This is a guest post by Johannes Freudenreich, Postdoctoral research fellow at the Geschwister-Scholl-Institut für Politikwissenschaft at the University of Munich. It is based on an recent article in Latin American Politics and Society
In the beginning of the 21st century, prospects of Latin American presidential democracies were good. The dictatorships of the 1970s and 1980s had vanished, economies were constantly growing, and comprehensive social welfare programs were implemented. Many political scientists link these successes to the ability of Latin American presidents to form, maintain and manage cabinet coalitions (Cheibub 2007). The differences between presidential and parliamentary systems of government seemed to have become rather marginal. Both presidents and prime ministers achieved legislative majorities by forming broad cabinet coalitions and critics of the presidential form of democracy, such as Juan Linz (1994), seemed to be proven wrong. However, soon presidential impeachments became the new pattern of political instability in the region (Pérez Liñan 2007). Cabinet reshuffling remains constantly high and broad corruption schemes, directly linked to coalition politics, have been disclosed, such as the Mensalão Scandal in Brazil, where the ruling party of President Lula da Silva used illegal side payments to secure the legislative support of members of the ruling coalition.
My recent article in Latin American Politics and Society takes a systematic look at the formation of cabinet coalitions in presidential systems over the past 25 years. It analyzes the extent to which presidents in 13 Latin American countries have formed coalitions that increase their law-making capabilities, and whether presidents form coalitions tailored to find majorities in Congress especially when presidents have low independent influence over policy based on their institutional law-making powers.
The study complements the perspective that cabinet coalitions are largely an instrument for finding legislative majorities with the idea that presidents use cabinet posts to honor pre-electoral support. The reason is the following: presidential elections provide strong incentives for electoral coordination because they tend to favor two-candidate competition. In a multi-party setting, this means that parties have incentives to form pre-electoral coalitions to present joint presidential candidates. When negotiating pre-electoral pacts, parties are likely to agree on how to share the benefits of winning including cabinet posts. After the election, presidents find it difficult to abandon these agreements as they need the trust and support of other parties within and outside of their coalition during their presidential term. Thus, it is expected that cabinet coalitions are likely to be based on the electoral team of presidents and that other legislative parties are invited to join the cabinet only additionally to parties of the existing pre-electoral coalition.
The study further argues that parties attractive as pre-electoral coalition partners are not necessarily the ones that would achieve cabinet participation if the negotiations of cabinet posts were an unconstrained post-electoral process. For example, in a one-dimensional policy space, extreme parties, parties more extreme than the president to the median legislator, are relatively unimportant for legislative decisions and thus unlikely to be included in the cabinet for legislative reasons. In a presidential race, however, extreme parties can provide valuable votes and campaign resources and therefore have far stronger blackmailing power. Furthermore, presidential contests produce a strong antagonism between the president and the parties of the president’s electoral rivals. Since the president’s survival in office is not contingent on the support of other parties in parliament, parties that present a strong presidential candidate are likely to be excluded from the cabinet, even if their inclusion is rational from a lawmaking perspective. It is therefore expected that the party of the runner-up is generally excluded from the presidential cabinet and that the overall explanatory power of variables of legislative bargaining increases once one controls for the effects of pre-electoral coalition formation and competition.
The study empirically evaluates this argumentation on the basis of so-called conditional logit models, presenting a new empirical strategy to analyze cabinet formation under this type of regime. The tests are conducted on a new dataset of 107 democratic cabinets in Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, the Dominican Republic, El Salvador, Honduras, Nicaragua, Panama, Uruguay, and Venezuela. Based on the new method and data, this study presents the most comprehensive test yet of the determinants of the partisan composition of presidential cabinets.
The most note-worthy empirical results are:
First, presidents try to form majority coalitions, but it is the upper house majority not the lower house majority which makes cabinet coalitions significantly likely to from. One potential explanation for this phenomenon is that there are generally fewer parties in the upper than in the lower chamber, due to the disproportionality of electoral systems used to elect upper chambers in Latin America. Thus, the president’s party is often overrepresented in the upper house, which makes it easier for presidents to find majorities. Furthermore, upper chambers are generally strong in Latin America (Nolte and Llanos 2004), and controlling an upper chamber is often sufficient for the president to prevent a veto override.
Second, contrary to expectations in the literature, extensive presidential decree powers decrease the probability of the occurrence of cabinets which control only a minority of seats in the lower house of congress. A potential explanation for this phenomenon is similar to the argument developed by Strøm (1990) for minority governments in parliamentary systems. Parties prefer to stay in opposition when the government has a weak independent influence on policy. The other explanation is that pre-electoral coalition formation is more prevalent when presidents’ institutional authority is high, as political actors make a relatively simple calculation about the benefits and the costs of coordination in presidential elections. The more powerful the president, the higher the incentives for pre-electoral coalition formation (Hicken and Stoll 2008; Freudenreich 2013). And if the a coalition is in power anyway, it is easier to extend this coalition to secure a majority in the lower house of congress.
Third, considerations of governability and pre-electoral bargaining describe two distinct yet compatible sets of factors that influence cabinet formation in presidential systems. Many cabinet coalitions in Latin America are congruent or extended versions of the pre-electoral coalition of the president and parties of the main presidential competitor are generally excluded from the cabinet, but these factors are distinct to the incentives of legislative bargaining. The explanatory power of variables associated with governability increases once variables of pre-electoral bargaining are included in the statistical model. For example, cabinet coalitions are more likely to form when they include the median party in the lower chamber of congress, but this effect is only statistically significant when one controls for the effects of pre-electoral bargaining.
Overall, the paper tries to show that an inclusive approach is necessary to study coalition dynamics in presidential systems. Pre-electoral commitments strongly affect cabinet formation and thereby also confound the relationship between cabinet formation, legislative bargaining and governability.
Cheibub, José A. 2007. Presidentialism, Parliamentarism, and Democracy. New York: Cambridge University Press.
Freudenreich, Johannes. 2013. Coalition Formation in Presidential Systems. Ph.D. diss., University of Potsdam.
Hicken, Allen, and Heather Stoll. 2008. Electoral Rules and the Size of the Prize: How Political Institutions Shape Presidential Party Systems. Journal of Politics 70, 4: 1109–27.
Linz, Juan J. 1994. Presidential or Parliamentary Democracy: Does it Make a Difference? In The Failure of Presidential Democracy: The Case of Latin America, ed. Linz and Arturo Valenzuela. Baltimore: Johns Hopkins University Press. 3–89.
Nolte, Detlef/Mariana Llanos. 2004. “Starker Bikameralismus? Zur Verfassungslage lateinamerikanischer Zweikammersysteme.” Zeitschrift für Parlamentsfragen 35: 113-131.
Pérez-Liñán, Aníbal 2007. Presidential Impeachment and the New Political Instability in Latin America. Cambridge University Press: New York.
Strøm, Kaare. 1990. Minority Government and Majority Rule. Cambridge: Cambridge University Press.