Electricity subsidies are popular among voters in many developing countries. However, electrical companies cannot always meet demand for electricity and in some locations must cut power to parts of the grid. This project asks whether such outages occur disproportionately in certain neighborhoods, such as poor neighborhoods or neighborhoods without political connections. If they do occur more frequently in poorer neighborhoods, this is evidence suggesting that certain electricity subsidies may be regressive policies.
The first major goal of this project is to determine whether presence of outages is correlated with wealth or political connections. Our primary dataset is a complete record of outages in Dhaka, Bangladesh over a two year period. The second major goal is to measure the cost of outages on economic welfare and productivity. We intend to instrument outages on weather. Weather variables can make for good instruments because temperature, for example, can alter how much electricity people use to cool their homes and thus how many outages occur.
Most work on this project will involve data analysis in STATA and translating analysis into clear and ordered graphs, tables, etc.
RAs demonstrating superior skill and motivation are encouraged to continue expanding their work with Professor Mobarak in the fall and beyond.
Requisite Skills and Qualifications:
The ideal summer candidate will be skilled in STATA. However, we will also consider applicants with experience in a similar statistical language such as R or MATLAB. Applicants with a quantitative background, especially in statistics or data analysis will be strongly preferred. Although learning from and collaborating with more senior researchers is an important part of the job, RAs should also be motivated and independent problem solvers.