Please answer all questions separately
Imagine a population where there is only one level of ability. The Department of Economics of
University A is asked to estimate the returns to schooling for this specific population. The research team
gathers data from all levels of educational qualifications and estimates the following specification:
log() = + ℎ
Where w is the worker’s wage rate, s is the number of years of schooling and b captures the returns to
schooling. The researchers find that the returns to schooling are equal to = 0.07 .
a. What do the findings suggest about the relation between years of schooling and wages?
Interpret the result. Do the findings agree with economic theory? [30%]
b. University B criticizes the methodology used by University A, claiming that important
determinants are omitted from the main specification. They suggest that the “Mincer equation”
is used instead. Explain what the “Mincer equation” is and why it might be a more appropriate
approach in this framework. [25%]
c. Imagine that the initial population has two levels of ability instead of one; High and Low.
Ability is unobservable. University A uses the Mincer equation to estimate the returns to
schooling. University B claims that the estimate is biased. What kind of biases does estimating
this equation in a population with two different ability levels cause? Is the final estimate likely
to be overestimated or underestimated due to those biases, and why? [25%]
d. University A hires you to find a way to properly estimate the returns to schooling. Suggest a
method you would use to correct the biases, and describe it. [20%]
General instructions about the assignment:
Presentation: The file submitted should be in word or pdf format.
Length: Between 3 and 5 A4 pages (Times New Roman, 11 or similar). Use the layout on this document
as a reference point.
Marks allocation: as described
