Search

Why do legislative parties emerge in democracies where elections are contested by individual candidates, rather than national party organizations? And can parties survive in the absence electoral pressure for their members to work on shared political goals? In this article, we examine the emergence and maintenance of party discipline in an atypical legislative context: California's 1878–79 constitutional convention. The unusual partisan alignments among the delegates at the California convention provide us with a unique empirical opportunity to test election- and policy-based explanations for legislative discipline. Our study combines a careful reading of the historical record with a statistical analysis of roll call votes cast at the convention to show how leaders of the “Nonpartisan” majority held together their disparate coalition of Democratic and Republican members in the face of conflicting preferences, ideological divisions, and well-organized political opponents. Our findings provide evidence that cohesive parties can exist even in the absence of broadly shared electoral pressures or policy goals.

In this paper, we study development in a panel of 87 countries from 1970 to 2005. We focus on characterizing institutionally driven heterogeneities in the development effects of macroeconomic policies and on comparing the development process as measured by GDP to that measured by the Human Development Index (HDI). We do so within a novel dynamic panel modelling framework that can account for crucial aspects of both the cross-sectional and intertemporal features of the observed process of development, and that can capture the dependence of the development effects of macroeconomic policies on differences in countries’ persistent characteristics, such as their institutions. Among our findings are that macroeconomic policies affect development with less delay than suggested by conventional econometric frameworks, yet impact HDI with longer delay and overall less strongly than GDP. Differences in countries’ persistent characteristics may even affect the sign of the long-run development effects of a given macroeconomic policy: Fiscal stimuli in the form of government consumption expansions positively affect long-run GDP in countries with low institutional quality, but negatively affect long-run GDP in countries with high institutional quality.

Research has suggested that states with citizen initiatives have higher turnout rates than states without them; however, these studies were conducted by analyzing elections decades after states have adopted and started using initiatives. Using data from 1870 to 2008, we attempt to resolve competing claims about the nature of how citizen initiatives affect turnout in the American states. We provide evidence that mobilization is the mechanism through which direct democracy increases turnout. Contrary to previous research, we show that the adoption of the initiative and past usage of the process does not lead to higher turnout in a given election. Citizen initiative campaigns mobilize the electorate in current elections and the number of competitive initiative elections has a greater effect on turnout than uncompetitive races.

The empirical evidence currently available in the literature regarding the effects of a country's IMF program participation on its output growth is rather mixed. To shed new evidence on this issue, in this paper we specify a state-dependent panel data model accounting in particular for program participation selection and the potential conditionality of the output growth effects of program participation on a country's degree of program implementation and institutional factors such as quality of governance, internal stability, health, and educational attainment. We find that the effects of IMF program participation on output growth vary systematically with the degree of program implementation as well as our index of institutional factors, and that these effects are positive only if the IMF program is implemented to a sufficient degree or if the program participation is coupled with sufficient progress in improving institutional quality.

Vote choice in initiative races is typically discussed in terms of "yes" and "no" votes, and confusion is widely regarded as contributing to an increase in "no" votes. Using data from two San Diego, California exit polls in 2008 and the Washington Poll in 2007, I present an analysis that provides no consistent support for the claim that confusion leads to more "no" votes. Another method of exploring vote choice investigates correct voting, defined as voting in line with one's preferences. Though confusion does not consistently lead to "no" votes, confusion does lead to more incorrect votes and inhibits the ability of citizens to match their votes with their preferences. On a positive note, endorsements are shown to enable voters to better align their preferences with the choices they face on the ballot.