John M. Carey

Incumbency and the Probability of Reelection in State Legislative Elections

We build on work estimating and explaining the incumbency advantage in state legislative elections. Our work makes advances in three ways. First, our model measures the effect of incumbency on the probability of reelection, rather than on candidate vote share or margin of victory. Second, we accommodate both multimember district (MMD) elections that are excluded from most previous studies and uncontested and partially contested (MMD) races. Third, we use an improved method of controlling for the underlying partisan makeup of districts. We calculate incumbency advantage using data from elections in 96 legislative chambers across 49 states in the 1992-1994 electoral cycle. We then model relative incumbency advantage across the states as a function of institutional characteristics. We find that district type, term length, and electoral formula have substantial effects on incumbent safety; incumbents in multimember post and free-for-all districts are more vulnerable than those in traditional SMDs, as are those with four-year, rather than two-year, terms Professionalization also affects incumbency safety, and salary rather than other resources best accounts for incumbency advantage.

Parchment, Equilibria, and Institutions

Institutions are rules that constrain political behavior. Although there is consensus that being written down is neither necessary nor sufficient for an institution to be effective, much research on comparative institutions focuses on formal, parchment, institutions. This article argues that parchment can contribute to the generation of shared mutual expectations among political actors, which are essential to the effectiveness of institutions. Next, the article distinguishes between research that emphasizes the role of institutions in aggregating preferences into political decisions and research that relies on coordination models to identify conditions favoring certain equilibrium outcomes when multiple equilibria are possible. The article notes the increasing prominence of such coordination models in research on comparative institutions and concludes with some reflections about the prospects for this trend to foster connections between institutional analysis and the field of comparative politics more broadly.

The Effects of Term Limits on State Legislatures

Legislative theory suggests that anticipatory effects of term limits would first affect the types of individuals elected to office and only later influence the legislature itself. Our results, based on a 1995 survey of nearly 3000 state legislators nationwide, indicate otherwise. There are no systematic differences between term limit and non-term limit states in the composition of the legislature (e.g., professional backgrounds). Yet with respect to legislative behavior, term limits decrease the time legislators devote to securing pork, and heighten the priority they place on the needs of the state and on the demands of conscience relative to district interests. At the

The Effects of Term Limits on State Legislatures: A New Survey of the 50 States

Term limits on legislators were adopted in 21 states during the early 1990s. Beginning in 1996, the limits legally barred incumbents from reelection in 11 states, and they will do so in four more by 2010. In 2002, we conducted the only survey of legislators in all 50 states aimed at assessing the impact of term limits on state legislative representation. We found that term limits have virtually no effect on the types of people elected to office�whether measured by a range of demographic characteristics or by ideological predisposition�but they do have measurable impact on certain behaviors and priorities reported by legislators in the survey, and on the balance of power among various institutional actors in the arena of state politics. We characterize the biggest impact on behavior and priorities as a �Burkean shift,� whereby term-limited legislators become less beholden to the constituents in their geographical districts and more attentive to other concerns. The reform also increases the power of the executive branch (governors and the bureaucracy) over legislative outcomes and weakens the influence of majority party leaders and committee chairs, albeit for different reasons.

Presidential Agenda Control and Spending Policy: Lessons from General Pinochet's Constitution

Formal institutions put in place upon the establishment of a new democracy can have profound effects on political bargaining. We demonstrate how the budgetary procedure bequeathed by the outgoing Chilean military regime affects policy choices available to elected officials. Chilean budget procedure should discourage deficits, allow for a reduction in the relative size of the defense budget, and facilitate cuts in executive proposals when the institutional interests of the legislature are at stake but not under conditions of coalitional conflict. We present a simple spatial model of bargaining over spending decisions between the executive and Congress that facilitates comparisons between the Chilean budget procedure and that of other presidential systems. The model suggests that, relative to other regimes, Chiles budget process should constrain spending and favor the presidents preferences over the legislatures. Comparative fiscal data from twelve other presidential democracies and from the first eight Chilean budgets since the transition to democracy, as well as interviews with key legislators and executive officials, all support our hypotheses.

Primary Elections and Candidate Strength in Latin America

Political parties throughout Latin America rely increasingly on primary elections to select candidates for public office. Where they are adopted, primaries are generally touted as moves toward openness and internal party democracy. Yet politicians and party leaders are concerned with winning elections, and there are reasons to expect that primaries select candidates who are weaker in general election competition than other methods. Using data from every democratic presidential election in Latin America since the late 1970s, we test whether primaries systematically affect candidate strength. We find evidence of a primary bonus—that is, other things equal, primary-selected candidates are stronger than those selected by other procedures.

Parties and Accountable Government in New Democracies

Political scientists have long associated accountability with strong political parties, and have also frequently noted the weakness of parties in new democracies. This article disaggregates the strong party ideal into two components - legislative discipline and programmatic platforms - and suggests that the former in the absence of the latter can undermine accountability. We describe levels of discipline in parties in various new democracies. Then we provide a taxonomy of political party origins, according to how origins affect the proclivity of parties toward both programmatic policy and legislative discipline.

Insurance for Good Losers and the Survival of Chile's Concertacion

In the transition from military rule to democracy, the government of Augusto Pinochet bequeathed to Chile a unique electoral law by which all legislative seats are contested in two-member districts. A key implication of this rule is that in order to secure legislative majorities, coalitions have to put their strongest candidates in the most precarious electoral list positions. This generates a divergence of interests between coalitions and politicians. Chiles largest coalition, the Concertación, has resolved the dilemma by providing appointed posts to unsuccessful congressional candidates who accept personal political risk on the coalitions behalf. This study argues that this insurance system has provided the critical glue to hold the coalition together since Chiles transition to democracy in 1990. Recent changes in the electoral environment could threaten the Concertacións control over the appointed posts that have sustained this informal institution. This could jeopardize the Concertacións cohesion during the process of negotiating coalition candidate lists for the 2005 legislative elections.

Transparency Versus Collective Action Fujimori's Legacy and the Peruvian Congress

During the 1990s, the Peruvian Congress was thoroughly dominated by the presidential administration of Alberto Fujimori. In 2000, during the scandals that brought down Fujimori and his associates, the legislature was exposed as riddled with corruption. In the wake of these events, one might expect the congress to be thoroughly discredited and institutionally weakened. I argue that the Peruvian Congress made significant strides toward legislative transparency and strengthening the accountability of individual legislators to voters and that, ironically, some of the key steps in this direction were taken during the Fujimori era. This same trend, however, presents potential obstacles to collective action in the legislature. I support these claims with analysis of changes in the legislative party and coalition system from the late 1990s through the transition, examination of legislative voting patterns, and interviews with key actors inside and outside congress.

State-Level Institutional Effects on Legislative Coalition Unity in Brazil

Research on political institutions suggests that sub-national factors affect the proclivity of legislators from the same party or coalition to vote together. We estimate the effects of such forces operating at the State-level - intra-list electoral competition, and alliance with governors. We propose that larger cohorts, in which the imperative for legislators to distinguish themselves from the group is stronger, should be less unified than smaller cohorts. We also derive from theoretical hypotheses that cohorts allied with governors may be either more or less unified than other cohorts. We analyze unity among coalition cohorts on recorded floor votes in the Brazilian Chamber of Deputies. We find support for the hypothesis that larger cohorts are less unified, but detect no net effect of alliance with governors on cohort voting unity. Governors are not dominant brokers of legislative coalitions, suggesting that the net gubernatorial effect is contingent on factors that shape their influence relative to national-level legislative actors.