Selected Papers

Foreign Aid and Undeserved Credit Claiming (with C. Schneider), American Journal of Political Science (2017).

Politicians in developing countries misuse foreign aid to get reelected by fiscally manipulating foreign aid resources or domestic budgets. Our paper suggests another mechanism that does not require politicians to have any control over foreign aid in order to make use of it for electoral purposes: credit-claiming. We analyze the conditions under which local politicians can undeservedly take credit for the receipt of foreign aid and thereby boost their chances of reelection. We theorize that politicians can employ a variety of techniques to claim credit for development aid even when they have little or no influence on its actual allocation. Using a subnational World Bank development program in the Philippines, we demonstrate that credit-claiming is an important strategy to exploit foreign aid inflows and that the political effects of aid can persist even when projects are designed to minimize the diversion or misuse of funds.

"Credit Claiming and the Politics of Relief Efforts" The Monkey Cage, The Washington Post (22 November 2013).


Politician Family Networks and Electoral Outcomes: Evidence from the Philippines (with J. Labonne and P. Querubin). Forthcoming, American Economic Review.

We demonstrate the electoral importance of politician family networks and provide evidence of the mechanisms behind the relationship. We use a 20 million person dataset, allowing us to reconstruct intermarriage networks for over 15,000 villages in 709 municipalities in the Philippines. We show that politicians are disproportionately drawn from more central families and that, controlling for candidate fixed effects, candidates receive a higher vote share in villages where their families are more central. We present evidence that centrality confers organizational and logistical advantages that facilitate clientelistic transactions such as vote buying and do not operate through popularity, name recognition or through the choice of policies more aligned with their constituents' preferences.

Online Technical Appendix
Family Name Matching Tools and Code (Beta Version)


Political Parties, Clientelism, and Bureaucratic Reform (with P. Keefer), Comparative Political Studies (2015).

The challenge of public administration reform is well-known: politicians often have little interest in the efficient implementation of government policy. Using new data from 439 World Bank public sector reform loans in 109 countries, we demonstrate that such reforms are significantly less likely to succeed in the presence of non-programmatic political parties. Earlier research uses evidence from a small group of countries to conclude that clientelist politicians resist reforms that restrict their patronage powers. We support this conclusion with new evidence from many countries, allowing us to rule out alternative explanations, including the effect of electoral and political institutions. We also examine reforms that have not been the subject of prior research: those that make public sector financial management more transparent. Here, we identify a second mechanism through which non-programmatic parties undermine public sector reform: clientelist politicians have weaker incentives to exercise oversight of policy implementation by the executive branch.

"Reform and the Power of Party Platforms," Ideas Matter, Inter-American Development Bank Research Department (4 March 2016).


Village Social Network Structures and Electoral Competition (with J. Labonne and P. Querubin).

In this paper, we test whether the structure of village social networks affects political competition in local elections. We use data on 20 million individuals in 15,000 villages of the Philippines. We take advantage of naming conventions to assess intermarriage links between families and reconstruct the family networks in all of those villages. Using data from the 2010 local elections we show that there is less political competition in villages with dense social networks, a result that is robust to controlling for a large number of village and candidate characteristics and to alternative estimation techniques. We then explore the mechanisms behind this effect and present evidence, from a detailed dataset collected in 284 villages after the 2013 local elections, that political influence is more concentrated in denser villages and that this is associated with higher turnout and higher vote margins for the winning candidate.

Winner, Best Conference Paper, Political Networks Section, American Political Science Association


Incumbent Advantage, Voter Information, and Vote Buying (with P. Keefer and J. Labonne). Under Review.

Results from a new experiment shed light on the effects of voter information on vote buying and incumbent advantage. The treatment provided voters with information about a major spending program and the proposed allocations and promises of mayoral candidates just prior to municipal elections. It left voters more knowledgeable about candidates' proposed policies and increased the salience of spending, but did not affect vote shares and turnout. Treated voters were more likely to be targeted for vote buying. We develop a model of vote buying that accounts for these results. The information raised voter expectations regarding incumbent performance, especially in incumbent strongholds. Incumbents increased vote buying in response. Consistent with this explanation, both knowledge and vote buying impacts were higher in incumbent-dominated municipalities. Our findings show that, in a political environment where vote buying is the currency of electoral mobilization, incumbent efforts to increase voter welfare may take the form of greater vote buying.

Pre-Analysis Plan:
American Economic Association (AEA) Registry for Randomized Control Trials
J-PAL Hypothesis Registry


Social Networks and the Targeting of Vote Buying. Under Review.

Social networks have been shown to facilitate political cooperation and information transmission in established democracies. These same social networks, however, make it easier for politicians in consolidating democracies to engage in illegal electoral strategies. In particular, voter social networks can be used to identify and monitor voters for vote buying. Using a survey of 864 households in the Philippines, this paper demonstrates that individuals with more friend and family ties are disproportionately targeted for vote buying. In addition, this paper exploits differences in the perception of vote secrecy to provide evidence consistent with the monitoring mechanism of networks, on the rationale that monitoring voters is only necessary when the vote is perceived to be secret.


Democracies Only: When do IMF Agreements Serve as a Seal of Approval? (with M. Bauer and B. Graham). The Review of International Organizations (2011).

Conditional lending by the IMF is predicated, in part, on the belief that IMF programs should be associated with increased capital inflows to participating countries. This belief is generally consistent with theoretical arguments in the academic literature (e.g. Bird and Rowlands 1997, Bordo et al 2004, an important exception includes Morris and Shin 2006). However, the empirical literature has produced ambiguous results (e.g. Jensen 2004). We argue that confusion arises because participating in an IMF agreement only catalyzes FDI inflows in democracies, but has the opposite effect on FDI inflows for autocracies. This is because access to FDI depends on a country's ability to credibly commit to implementation, and this ability varies systematically across regime type. In addition to explaining this puzzle in the literature, our work provides a new theory for how IMF participation signals investors about a country's macroeconomic health and policy intentions. Last, our paper features a new methodological contribution to the literature—the use of a treatment effects model with a Markov transition in the first stage. This allows us to model the decision to enter a new IMF program as distinct from the decision to continue an existing program—an innovation that improves the precision and accuracy of the estimates.