The following is a list of published and ongoing projects. Please email me at sarah_hummel@fas.harvard.edu for the latest version of any working papers.

Publications:

• Elections, War, and Gender: Self-Selection and the Pursuit of Victory (co-authored with Stephen Chaudoin and Yon Soo Park)

Why might female leaders of democratic countries commit more resources - money, equipment, soldiers, etc. - to interstate conflicts than male leaders? We argue that gender bias in the process of democratic election helps explain this behavior: since running for office is generally more costly for women than men, only women who place a higher value on winning competitions choose to become candidates for political office. After election, they also devote more resources to pursuing victory in conflict situations. To provide microfoundational evidence for this claim, we analyze data from an online laboratory game featuring real-time, group play in which participants chose to run for election, conducted a simple campaign, and represented their group in a contest game if elected. Women who placed a higher non-monetary value on winning were more likely to self-select into candidacy and, when victorious, spent more resources in intergroup contests than male elected leaders. The data suggest that electoral selection plays an important role in observed differences between male and female leaders in the real-world.

Forthcoming. International Organization

• The Election Effect: Democracy in Inter-Group Contests (co-authored with Stephen Chaudoin and Yon Soo Park)

Interactions between groups often depend on choices made by democratically selected group leaders. Using an online laboratory experiment, we show that democratic leader selection increases inefficient effort in inter-group contest games. We attribute a large portion of this increase to an election effect, wherein individuals behave differently after the experience of being elected by members of their group. Democratic election intensifies group identification and creates a sense of obligation to voters, causing leaders to exert more costly effort in competitive situations. We use a carefully specified decomposition strategy to distinguish the election effect from better known selection effects, wherein eventual leaders are non-randomly chosen. From a welfare perspective, our negative finding in an inter-group interaction is contrary to the near-universal positive effects of democracy found in intra-group experiments.

Forthcoming. International Studies Quarterly

• Leader Age, Death, and Political Liberalization in Dictatorships

This article examines how expectations about the likelihood of a dictator’s death affect the strategic calculations of regime insiders and potential challengers. On the one hand, would-be reformers are better positioned to plan and execute post-death challenges as dictators age. On the other hand, regime insiders anticipate these challenges and try to proactively solve the problem of political succession. The circumstances surrounding leader death determine which of these competing effects dominates. Accordingly, leader death is more liberalizing as leaders age in personalist regimes compared to nonpersonalist regimes, and in countries with high levels of economic development compared to those with low levels of development. Furthermore, preemptive actions in personalist dictatorships, such as coup attempts and irregular removals, are more likely as leaders age and their death becomes imminent.

2020. The Journal of Politics. 82(3): 981-995.

• Sideways Concessions and Individual Decisions to Protest

Sideways concessions to protest are policy reforms that decrease grievance among potential protesters, without being directly linked to the stated demands of the protest. By avoiding both the backlash effect of repression and the inspirational effect of direct concessions, they are theoretically powerful tools for quelling unrest. This article evaluates the effectiveness of sideways concessions at reducing individual mobilization potential using a survey experiment conducted in Kyrgyzstan in October 2015. The evidence shows sideways concessions are effective among respondents who were dissatisfied with the government and not optimistic about the future of the country. The article also demonstrates the plausibility of these results in other settings, drawing on observational data from the 2014 Gezi Park protests in Turkey and the 2013 Euromaidan protests in Ukraine.

2019. Comparative Politics, 52(1): 65-96.

• Relative Water Scarcity and Country Relations along Cross-Boundary Rivers: Evidence from the Aral Sea Basin

How do countries that share cross-border rivers respond to periods of abnormally low water availability? Existing research concerning water scarcity focuses on how cross-basin differences in absolute availability influence relations between countries. I argue that understanding whether countries react cooperatively or conflictually to within-basin shortages is important. I use the case of two major cross-boundary rivers in the Aral Sea basin of Central Asia to study the effects of within-basin relative scarcity. Employing original data on interactions among the Central Asian countries over the issue of water management, I find an association between, on the one hand, relative water scarcity and, on the other hand, an increased likelihood of both cooperative and conflictual interactions. By showing that relative scarcity affects when cooperative and conflictual events occur, my analysis highlights the fact that absolute scarcity is not the only type of water scarcity that influences international relations on cross-boundary rivers.

2017. International Studies Quarterly, 61(4): 795-808.

Selected Ongoing Projects:

• Digital Despots and Tweeting Democrats: The Influence of Regime Type on Leaders' Social Media Communication (with Philipp Zimmer)

How do the strategic political incentives of different regime types influence the structure and sentiment of social media communication by heads of government? Political leaders increasingly view social media as an essential way to connect with their citizens. However, differences in accountability structures and control over information, which vary with the level of democracy, systematically influence the form this connection takes. We analyze the structure and sentiment of over 30,000 French and English language tweets made by 24 political leaders in Sub-Saharan Africa between 2018 and 2021. We find evidence that leaders of more democratic countries tweet more frequently, are less likely to engage directly with specific users, and are more measured in their communication style.

R&R at Political Science Research Methods: Current Version (updated 04/22/2024)

• Mutual Restraint in Nondemocratic Legislatures

Nondemocratic regimes often benefit from tolerating the expression of critical opinions and competing policy visions in the legislative branch. Among other things, doing so enables them to co-opt potential challengers and gather important information about societal preferences. However, these systems only work when there is mutual restraint: deputies respect regime-set limits on their freedom to express critical opinions, and the regime refrains from punishing deputies for expressing negative views within these limits. By examining the responsiveness of individual deputies to regime-set limits, we are better able to identify both the function and strength of non-democratic legislatures. I illustrate these dynamics using data from the Kyrgyz Jogorku Kenesh about both voting behavior and legislative success (2016-2020).

Under Review: Current Version (updated 05/28/2024)

• Falling in Line: Legislator Behavior in the Russian State Duma under Putin

How does authoritarian regime consolidation affect legislator behavior? During periods of relative stability, authoritarian legislatures function according to a system of mutual restraint: legislators refrain from expressing opposing viewpoints outside of regime-set limits on form and timing, and regimes do not punish legislators who are critical within these limits. However, this system is harder to maintain during transitional periods when regime incentives are fluid and legislators have less information about acceptable behavior. In this paper, I argue that legislators respond to the uncertainty associated with regime consolidation by rapidly reducing the visibility of their negative opinions as they seek to identify the new normal. I use the case of Vladimir Putin’s Russia to illustrate these dynamics. Drawing on data from over 20,000 bills and 50,000 roll-call votes that span two decades of legislative activity, I document how individual legislators navigated the transition and discuss the implications for our understanding of how legislatures function in authoritarian regimes. .

• Buying the People or the Purse-Strings? The Allocation of Housing Projects in Turkey

How do political parties distribute economic benefits to maximize the expected electoral return on their investment? In this paper, I focus on whether parties accomplish this by appealing directly to citizens or, instead, focus on gaining the support of key financial actors in important electoral districts. To answer this question, I utilize data from the centrally-administered TOKI project on the allocation of housing projects in modern Turkey. I examine whether electoral considerations better explain the geographic location of the housing projects themselves or the geographic location of the construction firms hired to fulfill these contracts.