is not attained in spite of the legislature, but with its action. What is seen is that the legislative powers of the president are important in order to guide the agenda, to determine the topics that will be placed on the agenda. However, these powers are not sufficient to guarantee that the content is determined by the executive. The legislative strength of the president does not imply any delimitation of the legislative branch.

References

Publication Dates

History