Current Research

“Ethnically Biased? Experimental Evidence from Kenya”
with Lars Ivar Oppedal Berge, Kjetil Bjorvatn, Simon Galle, Edward Miguel, Bertil Tungodden, and Kelly Zhang.  (Online Appendix)

Ethnicity has been shown to shape political, social, and economic behavior in Africa, but the underlying mechanisms remain contested. We utilize lab experiments to isolate one mechanism—an individual’s bias in favor of coethnics and against non-coethnics—that has been central in both theory and in the conventional wisdom about the impact of ethnicity. We employ an unusually rich research design involving a large sample of 1,300 subjects from Nairobi, Kenya; the collection of multiple rounds of experimental data with varying proximity to national elections; within-lab priming conditions; both standard and novel experimental measures of coethnic bias; and an implicit association test (IAT). Our tests find very little evidence of explicit or implicit coethnic bias in the behavioral experiments and IAT. These results run against the common presumption of extensive coethnic bias among ordinary Africans and suggest that mechanisms other than coethnic bias must account for the strong associations we see in the region between ethnicity and political, social and economic outcomes.


Accountability Can Transform (ACT) Health: A Replication and Extension of Bjorkman and Svensson, 2009
with Doug Parkerson and Pia Raffler (interview with BYU Radio’s Julie Rose)

This project constitutes a scaled up replication of the celebrated “Power to the People” (P2P) health intervention in Uganda (Björkman and Svensson, 2009), which reported striking impacts on infant weights, under-5 mortality, community engagement, and other outcomes. As in P2P, our replication collects information about health delivery in local health facilities, distributes this information to citizens and frontline healthcare providers in a report card, mobilizes citizens and healthcare providers in light of this information, and organizes meetings in which citizens and providers can jointly discuss how to improve health outcomes in the community. However our study employs a factorial design to break P2P’s complex treatment into two of its principal components: 1) the provision of information about health delivery quality to citizens and clinic staff and the mobilization of these actors in light of this information and 2) the holding of interface meetings between community members and clinic staff in which the action plans they each develop can be discussed and coordinated. We randomize these treatments across 377 health facilities (and associated catchment areas), a more than seven-fold increase in the size of the sample used in P2P.
(Pre-Analysis Plan)


“(Under What Conditions) Do Politicians Reward their Supporters? Evidence from Kenya’s Constituencies Development Fund”
with J. Andrew Harris

We leverage innovative spatial modeling techniques and data on the precise geo-locations of more than 34,000 Constituency Development Fund (CDF) projects in Kenya to test whether MPs reward their supporters. We find only weak evidence that MPs channel projects disproportionately to areas inhabited by their political allies, once we control for other factors that affect where projects are placed such as population density, poverty rates, ethnic demographics, and distance to paved roads. Notwithstanding this result, we find evidence for significant cross-constituency variation in political targeting, driven in large part by the spatial segregation of the MP’s supporters and opponents. Our findings thus challenge the conventional wisdom about the centrality of clientelistic transfers in Africa and underscore how local conditions generate particular incentives and opportunities for the strategic allocation of political goods. Although based on substantive materials from an African case, the paper thus speaks to fundamental issues in distributive politics.


“Term Limits, Leadership, Political Competition, and the Transfer of Power in Africa”
forthcoming in Nicholas Cheeseman, Ed. Politics in Africa: The Importance of Institutions
with Daniel J. Young (coding of 3rd term cases)

We update our 2007 Journal of Democracy article with data through the end of 2015 on how African leaders respond to two-term limits, which we take to be the most telling indicator of whether political power is institutionalized. Through a presentation of general patterns and a discussion several key cases, we show that two broad patterns are emerging. In one set of countries, multiple presidents have faced term limits and have voluntarily stepped aside when those limits have been reached. In all likelihood, term limits will continue to be respected in these countries in the future. In another set of countries, however, term limits have been circumvented and the prospects of leaders abiding by them in the future are poor. Nonetheless, we underscore that, even in the latter set of cases, the fact that leaders attempt to evade term limits by working through, rather than around, existing institutional frameworks is a critical—and in the broad context of African post-colonial history, novel—development.


“What Messages Encourage Young South Africans to Register and Vote?”
with Adam Berinsky, Daniel de Kadt, and Kate Orkin (Business Day op-ed)

Widespread electoral participation is a foundational element of a healthy democracy and an important mechanism by which voters exercise governance oversight. Existing work attributes low voter turnout in developing countries either to poor voter knowledge or to disillusionment with flawed electoral institutions. Yet turnout often remains low and sometimes declines as democracy matures, despite voters gaining basic knowledge about, and trust in, electoral processes. This problem is particularly acute in young urban communities. This study considers a driver of participation not previously explored in developing countries—the motivation to vote stemming from the intrinsic value of the act itself. In this paper we present preliminary evidence from survey data, focus groups, and a large-scale pilot experiment in South Africa that, taken together, suggest that a lack of motivation may be an important constraint on the political participation of young urban voters. We examine the relative efficacy of informational messages about how, when, and where to vote; motivational messages about why voting is important; and a combination of both, on voter registration and turnout. We hope to expand this work into a field experiment targeting 6,000 youth in urban Gauteng, South Africa. In concert with South Africa’s Independent Electoral Commission (IEC), Activate!, a network of youth leaders, and JPAL-Africa, the broader study will produce actionable evidence to improve outreach, bolster participation, and improve governance in South Africa and similar developing countries.


“PAPs: A Stocktaking”
with George Ofosu

The evidence-based community has held up preregistration as a solution to the problem of research credibility, but—ironically—without any evidence that preregistration works. The goal of our proposed research is to provide an evidentiary base for assessing whether PAPs—as they are currently used—are effective in achieving their stated objectives of preventing “fishing,” reducing scope for the post-hoc adjustment of research hypotheses, and solving the “file drawer problem.” We aim to do this by analyzing a random sample of 300 studies that have been pre-registered on the AEA and EGAP registration platforms, evenly distributed across studies that are still in progress, completed and resulting in a publicly available paper, and completed but (as far as we can determine) not resulting in a publicly available paper. Given the significant costs of researcher time and energy in preparing PAPs, and the implications that adhering to them may have on opportunities for breakthroughs that come from unexpected, surprise results, it is critical to take stock of whether PAPs are working.


“Ethnic Bias in Urban and Rural Africa”

The project investigates whether coethnic bias (defined as differential altruism toward coethnics and non-coethnics) differs across urban and rural residents in Kenya. The project follows up on the results reported in Berge et al (2017), which investigates coethnic bias among residents of Nairobi and finds no evidence of bias—except among subjects who had lived in Nairobi for fewer than ten years. These findings suggest that, contrary to the expectations of modernization theory, urban residence may dampen rather than heighten coethnic bias. A key implication is that we should see differences in the levels of coethnic bias across urban and rural samples. To measure coethnic bias, I compare patterns of play in Dictator, Public Goods, and “Choose Your Dictator” games across coethnic and non-coethnic partners recruited in urban and rural settings in Kenya. Although prior work has employed behavioral games to measure coethnic bias, this research has been limited either to urban or rural domains. No research to my knowledge has been designed explicitly to compare levels of coethnic bias in urban and rural settings using these techniques.