Panel Data Analysis of the Impact of Larger State Size/Land Area on Voter Turnout in US Presidential Elections: 2000, 2004, 2008, and 2012

Abstract

This study has two objectives, both of which seek to provide insights into factors that influenced the voter participation rate in general elections in the US in recent years. The principal objective of this study is to proffer and then empirically investigate the following hypothesis: the voter participation rate of registered voters is reduced by greater state size (in squares miles) because, other things held the same, greater state size increases the transactions costs of voting for many of those persons who prefer to cast a ballot in person rather than by mail. A focus on this variable is unique in the literature. The secondary general objective of this study is to identify other factors that influenced the voter participation rate in recent years. The study adopts a state-level panel dataset and reports both Cross-section Random-effects estimations and dynamic panel estimations (Panel GMM estimates) for the first four US Presidential elections years of the 21st century, i.e., 2000, 2004, 2008, and 2012. After allowing for a variety of variables in the model, it is found that a 10,000 square mile greater size for a state in turn implies 0.148%-0.156% lower voter turnout, whereas a 100,000 square mile greater state size implies a 1.48%-1.56% lower voter turnout. Furthermore, the study finds that the voter participation rate of registered voters was an increasing function of six other factors: the unemployment rate, the percent of the population that earned a high school diploma or more, election competitiveness, median family income, and the percentage of the population that was either Afro-American/black or Hispanic/Latin.