Scribd
Upload
16 views2 pages

Ps 4

Uploaded by

Israel Martínez
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
16 views2 pages

Ps 4

Uploaded by

Israel Martínez
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

ECONOMICS OF EDUCATION

Bachelor in Economics
Problem Set 4: Topic 3 - The Return to Education

Table 1: Earnings Equations.

1980-81 1980-81 1990-91 1990-91


Constant 12,232 12,258 13,178 13,185
(432) (436) (365) (366)
Primary Education 0,310 0,300 0,286 0,283
(30) (30) (16) (16)
Primary Education (EGB) 0,580 0,570 0,408 0,405
(34) (34) (19) (19)
Upper Secondary Education 0,798 0,788 0,664 0,661
(41) (40) (26) (26)
Vocational Education 0,749 0,737 0,630 0,628
(24) (24) (22) (22)
2-Year College Degree 0,940 0,928 0,858 0,858
(42) (42) (31) (31)
Bachelor Degree 1,196 1,188 1,075 1,075
(51) (51) (37) (37)
Experience 0,042 0,041 0,037 0,036
(25) (25) (17) (17)
Experience2 -0,0006 -0,0006 -0,0004 -0,0004
(26) (26) (14) (14)
Part-Time Job -0,437 -0,293
(17) (6,9)
2
R 0,27 0,28 0,15 0,15
F 725 688 264 241
o
N of observations 15877 15877 12105 12105

Source: Encuestas de Presupuestos Familiares. Dependent Variable: Logarithm of labor earnings. OLS estimation
and t-statistics in parenthesis. San Segundo (1996).

1. Provide an explanation for the estimated earnings equations in Table 1. Does the fact that the
estimated coe¢ cient of the variable 2-Year College Degree in 1980-1981 (0,928) being higher
than the same coe¢ cient in 1990-1991 (0,858) mean that the estimated return to 2-Year College
Education (with respect to Vocational Education) is lower in 1980-1981 than in 1990-91?

2. Compute the private return to College Education using the estimated parameters of Table 1.
Explain the results obtained. Is there any evidence of overeducation?

3. Using the estimated parameters of Table 1, compute the economic return to an additional
year of labor experience for workers with experience=5, 10, and 20 years. Explain the results
obtained. Which is the level of labor experience that maximizes earnings?

4. Explain which type of biases may a¤ect estimators for the return to College Education com-
puted in Question 2 and provide alternative solutions for these biases.

5. Assume that each individual in the economy decides her optimal demand for education as in
Exercise 5 of Problem Set 3. Each individual lives T years and she chooses her demand for
consumption goods C and education E to maximize her utility subject to her budget and time

1
constraints. There is no uncentainty in the economy. Individual’s utility function is U = ln(C).
The unit price of the consumption good is p. The individual may devote her T years of life
either to invest in education E or to work (T E). Each year devoted to study has a cost
m h, where m is the monetary cost of education (tuition fees, transportation, housing, etc).
The cost of education is reduced by the innate ability of the individual h. For each year the
individual devotes to work she earns an annual wage wt equal to her marginal productivity,
wt = + E + 21 h, where is the minimum labor wage, which is augmented by the other
two factors a¤ecting labor productivity: acquired education E and innate ability h. is the
return to education. Apart from labor earnings, each individual is endowed with an initial
wealth R. It is known that for each individual in the economy = 1 = 20, T = 50, and
p = 1. However, there is heterogeneity across individuals in their innate ability h and in the
monetary cost of education m because individuals are located at di¤erent distances of schools.
Individuals may have high-ability h = 16 or low-ability h = 4. Education costs may be high
m = 14 if the individual is located far way from the school or low m = 8 if the individual
is close to the school. Both individual’s characteristics are independently distributed in the
population according to the following probabilities:
h P r(h) m P r(m)
16 0:4 14 0:5
4 0:6 8 0:5

(a) Compute the optimal demand for education E and the wage wt earned by each individual
in the economy. (Notice that you may use the generic optimal demand for education you
obtained in Part b) of Exercise 5 of Problem Set 3).
(b) Given this population and individuals’demand for education and wages de…ned in Part
5a, suppose that a researcher aims at estimating the “average return to education”, that
is, the parameter in the wage equation w = + E + 21 h. The researcher observes, for
the whole population, w, E y m, but she does not observe h.
1. Compute the estimated value for obtained by the researcher if she estimates the
wage equation using OLS with the observed data. That is, the researcher estimates
using OLS the equation w = + E + " where " is the error term due to unobservable
variables. Provide an explanation for the di¤erences observed between the estimated
value for and the true value of this parameter.
2. Compute other estimator which allows the researcher to obtain the true value of :
using observable data.
3. What if the “returns to education” were heterogeneous and we use the estimator
obtained in Part 5(b)2 to estimate the average return to education?

You might also like