Development economics2

Part 2: Aid, policy and economic growth In part 2 you will analyze issues to do with aid, policy and growth, using cross-country macro data. These data stored in the Stata file [login to view URL], have been analyzed in the following article: Easterly, W., Levine, R., Roodman, D. (2004), “Aid, Policies and Growth: Comment,”, American Economic Review, 94(3): 774-780, which is a comment on: Burnside, C., Dollar, D. (2000), “Aid, Policies, and Growth.” American Economic Review, 90(4): 847- 868. It is important that you note the distinction between the variables defined by Burnside-Dollar (these start with bd, e.g. bdssa is the dummy for Sub-Saharan Africa in the BD dataset) and those defined by Easterly-Levine-Roodman (these start with elr, e.g. elrssa is the dummy for Sub-Saharan Africa in the ELR dataset). Thus, when we replicate the analysis in BD we should use the variables starting with bd, and when we replicate the analysis in ELR we should use the variables starting with elr. Problems a) Describe the dataset [login to view URL] – e.g. how many countries are included, what time periods, what variables? b) Compare Sub-Saharan Africa (SSA with the rest of the world (RoW), with respect to the level and growth of per capita GDP using the BD measures (bdgdp an bdgdpg, respectively). Is there any evidence that SSA has been “catching up” with RoW with respect to per capita GDP over the sample period? c) How is the Burnside-Dollar measure of aid defined? Calculate the sample mean, median, and standard deviation of this variable for the entire data set, and for the subsample of countries in Sub-Saharan Africa. Do you think the numbers obtained are sensible (i.e. not obviously completely wrong)? d) How is the Burnside-Dollar measure of policy defined? Is the variable correlated with aid? Interpret your finding. e) Is aid correlated with per capita GDP? Is aid correlated with the growth rate in per capita GDP? Do you think you are learning anything about the casual effect of aid onto either the level or the growth of income from these correlations? f) Replicate the OLS results shown in Easterly-Levine-Roodman, Table 1, column 1 (these results are the same as those shown in BD, Table 4, column 5 OLS). g) Interpret the estimated coefficients on aid and aid*policy obtained in (g). h) Interpret the estimated coefficients, obtained in (f), on Sub-Saharan Africa. i) According to the results obtained in (f), what would be the effect on economic growth (in percentage points) of an increase in aid from 1% of GDP to 2% of GDP, for a country with “poor”, “average” and “good” policies respectively? j) Replicate the OLS results shown in Easterly-Levine-Roodman, Table 1, column 2. Interpret the results on aid and aid*policy. What role does aid play for growth? k) Why are the results in (f) so different from those in (j)? Which set of results should we believe, and why? All results i.e graphs and tables shall be presented in a word document together with an explanation. The data set file is appended.