Elections as Incentives: Project Completion and Visibility in African Politics (2017). Job Market Paper.
Abstract: This paper explores the disciplining effect of elections on policy in Sub-Saharan Africa. I test whether voters reward incumbents for completing development projects and whether governments expedite completion in response to electoral incentives. Using a dataset of investment projects funded by the World Bank between 1995 and 2014, I show that project completion yields large electoral benefits for incumbents in a sample of 61 African national elections. This effect is primarily driven by projects in visible sectors such as basic infrastructure and social services. The causal effect of completion is identified from an instrumental variables strategy that exploits exogenous variation in the workload of project team leaders at the World Bank. Using a second instrument that predicts the occurrence of elections based on pre-determined constitutional rules, I then show that governments expedite completion before elections, target their effort towards visible projects, and prioritize completing ongoing projects over initiating new projects to influence the electorate. These effects are not driven by intertemporal substitution of government effort across the electoral cycle. Finally, I provide evidence that democratic institutions accelerate project completion in visible sectors on aggregate. Even in Africa's hybrid democratic regimes, elections create incentives for politicians to implement policy.
There Is No Free House: Ethnic Patronage in a Kenyan Slum (2017), with Thomas Stoker and Tavneet Suri. Revise & Resubmit, AEJ: Applied.
Abstract: Using unique data from one of Africa's largest informal settlements, the Kibera slum in Nairobi, we provide evidence of ethnic patronage in the determination of rental prices and investments. Slum residents pay higher rents and live in lower quality housing (measured by satellite pictures) when the landlord and the locality chief belong to the same tribe. Conversely, rental prices are lower and investments higher when residents and chiefs are co-ethnics. Our identification strategy relies on the exogenous appointment of chiefs and is supported by several tests, including a regression discontinuity design.
The Perils of Voter Mobilization (2017), with Vincent Pons and Tavneet Suri. Submitted.
Abstract: Voter mobilization campaigns face trade-offs in young democracies. In a large-scale experiment implemented in 2013 with the Kenyan Electoral Commission (IEBC), text messages intended to mobilize voters boosted participation but also decreased trust in electoral institutions after the election, a decrease that was stronger in areas that experienced election-related violence, and for individuals on the losing side of the election. The mobilization backfired because the IEBC promised an electronic voting system that failed, resulting in manual voting and tallying delays. Using a simple model, we show that signaling high institutional capacity via a mobilization campaign can negatively affect beliefs about the fairness of the election.
A Signaling Theory of Distributive Policy Choice: Evidence from Senegal (2016), with Jessica Gottlieb, Guy Grossman, and Horacio Larreguy. Revise & Resubmit, Journal of Politics.
Abstract: A recent literature emphasizes political economy factors behind the wave of administrative splits across the developing world. While past studies have focused on why some groups are more likely to obtain new administrative units, they do not explain why vote-maximizing incumbents use this arguably wasteful policy in the first place. We contribute to this literature by embedding administrative unit splits within incumbents' broader electoral strategy of distributive policies. We argue that incumbents prefer to target local public goods to groups for whom this is a credible signal of commitment, namely those with a history of a reciprocal relationship. Other groups require a stronger signal which, we argue, is provided by the creation of new administrative units that entail an increase in stable fiscal transfers due to the stickiness of administrative boundaries. We test our signaling theory using the case of Senegal and find robust support for its core predictions.
Diversity and Team Performance in a Kenyan Organization (2016), with Vincent Pons and Tavneet Suri. Harvard Business School Working Paper No. 16-078.
Abstract: We present the results from a field experiment on team diversity. Individuals working as door-to-door canvassers for a non-profit organization were randomly assigned a teammate, a senior manager and a set of households to visit. This created random variation within the organization in the degree of horizontal diversity (between teammates), vertical diversity (between teammates and their manager) and external diversity (between teams and their clients). We find that horizontal ethnic diversity decreases team performance, while vertical (or hierarchical) diversity improves performance, and external diversity has no effect. The data on time use suggests that horizontally homogeneous teams organized tasks in a more efficient way, while vertically homogeneous teams exerted lower effort.
Research in Progress:
Discretionary Funds and Dynamic Corruption Incentives: Evidence from Africa (2017). Presentation slides available on request.
Abstract: Previous studies of corruption have found evidence of “golden goose” effects: higher expected future rents may result in a short-term decrease in theft. This theory implies that politicians choose an upward-sloping corruption profile over their careers, even in the absence of term limits. I provide preliminary evidence of this phenomenon using survey data on perceptions of MP corruption in Sub-Saharan Africa. Opportunities for corruption in this setting arise from the adoption of Constituency Development Funds – political funds allocated using rigid rules which are spent at the discretion of local MPs. To construct a measure of potential rents, I exploit the malapportionment of the voting population across constituencies, which creates variation in the amount of discretionary funds available per capita. I find that perceived corruption decreases for first-term MPs and increases for multiple-term MPs. Consistent with this, incumbent MPs with access to larger discretionary funds are more likely to win re-election.
A Market Equilibrium Approach to Reduce the Incidence of Vote-Buying (2016), with Chris Blattman, Horacio Larreguy, and Otis Reid. Description of the evaluation on the J-PAL website. Please email me for a draft.
Abstract: Politicians in many developing countries win elections by purchasing votes, undermining political accountability and public goods provision. We present the results from a randomized experiment designed to reduce the prevalence of vote-buying in the 2016 election in Uganda. The intervention consisted of a large-scale grassroots campaign aimed at convincing voters and communities to collective renounce vote-buying. Our design allows us to estimate how candidates and their political brokers respond to the campaign in treatment and spillover areas, how voters coordinate their actions as a result of the campaign, and how these effects vary with local treatment intensity. While the campaign did not reduce vote-buying, it had sizable effects on electoral outcomes, with opposition candidates benefiting from the campaign at the expense of incumbent candidates. We present evidence that on the supply side of the market for votes, the campaign convinced some voters to vote in their conscience, regardless of any gifts received. On the demand side, incumbent candidates and their brokers responded to the campaign by attempting to buy more votes in areas neighboring treatment villages (spillover areas), while opposition candidates increased both their campaigning efforts and vote-buying efforts in treatment and spillover areas. This led to an increase in vote-buying, campaigning and turnout in spillover villages in equilibrium, which account for the effects of the campaign on electoral outcomes.
Islamic Institutions, Elites, and Structural Change: Evidence from Indonesia (2015), with Samuel Bazzi and Gabriel Koehler-Derrick.
Abstract: We exploit the aborted Indonesian land reform of 1960 to estimate the long-term economic impact of an institution present throughout the Islamic world – the waqf. In Islamic law, the waqf (plural awqaf) is an inalienable religious endowment established to encourage the provision of public goods and charitable services in perpetuity. Kuran (2001) argues that this institution was a major factor leading to economic stagnation in predominantly Muslim societies, because it hindered the reallocation of capital in periods of structural change. We provide preliminary evidence that in anticipation of the planned land redistribution, landowners with holdings above the maximum threshold allowed by the 1960 Agrarian Law transferred a large fraction of their land into waqf status to avoid seizure by the local government. We use plausibly exogenous regional variation in the intensity of the planned reform to identify the effect of awqaf on long-term agricultural and economic development, and to test Kuran’s hypothesis on the relationship between awqaf and resource misallocation.
Pre-Colonial Institutions, Ethnic Voting and Clientelism in Sub-Saharan Africa (2015), with Horacio Larreguy.
Abstract: A recent literature shows that pre-colonial political centralization fosters public goods provision in contemporary Sub-Saharan Africa. In this paper, we show that differences in public goods provision across areas with varying pre-colonial centralization results from more transfers of resources from governments, in exchange for electoral support, in areas where chiefs have a historical advantage in mobilizing their communities. We first provide a simple model where an incumbent politician garners support through targeted transfers. The electoral efficiency of these transfers depends on the ability of chiefs to mobilize voters, and increases when these are co-ethnics of the incumbent. We successfully test the distinct predictions the model generates using a newly constructed constituency-level dataset combining electoral outcomes and satellite lights from 35 Sub-Saharan African countries.