Research

How do electoral rules affect the poor? How responsive are elected governments to the interests of low-income citizens? When do parties have an incentive to seek the support of the low-income citizens? These questions motivate a broadly comparative analysis of the relationship between antipoverty policy and electoral rules. Presenting a series of formal analytic examples, and using Luxembourg Income Study data in focused case studies and broadly comparative empirical analysis, this research demonstrates that electoral rules interact with the context in which elections are held -- specifically, the distribution of low-income citizens across electoral districts -- to create or limit legislators' incentives to be responsive to the poor. In this way, the very institutions of democratic government may undermine opportunities for a more equitable society. This dissertation project establishes the foundation of a research agenda motivated by broader questions about whether and how the institutions of contemporary democracies create incentives to build societies that reflect democratic ideals.


Electoral Politics and Poverty Relief: How Changing Electoral Incentives can Help the Poor.
*This working paper presents research included in Chapter 3. Measuring Poverty Relief, Chapter 4. Changing Electoral Incentives: Electoral Reform in Italy, and Chapter 5. Changing Electoral Incentives: German Reunification, of my dissertation project.


Other materials:

• Project Outline
  
  
• Project Summary
  
  
• Chapter 2. How Electoral Rules and Electoral Contexts Create Incentives to Represent Low-Income Citizens.
  Revised October 2, 2006.

In recent years, large sets of national surveys with shared content have increasingly been used for cross-national opinion research. But, scholars have not yet settled on the most flexible and efficient models for utilizing such data. We present a two-step strategy for such analysis that takes advantage of the fact that in such data sets each cluster (i.e., country sample) is large enough to sustain separate analysis of its internal variances and covariances. We illustrate the method by examining a puzzle of comparative electoral behavior - why does turnout decline rather than increase with the number of parties competing in an election (Blais and Dobrzynska 1998, for example)? This discussion demonstrates the ease with which a two-step strategy incorporates confounding variables operating at different levels of analysis. Technical appendices demonstrate that the two-step strategy does not lose efficiency of estimation as compared with a pooling strategy.


"Political-Economic Cycles."
with R. Franzese. Forthcoming in Witten, D., and B. Weingast (eds.), Oxford Handbook of Political Economy. New York: Oxford University Press.

This entry surveys theoretical and empirical work on such political-economic cycles, i.e., cycles in economic outcomes induced by electoral and partisan competition. In particular, we find the evidence to suggest that closer attention to the context in which economic policy decisions are made offers important and promising opportunities for future research. We review the literature on electoral cycles and emphasize how electoral context may heighten or inhibit incentives to electioneer. Then, turning to partisan cycles, we present a survey of the theoretical and empirical literature linking economic policy to partisan electoral motivations. Here too the opportunities for the incorporation of contextual factors, we suggest, seem very promising.

Theorists broadly agree that democratic policymakers respond to pressures from their constituents; however, heterogeneity prevails over exactly what comprises the constituency to which policymakers respond. We propose conceiving the bases of democratic representation as a continuum from the interests of the policymaker's geographic constituency, her electoral district, to those of her party's supporters, her partisan constituency. The effective constituency to which policymakers respond might then be represented by some convex combination of these partisan and geographic extremes, with the partisan weight summarized by the degree to which parties act and receive electoral support as units, party unity. Reexamination of the familiar Weingast-Shepsle-Johnsen (WSJ) model of distributive politics (the law of 1/n) undergirds empirical evaluation of this conception of the effective constituency. Postwar histories of public spending and distributive politics exhibit too much temporal variance to support either models based purely in partisan or electoral cycles or a pure-electoral-district WSJ model, but postwar public spending in the United States, where data best-suited to evaluate our argument exist, do support a WSJ model, once modified to reflect our effective constituency concept.

In an earlier collaborative project, Jusko & Shively (2005) illustrate a two-step strategy that is well-suited for the analysis of cross-national public opinion data. Working through a puzzle of comparative politics � that turnout does not increase with the number of parties competing, this discussion extends this earlier research, and contributes a two-step strategy for the analysis of binary response models. Technical appendices demonstrate that a two-step estimation strategy yields parameter estimates that are asymptotically consistent and efficient.

Does gun control policy effectively reduce violent crime? Measures that aim to limit the number, type and/or availability of firearms have been among the most controversial domestic policies considered by American legislators over the last decade. While it is indisputable that the presence of firearms contributes to the number of violent crimes committed in the United States each year-- more than half of the homicides committed in the US in 2000 involved firearms-- there has been little research that establishes gun control policy as an effective way to limit violence in American cities, perhaps because this research question poses a significant methodological challenge: It is difficult to evaluate the effects of policy that has been (more or less) uniformly implemented across the US. Simply the counterfactual-- what rates of crime in American cities would be if more restrictive gun control measures had been in place-- is unobserved. To overcome this challenge, this research presents an analysis in which data from Canadian cities, where gun control policy has been considerably more restrictive, are used as counterfactual evidence. By matching Canadian and American cities on factors that contribute to violent crime, this analysis generates an estimate of the effect gun control policy would have on limiting violent crime in the US.