Re-examining the effects of contribution limits on campaign expenditures in gubernatorial elections

Abstract

While much time has been spent delving into relationship between campaign spending and election outcomes, little has been spent on the question of campaign finance laws on elections. Until Eom and Gross addressed the issue in, "Contribution Limits and Disparity in Contributions between Gubernatorial Candidates," the effects of laws on campaign spending were always assumed to be uniform between all candidates. Running a standard OLS regression with the disparity between campaign spending of incumbent and challengers in gubernatorial elections as the dependent and contribution limits and other control variables as independents, they conclude that these laws do not impact the disparity between candidates in spending. I recognized flaws in their analysis in three areas: 1) small data set of 57 observations, 2) lack of adequate control variables, and 3) an ill defined law independent variable. Before proposing other models or doing further work, I determined it best to instead retest their model after fixing these three flaws. Using a data set of 368 observations, including state control variables and disaggregating the law variable into separate corporate and individual limitations yielded the opposite conclusion of their paper. Using the model they have proposed corporate contribution limits have a significant negative impact on spending disparity while individual contribution limits have a significant positive impact on spending disparity. Borrowing their terminology, this would imply corporate limits increase election competition while individual limits decrease election competition.