Core Research Areas

The main objective of the center’s research is to provide policy-relevant answers to major questions in the political economy of development. Research is likely to fall within three major clusters: political institutions and accountability, conflict and development, and the politics of development aid.

Political Institutions and Accountability
International interventions increasingly focus on providing support for the creation or reform of political institutions. A major question is when and how such interventions in fact lead to better decision making. What institutions render politicians more accountable to populations? Does transparency reduce corruption? When does broader participation in decision making improve the quality of decisions? When does decentralization produce more efficient decisions? A number of ongoing projects already address some of these questions. In one project we are undertaking a field experiment on political information and accountability. In this project we collect information on the behavior of parliamentarians in Uganda with respect to their behavior in committees, on the floor and in their constituencies. The publication of this information is then timed randomly relative to the re-election of MPs and the impact of the information on elite behavior is examined. The main aim is to assess (a) the extent to which transparency affects political accountability (positively or negatively) and (b) the mechanisms through which this works. In a second project in DR Congo we are examining the effects of affirmative action for enhancing political representation and accountability. Existing evidence on the effects of gender quotas for parliaments and in committees suggest that the gender of representatives does have a substantial effect on the kinds of decisions that are made and on subsequent women’s welfare. Causality Is hard to establish however since the existence of quotas is likely indicative of other features of a polity that matter for welfare. As part of an experiment on the question in DRC, IRC is requiring 50% gender quotas for decision making bodies in approximately 300 village councils that determine the allocation of significant development aid. This experiment provides a unique opportunity that is of relevance for program design but also for understanding the mechanisms of political accountability. There is much scope for further projects along these lines that exploit variation in decision making rules, organizational structures, and information environments to account for variation in the effectiveness of political institutions.

Conflict and Development
Violent conflict poses a special challenge to economic development. National governments and international organizations seek strategies to prevent violence and to build peace in its aftermath. How is political order established? How does poverty affect conflict risks? How are violent organizational structures formed and dismantled? Some of these questions are now being addressed in ongoing field projects. One project focuses on the reintegration of combatants and links in to a longstanding debate in the study of political violence on the relative importance of private economic incentives and social capital on the propensity to participate in armed conflict. To address the question the project works with the reintegration unit of the Colombian Presidency and involves the randomization of post conflict government investments between socially oriented investments (social clubs, sports clubs, community centers) and private economic investments (job creation, enterprise promotion). The outcome variables include participation in gangs and criminal activity as well as the return of ex-combatants to fighting groups. Another project is designed to examine the effects of economic interventions on peacebuilding. This project is being implemented together with one of the largest global NGOs doing this type of work, the IRC (Liberia and Congo) and uses public lotteries to select the communities that will receive development aid from IRC. The Liberia project involves a population of approximately 200,000 individuals in the northern county of Lofa. The Congo project is being implemented throughout East Congo from South Kivu to Katanga and involves a civilian population of approximately 4 million. In both cases we examine the effects of welfare interventions of the local distribution of crime and violence.

Politics of development aid
The third area of interest is the effectiveness of development aid. A core aim is to understand what are the political consequences of development aid and how different mechanisms for delivering aid interact with local political processes. When, for example, is it more effective to provide services through government or through the private sector? Does aid generate political dependency? Do financial contributions from beneficiaries lead to better project selection and implementation? The Congo and Liberia projects described above address some of these questions; in particular they examine how local political structures change in response to influxes of development aid. In addition, we are in the early stages of a project implemented together with the center for applied statistics and Columbia’s school of engineering to examine the merits of market and non-market mechanisms for distributing development technologies in Uganda. There is however considerable scope of more work in this area to better understand the logic of rent-seeking, aid dependency, and allocational mechanisms in developing areas.