Definitions of Regime Type in the Literature

CEUeTDCollection

Regionalism also plagues presidentialism, since apart from the United States presidentialism mostly appeared in Latin America and only in the developing world, which makes it difficult to disentangle regime type from socioeconomic, cultural and other factors.²⁰⁹ Small nations have an advantage in democratic stability, because they tend to be more homogeneous in ethnic and religious terms. Here parliamentarism has “a built-in advantage simply because Britain colonized many small island territories,” not giving researchers a chance to see how presidentialism might have operated within such settings.²¹⁰ It may well be that parliamentarism operates better in developing countries: stable democracies with low and medium income are all parliamentary regimes. They, however, are also former British colonies, which again points toward the importance of colonial heritage.²¹¹

CEUeTDCollection

and the prime minister is selected by the legislature. Later he added a third criterion:

parliamentarism requires a collective or collegial executive.²¹⁴

Shugart and Carey note that most contemporary definitions are very much alike, stressing the popular election of the chief executive, fixed term of office for the legislature and the chief executive, the chief executive’s selection of and authority over the cabinet.²¹⁵ They argue that all these aspects capture the same phenomenon, that is, the separation of powers. The two systems differ in this aspect, that is, with regard to the origins and survival of the executive. Thus while presidentialism is built on the idea of maximum separation of powers, parliamentary democracies lack this feature.²¹⁶ Shugart and Carey find Lijphart’s collective vs. single-headed executive criterion redundant, citing the empirical reality of Uruguay between 1919-1933 and 1954-1964, which suggests that presidential democracies may have collective executives. However, they believe that in spite of the necessity of separation of powers, presidents must be granted some law-making authority – at the least, veto power – constitutionally, because otherwise they are nothing more than mere executors.²¹⁷

Despite the general agreement between researchers about the defining difference between parliamentary and presidential regime types, seeing regime type by the parliamentary-presidential dichotomy is not without problems. First, Jones asserts that current definitions of presidentialism deceptively direct too much attention to the role of the president in the system; therefore he seeks to replace ‘presidentialism’ with the term, ‘separated system’.²¹⁸ Second, what is worse, arguments in favor of one or the other system of government often have different regime types in mind: presidentialism is sometimes measured

214 Lijphart 1999, 117-118; Shugart and Carey 1992, 20. See also, “Chapter: The Constitution” in Hudson, Rex A. and Sandra W. Meditz, eds. 1990.Uruguay: A Country Study. Washington: GPO for the Library of Congress, 1990. Available:http://countrystudies.us/uruguay/62.htm. Access: May 25, 2005.

215 Shugart and Carey 1992, 20.

216 Shugart and Carey 1992, 15.

217 Shugart and Carey 1992, 20-21.

218 Jones 1997, 20-22; Jones, Charles O. 1994.The Presidency In a Separated System. Washington, D.C.: The Brookings Institution, 1-2.

CEUeTDCollection

against majoritarian (Westminster) parliamentarism and, at other times, it is contrasted with parliamentary regimes with coalition governments – whichever serves the point to be made better.²¹⁹ This points to a more general problem, that is, if parliamentary (and presidential) regimes are capable of such diversity they do not form a single type. In other words, intra-class variation is bigger than inter-intra-class variation. According to Rockman, this makes it necessary to reformulate the presidential-parliamentary differentiation.²²⁰

Third, separation of powers may be present in parliamentary regimes in the form of coalition governments, federalism, or bicameral legislatures (if the two houses have reasonably symmetrical powers).²²¹ Definitions cannot tell much about the relative strength of presidents and prime ministers vis-à-vis the legislative or each other, either. The presidency in America is sometimes described as being very strong; at other times, as very weak.²²² To be sure, presidential strength may vary by issue area, depending on how much power the president has constitutionally or delegated by Congress. However, variation in power and influence exist within one issue area even across and within presidential administrations.

Sartori notes something similar with regard to prime ministers: a government’s dependence on the legislative branch for ‘election’ and survival says little about why a government is strong or weak.²²³ Jones observes the same about presidents whose power depends on their resources, advantages, strategic position and opportunities.²²⁴ This suggests that while structural features are a good starting point in investigating the workings of various democratic executives, deduction from institutional features alone results in misleading

219 Rockman 1997, 60.

220 Rockman 1997, 60. See also Sartori 1997, 105 on definitional problems.

221 Peters 1997, 69 and Haggard and McCubbins 2001, 3.

222 Reiter and Tillman 2002, 815; Jones 1997, 25.

223 Sartori 1997, 107-108.

224 Jones 1997, 2.

CEUeTDCollection

conclusions. In other words, changes in political conditions must also be taken into account.²²⁵

All this suggests that despite its problems the presidential-parliamentary division remains the most fundamental distinction between democracies.²²⁶ While pure types are rare and some intermediate regime types exist between these two, most regimes fall within these two categories.²²⁷ Moreover, the separation of powers is a viable concept to separate one from the other. However, minor modifications in the content of separation of powers are in order and, to be able to take the political dimension into account, the introduction of an additional dimension is necessary. The next section starts with the explication of this distinction so as to provide a starting point to describing the hypotheses about the effect of regime type on risk-taking.