What drives the economic development of a country? This question has been the center of debates for a long time and has had important economic and political implications. In a highly influential paper, Acemoglu, Johnson, and Robinson (2001, American Economic Review) inves- tigated the effect of political institutions on economic performance. Their theory is that good institutions, such as rule-of-law and property rights, should result in a country having higher long-term economic output than if the same country had poor institutions. To support their theory, they use a sample of 64 former European colonies. The key variables are loggdy (Log GDP per capita in 1995), risk (Measure of legal protec tion against expropriation, 0: lowest protection, 10: highest), and logmort0 (the logarithm of mortality rate of European settlers). The baseline econometric specification is log(GDP per capita) 8 0+8 1risk + E, (1) where & is the error term. The parameter of interest is 1. The authors argue that risk is likely to be endogenous. To handle this problem, they use log(mortality) as an instrument and set up the first-stage equation risk = To + Tilog(mortality) + u, (2) where u is the error term. Using the provided dataset, you are required to do your own analysis on the problem focusing on (but not limited to) the following questions. It is also expected that your group uses its own Stata do-file to generate the results. 1. Estimate (1) by OLS and 2SLS and estimate (2) by OLS. Discuss the results. Provide a formal test result whether risk is endogenous. 2. Compare the homoskedastic and heteroskedastic standard errors. 3. The authors use log(mortality) as the instrument, rather than, the level of mortality. Estimate the model using the level of mortality as the instrument. (This variable is not provided in the dataset so you need to take the exponential of log( mortality).) Explain why the authors preferred log(mortality). 4. Estimate the model using an additional instrumental variable, the square of log(mortality). How do the results change? What is the interpretation? 5. Discuss the relevance and the exogeneity of the instrument. Provide formal test results, whenever possible. 6. Add additional exogenous variables, latitude and africa, to the baseline specifications. Estimate the model by OLS and 2SLS. Are the included variables predictive of the level of GDP in OLS and 2SLS? How does the interpretation of the effect of latitude and africa change