EC2C1 Econometrics II
Empirical Project 2023-24
In this project you will be asked to present your own version of results reported in the paper by Fernanda Brollo, Tommaso Nannicini, Roberto Perotti, and Guido Tabellini “The Political Resource Curse,” American Economic Review, vol. 103, no. 5, August 2013. The authors investigate whether mayors in Brazil become more corrupt when they have access to more resources and whether these resources attract less skilled candidates to run in subsequent elections.
Write an individual essay that addresses the following points:
. Describe the research question and the research design.
. Present and discuss results of additional federal transfers on corruption episodes, the re- election of incumbent mayors, and the education of political challengers.
. The specifications in the paper are needlessly complex. Present results based on simpler specifications. Explain and motivate your choices.
. The data are for two periods. In addition to the RD analysis in the paper, carry out a DD
exercise and compare the results. Discuss the strength and weaknesses of each approach.
. The authors use some specifications that can be criticised on the basis of what you have learned in the course. Improve on their practice.
Keep the following in mind:
. Write a single coherent essay rather than point by point answers to the questions above. Your essay must be in your own words and not simply follow the paper.
. You receive a data set (BNPT_small.dta) that differs slightly from the one used in the original paper. Your results will therefore not be exactly the same as those of the authors. This is based on their “small” data; it doesn’t let you produce results as in Tables 8 and 9.
. Limit your essay to 2,500 words and up to 8 tables and figures. This means you will have to be focused and concise. Limit your analysis to a few key variables; don’t try to reproduce everything in the paper. Don’t give drawn out answers.
. The theoretical model and some econometric detail in the paper is beyond what we do in this course. You can safely ignore what you haven’t learned.
. You should consult the authors’ replication files to figure out details about the exact
specifications they have used (the published paper does not always have all the details). But remember that we don’t ask you to replicate their results. Make sure you don’t copy tables or figures directly from the paper.
. You can consult other literature if you feel it is helpful to you but this is not necessary for a successful report. Stick to the instructions above and don’t get sidetracked. Cite any sources you use carefully.
Deadlines:
Week 9, 13 March, 12 noon: Submit progress report of no more than 500 words outlining what you have done so far for the project, future plans, and areas you are struggling with. (10% of your mark).
The due date for the project is 8 May, 12 noon. You must submit the Stata log file producing your results together with your paper. (80% of your mark).
10% of your mark for the WT portion of the course is based on your homework submissions.