Great week: Dissertation proposal and grant.

Last week was busy but fantastic for two reasons: 1) I handed in my dissertation proposal, and 2) together with a colleague I received a grant to undertake a network experiment in the DR Congo.

Dissertation proposal

In their third year, Political Science PhDs at Columbia (a.o. me) have to hand in their dissertation proposal; a 12 page document outlining: dissertation topic, why we should care, contribution, the approach, literature review, etc. Before handing in one needs to obtain signatures of two faculty members that 'sponsor' the project. Then in May one defends the proposal to a committee of four professors. Yesterday I handed in my proposal after obtaining signatures from Kimuli Kasara and Macartan Humphreys.

Below you find a required one-page summary (if interested I can send you the 12-page document that elaborates on what is written below - especially on the approach):

Title: Forced Migration and the Provision of Public Goods: Evidence from the DR Congo

Much recent work has focused on the impact of diversity - whether in ethnicity, caste, race, language, income, age, occupation, education or religion - on public good provision. Diverse communities - compared to homogeneous communities - seem to do a worse job of producing public goods such as schools, widening roads and providing good health care. Also why this could be the case has been taken up.

Given the interest in the impact of diversity on public good provision it is surprising that little work has been done on the impact of migration on public good provision. First, migration by definition leads to diversity - at the very least the cleavage native versus immigrant is (temporarily) created. Second, migration also impacts public good provision in ways that go beyond diversity. This dissertation will focus on forced migration in the developing world. Currently over 26 million people are internally displaced (IDP) and over 10 million are refugee; the far larger majority of these forced migrants reside in the developing world (UNHCR 2009). Especially these migrants - because they have a home they would like to return to, and host communities are likely to be very fragile - are likely to impact public good provision negatively.

However, from a unique and representative survey undertaken in Eastern Democratic Republic of Congo we obtain an extraordinary result. Contrary to the diversity and public goods literature – and the additional negative bias that we expect from migration – we find that migration and public goods provision is positively correlated.

How can this be the case? This dissertation takes this puzzle head on, and argues that this could be the case for four reasons:
1. migrants' opportunity cost to contribution is likely to be lower;
2. migrants are often dependent on and therefore feel grateful to the host community;
3. migrants are easier forced than natives to contribute to the public good;
4. migrants may want to settle and be accepted, and therefore are willing to send a costly signal.

First, because of the strategic nature of public good provisions, the dissertation’s formal part will work out these channels in more detail by making use of game-theory. Second, the dissertation's empirical part will be based on four pillars: 1) interviews with both natives and migrants, 2) survey-work, 3) geographic mapping, and 4) behavioral games.

The stated hypotheses will be tested in Eastern DR Congo; one of world's regions most affected by forced migration. First, it experienced one of the world's largest refugee inflows when in 1994, during the Rwandan genocide, over 2 million Hutu refugees moved into the country; many of them are still present. Second, sustained fighting over the last decades has put the IDP number at over 2 million; most of them are to be found in the East One of world's regions most affected by forced migration.

If you would like to read the 12 page document please send me an email. I will be distributing this proposal widely - to a wide range of different professors, policy makers in the Congo and outside the Congo, but also friends and family. If you have any suggestions or remarks on the proposal, please let me know!

Grant for network project.

Together with a friend and colleague (Neelanjan Sircar) I applied to a summer research grant by Columbia’s Applied Statistics Center and Columbia's Center for the Study of Development Strategies. Yesterday we heard that we got the grant! Below you find the our grant proposal's introduction:

Title: Latent Network Structures and Public Goods Games

Many aspects of our lives are governed by social networks, making it critical to understand how these networks impact human decision-making. While for many years much work on networks has been done in fields such as computer sciences and sociology, it has been largely absent in the economics and political science literature. In addition, an individual’s strategic decisions about how to interact with another individual are impacted not only by previous interactions with the other individual, but also by the other individual’s previous interaction with friends and acquaintances. Whether I contribute to a public good depends on my expectations of what other people do, and I use these. Work that makes use of experiments to understand social networks, while potentially very informative, has been limited. We propose an experiment that addresses the role of existing network structure and observable individual attributes upon public good provision. The experiment will be conducted in a lab at NYU (where individuals are taken from a single class) and in 5 villages in Eastern Congo. In both cases the samples have previous social interaction and developed interpersonal preferences, what we will refer to as a latent social network.

Networks will take a central role in both mine and Neelanjan's dissertation. With regards to my dissertation obtaining a clear understanding of the formation and the importance of networks (think of migrants vs natives in Congolese villages) in strategic situations (think of contributing to public goods) is crucially important. This experiment is therefore the first empirical step in our dissertations.