Campaigns, Strategic Voting and Rational Abstention: Rolling Elections in India as a Quasi-Experiment
In this project, I look into the ways campaigning affects the opportunities and incentives for strategic voting and rational abstention in multi-party elections under FPTP. I capitalize on the quasi-experimental settings created by the institution of rolling elections in India: in this country, voting in general elections is organized in a series of phases spread over almost a month, with different districts (even within a single state) voting on different days. I treat the timing of the district vote as an exogenously determined endpoint of campaigning and examine the effect of the length of campaigning on the absolute size of the vote for leading candidates, the absolute size of the vote for trailing candidates, intra-district coordination and aggregate turnout. This project is a part of a broader agenda that looks into the ways campaign activities condition the effects of electoral institutions.
Campaigns, Information, and Coordinated Voting in Multi-Party Elections under Plurality Rule
This work stems from two conflicting observartions. On the one hand, evidence abounds that voting in single-member districts produces tight party systems. On the other hand, surveys show that voters have hard time identifying leading candidates in their single-member districts (e.g., Blais and Turgeon, 2004), let alone casting a strategic vote. When strategic voting is limited, behavioral factors of sincere voting play a large part in coordinating uninformed voters and inform the expectations of potentially strategic voters about the patterns of voting. Using the 2011 Canadian Election Survey, I found strong effects of the density of campaign contacts and the asymmetries in the campaigns spending and party identification on the predictability of the patterns of intra-district competition. A comparison of the effects of behavioral factors on the uninformed and informed voters confirms that centrifugal spending and party identification influence non-strategic clustering of voters.
Political Constraints on Corruption at Subnational Level: Evidence from India
with Dina Rosenberg
A large share of corruption in federations takes place at subnational level; thus, the aggregate level of corruption in politically decentralized polities to a large degree depends on the efforts of subnational politicians. Dina Balalaeva and I examine the incentives of subnational office-holders to invest efforts in control over corruption. The electoral competition within subnational units, the pressure of the national party leadership and the competition among subnational units are three complementary mechanisms at work. Subnational incumbents are motivated to invest efforts in control over corruption to the extent to which their performance affects their re-election prospects (1), to the extent to which the national party leadership presses them to improve their performance (2), and to the extent to which the investors factor in the level of corruption when choosing the location of their firms (3). We make use of the indicators from the World Bank's "Enterprise Surveys" project, as well as institutional and electoral data, to disentangle these effects on the data from the states of India.
Issue Voting as a Constrained Choice Problem
with Mert Moral
Studies of issue voting (e.g., the proximity and directional voting theories) typically treat all parties as equally likely to be included in the voters' choice sets and treat voters as having the same choice sets as they make a choice among political parties. in this project, Mert Moral and I relax this assumption by building a maximum likelihood model of a vote choice that incorporates the elements of the constrained multinomial logit (Martinez et al. 2009, Castro et al. 2013) into the conditional logistic regression.
Here is a link to the poster about it, which we prepared for the Polmeth Summer meeting of 2015.
Perils and Pitfalls of Ignoring Disproportionality’s Behavioral Component
with Robin E. Best
The disproportionality of electoral institutions is often measured by the discrepancy between the distribution of votes among political parties and the resulting allocation of seats (e.g., via Michael Gallagher's least squares index, the Loosemore-Handby index and Lijpharts' index). In this project, Robin E. Best and I analytically disaggregate the observed disproportionality in the vote-to-seat translation into its institutional and behavioral components and show that behavioral elements (mainly, strategic behavior or lack thereof) do not only affect the observed disproportionality, but also condition the connection between electoral institutions and observed discrepancy between vote and seat distributions. For this reason, we argue, the measures of observed disproportionality should not be used as institutional measures.
Institutional Neutrality in Party System Development: The Mixed Member System
with Olga Shvetsova
It is often said that electoral institutions shape the development of party systems: they create incentive for the competition and/or cooperation of political elites in the electoral process. In this project, we are looking for an electoral system that leaves the parties unaffected. Specifically, in this essay we analyze the nature of incentives for the consolidation and/or fragmentation in party systems that come purely from election laws, and make an argument that these incentives are not as rich and varied as the empirical literature would have us believe. We support our argument with the discussion of the development of party systems in Germany circa the 1950s — under the mixed-member proportional system — and in Russia circa the 1990s — under the mixed-member majoritarian system. — Our conclusion is that other, non-election laws formal institutions, as well as the structural and behavioral factors, play a much stronger and possibly a decisive role in many cases of party system development.