Since the early 1990s the study of presidential democracies has become a fast-
growing industry in the sub-discipline of comparative politics. The original questions that
motivated the first works on the topic (see, among others, Lijphart 1992; Linz 1994;
Shugart and Carey 1992) were essentially two: (1) what are the main differences between
presidentialism and parliamentarism?, and (2) does the former foster less stable and
effective democratic regimes than the latter?
Since then the research agenda on presidentialism has expanded, and nowadays its
main concerns are, not surprisingly, very similar to those that have been driving students of
parliamentary democracies since the early 1970s, namely, the identification of patterns of
coalition formation, cabinet selection, cabinet durability, and ministerial turnover. Most of
the works on the new agenda on comparative presidentialism have focused on the first two
aspects (Altman 2000a; 2000b; Amorim Neto, 2002; 2006; Cheibub, 2002; Cheibub,
Przeworski, and Saiegh, 2004; Deheza, 1997; Geddes, 1994; Lanzaro, 2001; Negretto
2003). It is now time to devote more research effort to the investigation of the third and
fourth aspects. This paper purports to analyze the third one, cabinet durability.
In a forthcoming piece (Amorim Neto 2006) I propose a decision-theoretic model to
explain how cabinets help presidents implement their policy-making strategies. Presidents
are assumed to have two policy-making strategies: a strategy based on the use of statutes or
a strategy based on executive prerogatives. If the president’s preferences and the
institutional incentives and economic conditions he faces favor a statutory strategy, he is
more likely to appoint a majority cabinet, select more partisan ministers, and distribute
portfolios to parties on a proportional basis (the measure of proportionality will be defined
below). Econometric analysis of 106 cabinets appointed in 13 countries of the Americas
1
On ministerial turnover, see Martinez Gallardo (2005).
2
Convention