Essays on information revelation and political institutions
How Choon, Marie Thea J. S. S.
MetadataShow full item record
This dissertation examines how political institutions shape incentives for the transmission of policy-relevant private information. The first two chapters present theoretical frameworks to examine strategic information transmission to political candidates by a biased adviser. The third explores the causal impact of gender representation on municipal outcomes in the United States. In Chapter One, I examine a model of "cheap talk" lobbying with no commitment. A biased adviser seeks to influence the policy outcome of a Downsian election by sending messages to the two candidates before they announce their policy platforms. I show that the adviser may credibly reveal some information on voter preferences, but only privately to one candidate. Political competition has a disciplining effect; the adviser prefers extreme policies, but instead recommends a pragmatic policy --- one that is just close enough to voters' preference. In some situations, the presence of the biased adviser benefits the median voter. The second chapter presents a model of informational lobbying with full commitment. The biased adviser strategically designs informative signals on voter preferences that will be observed by each of the two candidates. In contrast to the cheap talk context, the optimal signal structure is shown to involve only public signals that are observed by both candidates. In particular, candidates receive precise information about how extreme voter preferences are, but not whether voters lean right or left. Consequently, both candidates choose the same biased policy, as a result of which the median voter is always worse off. Chapter Three investigates the effect of gender representation on municipal outcomes in the United States between 2008 and 2016. Using novel data, the analysis exploits close elections between male and female candidates to measure the impact of an exogenous increase in the number of female council members. Consistent with the existing literature, we find evidence of decreased per capita expenditure, which, we argue, is not driven by revenue constraints but by increased disagreement or "gridlock" within the council. We also find no significant effect of gender representation on the composition of municipal spending or on other women's political aspirations.
RightsAttribution-NonCommercial-ShareAlike 4.0 International