In this research assignment, you need to conduct an empirical research about the determinants of cross country heterogeneity in GDP per capita .

Your research should include:

-     A title and a short introduction.

-     A literature review, focusing on growth empirics literature.

-     An empirical section that describe data and methodologies and regression results.    Hint: the choice of variables in your regression should be supported by the literature review in the previous section.

-     A short conclusion.

The suggested word limit is 2000 words (references and appendices are not included in the word count).

In the empirical section, you are asked to use the Macrohistory database. These data have a much smaller number of countries but a much larger number of years per country. First of

all, familiarize yourself with the dataset and variables. All the relevant information is here:

https://www.macrohistory.net/database/ .

•   After familiarizing with the dataset, select a subgroup of variables that you consider are  relevant to understand the drivers of countries’ GDP per-capita. To make this decision     use the models we used in lectures as a guide. However, you are free to think out of the box and look for relevant relationships we have missed so far1 . You can collect additional data for your empirical section.

•   Provide descriptive statistics on your variables (number of observations, mean, median, standard deviation).

•   Proceed to estimate the relationship between (log) GDP per-capita or GDP growth (your choice) and the variables of interest that you selected. Start by estimating a simple OLS  regression with all the countries pooled together. Interpret the coefficients of the regression and discuss econometric and economic issues behind the results. Can you      make sense of the results using any existing theory? Are your results consistent with previous literature?

•   Discuss the econometric issues of your OLS estimates.

•   Extend the previous analysis and estimate a panel fixed-effect regression to control for countries’ unobserved heterogeneity that might shape the observed heterogeneity in GDP per-capita (growth)2 . How different are the results compared to OLS? Are your results consistent with previous literature?