Date of Award

Spring 5-15-2017

Author's School

Graduate School of Arts and Sciences

Author's Department

Political Science

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Democratically elected governments may sometimes give in to pressures from extreme constituents thereby failing to adopt policies congruent with the median voter's preferences. This situation is exacerbated by the inability or the reluctance of voters to acquire information about the consequences of policy alternatives. Opposition parties, in principle, could help remedy this problem using the prerogatives at their disposal such as participating in legislative debates and proposing bills, amendments or no confidence motions. In practice, however, they may have incentives to mislead the voters for policy or election purposes. Given the critical role of opposition parties to remedy problems of incongruence, it is essential to gain theoretical clarity and empirical knowledge as to when and how opposition parties use the tools available to them in legislatures. My dissertation contributes to the recently growing literature that studies the behavior of opposition parties in different institutional contexts, and the implications of that behavior for democratic representation. In particular, I examine the conditions under which opposition parties provide the voters with accurate information, thereby inducing the government to act in accordance with voters' demands and interests. My analysis reconsiders the conventional wisdom that the influence of opposition parties on policy making is commensurate with their electoral strength or seat share in the legislature.In the first paper, I develop a formal model that examines the interaction between a government with control over a policy proposal and an opposition party faced with the decision whether to attract voters' attention to the proposal by objecting to its adoption. The model assumes that the government has incentives to pursue policies divergent from voter preferences and that voters face non-trivial costs in monitoring the government's legislative proposals. Voters can use the objection of the opposition party as a signal to identify adverse government proposals. However, for the mechanism to work, the opposition party should be responsible, i.e. not alarm the voters too frequently. I show that confrontation between the government and the opposition party is more likely when the chances of election for the opposition party are sufficiently low and the opposition activists are sufficiently distant in terms of their ideological positions from the government (as well as the voter and the opposition party itself). In addition to providing useful signals, the presence of a responsible opposition party may increase voters' welfare by inducing the government to preemptively adjust its policies.In the second paper, I propose a formal model that explores whether the voters can receive accurate information from the opposition party under further constraints. More specifically, the model assumes that the voters do not have the means to verify the accuracy of information they receive about the policy and furthermore they have limited information about the preferences of the government and the opposition party. I show that the opposition party can discipline the government to choose policies congruent with voters' preferences if the reputation of the opposition is high and the benefit of policy to the government is small relative to the benefit of winning elections. Under the same conditions, however, misleading messages of the opposition party may cause a good government to implement policies that bring about bad outcomes for the voters.In the the third paper, I examine when legislators withdraw their support from the bills that they cosponsored using an original dataset of cosponsorship in the US House of Representatives. I argue that the legislators take into account the preferences of their constituency in their cosponsorship decisions. Since acquiring information about the bills is costly, legislators use the cosponsorship decisions of their counterparts as signals for the acceptability of the bill for their constituency. My analysis shows that there is a relationship between the partisan composition of cosponsors and the likelihood that he/she decides to remove his/her name from the bill later. More specifically, a legislator is less likely to remove his/her name from a bill that he/she cosponsored as the difference between the number of his/her partisans that cosponsored the bill and the number of his/her opponents that cosponsored the bill increases.


English (en)

Chair and Committee

Matthew J. Gabel

Committee Members

Randall L. Calvert, Brian F. Crisp, Justin Fox, John W. Patty,


Permanent URL: https://doi.org/10.7936/K7PZ578G