Testing the anti-Incumbency Hypothesis in India: A Natural Experiment (with Devesh Tiwari)
Abstract: Studies of congressional elections in the United States suggest that high voter turnout is usually a bad sign for incumbents, either because the factors that cause voters to reject incumbents also cause them to turn out at polls or because non-core voters, who contribute to high turnout, are less invested in the political status quo. Demonstrating the effects of turnout on incumbents, however, is complicated because voters are more likely to turn out when they perceive election outcomes to be very close. To deal with this problem, we test the anti-incumbent hypothesis on parliamentary elections in India—the world’s largest democracy—where election schedules have an exogenous effect on election uncertainty. By taking advantage of election timing across different states we show that high turnout significantly reduces the likelihood that an incumbent party candidate is elected and that this effect is more pronounced in elections occurring later in the election schedule, when uncertainty over outcomes is less severe. These findings provide strong empirical support for the anti-incumbent hypothesis and contribute to the study of comparative elections by shifting focus away from vote share, which is highly context specific, and towards a more general measure of election turnout.
Minorities in a Majority Muslim State: Examining Indonesia’s 2014 Presidential election (with Kai Ostwald and Paul Schuler)
Abstract: The spread of democracy in the Muslim world has proven slower and more tumultuous than in other setting. The appeal of conservative Islamic parties, which espouse non-democratic policy preferences, has been identified as a primary obstacle. But how appealing are Islamic political platforms and how do non-Muslim minorities respond to populist campaigns? To answer this question, we examine electoral patterns in Indonesia?s 2014 presidential election. While many have speculated that religion was a major factor in determining the outcome, data limitations have thus far prevented a comprehensive empirical assessment. Using a novel crowd-sourced dataset on voting outcomes across 480,000 polling stations combined with village and district-level data on income, education, religion, and infrastructure, we offer the first comprehensive assessment of the 2014 Indonesian election. Our results confirm that religion did play an important role in the election, but not in the way most observers have predicted. Specifically, we find the conservative vote was not a deciding factor. The most salient demographic were non-Muslim minorities, which voted almost en-masse for the moderate candidate, so much so that the results from these regions almost appear fraudulent.
Far from Home: China’s Economic Foreign Policy in Sri Lanka (with, Janeen Fernando)
Abstract: There is little question as to China’s recent push for greater global influence. Yet, it is unclear how target states and their respective populations respond China’s efforts. In this study, we examine China’s efforts to cultivate political and economic relations in Sri Lanka, a key link in China’s maritime ambitions. Using evidence from survey experiments executed during Sri Lanka’s most recent parliamentary elections, in which relations with China featured as the top foreign policy issue, we examine the differences between voters who are pro and anti-China. We also examine how perceptions and misperceptions over China’s economic prospects are conditioning these preferences. We find that leftist economic orientation and social traditionalism are strong predictors of a pro-China bias, as are overblown estimates of China’s economic prowess. Likewise, we find that China’s reputation for corruption and environmental destruction are the most salient predictors of anti-China sentiment. Finally, we find that preferences over relations with China had an independent and significant effect on vote choice.