Cabinets, the Bureaucracy, Subnational
Governments, and the Judiciary

This chapter focuses on four more actors that have formally ascribed roles in the policymaking process (PMP): the cabinet, the bureaucracy, subnational governments, and the judiciary. As with the actors examined in Chapter 3, the extent and nature of the roles performed by these actors vary considerably across countries, in part because of differences in the formal and informal rules associated with the actor, and in part because of interactions with characteristics of the broader policymaking process. Given the focus of this report, the discussion of subnational governments is restricted to an analysis of their potential role in influencing national policy decisions or their implementation.

Cabinets1

The policymaking role of cabinets is somewhat more ambiguous in presidential systems than in parliamentary systems. Executive power in “pure” presidential systems is concentrated in one person: the president. Thus cabinet ministers tend to have a narrower role as aides to the president. By contrast, in parliamentary systems, cabinet ministers typically also serve as members of parliament and, along with the head of government, are collectively responsible to it. In part because of the cabinet’s ambiguous status, the cabinet as a collective body in most presidential systems in Latin America does not appear to have an important influence on policy. But given the informal way cabinets generally function and the private nature of meetings in Latin American countries, little is known about how decisions are actually made, including who participates and how often the cabinet meets collectively.

Nevertheless, it is clear that cabinet ministers, at least as individuals, play key roles in every stage of policymaking. Cabinet ministers and those directly under them in the bureaucracy provide the knowledge and expertise necessary to formulate policy. The legislature and political parties rarely have comparable resources at their disposal that allow them to participate as centrally in the formulation of public policy. Cabinet ministers also advocate and defend specific policy proposals and decisions and guide bills through the legislature. Cabinet ministers are involved in interpreting and putting into practice enacted legislation through the executive’s rulemaking authority. They direct large bureaucratic agencies and are in charge of evaluating and supervising the details of policy implementation. Cabinets additionally serve as a tool for building governmental and legislative coalitions. Presidents whose parties lack a majority in the legislature can help secure the support of members of other parties through cabinet appointments.

Features of Cabinets and Policymaking

Characteristics related to the stability, coordination, and formation of cabinets are likely to have important effects on the key features of public policies.

Stability of Cabinets

To promote long-term policies and to allow ministers to see programs and policy implementation through to completion, a certain degree of stability for officials in cabinet posts is likely to be necessary. Frequent turnover of cabinet ministries is likely to promote a more short-term orientation to policy and more frequent changes in policy direction. Longer tenures also facilitate and improve the efficiency of delegating responsibilities and tasks to bureaucrats, which is essential for effective policy implementation. Frequent changes in the cabinet can leave leadership vacuums that may contribute to bureaucratic inertia and even corruption. In addition, longer tenure allows ministers to accumulate valuable expertise specific to the policy area in which they work and to develop political and managerial skills that are likely to improve the quality of their performance in their different policymaking functions.

Cabinet stability is generally low in Latin America. Between 1990 and 2000, almost a quarter (22 percent) of all ministers in a sample of 12 Latin American countries remained in the same portfolio for less than six months. Three-quarters (75 percent) of all ministers had tenures of less than two years (while presidential terms vary between four and six years). Figure 4.1 compares countries in terms of the average number of different individuals who served in a given ministry from 1988 to 2000. The countries with the greatest cabinet instability are Ecuador, Brazil, Venezuela, and Colombia. Cabinets are most stable in Uruguay, Costa Rica, and Chile.

Capacity for Coordination

The ability of governmental ministries to coordinate among themselves and to cooperate with other institutions, including the bureaucracy, is likely to have a large effect both on the policymaking process itself and on the characteristics of policy outcomes. One factor affecting the capacity for coordination has to do with the partisan make-up of the cabinet. The more parties that participate in the cabinet and the greater the divergence of ideologies and interests among them, 2 the more likely it will be that difficulties in coordination will arise, and the greater those difficulties are likely to be. Single-party majority governments tend to be able to minimize the costs of reaching agreements and making decisions, while coalition governments may find it more difficult to coordinate policy.

The contrast between the Partido Revolucionario Institucional (PRI) governments in Mexico and coalition governments in Argentina is illustrative. During the period of PRI hegemony in Mexico, cabinet coordination was aided by the centralization of power in the president, who was the de facto leader of the party, and by the high degree of ideological cohesiveness in the cabinet. By contrast, during the De la Rúa administration in Argentina, although most ministries were headed by politicians from the president’s party, the Radical Party, politicians from a rival party, Frepaso, were given the labor and social affairs ministries. Power was further divided within ministries through the appointment of vice-ministers from a party different from the minister’s. This presented problems of coordination, since vice-ministers sometimes answered to their party leader instead of to the minister and Frepaso ministers in some cases answered to their party leaders rather than to the president.3

Coordination with the bureaucracy is likely to be enhanced by relative cabinet stability, as well as by the appointment of senior (nonpolitical) civil servants to cabinet and undersecretary positions.

The most common institutional instrument used for fostering cabinet coordination across Latin America is thematic cabinets or interministerial committees. In Mexico, for instance, the Zedillo administration had a Security Cabinet, an Economic Cabinet, and an Agrarian Affairs Cabinet. During the Fox administration, the thematic cabinets have been Growth with Quality, Order, and Respect; and Human and Social Development. In Argentina, several attempts have been made since 1993 to establish a special cabinet for social policy. In both Mexico and Argentina, evidence suggests that these thematic cabinets have had limited effectiveness in promoting improved coordination among the ministries.4 The frequency of meetings has depended upon the president and the priorities of the political agenda; at some points in an administration’s term the special cabinets have hardly met at all.

An important feature of Latin American cabinets has been the central role of the finance ministry and, more widely, of a limited number of technical ministries. The finance ministry has dominated the budget process and been able to prioritize its goals on the presidential agenda. Thus it has been in a privileged position in respect to other ministries and, in some cases, in relation to regional governors. This dominant role has been supported in some instances by constitutional provisions, by the relative weakness of the legislature in the budget process, and by the more competitive recruitment and greater stability of personnel in the finance ministry and its superior institutional capacity. The predominance of one ministry in the cabinet may improve coordination in respect to some policy goals, while weakening it in relation to others.

Another potential source of poor coordination relates to the structure of the cabinet. An excessively large number of cabinet ministries or policymaking units with jurisdiction over policy in a particular area could be expected to limit the capacity to develop policy that is consistent over time and implemented coherently. However, the sheer number of ministries might not accurately reflect the potential for an overlap of jurisdictions or the tendency for frequent changes to be made to their areas of competence.

A case in point is Argentina. Although the number of ministries is fixed by law, presidents have worked around these legal restrictions by creating secretaries with ministerial rank. On top of this, there has been frequent ministerial reorganization, in many cases involving a redistribution of jurisdictions at the level of secretaries and undersecretaries.5 During President Carlos Menem’s term, for example, the ministries of agriculture and infrastructure were placed under the ministry of the economy as a way for the latter to enhance control over spending. President Eduardo Duhalde, on the other hand, created a ministry of production as a signal of the importance of industry. The existence of multiple agencies with overlapping jurisdictions and the existence of a parallel bureaucracy composed mainly of temporary officials typically has made coordination difficult—among ministers, and among ministers and the bureaucrats in their agencies—and has made it harder to attain coherent policy.6

Cabinet Formation

How presidents use their power to appoint and dismiss members of the cabinet can have important effects on their ability to obtain sufficient legislative support to enact policies and on the stability and the effectiveness of government policymaking.

Cabinet appointments can be effective in securing political support, since they provide a means for political parties as a whole and politicians as individuals to influence policy to a degree that is not possible through positions in the legislature. Cabinet positions potentially give parties and politicians control over important patronage and budgetary resources that can be used for political ends. However, incentives for sustaining inter-party coalitions and for coalition legislators to adhere to the government’s position are likely to be weaker in presidential systems than in parliamentary systems. This is the case since the head of government and legislators are elected separately and their terms of office are fixed; thus their survival in office does not depend upon maintaining majority support for the government.

In choosing who will serve in their cabinets, presidents must balance the objectives of maintaining political support with those of ensuring that the individuals appointed have sufficient policy expertise, will follow policies reasonably consistent with the president’s thinking, and will be able to work sufficiently well together. The latter two concerns are especially relevant when individuals from different parties are appointed for the purpose of building coalition governments. The case of Bolivia, discussed in Box 4.1, illustrates how cabinet appointments have been used to sustain governmental coalitions.

The capabilities of cabinet officials as experts in their field, as managers, and as advocates of the government’s policies are likely to have an important effect on the characteristics of policy outcomes. To the extent that cabinet members are selected purely either on the basis of their personal loyalty to the president or because of their political party connections, this may detract from their ability to design sound policies and manage their bureaucratic agencies effectively. But, since policymaking is not a purely technical matter, and political and managerial skills contribute to a minister’s effectiveness, broad criteria of competence, rather than narrow training in the policy field, may be most suitable.

The Bureaucracy 7

The bureaucracy is one of the institutional anchors for the effective functioning of the democratic system and the enforcement of the rule of law. The bureaucracy is more than a stock of human resources, an organizational apparatus, or an employment system. Rather, it is an articulated set of operating rules and guidelines regulating the executive branch that aims to give continuity, coherence, and relevance to public policies, while ensuring a neutral, objective, and non-arbitrary exercise of public authority. The bureaucracy is a key actor for encouraging inter-temporal agreements, especially through its role in putting such agreements into practice. A neutral and professional bureaucracy limits the scope for the adoption of opportunistic policies and enhances the trust of actors that commitments made as a part of policy agreements will be fulfilled.

Latin America has traditionally been viewed as a region with large but weak States, with little capacity to respond to the needs of citizens. Historically, much of this weakness has been associated with the lack of a stable professional bureaucracy. The bureaucracy has been perceived as an employment system or a resource in the hands of politicians and corporate interests—and thus far removed from the Weberian ideal type characterized by regularized and impersonal procedures and employment decisions based on technical qualifications and achievements.

The weakness of the bureaucracy has contributed to the weakness of the executive branch in Latin American countries, particularly in its relations with other political actors and with special interests. Bureaucratic weakness has been one cause of the ineffectiveness of development policies in the region, for which the State historically has been an important actor. This contrasts sharply with the development experience of the Asian “tigers.” A strong and autonomous State, supported by a highly professional and meritocratic bureaucracy, has been considered a key factor in the success of the Asian tigers. The Southeast Asian countries rank considerably higher in cross-national indicators of the quality of the bureaucracy, even relative to the more advanced countries of Latin America.
Despite its role as an institutional foundation of the rule of law, the bureaucracy holds a subordinate position to the government in the constitutional system. The bureaucracy is not legally an autonomous and responsible agent of the political system, but rather a part of the executive branch and responsible to the government.

The transition from authoritarian to democratic regimes has been linked to a certain tendency to further subordinate the bureaucracy to political control. In many cases, this has led to a reversion to clientelistic practices. This situation may arise from the need to use public employment as a resource to reward the members of the winning party with jobs (the spoils system) and from the need to reduce the technical autonomy that bureaucratic officials frequently obtain in certain sectors, which limits the scope for political action by the government. Historically, democracy has been slow to overcome this dilemma.

The bureaucracy plays varied and contradictory roles in the policymaking process, depending on how closely it approaches the ideal of being a neutral and professional actor that guarantees the stability, adaptability, and public interest of policies or mainly functions as a private resource: either of political parties, which use it to obtain votes, or of civil servants, who defend their own interests while being protected by job security. The degree to which the bureaucracy fills one or the other role affects some of the characteristics of public policies.

Characterizing Latin American Bureaucracies

An analysis of the bureaucracy requires an analysis of public employment, its main element. Public employment has both a quantitative and a qualitative dimension. The qualitative dimension is the more relevant for understanding the capacity and effectiveness of the bureaucracy. Thus this study does not devote much attention to analyzing quantitative measures of the bureaucracy, although the evidence presented below indicates that the size and quality of the bureaucracy are not closely correlated.

A qualitative approach examines the extent to which bureaucracies are endowed with the institutional attributes required to perform the regulatory roles assigned to them in a representative democracy. In carrying out this analysis, this study uses the data, indices, and conclusions of an institutional diagnostic study of the civil service conducted in 18 countries of the region.8

The degree of autonomy of the bureaucracy can be measured using the merit index (Figure 4.2), which evaluates the degree to which effective guarantees of professionalism in the civil service are in place and the degree to which civil servants are effectively protected from arbitrariness, politicization, and rent-seeking.9

Three groups of countries may be distinguished. The top performers, Brazil, Chile, and Costa Rica, have indices between 55 and 90 (out of 100), reflecting a widespread acceptance of the principles of merit in decisions regarding the hiring, promotion, and dismissal of public officials. A middle group of countries, with indices between 30 and 55, includes Argentina, Colombia, Mexico, Uruguay, and Venezuela. Practices based on merit coexist with traditions of political patronage. A third group of countries, comprised of Bolivia, Paraguay, Dominican Republic, Peru, Ecuador, and all of the Central American countries except Costa Rica, have indices below 30, reflecting a strong politicization of decisions on hiring, promotion, and dismissal.

To be able to perform substantive roles in designing and implementing public policies, the bureaucracy also requires adequate technical capacities and incentives for effective performance. The functional capacity index (Figure 4.3) is a good approximation of these characteristics. It rates the characteristics of salary compensation systems and systems for evaluating the performance of public officials.10

Brazil and Chile stand out, with indices near 60 out of 100. Their scores reflect rational systems of salary management, with relative internal equity and processes to improve salary competitiveness, as well as evaluation processes that begin to relate individual performance to group and institutional performance. The next group of countries, with indices ranging between 35 and 50, includes Costa Rica, Colombia, Argentina, Uruguay, Mexico, and Venezuela. These countries have gone through the process of rationalizing the salary system, although internal inequity continues and problems of salary competitiveness persist at managerial levels. The group with the worst results has indices between 10 and 25 and is made up of the Dominican Republic, Ecuador, Bolivia, El Salvador, Guatemala, Nicaragua, Peru, Panama, Paraguay, and Honduras. These countries are characterized by diversity of payment criteria, lack of information on compensation, high levels of inequity, and an absence of performance evaluation.

Considering both indices together, the countries analyzed can be grouped into three levels of bureaucratic development. The first group has bureaucracies with only a minimum level of development, in which the civil service system cannot guarantee the attraction and retention of competent personnel, and lacks the management mechanisms necessary to promote efficient performance on the part of civil servants. This group includes countries with low scores on both indices: Panama, El Salvador, Nicaragua, Honduras, Peru, Guatemala, Ecuador, the Dominican Republic, Paraguay, and Bolivia. The second group has civil service systems that are fairly well structured but that have not been consolidated in terms of merit guarantees and management tools that would allow for an effective utilization of capabilities. This group consists of Venezuela, Mexico, Colombia, Uruguay, Argentina, and Costa Rica. The countries in the third group, Brazil and Chile, excel on both indices and are more institutionalized relative to other countries, despite having different profiles in terms of their civil service systems.

Figure 4.4 relates the bureaucracy’s size to its quality. To measure quality, the indices of merit and functional capacity are combined with an index of efficiency into an aggregate civil service index.11 The size measure is the percentage of the total population employed in the public sector. The result highlights the fact that quantity and quality are barely correlated (0.26), since among the countries whose bureaucracies are dysfunctional, some are large and some are small.

Bureaucratic Configurations
and Prevailing Roles

The bureaucracies in the Latin American countries are not homogeneous actors. In reality, they are a set of complex and interdependent organizations that vary in terms of their autonomy and capacity. This heterogeneity may provide some keys to understanding the internal dynamics of the State apparatus and the degree to which different parts of the bureaucracy may fulfill different and even contradictory roles in the same country.

Starting with this assumption, Figure 4.5 presents a basic categorization of the different types of bureaucracies found in the countries of the region. The placement of the bureaucratic types (and country examples), though based on qualitative case studies of civil service systems in the region, is not precise, and is performed for illustrative purposes only. The prevalence of these types of bureaucracies varies across countries, and the types also coexist within countries. Each is shaped within historic contexts and by predominant political practices. In addition, each type of bureaucracy tends to perform certain roles in the policymaking process, although exceptions may exist that exceed the scope of this analysis.

Administrative bureaucracy. This classic form of bureaucracy is characterized by low capacity and a relatively high degree of autonomy. It includes the apparatus that exercises administrative functions in the various ministries and sectors of the State. These areas are normally covered by formal standards of merit, which are not applied in practice. In effect, they represent frustrated or partial attempts to develop a traditional Weberian bureaucracy (that is, one that is rational, hierarchical, and merit based). Civil servants are hired on the basis more of political than meritocratic criteria, but they have some job security. The degree of their technical competence and orientation toward good performance is low. These are the areas most affected by budget cuts (in some countries, such as Peru or Uruguay, funding to these bodies has been frozen). The bureaucracies of Peru, Venezuela, and Ecuador (on the lower end in this category in terms of autonomy and capacity) and of Argentina, Colombia, Costa Rica, and Uruguay (at the higher end) fall into this category, with varying degrees of autonomy and capacity.

This type of bureaucracy has a limited ability to play an active role in the different stages of policymaking. This lack of ability prevents it from effectively exerting influence at the decision making stage. Decisions are usually made in the policy section of the ministries. It may perform some role in policy implementation, although with a tendency toward formalism and control through conformity with bureaucratic procedures, rather than through the effective management of services. Its potential as a resource for political exchange is limited because of the relative lack of regular turnover of staff and the progressive decline of its importance due to reductions in overall levels of public employment.

Clientelistic bureaucracy. This form of bureaucracy is characterized by low autonomy and low capacity. It is made up of public officials who temporarily enter government because of loyalty or party affiliation. Ministerial rotations or changes of government are likely to affect the stock of human resources and may even result in large-scale turnover of civil servants. For a subset of these bureaucracies, control over access and permanence lies not with the party system but with trade unions. The most notable cases occur in the Central American countries (with the exception of Costa Rica) and in the Dominican Republic, Paraguay, and Bolivia (except in some meritocratic enclaves). In Mexico, control of the government apparatus has traditionally been divided between the governing party for managerial positions (until recently, when the Career Law was approved), and the trade unions for the remainder of the jobs. Other countries exhibit some of these characteristics as well. Examples include temporary appointments or contracted employees in Argentina, Colombia, Peru, and Uruguay, under the protection of transitional or special regimes of employment that grant the government greater flexibility in appointing and dismissing public officials.

The roles performed by clientistic bureaucracies are related to their nature primarily as a political resource of the governing party to exchange jobs for votes or political support. This type of bureaucracy is an extension of the political party. It has some veto power over the professional or meritocratic segments of the bureaucracy, with which it may come into conflict. Its role in designing and implementing policies is virtually irrelevant, except in relation to the operational level of the most simple and routine tasks.

Parallel bureaucracy (“technical teams” or “project teams”). This form of bureaucracy is characterized by low autonomy and high capacity. It is comprised of managers hired under flexible contractual agreements, a practice that has spread to a majority of Latin American countries since the early 1990s. The employment system is usually governed by laws relating to service contracts or other legal arrangements. Those hired are not part of the permanent structure of the civil service, although in several countries the contracts of these officials are routinely renewed. These groups of officials do not, strictly speaking, owe allegiance to any political party, and they possess expert knowledge in certain areas of policy. In the majority of cases they have been hired to cover certain technical needs. In some cases they have developed political as well as technical skills (“technopols”). Parallel institutions, or what have come to be called “parallel ministries,” are usually configured either outside or within the regular agencies. These structures encounter varying degrees of resistance from other internal bureaucratic actors and achieve varying degrees of success.

In some cases, teams of technical advisors perform a key role in designing policy alternatives, working closely with the chief executive. In other cases, they are more centered on ensuring the execution of policies or projects or the effective delivery of certain public services. This category includes the diverse organizations that manage projects with international financing or social funds. They are characterized by different degrees of autonomy and capacity. However, parallel bureaucracies do not contribute to the strengthening of the inter-temporal capacities of the public sector, since they remain isolated and concentrated in very limited roles in the policymaking process and do not become institutionalized in the broader public sector.

Meritocratic bureaucracy. This form of bureaucracy is characterized by high autonomy and high capacity, in various combinations. It is composed of permanent civil servants recruited on the basis of merit and incorporated into professional careers, with varying incentives favoring professional job performance. It is made up of administrative bureaucracies in which merit and capacity have been preserved, as in the case of Chile or Brazil (government careers and posts), specialized agencies connected to the fiscal or economic bureaucracy (such as central banks, regulatory agencies, or tax administrations, including the Superintendencia Nacional de Administración Tributaria [SUNAT] and the Internal Revenue Service in Chile and Peru), and professional corps that have established their own personnel regulations based on merit and capacity (the diplomatic service in various countries, including Brazil and Mexico, government administrators in Argentina, and some expert professionals in other countries, such as economists, lawyers, and engineers). Social sector bureaucracies (personnel from the education and health sectors) can be placed midway between the administrative and meritocratic bureaucracies, depending on the countries and sectors involved.

These bureaucracies express opinions and act. Most have a specific purview: thematic areas of public policy that require a degree of training or specialized knowledge, which gives them a voice in and influence over the area in which they are acting. This makes them a major actor in maintaining the stability and public interest orientation of policies. This type of bureaucracy constitutes an actor with a specific culture, sometimes insulated, and with a strong esprit de corps, which can result in corporative biases. This characteristic can make such bureaucracies prone to participate more in the design than in the implementation of policies and to demand decision making autonomy that in many cases brings them into conflict with other governmental agencies. They may make alliances with other agencies and even with external societal interests. While this may give them a capacity to intermediate among these different interests, it also makes them susceptible to capture by such interests.

In recent years, numerous countries have sought to expand and strengthen their meritocratic bureaucracies. The reform experiences have varied greatly, and in some cases significant advances have not resulted. Although many countries have adopted new civil service laws, these laws have not always been implemented effectively because of a mix of political factors (related to the value that public employment continues to have as a currency of political exchange), and of fiscal factors (related to the inflationary risk associated with public employment reform). The countries that have succeeded in putting the reform on a steady trajectory have done so in situations in which the strengthening of State capacity is made a priority and when the reforms are undertaken in a manner consistent with broader fiscal policy.

Subnational Authorities 12

Most Latin American countries have undergone a process of political and fiscal decentralization over the last several years. Mayors are now elected in most countries in the region, rather than being appointed by the national government. The four federal countries—Argentina, Brazil, Mexico, and Venezuela—along with Colombia, Paraguay, and Peru, also elect regional (provincial) authorities. A number of important government functions and fiscal resources have additionally been decentralized to lower levels of government. This process has induced important democratizing dynamics in many of the countries, as well as important changes in the provision of local public goods.

Subnational officials, elected or not, have always had an important role in the implementation of certain public policies. The trends of political and fiscal decentralization have only increased their importance. Whether through their role in policy implementation or through their political role in parties and the party system, subnational officials are also players in the national policymaking game. In keeping with the focus of this report, this section briefly analyzes the role of subnational political authorities in national policymaking, concentrating on the role of governors.

Subnational actors can affect the national PMP through different channels. At the implementation stage, they can exert influence by obstructing, delaying, or reshaping national policies. They can also use some of their “local” policies as strategic weapons in negotiations with national authorities. For instance, subnational governments can take actions that may compromise national macroeconomic stability. This provides them with bargaining power with respect to the national government. In some countries, subnational officials are important political players within parties and have an important say in the nomination of candidates to the national legislature. National legislators in those countries thus view provincial governors as their political patrons. In those cases, subnational actors can have an effect on the drafting or vetoing of the national legislative agenda.

The exact role that subnational actors have in the national PMP depends on their incentives, and on the formal and informal rules of their engagement in the national PMP. Governors typically have incentives to obtain resources for their provinces or regions, as well as to try to build their bases of political power. Alongside the beneficial effects of giving more voice to local interests that might otherwise “get lost” in the national policymaking arena, this role of governors can generate a coordination or “commons” problem, given their excessive focus on local and regional issues rather than on broader public goods. In some Latin American countries, such as Argentina, the province is one of the key arenas for the accumulation of party power and influence, and provincial party elites, particularly governors, are crucial players both in provincial politics and in the articulation of national political coalitions.

Rules That Affect the Role of Subnational Authorities
in National Policymaking

The willingness and ability of subnational political actors to influence national policymaking depends on a configuration of rules determining their incentives and capabilities. Some of the most salient such rules include the following:

Method of selecting subnational authorities. Governors play a more important role when they are popularly elected, rather than appointed.

Federal constitutional structure. In general, governors have played a more significant role in the national policymaking process in the four federal countries, compared to the other countries in the region. The influence of governors has arisen more gradually and recently in Mexico and Venezuela than in Argentina and Brazil because competitive elections have been institutionalized more recently as the means for selecting subnational authorities.

Territorial bicameralism. Some countries, especially federal ones, have territorial chambers (senates) that are based on the representation of subnational political units, such as provinces, rather than on population. These territorial chambers tend to increase the power of subnational political actors.

Overrepresentation of underpopulated provinces in the legislature (malapportionment). Malapportionment strengthens the political power of the majority of the provinces in the national legislature relative to the few, more populated provinces, including the capital. As a consequence, overrepresented states typically receive disproportionately higher resources per capita, controlling for other factors. By strengthening the presence of smaller provinces in the legislature and by giving them a disproportionate share of resources, malapportionment tends to increase the power of regional authorities in national policymaking.

Electoral and party system. Characteristics of the system used for selecting representatives to the national legislature are important because they shape legislators’ incentives. Among the relevant factors are the nature of the mechanisms used by parties for nominating candidates, particularly the role that regional authorities play in the nomination process; the extent of influence of regional executives on the political careers of national legislators, including their likelihood of obtaining positions in regional government; whether provinces/states are used as the electoral districts; and whether presidential elections coincide with national legislative elections and with regional elections. When regional authorities influence the nomination, election chances, and future political career possibilities of legislators, this enhances their influence on legislators from their provinces—and thus on national policymaking. Likewise, when legislative elections are held simultaneously with elections for governors but not necessarily the presidency, when legislative electoral districts coincide with the provinces/states, and when other aspects of the electoral system favor denationalized and decentralized party structures, the influence of subnational officials in policymaking will be encouraged.

Federal fiscal arrangements. The incentives of subnational political actors are strongly affected by their dependence on national funding decisions to finance their spending responsibilities. Many details of federal fiscal arrangements are important in determining the overall “federal fiscal game.” These details include the distribution of taxing authorities and spending responsibilities of provinces; the way in which transfers to cover “vertical fiscal imbalances”—imbalances between the amount provinces spend and the amount they collect in revenues—are decided and implemented; and the borrowing authority of provincial governments.13

Table 4.1 focuses on five Latin American countries where subnational actors are important, and summarizes some of the main factors affecting the role of subnational actors in the national PMP. In each of these cases, governors and mayors are now elected by popular vote.

These institutional factors interact, so that a given feature can have different effects depending on the full institutional context. For instance, the role of regional leaders in the senate might be increased by malapportionment and by the fact that senatorial candidates are nominated at the subnational level by regional leaders. Factors addressed in other parts of this report, such as the budgetary powers of the president, also affect the equilibrium behavior of key players. The actual effects of any of these features are conditional on the dynamics of each particular country case. To illustrate how these institutional factors combine to affect the role of governors, the cases of Argentina, Brazil, and Mexico are discussed below. In Argentina, governors play a very important role. In Brazil, the role has been significant but has declined over the past decade. In Mexico, governors are still secondary players, but their role is growing because of political and fiscal decentralization.

The Role of Governors in Argentina, Brazil, and Mexico

Argentina

In Argentina, subnational political actors, particularly provincial governors, are quite important in national politics and in national policymaking. The particular way in which governors enter the game, together with other institutional features of the country, helps to explain why Argentina has a relatively noncooperative policymaking process oriented to the short term. This lack of cooperation and focus on the short term has repercussions for some of the characteristics of public policies identified in Chapter 6.

Argentina has a long federal tradition. As in the United States, the power of the State is defined in the constitution as being derived from the provinces, not the central government. The political importance of Argentina’s provinces has been somewhat obscured during periods of strong national leadership (like that of Perón) or during the many military dictatorships from 1930 to 1983. Since the latest transition to democracy, however, subnational political actors have reemerged as crucial players in the national political arena.
Governors, who have been elected for most of Argentina’s democratic history, play a central role in Argentina’s national policymaking process. Provincial governments have authority over significant policy areas, and their authority has increased over the last two decades, as Argentina’s expenditure decentralization is the highest in Latin America. However, tax collection is fairly centralized, creating a large vertical fiscal imbalance, which makes provincial governments heavily dependent on national funding. The situation is further complicated by Argentina’s bicameral congress, with a territorially based senate. Argentina has the highest level of malapportionment in the upper chamber in the Latin American region, and the third highest in the lower chamber. This confers a disproportionate degree of influence on the governors of small provinces.

The organization of party politics is heavily centered on the provincial arena. Political careers, especially successful ones, are often built from provincial bases of power. The overrepresentation of small provinces in both chambers of the legislature, the methods used for nominating candidates, the characteristics of federal fiscal arrangements, and the clientelistic nature of local political loyalties all contribute to making the province a crucial arena of political exchange. The importance of governors is reflected in the fact that three out of the five presidents since the return to democracy have been governors (Menem, Duhalde, and Kirchner).14

The electoral system is based on closed lists with the provinces as electoral districts. This allows provincial party leaders in general and governors in particular to play a very influential role in the nomination process for legislative elections. As mentioned in Chapter 3, Argentina has a fairly denationalized party system, and the main parties in the country can be described as confederations of subnational bases of power. Subnational party leaders are crucial players in the nomination of candidates for the national congress, and national legislators tend to view these local leaders as their bosses. This and other factors of the Argentine institutional system help to explain why the national congress has a relatively weak role with respect to the adoption of national public policies. Scholars have characterized Argentine legislators as professional politicians but amateur legislators, since they tend not to get reelected and have relatively weak incentives to invest in the policymaking capabilities of congress.15 Instead, they tend to focus on the short term, which has an adverse effect on the quality of public policies.

Several of the crucial political and policy decisions of the last decade have been negotiated in meetings of the president or some key minister with provincial governors. The arenas for these political and policy exchanges, in turn, are not very well institutionalized. This, together with a judiciary that plays a limited enforcement role and a relatively weak bureaucracy, tends to create a policymaking process that concentrates on the distribution of short-term benefits, to the detriment of longer-term, broader concerns.

These tendencies are exemplified by the frequent exchanges of provincial support of national policies for fiscal benefits to the provinces. This contributes to creating a political system that tends to operate more on the basis of clientelistic exchanges than of policy debates and consensus-building. As a consequence, policies tend to be volatile and obtain credibility only if they are “hard-wired.” Policies also have a tendency not to be well coordinated or enforced, and often tend to favor specific constituencies instead of more public-regarding objectives.

Brazil

In Brazil, several factors inherited at the time of the return to democracy in the mid-1980s at first contributed to giving state governors a significant role in national policymaking. Yet recent trends have tended to water down the significance of this role. These trends have followed from the institutionalization of democracy, as well as from some institutional reforms adopted in response to the severe fiscal crisis the country experienced in the early 1990s.

During the transition to democracy, governors were directly elected seven years before the first direct election of the president. This conferred legitimacy on the governors and empowered them to play a crucial role in the elite’s bargaining game. This particular distribution of power and legitimacy resulted in a constitution that reflected the relatively strong position of the governors.

The Brazilian Constitution vests a number of policy domains as well as revenue authority in the states, which at the same time enjoy administrative autonomy. In this context, if the policy preferences of governors and the executive diverge, governors have some power to advance their preferences. Because governors are not primarily concerned with fiscal stability at the national level, and they have a preference for higher federal public spending and geographically concentrated investments, fiscal policy usually has been a matter of conflict.

The influence of governors has always been indirect, stemming from their role during the democratic transition and the substantial tax authority that the constitution grants them. Brazil’s version of the value-added tax—the ICMS—is collected by the states and represents the single most important tax in the country, accounting for one-third of tax revenues.

Given the influence that they could have on the political careers of legislators at the state level, governors had the capacity to influence the behavior of federal deputies and senators in congress. The relative dependence of legislators on support from governors was reinforced by the fact that until 1994 legislative elections coincided with elections for governors and state legislators, but not for the presidency. This meant that congressional candidates coordinated their campaigns around gubernatorial candidates and the electoral and other resources they could provide, not presidential candidates or national parties, which did not fund legislative candidacies. The economic autonomy enjoyed by governors allowed them to provide valuable support to candidates.

However, the influence of governors faded as political actors at different levels began sharing in the legitimacy conferred by direct popular election, and with the shift to simultaneous presidential and congressional elections. The inflation crisis, and the Real Plan implemented to solve it, allowed the federal government to impose certain conditions on the states. This undercut the power of governors, who were forced to privatize banks and public enterprises. The governors also lost vital resources with which to influence national and subnational politics.

In sum, throughout most of the last decade, the national executive has been able to implement its agenda by recentralizing the political game. This has included passing legislation that adversely affected state governors, such as the Social Emergency Fund (FSE). This constitutional reform, passed in 1993, allowed the Franco administration to use revenues to reduce the central government deficit that had previously been earmarked in the 1988 Constitution for specific expenditures that ultimately benefited the states. As the financial sovereignty of subnational governments has weakened, regional authorities have also lost much of their influence on the nomination process for national legislators.

Mexico

Two of the most relevant features of Mexico’s PMP during the 20th century have been the hegemonic position attained by the PRI in the party system and the federal nature of its regime. Governorships have always been important in a federal country with a political system that could not resolve its political and power struggles within the electoral system. In fact, state governorships were the most valuable positions delivered by the president to party leaders; they were even preferred to senate seats. Notwithstanding their importance in the PRI hierarchy, governors were not autonomous agents in the PMP.

The importance of these positions motivated legislators to compete for the PRI nomination for governor alongside prominent federal politicians and mayors of large municipalities. Governors had control over a considerable system of patronage, which became even more important as the process of decentralization deepened.16

As in many other countries, because the decentralization process developed gradually, the degree of political, fiscal, and functional decentralization was not homogenous. Fiscal decentralization was the starting point of a broader process of decentralization. The federal government’s decision to initiate the process of decentralization was probably based on the desire to stave off the PRI’s political decline.17 As the PRI lost its ability to guarantee economic growth, the federal government was forced to relinquish some control over expenditures, taxes, and policymaking to local governments, which, in turn, were better able to cultivate support with these added resources.18 Control of these resources is directly related to the role governors acquired on the national stage.

Decentralization fundamentally changed the once centralized and opaque PMP because it opened the door to the insertion of local interests into national politics. The role of local leaders in national politics gained importance. This process was strengthened when, for the first time, the PRI lost a governorship in 1989 and regional elections started to become competitive. Since then, governors have increasingly played a relevant role in national policymaking. The position of governor is now a key stepping-stone to the presidency, in contrast with the PRI-dominated era, when national executive positions were key. Furthermore, the patronage network that governors control has allowed them some influence over legislators. This influence has been small but growing, particularly with respect to the nomination of legislative candidates and to some extent with respect to determining the future career of legislators, who are barred from reelection.

These decentralizing processes generated a fundamental realignment of the interests represented in the national congress, which then had broader implications. First, states have gained political leverage over the federal government. Additional taxes have been placed within the national tax revenue sharing system (SNCF)19 and the percentage of total tax revenues allocated to state governments has increased. Total transfers to states increased by nearly 20 percent, on average, in the early 1990s.20 Second, within Mexico, opposition-controlled states and poorer states appear to have gained additional leverage over the federal government during negotiations on the distribution of federal revenue transfers. Thus the political and fiscal components of the decentralization process have mutually reinforced one another.

The Judiciary 21

Another actor that can potentially play an important role in policymaking is the judiciary. Historically, in much of the region the judicial branch has been characterized by dependence on the executive and a lack of activism in interpreting the law, in challenging the legality of executive actions, or in reviewing the constitutionality of laws. Along with the broader process of democratization, the judicial reforms adopted over the last two decades have changed the structure and the operation of the judicial system in the region. The potential for the courts’ involvement in the policymaking process (reflected in the levels of judicial independence) has increased significantly. As a result, in some countries the judiciary is assuming a more active and independent role in the adoption, implementation, and enforcement of public policy.

With the establishment of institutional structures that are conducive to higher levels of judicial independence, courts have become less subservient and the judiciary has repositioned itself in relation to other branches of government. The number of judicial rulings against the executive’s preferences has generally increased in many Latin American countries. Overall, court decisions are assuming greater importance in both public policy and politics.22

The Potential Policymaking Roles of the Judiciary

Judiciaries can assume a number of different roles in the policymaking process. These potential roles can be divided into four categories: veto player, policy player, impartial referee, and societal representative. The role of judiciaries in Latin America varies in terms of which of these roles is performed and the degree of activism and independence with which they are carried out.

Veto player. In this role, the judiciary vetoes legislation on constitutional grounds, but also partly on the basis of its own preferences. If the judiciary is independent and performs the veto player role, it may be more difficult to change policy arbitrarily, and the legislature and the executive may have to take into account the preferences of the judiciary. The veto role is more extensive and has a more profound impact when the courts’ decisions apply to the law in general rather than to just a specific case. Given the trend toward “hard-wiring” many aspects of public policy into constitutions, the role for the judiciary in this regard has expanded.

Policy player. When the courts review laws to determine legislative intent or when they give new interpretations to legislation on the basis of their views of the legal system as a whole, they are imposing their own policy preferences on policy outcomes. Institutional features favoring this kind of role include those in which the courts can make rulings in the absence of a concrete case, before the law is even adopted, and when the courts’ rulings on the constitutionality of laws are applicable in general, rather than to just a specific case.

Impartial referee. Courts can be called upon to ensure that public policies are effectively applied by acting as an external enforcer of agreements undertaken by others (including those encoded in the constitution) and a mediator between contracting parties. This role is being played when the court makes sure the executive does not exceed its powers, for example, by legislating by decree on matters to which this power is not supposed to apply; when it stops the government from enforcing taxes it is not supposed to levy; or when it protects the autonomy of an independent central bank in monetary policy. An effective performance of the impartial referee role can facilitate the forging of political agreements needed to adopt policy reforms and contribute to the fair and efficient implementation of those reforms.23

Societal representative. It is often difficult for certain sectors of the population (such as the poor and rural farmers) to influence the formation of policies. Because judiciaries can help enforce constitutional stipulations for equal protection under the law and for the defense of civil and social rights, they can provide a voice for marginalized groups and an alternative channel for societal representation.

The scope of judicial activism in ten Latin American countries in relation to these four potential roles is portrayed in Table 4.2.24 The impartial referee role is performed especially actively in Brazil and Chile. In these countries, the courts also actively perform the veto player and policy player roles. Particularly as a result of the 1991 constitutional reform, the newly created Constitutional Court in Colombia has played an increasingly active role in blocking and/or interpreting legislation (veto and policy player roles). Conversely, in Argentina, Ecuador, Paraguay, Peru, and Venezuela, the judiciary has tended to be more dependent on the executive. This has limited its ability to play a significant role as an impartial referee or as a policy player.

An active and independent judiciary is likely to affect the policymaking process and the key features of public policies in several ways. First, as an impartial referee, the judiciary can help ensure the effectiveness of the implementation and enforcement of policies. Second, as an external enforcer of agreements, it can help facilitate inter-temporal political transactions, such as commitments by the executive to regulate privatized firms effectively or enact pension reforms that reduce benefits now in exchange for the greater security of benefits in the future. This can encourage greater stability of policies, since some agreements (or policy equilibria) in the public interest will be possible that would not be otherwise. Third, in performing the societal representation role, the courts can contribute to making policies more public-regarding.

Fourth, the judiciary can contribute to adaptability. For example, when policies have been delegated to an autonomous agency (like the central bank), and the judiciary upholds the terms of that delegation, the agency can then manage some area of policy as it sees fit, without the need for safeguards that would prevent opportunism at the cost of rigidity.

Fifth, when judiciaries are active as veto players, policies are likely to be more stable, since policy changes must be consistent with the preferences of another institutional actor, with the constitution and existing laws, and with the preferences of other actors that may act through the judiciary to challenge policy changes. However, in performing this role, the judiciary might make policies less adaptable, depending upon how its preferences match with those of the executive and the legislature. As policy players, the courts could potentially make policies more volatile.

Factors Affecting the Judiciary’s Policy Independence and Activism

The level of independence of the judiciary in a given country is related to both the potential for individual judges to act according to their own policy objectives and the potential for courts—collectively—to be involved in the policymaking process.

Judicial independence can be defined as the judiciary’s ability “to make decisions according to the law and not based on external or internal political factors.”25 It has four inter-related dimensions: substantive independence, or the power to make judicial decisions and exercise official duties subject to no other authority but the law; personal independence, or stability of tenure and freedom from intimidation or threats; collective independence, or judicial participation in the central administration of courts; and internal independence, or independence from judicial superiors and colleagues.26

Among the characteristics of the judiciary that affect judicial independence, several stand out: the extent of budgetary autonomy; the level of transparency and the extent of the use of meritocratic criteria in the process for nominating and appointing judges; the stability of the tenure of judges; and the reach of judicial review powers. Effective judicial independence also depends upon the behavior of other actors, such as whether the president or political parties regularly interfere with the courts. This, in turn, depends upon these actors’ incentives.

Judicial Budget Autonomy

In the context of public budget scarcity, the reliability and autonomy of the judicial branch’s budget is essential to judicial independence. Although in many countries, the judicial branch itself prepares the annual budget (Costa Rica, Guatemala, and Uruguay, for example), in most cases the budget’s execution still depends on the approval of either the legislature or the executive, thus hindering the branch’s ability to administer its own resources. In some countries—mostly in Central America—the minimum budget dedicated to the judicial branch is stipulated by law. However, that does not necessarily mean that this amount is allocated in practice.

The share of the public sector’s budget dedicated to the justice sector varies significantly across countries, from over 3 percent in Costa Rica to less than 1 percent in Ecuador and Peru.27 Although the size of the budget alone does not ensure efficiency, it is an important indicator of the amount of funds available to the judiciary to carry out its operations.

Appointment System

Beyond financial resources, the judiciary needs to have qualified judges with appropriate incentives to ensure efficiency and the rule of law. The ways in which judges are chosen affects the extent to which judges are motivated to apply the law independently and impartially and to perform their role efficiently.

A wide variety of systems are used to select supreme court judges. In Argentina, Brazil, and Mexico, candidates are nominated and appointed by the president with the approval of the legislature (or one of its chambers).28 In other countries, such as Bolivia, El Salvador, Guatemala, and Honduras, the candidates are nominated by a judicial council or nominating board and then appointed by the legislature (or one of its chambers). In another group of countries, including the Dominican Republic and Peru, supreme court judges are appointed by a judicial council. Other things being equal, nonpartisan and multi-member councils would be expected to be more effective in promoting judicial independence than processes in which the executive and legislature play a substantial role in nominating and appointing judges.

The degree of transparency with which the nomination and appointment process takes place is also important for favoring or hindering judicial independence. Some countries, such as Chile and Guatemala, have put in place recruitment procedures that encourage competition, transparency, and the participation of renowned magistrates and academics in the selection of new judges.

Nonetheless, the simple creation of laws, judicial councils, and training programs may not correspond to the reality of the appointment system in the judicial sector. Many of the new requirements for choosing judges may be completely bypassed or implemented poorly.

Terms and Tenure

The reasoning behind establishing longer (or lifelong) terms for judges is to provide greater job security so that judges’ concerns about holding their position or being promoted do not influence their decisions. Thus, moving away from short terms that coincide with presidential and congressional elections is likely to favor judicial independence.

Some countries, including Argentina, Brazil, Chile, Ecuador, and El Salvador, have adopted lifetime tenure for judges (in some cases limited by a mandatory retirement age). For other countries, such as Bolivia, Colombia, and Costa Rica, only lower-court judges have lifetime tenure, while supreme court judges have fixed terms. In Paraguay, supreme court judges have life terms, while first- and second-tier judges can remain in their jobs for only five years. Finally, in some countries, such as Guatemala and Peru, judges have fixed terms, which must be periodically renewed.

It is important to highlight the gap between the letter of the law and reality. Despite the existence of lifetime tenure for judges in Ecuador, for instance, President Lucio Gutiérrez removed 27 of 31 supreme court judges in December 2004. Thus longer de jure tenure for supreme court judges has not necessarily translated into the lengthening of tenures in many countries. Figure 4.6 shows the average tenure of supreme court judges in selected Latin American countries from 1960 to 1995.

Judicial Review Powers

Judicial review is the power of a court (generally the supreme court, or in some cases, the constitutional court) to declare laws and other administrative acts unconstitutional. This power can vary according to whether it can be applied only to a specific case or to a hypothetical case (concrete or abstract), whether it can be applied before or only after a law is enacted (a priori or a posteriori), and whether only the supreme court can engage in judicial review or lower courts can do so as well (centralized or decentralized authority).29

Despite the continuing short­comings in constitution­­al adjudication provisions, there is a consensus that they have been strengthened considerably in the past two decades. (Box 4.2 presents the case of Mexico, where the supreme court has emerged as an important veto player.) The control of the constitutionality of laws is especially relevant in democratic contexts in which the constitution is detailed and there is a jurisdictional overlap of government responsibilities (as in the case of federal democracies). In these circumstances, conflicts and questions regarding constitutional issues arise frequently.

Despite recent reforms, there is still considerable variation in the region in respect to the scope of judicial review powers, in terms of the breadth of mechanisms for bringing issues to the attention of the courts, the extent of the applicability of judicial decisions based on individual or institutional complaints, and whether individual complaints can provide a basis for rulings on the general constitutionality of laws.30

The structural characteristics described above suggest that the impact of Latin American judiciaries on the policymaking process and public policy is uneven across the region. This variation in the potential for courts’ involvement in politics is corroborated by available indicators of de facto judicial independence. Table 4.3 presents two such indicators. The first one, published by the World Economic Forum in 2004, is a subjective indicator based on survey responses of business executives. According to this index, Uruguay, Chile and Brazil appear at the top of the ranking. The second one, developed by Feld and Voigt,31 is based on objective criteria, such as the actual tenure of supreme court (or constitutional court) justices and its deviation from de jure tenure, whether justices were removed before the end of their term or the number of justices was increased, and whether the budget of the supreme court or the income of the justices was reduced in real terms, among others. According to this index, Costa Rica, Mexico and Chile appear at the top of the list.32