How Strategic Purges of State Security Personnel Protect Dictators Security Studies 2025 Link.
How can authoritarian leaders employ purges to maintain control over their security services? How can they prevent adverse collective action, such as coups, collusion, and cover-ups? The fundamental problem of repression is that the agents of repression, once empowered, can turn on their leaders, usurping their power or even deposing them. I argue that leaders can use purges to undermine collective action capacity within cliques by targeting both high and low-ranking individuals. I use newly compiled data on 36,896 low-ranking NKVD (People’s Commissariat for Internal Affairs of the Soviet Union) officials under Stalin to show that officials with connections to purged officials were more likely to be purged themselves. This paper sheds light on the obscure workings of secret police organizations and how leaders control them. It also illuminates how authoritarian leaders prevent challenges and use repression to consolidate power.
Security and Economic Legacies of Population Resettlement: Evidence from Manchuria (With Xu Jing and Anna Zhang) Forthcoming at Journal of Historical Political Economy Link.
Why do post-colonial states engage in population resettlement in their frontier territories? In this paper, we shift away from the motivations for resettlement by advancing a cost-centric theory for resettlement. We contend that states may use the resettlement policy because they inherit the infrastructural capital to do so from settlers sent by former colonial powers seeking to consolidate their frontiers. We test the observable implications of the theory using a unique geo-coded archival dataset and in the context of Manchuria, a northeastern border region of China. We find that Manchurian areas that once received more Japanese settlers during the colonial period are associated with greater proximity to Chinese settlers in the post-colonial era. We also show that, contrary to most findings about the pro-growth institutions associated with colonial settlements, Japanese settlements led to slower economic development in the long run. By focusing on the costs rather than motivations of resettlement, our paper expands our understanding of the rationale for state-sponsored resettlement policies and uncovers an alternative relationship between colonial settlement and economic development.
How Settlement and Inter-Ethnic Conflict Shapes State Capacity. Forthcoming at Conflict Management and Peace Sciences Link.
How does ethnic diversity and conflict affect the capacity of a state to govern its subjects in the context of settler migration? In a multiethnic society where settler groups are displacing natives with traditional collective land ownership institutions, the conflict shapes the distribution of state power. Whereas conflict is normally associated with weaker state power, this paper shows that states pursuing expanding territorial control through land registration can expand their authority over settler groups through opportunistically intervening in conflicts between settlers and natives. On the one hand, such conflict increases the settlers' reliance on state authority, as they must rely on state power to assert individual property rights over parceled plots for intensive cultivation. On the other hand, established native populations with collective property rights over nonparceled land for extensive use have recourse to their own established authorities such as local ethnic and sectarian elites and are therefore less directly reliant on state power. This proposition is tested using data from the 1940 Manchukuo Census, using age heaping as a measure for state capacity as legibility. I find evidence consistent with my theory in the case of Han Chinese settlement into Mongol lands during the period of Japanese rule.
Mass receptiveness to unconstrained emergency legislation during crisis: survey experiment in pandemic-era Japan (with Charles Crabtree)Humanities and Social Sciences Communications 2025 Link.
To what extent does government messaging influence the willingness of citizens to accept constitutional amendments that empower the executive during crisis? Leaders trying to increase their power often attempt to mobilize public opinion for emergency legislation by emphasizing institutional constraints and crisis severity. To test the extent to which the public is swayed by such rhetoric, a vignette survey experiment was conducted with a national sample of 2569 Japanese during the COVID-19 pandemic. The experiment asks respondents to consider the tradeoff between executive power and their own safety, in a realistic setting. We find robust null effects, suggesting that such messaging does little to sway respondents.
Which Policy Attributes Affect Assisted Voluntary Return and Reintegration Uptake Among Ukrainian Evacuees in Japan? (with Ghulam Dastgir Khan, Alina Repeshko, Truong Pham, and Ichihashi Masaru)Humanities and Social Sciences Communications 2025 Link.
What attributes of assisted voluntary return and remigration (AVRR) policies are likely to increase acceptance? Governments have put increasing efforts into remigration, or the return of migrants to their countries of origin. Whereas existing research has focused extensively on the determinants of policy choices, the efficacy of particular policies enacted by governments, as well as other push/pull factors in the host and origin societies that influence remigration choices, there is little work on what attributes of an AVRR policy influence policy uptake. To this end, we conducted a randomised conjoint survey on 242 Ukrainian evacuees in Japan from June to July 2024 which at the time accommodated approximately 2600 Ukrainians fleeing the 2022 Russian invasion. We find that factors such as guarantee of housing in an area less affected by conflict and the identity of the policy provider increase policy acceptance probability. Surprisingly, even large increases in one-time payments had little influence on policy uptake. This paper is the first application of conjoint experiments to study which features of AVRR policy affect acceptance probability. This approach would be useful globally to understand the preferences of potential AVRR policy recipients in diverse contexts.
Field Research When There Is Limited Access to the Field: Lessons from Japan, PS: Political Science & Politics 2022 Link. (with Amano, Dominiguez, et. al.)
How can scholars conduct field research when there is limited access to the field? This article first identifies how limited and uncertain field access can affect field research and then provides recommendations to address these challenges. We focus on conducting field research in Japan because of our substantive expertise, but we believe that the problems and solutions outlined in this article are applicable to a broad range of countries. Our hope is that this article contributes to the developing literature on conducting research during times of emergency and to the larger literature on best practices for field research.
The Effects of Electoral Anticipation on Portfolio Allocation, Party Politics 2020 Link. Suppl. materials.
Abstract: Whereas much of the existing literature on coalition formation has focused on the disparity between Gamsonian payoffs and the outcomes of bargaining models, dangers of ministerial drift, party ideal points, issue salience/emphasis, and negotiation complexity, very little has been said on the role of the way public opinion affects political behavior via expectation of future electoral returns. Following the logic of “issue yield,” this article argues that politicians allocate ministerial portfolios according to the distribution of public opinion in the hopes of obtaining better returns in subsequent elections and find compelling evidence linking public opinion with portfolio allocation.
Are Women Targeted for Corruption? Evidence from Randomized Conjoint Experiments on Traffic Police in Nigeria (With Aliyu Ali Bawalle, Diego Romero, Maria Nagawa, and Ghulam Dastgir Khan) Link.
While debates suggest women are disproportionately targeted for extortionary bribery, empirical findings remain mixed. Building on research showing that gender shapes access to political connections, wealth, and retaliation capacity, we hypothesize that women are more vulnerable to extortion. To test this, we employ a series of randomized conjoint experiments on approximately 600 traffic officers in Kano and Kogi states in Nigeria, a context characterized by widespread petty corruption and police predation. The experiments isolate the causal effect of a hypothetical driver’s gender on the likelihood of solicitation by an officer, while probing potential mechanisms. We specifically examine whether women are more likely to be targeted due to inferences officers make about drivers based on gender. We specifically examine whether officers interpret gender as a proxy for social standing, wealth, political connections, or legal knowledge, and whether “ambivalent sexism” norms reduce targeting of women despite structural disadvantages. Contrary to expectation, we find that police are less likely to target women. Furthermore, we find that women are not necessarily more difficult or less profitable targets for extortion. Qualitative evidence collected from officers and drivers suggest that prevailing social norms likely drive our outcome. Our findings contribute to debates on corruption, gender, and state-society relations by advancing a more precise understanding of how gender shapes vulnerability to bureaucratic extortion.
Traditional Gender Norms, Clientelism, and Corruption: Survey of African Entrepreneurs (With Ghulam Dastgir Khan, Aliyu Ali Bawalle, Diego Romero, and Maria Nagawa) Link.
How do differences in traditional gender norms impact gendered attitudes to wards clientelism and corruption? How do these norms impact perceptions of women engaged in clientelistic and corrupt exchanges? What, then, are the p ossible implications of these n orms for female entrepreneurs in Africa? Findings on women’s engagement in clientelistic networks and corruption ar e mixed. Some research shows women are less likely to engage in clientelist ic networks because of their position in social structures and t he resultan t expectations on their behavior (Alexander et al 2023; Daby 2021) while ot her research argues women are just as likely as men to engage in corruption (Brierley and Pereira 2023). We argue that women’s engagement in, and perc eptions of women’s e ngagement in corruption and clientelism, can be linked to cultural norms in society. Primarily, we expect highly patriarchal socie ties to exhibit larger differences in attitudes towards corruption and clie ntelism between men and women. We further expect mo re punitive approaches t owards women who engage in corruption and clientelism in more highly patria rchal societies regardless of gender. To test our expectations, we use the recently developed Male Dominance Inde x (MDI) (Guarnieri and Tur - Prats 2024) in a survey of male and female entre preneurs in Africa. Our population of interest is African entrepreneurs. We will partner with local enumerators in various countries across Africa incl uding Botswana, Mozambique, Zambia, Democratic Republic of the Congo, Eg ypt , Kenya, Malawi, Uganda, Rwanda, Ethiopia, Djibouti, Cameroon, Ghana, Seneg al, Nigeria, and Niger to generate a convenience sample of around 2000 to 4000 entrepreneurs in various localities. Selected individuals will be contacted face - to - face, by email , or by WhatsApp and invited to participate in a 15 - minute online survey.
How Balanced Historical Memory Shapes Outgroup Attitudes: Evidence From Japanese Settlement in Manchuria (With Xu Jing) Link.
Past studies show groups victimized by collective violence exhibit heterogeneous outcomes such as increased hostility or greater tolerance towards outgroups. Yet, victimized groups can also sometimes be perpetrators. We propose that narratives that emphasize balanced framing, or recognizing victimhood while acknowledging perpetration may be more effective in reducing hostility than either alone. We will test this proposition with a vignette experiment on Japanese respondents using examples from the case of Japanese settlers in Manchuria who faced violence during evacuation. We also test whether regions that have adopted balanced framing on this issue in public education exhibit different legacies of exposure to collective violence than regions that did not by conducting surveys of two prefectures with the highest settler emigration rates, where one teaches extensively with balanced framing (Nagano), and another does not (Yamagata).
Impact of aid information on recipients' stances on donor-recipient territorial dispute: experimental evidence from Timor-Leste Link.
Can foreign aid donors curry favor among citizens of the recipient country to obtain substantive policy concessions? The literature on the impact of aid on attitudes in recipient countries shows that disbursement can increase alignment with, and consumption of imports from, the donor. Others have shown mixed results, with null findings for the impact of aid or aid information on attitudes towards the donor and policy alignment, as well as donor misidentification, and negative effects on attitudes towards the donor for autocratic donors. A limitation of the literature is that it examines the impact of aid disbursement on general attitudes and relatively abstract ideological alignment, rather than concrete, substantive disputes between the donor and the recipient, partially due to a focus on North-South aid flows. To assess the impact of aid information on substantive disputes, we show how information on diplomatic and development assistance from Indonesia to Timor-Leste affects the stances on a bilateral territorial dispute among Timorese through a classroom experiment on 213 students in the National University of Timor-Leste. We find that aid information moderates the stances of the students regarding the dispute and decreases the magnitude of the claims made in the dispute.
...
Back