UC Berkeley

This dissertation confronts several important questions about the institutional regulation of elections at the intersection of law and political science. I begin by showing how recent political reforms linking responsiveness to legislative action are misguided. State lawmakers that are elected by very narrow margins are no more likely to vote with the opposing party on legislative bills than are state lawmakers elected by huge margins. In fact, lawmakers elected by more than 85% of the vote are more likely to engage in cross-party voting. This empirical finding turns the traditional view of gridlock on its head; electoral responsiveness and political gridlock are complementary and not competing values. The median voter theorem is actually agnostic about polarization and gridlock, but my findings suggest that the conventional predictions about close elections and gridlock attributed to the median voter theorem may be incorrect.

I also show how independent political spending has been driven underground in the wake of Citizens United v. FEC. Because all of the "new" money that was permitted by Citizens United was funnelled to nonprofit organizations that are not required to disclose their donors, shareholders and the public cannot regulate the process as the majority expected they would. On the other hand, new independent political expenditures after Citizens United were not disproportionately large. In other words, the fear that large corporations would drown out the voice of individuals appears to be misplaced.

Finally, we see that the decision to delegate is not just about expertise or efficiency. Instead, delegation requires an analysis to match policy domains to the most appropriate regulatory body. In some cases delegation can mitigate problems, in other cases delegation can exacerbate problems. I show that delegation is not advisable for hot-button partisan issues (e.g., voter I.D. laws, voter registration requirements, recount procedures, redistricting), but is very valuable for the regulation and administration of technical, nonpartisan issues (e.g., accessibility for disabled voters, voter intent standards, voting technology).

The main goal of this dissertation is to show how careful empirical analysis can orient effective political reform and improve the election law jurisprudence. I hope that this thesis will inspire other scholars to engage in empirical research about the political process and that my analysis will be a valuable resource for them.