Research Interests

My current research focuses on three interconnected areas: electoral authoritarianism, democratization, and political methodology. My dissertation argues that autocratic elections reveal information by channeling popular demands to the ruling party, leading to policy concessions and regime change. In an ongoing project, I use international regime diffusion as an instrument to test the causal effect of electoral authoritarianism on policy, human development outcomes, economic growth, and conflict behavior.

A current book project investigates the link from coups and contested elections to democratization, showing that these events mediate the effect of structural economic and political variables. In a recent AJPS article, I argue that the effect of economic development is conditioned by autocratic regime strength and violent leader change. Among other implications, this explains the well-known puzzle of why average income predicts democratic survival, but not democratization. I also have an active research interest in the international dimensions of democratization.

My methodological research includes an analysis of how unit fixed effects can induce bias in panel data, combined with a proposed fix, as well as critiques of matching and current approaches to causal inference. Several of my papers employ formal models, with a focus on elections and regime change.



Electoral Authoritarianism

“Electoral Authoritarianism and Human Development.” Under review. PDF


Abstract: Do autocratic institutions matter for the welfare of average citizens? Despite the large literature comparing democracies and autocracies, we know little about how human development outcomes differ among autocratic types. This paper argues that contested autocratic elections promote human development by improving state accountability and capacity. Using an instrumental
variables setup, I show that the presence and history of multiparty autocratic elections predict significantly better outcomes on health, education, gender equality, and basic freedoms relative to non-electoral autocracy. In fact, the effects on health and education are as strong as the effects of democracy. In contrast, legislatures and parties without multiparty elections produce slightly negative outcomes since these institutions chiefly concern elite cooptation. The results have major implications for the study of autocracy, the political economy of development, and the welfare effects of international election promotion.

Elections, Information, and Policy Responsiveness in Autocratic Regimes.” Revise & Resubmit, Comparative Political Studies. PDF


Abstract: The responsiveness of policy to election results is a central component of democracy. Do the outcomes of autocratic elections also affect policy choice? Even when the threat of turnover is low, I argue that autocratic elections influence policy by allowing citizens to signal dissatisfaction with the regime. Supplementing existing work, this study explains how this opposition is communicated credibly and then shows that ruling parties use this information to calibrate policy concessions. In the first cross-country analysis of autocratic election outcomes and policy choice, I find that negative electoral shocks to ruling parties predict increases in education and social welfare spending and decreases in military spending following elections. In contrast, there is no policy effect leading up to elections, in response to violent contestation, or in resource-rich regimes, illustrating a potential mechanism for the resource curse.

“Electoral Authoritarianism and Democracy: A Formal Model of Regime Transitions.” 2013. Journal of Theoretical Politics 25(2): 153-81. PDF


Abstract: Building on the formal literature on democratization, this paper models a dictator's choice between closed authoritarianism, electoral authoritarianism, and democracy in the shadow of violent revolt. Under autocracy, the dictator controls policy but lacks information on the policy demands of citizens and thus the likelihood of popular revolt. Electoral authoritarianism enables the dictator to tie policy choice to an electoral signal from citizens, which may be advantageous even if elections make revolt more likely to succeed. Implications are drawn for how economic inequality, regime strength, and uncertainty predict regime type, policy concessions, and political violence. A key result is that electoral authoritarianism is chosen for middle values of inequality and uncertainty.

“Democratic Pieces: Autocratic Elections and Democratic Development since 1815.” Forthcoming. British Journal of Political Science. PDF


Abstract: This paper overviews the history of autocratic elections since 1815 and then tests how a country's experience with autocratic elections influences both democratization and democratic survival. To comprehensively capture this history, I employ original measures of Robert Dahl's electoral dimensions of contestation and participation. First, I show that autocratic elections have been common for centuries, but their character has changed dramatically across time periods. Whereas high contestation almost always preceded high participation prior to 1940, the opposite occurs in modern regimes. Second, I demonstrate that a country's history of contestation predicts both democratization and democratic survival, whereas participation is positive for survival but generally negative for democratization. Thus, democracies are more likely to survive if they experience autocratic elections prior to democratizing, with implications for democracy promotion and future political development.

“The Origins of Electoral Authoritarianism and Democracy.”


Abstract: Despite the global spread of autocracies with multiparty elections, we know little about what predicts electoral authoritarianism (EA). To fill this gap, I use multinomial logit to simultaneously predict transitions to EA and democracy from non-electoral autocracy between 1946 and 2007. Because autocrats retain power under EA, I argue that EA transitions follow from a strategic calculus that balances incentives to hold elections against the costs of controlling them. As a result, socioeconomic factors that make voters easier to buy off, such as low average income and high inequality, predict EA adoption. In contrast, autocrats lack significant power under democracy, so democratization is predicted by regime weakness, but not socioeconomic conditions. Lastly, I find parallel regional diffusion effects: Democratic neighbors predict democratization and EA neighbors predict EA transition.



Democratization

“Economic Development, Violent Leader Removal, and Democratization.” American Journal of Political Science 56(4): 1002-20. PDF


Abstract: This paper argues that autocratic regime strength plays a critical mediating role in the link between economic development and democracy. Looking at 167 countries from 1875-2004, I find that development strengthens autocratic regimes, as indicated by a reduced likelihood of violent leader removal. Simultaneously, greater development predicts democratization, but only if a violent turnover has occurred in the recent past. Hence, development can cause democratization, but only in distinctive periods of regime vulnerability. Although development's stabilizing and democratizing forces roughly balance out within autocracies, they reinforce each other within democracies, explaining the puzzle of why economic development has a stronger effect on democratic stability than on democratization. Further, the theory extends to any variable that predicts violent leader removal and democracy following such violence, pointing to broad unexplored patterns of democratic development.

“Democracy by Example? Why Democracy Spreads When the World's Democracies Prosper.” Under review. PDF


Abstract: Does a positive association between democracy and economic growth around the world encourage the spread of democracy? Although this intuitive relationship has been linked to the global ebb and flow of fascism and Communism, no study has empirically tested this question. I argue that democracy's relative economic success influences perceptions of its domestic and international advantages. Looking at 166 countries from 1822-2004, I show that the world-level correlation between democracy and economic growth robustly predicts the spread of democracy and represents a major source of its historical advance. The results provide new insight into the foreign influences on democratization and China's potential challenge to the liberal democratic order.

“A Complete Data Set of Political Regimes, 1800-2007,” with Carles Boix and Sebastian Rosato. 2013. Comparative Political Studies 46(12): 1523-54. PDF


Abstract: This paper updates and describes a widely used data set on democracy. Covering 1800-2007 and 213 countries, it represents the most comprehensive dichotomous measure of democracy currently available. We argue that our measure’s distinguishing features—a concrete, dichotomous coding and a long time-span—are of critical value to empirical work on democracy. Inspired by Robert Dahl, we define a country as democratic if it satisfies conditions for both contestation and participation. Specifically, democracies feature political leaders chosen through free and fair elections and satisfy a threshold value of suffrage. After comparing our coding to other popular measures, we illustrate how democracy's predictive factors have evolved since 1800. In particular, we show that economic modernization variables have steadily declined in their correlation with democracy over time.

“The Patron's Dilemma: The Dynamics of Foreign-Supported Democratization,” with Michael K. McKoy. Journal of Conflict Resolution 56(5): 904-32. PDF


Abstract: We analyze an understudied mode of democratization in which the acquiescence of an autocratic regime's external ally, or patron, is pivotal to the success of a democratic movement. Although a democratic patron may prefer having democracy in its dependent allies, regime change threatens the economic and security benefits associated with the alliance. We formalize this dilemma through a repeated principal-agent model and demonstrate that the critical dimension is the patron's beliefs about the potential democracy's policies, rather than its value for democracy or the alliance goods. Patron support hinges on democratic movement signaling of its capacity to rule, popular support, and commitment to preserving the alliance. To test our theory, we analyze 25 democratic openings in American Cold War clients, followed by case studies of U.S.-aided democratization episodes in the Philippines and South Korea. We conclude with an analysis of the recent Egyptian revolution.



Quantitative Methodology

“The Uses and Abuses of Matching.” Under review. PDF


Abstract
: Matching has become a common technique in the social sciences, but there is widespread confusion regarding its purposes. This article evaluates the three most prominent justifications for matching: to improve causal inference, to reduce model dependence, and to prevent bias from model misspecification. First, I argue that matching offers no causal leverage or advantage in dealing with selection relative to regression alone. Second, I show that matching often increases model dependence and considerably widens the opportunity for data-mining. Claims to the contrary mistakenly ignore the sensitivity of estimates to choices concerning the match itself. Third, matching's real benefit is the prevention of treatment effect bias from uncorrected covariate nonlinearities. I introduce a test for when this use is supported by the data. However, non-parametric estimators that directly model these nonlinearities are usually a better solution. Thus, we should view matching as a non-ideal response to a specific analytical problem.

“The More Things Change: An Overlooked Problem with Fixed Effects.” Under review. PDF


Abstract: Unit fixed effects are widely used in the social sciences to account for unmeasured individual factors and thereby reduce omitted variable bias. This paper shows that the addition of fixed effects can actually induce omitted variable bias. This occurs when an omitted variable simultaneously predicts the dependent variable and within-unit changes of an independent variable. In such a case, a pooled model is consistent and negligibly biased, while the fixed effects model is inconsistent and seriously biased. I provide evidence that this is a common situation in political science research. After calculating the resulting bias, I introduce a simple method of testing and correcting for it, which is illustrated by examining the effect of oil wealth on democracy. Thus, I demonstrate how to improve the fixed effects model in the face of an overlooked, but likely endemic, statistical problem.


Voting and Elections

“For the Win! The Effect of Professional Sports Records on Mayoral Elections.” 2013. Social Science Quarterly 94(1): 59-78. PDF


Abstract: Voters are more likely to reelect incumbents when political outcomes are positive. Is this because they explicitly credit politicians for these outcomes or because they simply opt for the status quo when happy? To distinguish these two voting models, this paper proposes professional sports records as a proxy for electorate happiness that is unrelated to political performance, and looks at their effect on incumbent mayoral elections. For 39 American cities from 1948-2009, the study finds that winning sports records boost incumbent vote totals and the likelihood of reelection, exceeding in magnitude the effect of unemployment. To account for omitted variables, it is shown that sports records from the year following the election display no such relationship. The implication is that retrospective voting is partly driven by feelings of happiness unrelated to political evaluation. The paper concludes by discussing the implications for democratic accountability, which are not as dire as many authors claim.

“Citizen Forecasts of the 2008 U.S. Presidential Election,” with Guanchun Wang, Sanjeev R. Kulkarni, H. Vincent Poor, and Daniel Osherson. 2012. Politics & Policy 40(6): 1019-52. PDF


Abstract: We analyze individual probabilistic predictions of state outcomes in the 2008 U.S. presidential election. Employing an original survey of more than 19,000 respondents, we find that partisans gave higher probabilities to their favored candidates, but this bias was reduced by education, numerical sophistication, and the level of Obama support in their home states. In aggregate, we show that individual biases balance out and the group's predictions were highly accurate, outperforming both Intrade (a prediction market) and fivethirtyeight.com (a poll-based forecast). The implication is that electoral forecasters can often do better asking individuals who they think will win rather than who they want to win.

“Seizing the Mantle of Change: Modeling Candidate Quality as Effectiveness Instead of Valence.” 2011. Journal of Theoretical Politics 23(1): 52-68. PDF


Abstract: In spatial models of electoral competition, candidate quality is typically modeled as valence, a measure of general appeal assumed to be constant across voters. This paper introduces and formally models an alternative conception of candidate quality according to which candidates differ in their effectiveness, or likelihood of changing policy from the status quo. Although more effective candidates are electorally favored, voters' benefits from effectiveness are contingent on their policy preferences. The effectiveness model shares many qualitative features with the valence model, but adds several testable implications related to the position of the status quo and gives rise to non-monotonic voting. When valence and effectiveness are combined, valence dominates effectiveness in determining the winner if and only if the status quo policy is sufficiently close to the political center.




Published Papers

  • “Democratic Pieces: Autocratic Elections and Democratic Development since 1815.” Forthcoming. British Journal of Political Science. PDF

  • “Public Opinion towards Democracy: Citizens in Transitional Regimes,” with Juliet Pietsch and Jeffrey Karp. Forthcoming. Journal of Elections, Public Opinion & Parties.

  • “A Complete Data Set of Political Regimes, 1800-2007,” with Carles Boix and Sebastian Rosato. 2013. Comparative Political Studies 46(12): 1523-54. PDF

  • Electoral Authoritarianism and Democracy: A Formal Model of Regime Transitions.” 2013. Journal of Theoretical Politics 25(2): 153-81. PDF

  • “For the Win! The Effect of Professional Sports Records on Mayoral Elections.” 2013. Social Science Quarterly 94(1): 59-78. PDF

  • “Economic Development, Violent Leader Removal, and Democratization.” 2012. American Journal of Political Science 56(4): 1002-20. PDF

  • “The Patron's Dilemma: The Dynamics of Foreign-Supported Democratization,” with Michael K. McKoy. 2012. Journal of Conflict Resolution 56(5): 904-32. PDF

  • “Citizen Forecasts of the 2008 U.S. Presidential Election,” with Guanchun Wang, Sanjeev R. Kulkarni, H. Vincent Poor, and Daniel Osherson. 2012. Politics & Policy 40(6): 1019-52. PDF

  • “Seizing the Mantle of Change: Modeling Candidate Quality as Effectiveness Instead of Valence.” 2011. Journal of Theoretical Politics 23(1): 52-68. PDF

  • “Methods for Distance-Based Judgment Aggregation,” with Daniel Osherson. 2009. Social Choice and Welfare 32(4): 575-601. PDF

  • “Social Choice Theory Without Pareto: The Pivotal Voter Approach.” 2009. Mathematical Social Sciences 58(2): 251-5. PDF

  • “Judgment Aggregation and Subjective Decision-Making.” 2008. Economics and Philosophy 24(2): 205-31. PDF

Papers in Physical Sciences



Master's Thesis, LSE

Judgment Aggregation, Democratic Theory, and the Impossibility of Faithful Representation.

  • Received highest grade on Master's Thesis in the history of LSE's Government department.