ECO 330: Problem Set 2 Name: Student ID: Problem 1 Answers 1.a) The estimated coefficients and confidence intervals are (report to 2 decimal places) Coefficient 𝛼 đť›˝inst_qual Variable Constant Institution Quality Estimated Value #### #### Lower Bound for 95% Confidence Interval #### #### Upper Bound for 95% Confidence Interval #### #### The correlation between log GDP per Capita (PPP) in 1995 and Institution Quality in 1900 is: #### The interpretation of đť›˝inst in this regression is: [Write the interpretation here]. 1.b) Fill in the below table. Only the last column will be graded for correctness. Institution Quality in 1900 3.5 10 Predicted log GDP per Capita in 1995 #### #### Predicted GDP per Capita in 1995 (PPP, $/person) #### #### 1.c) [Write the reasoning behind using Settler Mortality as an instrumental variable for Institution Quality] The correlation between Settler Mortality is: #### The new estimate of đť›˝inst_qual is: #### [Write something about why the estimates differ between 1.a and 1.c] 1.d) The original regression in 1.a had ### observations. The IV regression in 1.c had ### observations. The number of observations [increased/decreased/stayed the same] for the IV regression in 1.c compared to the regression in 1.a. If we re-run the regression in 1.a for only the observation included in the IV regression, we find the following estimate of đť›˝inst_qual : #### [Write something about how much this estimate changes or does not change relative to 1.a] 1.e) Fill in the below table. Report all numbers to two decimal places, except report the đť›˝indep_time estimate to 3 decimal places. P-value column isn’t graded, but is useful for determining statistical significance. Coefficient 𝛼 đť›˝inst_qual đť›˝indep_time đť›˝africa đť›˝latitude đť›˝landlock Statistical Significance P-value at 0.05 level? #### [Yes or No] #### [Yes or No] #### [Yes or No] #### [Yes or No] #### [Yes or No] #### [Yes or No] Variable Estimated Value Constant #### Institution Quality #### Years Since Independence #### Africa Dummy #### Latitude (Dist. from Equator) #### Landlocked Dummy #### The confidence interval on đť›˝inst_qual [increases/decreases/stays the same] relative to 1.c. 1.f) The estimate for đť›˝africa becomes: #### and it [is or is not] statistically significant. Based on our results from 1.e, [write about whether institutions or geography seems more important for development and why] Further, our results from 1.e, [write about whether it seems there is something unexplained about the development of countries in Africa]. The reason we need to use a IV approach to conclude this is because [discuss how our results change between 1.e and the regression we ran for 1.f to find đť›˝africa, and how we should interpret these differences]. Problem 2 Answers 2.a) The summary statistics are Summary Statistics Average Value Standard Deviation First Quartile Third Quartile Min Value Max Value Fixed Capital Share Labor Share #### #### #### #### #### #### #### #### #### #### #### #### Intermediate Share #### #### #### #### #### #### Sales/Costs #### #### #### #### #### #### The most important factor of production is [Fixed Capital/Labor/Intermediate Inputs]. 2.b) Our estimated value of 𝜎 is #### and the variables for the establishment with Factory ID 64082 are Factory ID Code 64082 Capital Wedge #### Revenue Wedge #### Productivity (TFPQ) #### TFPR #### 2.c) [Delete the density chart below and replace it with your own for India] 2.d) The value for đť‘Ś/đť‘ŚPotential is: #### The percent gain from eliminating misallocation is: #### 2.e) The potential gains from eliminating misallocation in Table IV of the paper are: Country and Year China in 2005 India in 1994 United States in 1997 Percent Gain from Reducing Misallocation #### #### #### [Give a few examples of misallocation or things that may cause misallocation] [Write about why misallocation might be lower in the United States compared to India]