2006 University of Michigan Poverty Research Grants
Funded research
Karen Long Jusko, Department of Political Science, University of Michigan
The Political Representation of the Poor
Abstract
How well do contemporary democratic governments represent the interests of the poor? Under what conditions are elected officials likely to be more or less responsive to the poor? What are the distributional consequences of partisan and electoral incentives to be more or less responsive to the poor?
This dissertation research argues that some electoral institutions may undermine democratic opportunities for fair and equitable societies. It is well-established that electoral rules structure incentives for the distribution of publicly funded goods: Elected officials and political parties often structure distributive policy to their electoral advantage, resulting in electoral and partisan business cycles in the public spending. If the poor rely on distributive policies that benefit few other citizens, then antipoverty policy may be especially susceptible to this sort of manipulation. This dissertation research, therefore, is about how electoral rules affect the poor.
Electoral rules vary across countries in two important respects: the number of legislators elected in each electoral district (district magnitude), and whether the same number of legislators is elected in each district (variation in district magnitude). By determining the share of votes a party needs to secure in order to meet its electoral goals, district magnitude may make the support of low-income citizens necessary. And, by determining the number of viable electoral coalitions, variation in district magnitude may create incentives for parties to seek the support of low-income citizens as a partisan constituency.
The effects of electoral rules, however, may be modified by the political behavior of low-income voters in each political system. When in the electorate under-represent the poor, there may be no incentive for political parties or legislators to pay attention to their interests. However, some electoral rules may make the support of low-income voters necessary for parties to meet their electoral goals, even when the poor comprise a small proportion of the electorate. To the extent, then, that electoral rules create incentives for political parties to pay attention to the poor, distributive policy should be more favorable to low-income citizens.
The proposed research will investigate how electoral rules affect the poor in a broadly comparative analysis, and several comparative case studies. Special attention will be given to variation in the distributive policy across the U.S. states. The Luxembourg Income Study (LIS) data are especially well-suited for this project: For a large number of mature democracies, with considerable variation in electoral rules, the LIS data report sources of income for a large sample of individuals in each country. Specifically, the LIS data report the extent to which low-income citizens are reliant on different distributive policies. This is important because in order for distributive policy to be attributed to electoral incentive structures, policy outcomes must be visible to would-be voters.
Another feature of the LIS data that is well-suited to this research is that the LIS data are comprised of income study data collected as part of long-standing projects within each country (i.e., the U.S. data are drawn from the Current Population Survey). The LIS data, therefore, naturally bridge broadly comparative analysis and more detailed case studies, and provide a nested framework for an analysis of how electoral rules affect the poor.
