Abstract
Studies on coalition management in presidential systems usually focus on two types of goods used by the president and formateur party to hold together coalitions: exchange goods (such as individual budget amendments) and coalition goods (such as ministries). This research notes analyzes, with an original dataset of party members and political appointees in Brazil, a different type of good: presidential political appointments. Our study shows that partisan political appointees vary greatly among Brazilian ministries and within them. We also found that there is a disconnect between how many seats a political party holds in Congress and the number of political appointment offices it controls. This has implications for the literature on bureaucracy and politics and the literature on coalition management.
Introduction
Coalition management in presidentialist open list-proportional representation systems is no simple task. The heterogeneity of parties and their ideological positions force presidents to negotiate with a myriad of political actors, in various ways, in order to create and sustain a legislative coalition that approves a large piece of the president's legislative agenda.
Studies on coalition management usually focus on two types of goods used by the president and formateur party to hold together coalitions: exchange goods (such as individual budget amendments) and coalition goods (such as ministries). Analysts of Brazilian presidentialism have made important contributions that allow us to understand how both these goods are used. There remains, however, a lacuna to be filled regarding second tier coalition goods. How are political appointments in the federal bureaucracy used in the negotiation with parties that belong to the coalition? This research note 1 answers this question with an original dataset of 21,681 members of the Brazilian federal bureaucracy that occupied political appointments in December 2010.
A first version of this essay was presented at the IPSA-ECPR Joint Conference in São Paulo (February 16 to 19, 2011). We thank participants for suggestions, as well as Susan Scarrow, Fernando Abrucio, Regina Pacheco, Steve Finkel and Barry Ames for comments on the project and/or the essay. Excellent research assistance was provided by Thiago Belmar and Samir Almeida. A special thanks goes to Helio Gurovitz and Guilherme Evelin for the initial push.
The main purpose of the research note is to investigate partisan bargaining in the occupation of these offices in Brazil. Much work has already been done regarding the first tier of executive payoffs in presidentialist systems, such as execution of congressional budget amendments (Alston and Mueller 2005; Figueiredo and Limongi 2005) and portfolio allocation (Amorim Neto 1994, 2006; Martínez-Gallardo 2010) for coalition members, but the puzzle regarding payoffs in the second tier of the bureaucracy still remains. Our study shows that partisan political appointees vary greatly among ministries and within them. In some ministries, political appointment offices are occupied by as much as 25 percent of partisan appointees, while in others this score is lower than 5 percent. We also found that there is a disconnect between how many seats a political party holds in Congress and the number of political appointment offices it holds.
Why does this discussion justify a research note? First of all, the two databases that have made the research possible – made available by the Portal da Transparência of the Brazilian federal government and the Brazilian Tribunal Superior Eleitoral – have only recently been made public, not having yet been used in full by political scientists. In second place, our findings are innovative and run counter to much of the literature on bureaucracy in Brazil, such as Evans (1995) and D'Araújo (2009); finally, analysts have cried out for descriptive data regarding the occupation of political appointment offices in Brazil (Ames 2003: 362; Figueiredo 2010: 212). This is the first quantitative effort, which does not rely on surveys with potential political appointees, in this direction.
There are four sections to this research note. The first briefly reviews the literature on politics and bureaucracy in the presidentialist systems of United States and Brazil. The second section quickly describes the basic functioning of the Brazilian federal bureaucracy and the original dataset constructed in order to assess the extent of political appointments in Brazil. The third section analyzes three questions. The first is how the government distributes patronage to political parties and whether this distribution is proportional to each party's parliamentary and ministerial power; the second is whether the allocation of political appointment offices to party members obeys patronage and/or policymaking logics; the third is about the possible use of very high-level offices in the same rationale as junior ministers are hired in parliamentary systems. Section four concludes.
Considerations on the relationship between the bureaucracy and the exercise of power date back to the nineteenth century in political science. Wilson (1887) was one of the first to discuss the issues of civil service reform and political appointments by the executive. According to him, bureaucratic administration and politics should be separate arenas.
Weber (1968) supported the same argument when discussing his three types of authority and domination: traditional, charismatic and rational-legal. The latter would be characterized by the exercise of power under limited boundaries delimited by legal rules through an unpoliticized bureaucracy, which would guide itself by statutes and not by particularistic political considerations. Fully-developed bureaucracies are characterized as “impersonal”, “objective”, and indispensable (Weber 1968: 956-1005).
This understanding of the interaction between bureaucracy and politics changed during the twentieth century. The government became bigger in size and scope of its functions and a different way of seeing the politicization of the bureaucracy emerged. In his assessment of the “administrative state”, James Q. Wilson (1975) states a separation between politics and bureaucracy but with the bureaucracy being the dominant force, establishing a stark contrast to Weber and Woodrow Wilson.
Although these assessments on politics and bureaucracy are important they do not necessarily try to explain why executive chiefs try to politicize the bureaucracy. Terry Moe (1985) was one of the first scholars to develop a rationale for presidential efforts to politicize the bureaucracy. Presidents would be driven by the formidable expectations that citizens, politicians, and the media have about their office. In order to measure up to them, they would “seek control over the structures and processes of government” (Moe 1985: 239). Presidents would not be solely interested in efficiency or effectiveness, nor in “neutral competence”. Instead, they would be mainly concerned with the dynamics of political leadership, with political support and opposition etc. Presidents would seek an institutional system responsive to their needs as political leaders. “Responsive competence” would be preferred over “neutral competence”.
Since Moe's work, politicization started to be seen as an instrument for political control of the bureaucracy. One example of research done through this perspective is by Wood and Waterman (1991). Analyzing seven different public bureaucracies for responsiveness and the use of political tools applied in the late Carter and early Reagan administrations, the authors found responsiveness in all cases. Political appointments were considered the most important instrument of political control.
But politicization is not only viewed as an instrument of control. It is also viewed as a means for distributing patronage. Brown (1982) attempted to determine patterns of party affiliation among top-level presidential appointees from 1961 to 1980. There was evidence that extensive use of scarce presidential appointments to reward party stalwarts or to bolster party organizations was no longer viewed as a profitable exercise by the White House.
David Lewis (2007, 2008) contests the notions that presidential politicization of the executive branch is intended only to enhance political control of the bureaucracy and is successful at doing so, although he agrees with Brown's thesis that the use of political appointments as a way of rewarding party members loyal to the president was in decline. He argues that politicization choices are driven by patronage concerns, and politicization of the bureaucracy can ultimately make it harder for presidents to control the bureaucracy.
The literature reviewed so far is highly centered in U.S. presidential studies. This is because authors who study bureaucracy and partisan politics in Brazil have all too often ignored crucial aspects of this relationship that have been taken into account by analysts of the American presidency. Patronage has been ubiquitously considered as the single motivation behind partisan political appointments in Brazil, either because of cultural and historical reasons (Faoro 2000) or as a rational response to an uncertain environment (Schneider 1994). Patronage would entail bureaucratic inefficiency and lack of political representation, since political parties would be formally excluded from policymaking (Campello de Souza 1983: 32-33; Diniz 1997: 19; Nunes 1997). It was the worst of both worlds.
Alongside the supposedly patronage-ridden strategy – never backed up by the literature with more than a few case studies and/or anecdotal evidence –, “pockets of efficiency” were created within the bureaucracy by denying posts in certain ministries to partisan appointees and thus ensure better management and results (Evans 1995: 61-65; Geddes 1994; Gouvêa 1996), sacrificing political representation (Diniz 1997; Nunes 1997). The ministry of Foreign Relations, for example, has long been touted as a bureaucratic organization insulated from partisan interference and from the other ministries (Cheibub 1985), though some scholars argue that this has changed in recent times (Figueira 2010).
Proportional open-list representation ensures that congressional power in Brazil is very much fragmented, with no party holding more than 20 percent of the seats in the lower chamber (Kinzo 2004). It is undeniable then that Brazilian presidents and political parties need to find ways to negotiate, and partisan patronage for lower-level bureaucratic appointments is but one of the possible strategies. Political appointment slots can also be arguably used for policymaking purposes, as we shall argue below, after a brief presentation of the Brazilian executive federal bureaucracy.
Data and Method
The Brazilian federal bureaucracy consists of 537,095 active employees, as well as about the same number of retirees. As of December 2010, 21,681 (4 percent) of these employees were potential partisan appointees. They occupy the so-called “DAS” (an acronym of Direção e Assessoramento Superior, or High Level Execution and Advisory) offices, first implemented during the military dictatorship in 1970 and kept alive in the 1988 Constitution (see D'Araújo (2009: 17-18) for more details). DAS appointees are responsible, along with the minister, for the most important decisions taken in each ministry. One can divide the DAS appointees in two groups: DAS-1 to 3 and DAS-4 to 6. There are 17,114 (79 percent) DAS-1 to 3 appointees, and 4,567 (21 percent) DAS-4 to 6 appointees.
The first group is formed by low-level positions, with little policy formulation competence. Their monthly salary differs from 2,115 to 4,042 BRL (1,200–2,200 USD). Higher-level DAS appointees – that is, DAS-4 to 6 – have a monthly income from 6,843 to 11,179 BRL (3,880–6,351 USD) and possess control, influence and implement policies according to directives put forth by the minister and/or political parties. If the appointee is a career bureaucrat, he can opt to receive the full salary given to the position he gained by merit plus up to 60 percent of the DAS wage, a comfortable choice which some analysts consider excessively generous (De Bonis and Pacheco 2010: 359-360).
Since July 2005, DAS appointees are formally nominated by the minister of the Planejamento (D'Araújo 2009: 20), benefiting from informal consultation with the minister of the Casa Civil. 2 The Casa Civil is the most important bureau within the Brazilian presidency, also responsible for gathering legislative propositions from the other ministries and turning some of them into government bills and decrees (Queiroz 2009: 94-96).
We thank Fernando Abrucio for bringing this to our attention.
Also in 2005, a decree established that 75 percent of the lower-level DAS appointees and half of the DAS-4 appointees had to be occupied by civil service career bureaucrats. 3 This is an indication that the diagnostic once held by some analysts – such as “[In Brazil], unusually extensive powers of political appointment complement lack of meritocratic recruitment” (Evans 1995: 61) – nowadays need to be taken with a large pinch of salt.
A similar idea was proposed in the mid-nineties by minister Luiz Carlos Bresser-Pereira, but lacked support from the coalition (Pacheco 2010a: 194-195). The proposal probably only passed in July 2005 due to the unveiling of the “Mensalão” corruption scandal the month before. Santos (2009: 14) points out that a few ministerial units enacted even stricter rules for the appointment of DAS offices.
Nevertheless, DAS appointments are widely recognized as occupied according to political patronage criteria. In order to assess how true this supposition is, we checked the names of the members of all of Brazil's political parties, obtained from the Tribunal Superior Eleitoral, against the names of all the DAS appointees, available from the federal government's Portal da Transparência.
We expect parties that belong to the coalition to benefit from these appointments, so we analyze the presidential coalition from 2007 onwards, president Lula's last. In December 2010, the date when we gathered the data on political appointees, 4 the parties in the coalition were the PT (Partido dos Trabalhadores), PMDB (Partido do Movimento Democrático Brasileiro), PDT (Partido Democrático Trabalhista), PR (Partido Republicano), PSB (Partido Socialista Brasileiro), PC do B (Partido Comunista do Brasil), PP (Partido Popular), PTB (Partido Trabalhista Brasileiro) and PRB (Partido Republicano Brasileiro).
We also have data for April 2010, but the results were so similar that we opted to report only the most recent.
An important methodological caveat is in order. The dataset on party membership does not offer identification of the party member besides his/her full name, and the same is true for the dataset on political appointment occupants. We excluded all homonyms within datasets, as well as homonyms within parties (for example, a “Sérgio Praça” member of the PMDB in the state of Pará and a “Sérgio Praça” member of the PMDB in the state of São Paulo) and across parties (for example, a “Bruno Hoepers” member of the PP in the state of Rio de Janeiro and a “Bruno Hoepers” member of the PMDB in the state of Minas Gerais). Our analysis below thus potentially underestimates the number of political appointees, but neither the proportion between parties nor the distribution within ministries. 5
See Appendix I for a more detailed methodological explanation on this matter.
As mentioned before, studies on coalition management usually focus on two types of goods used by the president (and formateur party) to hold together coalitions: exchange and coalition goods. In the words of Raile, Pereira, and Power (2011), coalition goods – such as ministries – establish an exchange baseline, while exchange goods – such as individual budget amendments – cover the ongoing costs of operation. Can political appointments – specifically of the DAS type – be added to the first category? In order to begin answering this question, we must first show descriptive data in Table 1 about partisanship and political appointments in the Brazilian federal bureaucracy.
Partisan Political Appointees by Ministry in Brazil, December 2010 (in percent)
Partisan Political Appointees by Ministry in Brazil, December 2010 (in percent)
Note:
Some ministries were controlled by more than one party from 2007 onwards. The Defense Ministry was held by the PT until July 25, 2007, when it became controlled by the PMDB. The ministries of Environment and Tourism were held by the PT until March 31, 2010, and from that date until the end of president Lula's second term in office they were controlled by non-partisan ministers. The same happened to the ministry of Justice (held by the PT until February 10, 2010), Mines and Energy (held by the PMDB until March 31, 2010) and Transports (held by the PR until March 31, 2010).
Source: Authors’ own calculation based on original data in the Portal da Transparência do Governo Federal and the Tribunal Superior Eleitoral.
13.6 percent of DAS offices are filled by party members. It is striking how some ministries – such as the Ministry of Foreign Affairs and the Ministry of Finance – have proportionately few political appointees, whereas in other ministries – such as the Ministry of Fishing Affairs and the Ministry of Agrarian Development – they comprise almost a quarter of the DAS offices. 6
Table 2 in Appendix II shows how the DAS offices in each ministry are allocated to each of the parties in the coalition.
Coalition Goods and Proportionality for Coalition Partners, December 2010
Note:
Percentage calculated within the 2,946 DAS offices occupied by party members, comprising 13.6 percent of the total DAS offices.
Source: Authors’ own calculation based on original data in the Portal da Transparência do Governo Federal, the Tribunal Superior Eleitoral and the Cebrap congressional database.
It is also worth noting that most of the ministries created recently, such as the Sports, Tourism, Fishing Affairs and Urban Affairs Ministries, have higher levels of partisanship in their DAS offices than older ministries such as the Health and Finance Ministries. One factor that may account for this is the fact that bureaucratic careers have not yet been established in the younger ministries, and their DAS offices would thus have no way of being occupied by career civil servants associated to the ministry's topic. 7
We thank Regina Pacheco for this observation.
In ministries controlled by the PT, no other party rivals them in the appointment to DAS offices. In the four ministries controlled by the PMDB, PT shares most of the DAS offices with the PMDB, and a few of them are allocated to the smaller parties. The same pattern holds for ministries in which the minister is member of a smaller party. When the minister himself is not a political appointee, the PT and PMDB appoint a large percentage of the DAS offices. This is an indication that the PMDB is, indeed, the most important coalition partner not only on the floor of the Chamber of Deputies and the Senate, but also in the federal bureaucracy.
In a related line of argument, one can divide the ministries into over 200 “ministerial units” (unidades orçamentárias) and check their potential and de facto partisan occupation. The former is calculated dividing the number of DAS offices in the unit by the total number of employees, while the latter is the number of partisan DAS appointees divided by the total of DAS offices in the ministerial unit. There is a lot of variation among the 76 ministerial units that have a potential partisan occupation above zero, 8 as Table 3 in Appendix II shows.
Party Membership in the DAS Offices According to DAS Level, Selected Ministries, December 2010 (in percent)
Note:
Some ministries were controlled by more than one party from 2007 onwards. The Defense Ministry was held by the PT until July 25, 2007, when it became controlled by the PMDB. The ministries of Environment and Tourism were held by the PT until March 31, 2010, and from that date until the end of president Lula's second term in office they were controlled by non-partisan ministers. The same happened to the ministry of Justice (held by the PT until February 10, 2010), Mines and Energy (held by the PMDB until March 31, 2010) and Transports (held by the PR until March 31, 2010).
Source: Authors’ own calculation based on original data in the Portal da Transparência do Governo Federal and the Tribunal Superior Eleitoral.
123 ministerial units do not have DAS offices. See D'Araújo (2007), Câmara (2009) and Pacheco (2010b) for a review of other types of bonuses and gratifications offered to career civil servants in the Brazilian federal bureaucracy. We focus on the DAS offices here because they may be filled by appointees who are not career civil servants, unlike the other positions Pacheco (2010b) mentions.
Dividing the ministries into units allows us to propose a contribution to the literature on presidentialism in Brazil. The first is that ministries may not, for certain purposes, be treated as uniform analytical units. There are profound differences, to cite one example, between the Fundação Nacional do Índio (2,931 employees; potential and de facto partisan occupation around 22 percent) and the Departamento de Polícia Federal (14,340 employees; 2 percent potential partisan occupation and 7.5 percent de facto partisan occupation), though both belong to the Ministry of Justice. It is hardly safe to say anything about partisan affiliation in this ministry without observing the differences within ministerial units.
Another important question is the criteria used by the government in allocating control over DAS offices to parties belonging in the coalition. Are more of these offices given out to parties that control a larger proportion of parliamentary seats, in the manner of portfolio allocation in parliamentary systems (Browne and Franklin 1973)?
There are mainly two ways in which to assess the proportionality of the distribution of political appointments between parties that belong to the coalition: how they look according to the congressional representation of each party and how they look according to the ministries controlled by each party. Table 2 shows how parties are represented at the congressional level (seats in the Chamber of Deputies and the Senate) and at the executive level (ministries and DAS offices).
The president's party, the PT (Worker's Party), holds a disproportionate amount of the political appointments (or DAS offices). More than 30 percent of these offices are held by members of the PT, while 13 percent are held by the PMDB, the party's main coalition partner. This is partly in line with studies on coalition formation that predict proportional payoffs with a bonus for the formateur party (Baron and Ferejohn 1987; Ansolabehere et al. 2005). The bonus for the PT exists, even though the allocation of political appointments is not proportional.
Except for the PC do B and the PDT, the rest of the coalition partners hold less DAS offices than their congressional representation would entitle them to. This may occur for two reasons. The first is that the PT is a very fragmented party, with lots of rival factions (Lacerda 2002). In order to accommodate them all, the party has had to take control of more DAS offices than usual. The second reason is that the distribution of DAS offices mirrors, with minor differences, the distribution of ministries.
Why do the PC do B and the PDT have better luck in appointments than other coalition parties? Two factors may account for this. The first is the fact that the ministries they control – Labor and Sports – are not as important as others. Their budgets are relatively limited and their policy domains are less important than those of other ministries. It is no coincidence that as the 2014 World Cup and 2016 Summer Olympics loom closer, a new ministerial unit has been created in order to empty the PC do B-controlled Ministry of Sports of relevance (see Executive Decree 503/2010 and Bill 12.936/2011). The second factor is that the DAS offices in these ministries may be enough to secure the PDT and PC do B's support for government bills on the floor. In other words, the PT may not need to use all tools in its “presidential toolbox” to ensure these two parties’ support on the floor.
The PMDB is usually touted as the PT's most important coalition partner. Recent studies have argued, implausibly, that the PMDB does not control any of higher-level DAS offices (D'Araújo 2009: 51). Our data show that this party holds about half the political appointment offices it would be entitled to according to parliamentary weight. This may occur for three reasons. The first is that the PMDB simply does not have enough party members to fill the offices, even though they have more than two million members all over Brazil. 9 The second reason, more plausible, is that the PT prefers to leave room to bargain with the PMDB over time, indicating that political appointments work much in the same vein as the “exchange goods” noted by Raile, Pereira, and Power (2011). Also, at the end of president Lula's term in 2010, the PT would have little reason to appoint PMDB members to DAS offices, since all major government-initiated bills had already passed. More research on the relationship between different “presidential times” (Lewis and Strine 1996) and political appointments is needed to support this argument.
See <http://www.tse.gov.br/internet/eleicoes/eleitores_filiados.htm> (6 June 2011).
The second argument put forth in this essay is that political appointment offices in Brazil are not only used for patronage ends. This argument will be all the more convincing if there are different patterns of partisan occupation in the lower-level DAS offices and higher-level DAS offices.
The higher-level DAS offices (levels 4 to 6) are considered the “managers” of the ministries, responsible for a large part of the policymaking that takes place within the bureaucracy (D'Araújo 2009: 15). The DAS-1 to 3 offices, in turn, may be considered typical patronage appointments, with low salaries and less policymaking responsibility than the DAS-4 to 6 offices. 10 So if we assign political parties policy goals as well as patronage goals, we should expect the DAS-4 to 6 appointments to have a larger proportion of party members. Table 3 shows that this is true for the ministries considered as a whole, but varies a lot within the executive.
In some ministries, lower-level DAS offices are considered important because they entail work for regional branches of the ministries in sensitive areas such as tax collection (Loureiro, Abrucio, and Rosa 1998: 77).
An average of 18.5 percent of higher-level DAS offices is occupied by party members, while the same occurs for 12.3 percent of the lower-level offices. This indicates that political appointment offices in Brazil are used not only for patronage, but also for policymaking ends. This is very much in line with Limongi and Figueiredo's (2009) recent analysis of the choice of sectoral rapporteurs in the federal budget process. They argue that if a party controls ministry X, responsible for a certain policy domain in the Executive, the same party will appoint budget rapporteurs for policy domain X in the budget process. In other words, a pattern somewhat like the one found in parliamentary systems – in which certain parties are programmatically attracted to certain ministries – is arguably going on in Brazil for at least part of the cabinet.
A final point worthy of note regards the delegation strategy implemented by the PT in order to share power with its coalition partners. Thies (2001) tests two theories about how parties delegate power. The first, known as “ministerial government”, supposes that parties delegate ministerial portfolios in toto to one of their members (the party whose minister holds the portfolio). The second theory, dubbed “managed delegation”, supposes that members of a multiparty coalition delegate power, but also actively monitor the activity of ministerial posts by other parties. The critical piece of evidence used to test these rival theories is the appointment of junior ministers. If junior ministers are from the same party as the minister, the author assumes that the ministerial government model is in operation. If the junior ministers are from different parties, Thies infers that a managed delegation model is in operation, where the junior minister is assumed to perform an oversight function regarding the activity of the bureau in question. 11
See also Giannetti and Laver (2005) for an assessment of junior ministers and intracoalition bargaining in parliamentary systems.
Table 4 shows how the DAS-6 and Special Nature 12 Offices – the highest level possible below the minister, which entails a lot of policymaking responsibility – are distributed in the Brazilian federal bureaucracy. They are the Brazilian equivalent of “junior ministers”. The only party in the coalition not to appoint one is the PR (Partido da República).
Partisanship at the Highest Level (DAS-6 and Special Nature Offices), Selected Ministries, December 2010
Note:
Some ministries were controlled by more than one party from 2007 onwards. The Defense Ministry was held by the PT until July 25, 2007, when it became controlled by the PMDB. The ministries of Environment and Tourism were held by the PT until March 31, 2010, and from that date until the end of president Lula's second term in office they were controlled by non-partisan ministers. The same happened to the ministry of Justice (held by the PT until February 10, 2010), Mines and Energy (held by the PMDB until March 31, 2010) and Transports (held by the PR until March 31, 2010).
DEM, PSDB, PPS, PRP.
Source: Authors’ own calculation based on original data in the Portal da Transparência do Governo Federal and the Tribunal Superior Eleitoral.
The Special Nature Offices (NES, Cargos de Natureza Especial) are an intermediate level between the highest-level DAS offices (DAS-6) and the minister.
Most of these offices (68 percent) are allocated to non-partisan bureaucrats. The “managed delegation” strategy takes place in a few ministries, such as the Health, Sports, Labor, Urban Affairs etc. This piece of information allows us to suggest that coalition management in Brazil is less clientelistic and more policy-oriented than is usually assumed in the literature. 13
Authors such as Olivieri (2010) and Loureiro, Olivieri, and Braga Martes (2010) offer similar arguments.
The data and arguments presented in this research note open up an exciting research avenue in the agenda of bureaucracy and politics in presidentialist systems. A few tentative conclusions may be offered. The first regards the question of whether political appointment offices in the federal bureaucracy are “coalition goods” (that is, they establish an exchange baseline between the presidential party and other parties in the coalition) or “exchange goods” (in this view, DAS appointments would be made in order to cover the ongoing costs of holding together the coalition), to cite Raile, Pereira, and Power's (2011) typology. The evidence presented so far shows that DAS offices are easier to allocate than ministries, even though we are working on the construction of a dataset with at least three points in time in order to better support this argument. At least for now, we strongly suspect that political appointments may be used in order to adjust minor bargains within the coalition, therefore being similar to “exchange goods”.
The second main conclusion that we put forth in this essay is that, contrary to almost all of the literature on bureaucracy and politics in Brazil so far, political appointments are not used exclusively for patronage ends. The evidence presented above shows that a very large percentage of political appointments are filled out by career civil servants. Even though they may be party members, it is safe to say that they are also policy experts (though of course the level of expertise and specialization may vary). It also shows that the higher the level of the appointment is, the more important are the criteria of expertise and party policy domain.
Much research remains to be done, and we point out two questions as particularly important. The first is the dimension and impact of bureaucratic turnover on policymaking. If we find that occupants of lower-level DAS offices have very stable careers and stay in their positions for a long time, it would be an indication that ministerial turnover does not affect the bureaucracy in as significant a manner as scholars like Martínez-Gallardo (2010) suspect, and this has clear implications for the quality of policymaking (Spiller, Stein, and Tommasi. 2008).
The second question regards government performance. Lewis (2007) analyzes the impact of political appointments on bureaucratic performance by analyzing the relationship between political appointees and management through a measure of federal program performance implemented by the George W. Bush administration. He argues that federal programs administered by politically appointed bureau chiefs get systematically lower evaluations than programs run by bureau chiefs drawn from the civil service. Based on these findings he concludes that reducing the number of appointees or increased sensitivity to appointee selection based upon certain background characteristics could improve federal bureau management. An application of this rationale to the Brazilian federal bureaucracy is urgent, and this research note is one of the first steps in this direction.
Footnotes
Data and Methodological Considerations
The dataset that reveals the partisanship of occupants of political appointment offices in the Brazilian federal bureaucracy has two sources. The first is the federal government's Portal da Transparência do Governo Federal (http://transparencia.gov.br), which brings the name, DAS level, date of entry and exit and partial CPF (Cadastro de Pessoa Fisica) of all DAS appointees who were active employees in December 2010. Table 1 shows the basic distribution.
The second source of data is the Tribunal Superior Eleitoral, which makes available the names and date of entry of all party members to all of Brazil's 27 political parties. Table 2 shows the distribution of party members.
The only variable that can link both datasets is the full name of DAS appointees and party members. This entails an obvious task: sorting out, or at least estimating the presence of, homonyms. Scientific studies about homonyms have been mainly concerned with separating and identifying authors with the same name (either the first and second names or the full name) in databases of scientific publications. Homonyms may be an important source of error and can impact the quality and validity of the results. Asknes (2008) studied how frequently homonyms occur in a population of more than 30,000 Norwegian scientists and found that its overall occurrence is relatively low: 14 percent. They would be more frequent in the cases of common surnames. Others state that problems of homonyms are more frequent in countries such as China, Japan, South Korea, and Taiwan (Onodera et al. 2011) and point out to cultural differences that can be troubling when identifying names in databases. Fortunately, compound names in Portuguese are very common, and this may minimize the possibility of homonyms (Black 2003).
In this research note, the problem of homonyms may have important consequences because its occurrence across governmental agencies may overestimate the number of political appointees in these agencies. Therefore selection bias could plague our data analysis. In order to account for this problem we took some measures with the purpose of identifying homonyms and minimizing its presence on our data.
First we looked at the list of Brazilian party members and identified how many homonyms were present in each party. We cannot ascertain the true identity of the homonyms, since the Tribunal Superior Eleitoral lacks their Registro Geral and Cadastro de Pessoa Física (roughly equivalent to the North-American Social Security Number). So we opted to leave them out of the dataset, thus checking against the dataset of political appointees only the names of party members that are non-homonyms. Around 30 percent (4,177,131 names in total) of the members of each party are homonyms. This is encouraging because it shows that the distribution of homonyms is fairly homogeneous across political parties. This leaves us confident that some parties do not have more homonyms than others. The final dataset of party members thus comprises 9,710,533 names. We also found 86 homonyms within the DAS appointees database, leaving a total of 21,595 occupants of DAS offices.
Two other problems might occur. The first is the possibility of not attributing a party to a DAS appointee who is, in fact, member of a certain political party. This may occur because his name is present in more than one party membership list. To illustrate this scenario, one of the DAS appointees is called José Antônio da Silva – a very common Brazilian name present in all of 27 political party membership lists. In order to estimate the occurrence of similar problems, we ran a test, checking the dataset of DAS appointees against only the homonyms present in the party lists. We found that 2,229 (about 10 percent) of DAS appointees could have been attributed a political party but were not, due to uncertainty about whether or not their name corresponds to the party member's name. This percentage holds across ministries and ministerial units. In the worst case scenario, then, we underestimate political appointees of all parties by 10 percent, but only if all DAS appointees who were not attributed a political party turned out, in fact, to belong to one. 14
Although this may hurt some of our descriptive data, any future analyses we carry out about the determinants or effects of political appointment partisanship will not be biased because of any party differentials or different processes by party that may impede causal inference. We thank Steve Finkel for pointing this out.
Finally, there is the possibility of attributing a political party to a DAS appointee who actually belongs to no party at all, due to the fact that a homonym of his is present in one of the party membership lists. In order to estimate the occurrence of this problem, we looked at all 208 DAS-6 office holders. Our initial estimation indicated that 92 of them belonged to a political party. Checking the Internet for their political biographies and news related to their party affiliation and work, in only one case the information we gathered diverged from the information we had in our dataset. It was Alexandro da Anunciação Reis, who we believed was a member of the PT but showed up in numerous websites as a member of the PC do B. Although we cannot be sure, we think this is due to a recent party change, since our data was collected in January 2011. This leads us to believe that incorrect attributions of political membership to DAS appointees were rare.
Appendix II
Party Membership at the Highest Level: DAS-6 and NES Offices, December 2010 (in percent)
| Ministry (and party in control) * | PT | PMDB | PC do B | PSB | PDT | Total (N) |
|---|---|---|---|---|---|---|
| Presidency (PT) | 33.3 | - | - | 1.5 | 1.5 | 66 |
| Fishing Affairs (PT) | 60.0 | - | - | - | - | 12 |
| Planning, Budget, and Management (PT) | 45.5 | - | - | - | - | 11 |
| Agrarian Development (PT) | 42.9 | - | - | - | - | 5 |
| Social Development (PT) | 42.9 | - | - | - | - | 6 |
| Education (PT) | 41.7 | 8.3 | - | - | - | 9 |
| Social Security (PT) | 16.7 | - | - | - | - | 7 |
| Finance (PT) | 9.1 | - | - | - | - | 7 |
| Foreign Affairs (PT) | - | - | - | - | - | 11 |
| Health (PMDB) | 12.5 | 12.5 | - | - | - | 6 |
| Agriculture (PMDB) | - | - | - | - | - | 8 |
| Defense (PT; PMDB) | - | 12.5 | - | - | - | 9 |
| National Integration (PMDB) | - | - | - | 11.1 | - | 8 |
| Communications (PMDB) | - | - | - | - | - | 5 |
| Urban Affairs (PP) | 33.3 | - | - | - | 66.7 | 6 |
| Science and Technology (PSB) | 22.2 | - | - | - | - | 3 |
| Sports (PC do B) | 20.0 | - | 40.0 | - | - | 9 |
| Labor (PDT) | 16.7 | - | - | 33.3 | - | 12 |
| Culture (PV) | 8.3 | - | - | - | - | 19 |
| Mines and Energy (PMDB; none) | 28.6 | - | - | - | - | 3 |
| Justice (PT; none) | 21.1 | 5.3 | - | - | - | 7 |
| Development, Industry, and Commerce (none) | 11.1 | 11.1 | - | - | - | 9 |
| Environment (PT; none) | 10.0 | - | - | - | - | 10 |
| Tourism (PT; none) | - | - | - | - | - | 2 |
| Transports (PR; none) | - | - | - | - | - | 6 |
| Total | 22.3 | 1.9 | 0.7 | 1.5 | 1.1 | 256 |
| (58) | (5) | (2) | (4) | (3) |
| Ministry (and party in control) * | PP | PTB | PRB | Other parties ** | Nonpartisan | Total (N) |
|---|---|---|---|---|---|---|
| Presidency (PT) | - | - | - | - | 63.6 | 66 |
| Fishing Affairs (PT) | - | - | - | - | 40.0 | 12 |
| Planning, Budget, and Management (PT) | - | - | - | - | 54.5 | 11 |
| Agrarian Development (PT) | - | - | - | - | 57.1 | 5 |
| Social Development (PT) | - | - | - | - | 57.1 | 6 |
| Education (PT) | - | - | - | - | 50.0 | 9 |
| Social Security (PT) | - | - | - | - | 83.3 | 7 |
| Finance (PT) | - | - | - | - | 90.9 | 7 |
| Foreign Affairs (PT) | - | - | - | - | 100.0 | 11 |
| Health (PMDB) | - | - | - | - | 75.0 | 6 |
| Agriculture (PMDB) | - | - | 16.7 | 33.3 | 50.0 | 8 |
| Defense (PT; PMDB) | - | - | - | - | 87.5 | 9 |
| National Integration (PMDB) | - | - | - | - | 88.9 | 8 |
| Communications (PMDB) | - | - | - | 33.3 | 66.7 | 5 |
| Urban Affairs (PP) | - | - | - | - | - | 6 |
| Science and Technology (PSB) | - | 22.2 | - | 11.1 | 44.4 | 3 |
| Sports (PC do B) | - | - | - | - | 40.0 | 9 |
| Labor (PDT) | - | - | - | 16.7 | 33.3 | 12 |
| Culture (PV) | - | - | - | 8.3 | 83.3 | 19 |
| Mines and Energy (PMDB; none) | - | - | - | - | 71.4 | 3 |
| Justice (PT; none) | - | - | - | - | 73.7 | 7 |
| Development, Industry, and Commerce (none) | - | - | - | - | 77.8 | 9 |
| Environment (PT; none) | - | - | - | - | 90.0 | 10 |
| Tourism (PT; none) | - | - | - | - | 100.0 | 2 |
| Transports (PR; none) | 16.7 | - | - | - | 83.3 | 6 |
| Total | 0.3 | 0.7 | 0.3 | 3.8 | 66.9 | 256 |
| (1) | (2) | (1) | (10) | (174) |
Note:
Some ministries were controlled by more than one party from 2007 onwards. The Defense Ministry was held by the PT until July 25, 2007, when it became controlled by the PMDB. The ministries of Environment and Tourism were held by the PT until March 31, 2010, and from that date until the end of president Lula's second term in office they were controlled by non-partisan ministers. The same happened to the ministry of Justice (held by the PT until February 10, 2010), Mines and Energy (held by the PMDB until March 31, 2010) and Transports (held by the PR until March 31, 2010).
DEM, PSDB, PPS, PRP.
Source: Authors’ own calculation based on original data in the Portal da Transparência do Governo Federal and the Tribunal Superior Eleitoral.