1. Labor Quality: Investing in Human Capital
Chapter 4
Labor Quality: Investing in Human Capital
McGraw-Hill/Irwin
Copyright © 2013 by The McGraw-Hill Companies, Inc. All rights reserved.

2. 1. Investment in Human Capital: Concept and Data
12/27/2017
1. Investment in Human Capital: Concept and Data
4-2

3. Investment in Human Capital
12/27/2017
Investment in Human Capital
Expenditures on education and training can be treated as an investment in human capital just like investments in physical capital.
Human capital yields a rate of return (higher earnings) like physical capital.
Educational attainment has been rising in the U.S.
In 1970, 36% of the labor force was a high school dropout. In 2008, it was 12%.
4-3

4. Age-Earnings Profiles, by Education
12/27/2017
Age-Earnings Profiles, by Education
The male age-earnings profiles indicate those with more education have higher earnings.
The age-earnings profiles are steeper for those with more education.
Females have flatter age-earnings profiles.
4-4

5. 12/27/2017
2. Human Capital Model
4-5

6. 12/27/2017
Human Capital Model
The decision should be made by comparing the costs and benefits (higher earnings) of college.
Costs of attending college
The direct costs are the cost of tuition, fees, and books.
Room and board are not included since they are needed regardless of whether you go to college.
The indirect cost is the forgone earnings you give up while you attending college.
4-6

7. Age-Earnings With and Without College
12/27/2017
Age-Earnings With and Without College
The HH curve is the age-earnings profile if a person does not attend college.
The CC curve is the cost-earnings profile if one attends college.
The total cost of attending college is the sum of the direct costs (area 1) plus indirect costs (area 2).
The benefit of attending college is the increase in earnings due to the college degree (area 3).
Whether it is rational to attend college depends on whether the present value of the benefits exceeds the present value of the costs.
Annual Earnings C
Incremental Earnings (3) H
Indirect Costs (2) H 18 22 65
Age C
Direct Costs (1)
4-7

8. 12/27/2017
Present Value
Discounting converts the value of future dollars into today’s dollars through the interest rate.
The present value (Vp) of a payment received one year from now is: Vp =
Def:
Payment 1 year from now
1+Interest rate Vp =
Ex:
where i = 10%
$ 110
1.10 = $
4-8

9. 12/27/2017
Present Value
The present value of a future stream of incremental earnings or costs (E) : Vp = E0 + E1
(1 + i) + E2
(1 + i)2 + En
(1 + i)n +
Costs are represented as negative values of E.
A person should attend college if the net present value (Vp) is greater than 0.
4-9

10. Discounted Present Value
12/27/2017
Discounted Present Value
Present Value of Earnings (4)
Discounted Value (10 Percent Rate) (3)
Incremental
Earnings
(2)
Year (1)
PV of $8,000 Investment in Webmaster Training Program (Interest Rate = 10 Percent)
-$ 8,000
1.000
$ -8,000 1
$ 3,000
0.909
$ 2,727 2
$ 4,000
0.826
$ 3,305 3
$ 5,000
0.751
$ 3,755
$ 1,787
Suppose Melinda is considering taking a webmaster training program that involves direct costs of $3,000 and forgone earnings $5,000. The training program will increase Melinda’s earnings by $3,000, $4,000, and $5,000 for the 3 years she plans on working.
Because she can borrow the funds at an interest rate of 10%, we will discount the future expected income at an 10% rate.
What is the present value (PV) of this training program?
The PV of the training program is positive, Melinda should take the training program.
4-10

11. Internal Rate of Return
12/27/2017
Internal Rate of Return
The internal rate of return, r, is the rate of return at which Vp = 0: Vp = E1
(1 + r) + E2
(1 + r)2 + En
(1 + r)n E0 +
0 = +
A person should attend college if the rate of return (r) exceeds the market interest rate (i).
4-11

12. Generalizations Length of income stream Costs of attending college
12/27/2017
Length of income stream
The longer the stream of positive incremental earnings, the more likely the net present value will be positive.
As a result, younger people are more likely to attend college
Costs of attending college
The lower the cost of attending college, the more likely the net present value is positive.
Older people have a higher opportunity cost of attending college, less likely to attend.
4-12

13. Generalizations Earnings differential
12/27/2017
Earnings differential
The larger the college-high school earnings differential is, the more likely the net present value is greater.
College attendance rose in the 1980s as the college-high school premium increased.
4-13

14. Rate of Return by Country
12/27/2017
Rate of Return by Country
The rate of return per year of college education varies substantially across countries for males.
4-14

15. College-High School Wage Premium
12/27/2017
College-High School Wage Premium
The college-high school wage premium fell in the 1970s for both men and women.
The premium fell because of an increase in supply of college graduates due to the baby boom.
The premium rose after 1979, due to increases in the demand for college-trained workers because of technology improvements.
4-15

16. Caveats
12/27/2017
We can’t predict the college-high school wage premium for future graduates.
The charts report past differentials.
The future differential may be smaller as the high differential may increase future supply.
These are average earnings of college and high school graduates; the distribution of earnings around the mean is wide.
The quality of schooling matters as well as the quantity of schooling.
4-16

17. Quality of education
Dale and Krueger(2001) examined career earnings of individuals between 1976 and 1989 who were 1) accepted by a more selective college but decided to attend a less selective college and 2) who attended a more selective college.
- Controls for the ability problem.
Results indicate it does not pay to attend a more selective college as measured by average SAT score of entering freshmen except for minorities and students with poorly educated parents.
Using data pertaining to the elite public exam schools in the centrally administered Turkish system and a regression discontinuity design, Sinan Sarpca and Ahmet Alkan document that otherwise similar students who get placed in high schools with different levels of peer quality—or desirability—do not differ in their performance in the high-stakes university entrance exam or the resulting placement.

18. Private vs. Social Perspective
12/27/2017
Private vs. Social Perspective
Education yields substantial external or social benefits that society reaps.
More educated workers have lower unemployment rates.
Education raises the amount and quality of participation in the political process.
Children grow up in a better home environment if the parents are more educated.
The research discoveries of more educated persons yield benefits to society.
4-18

19. Private vs. Social Perspective
12/27/2017
Private vs. Social Perspective
The social rate of return is higher (lower) than the private rate of return, resources will be underallocated (overallocated) to human capital investments.
The private and social rate of return are quite similar.
4-19

20. The Effects of Compulsory Schooling Laws on Teenage Marriage and Births in Turkey
Kirdar, Dayioglu and Koc estimate the impact of the extension of compulsory schooling from 5 to 8 years in Turkey on marriage and birth outcomes of teenage women .
The law changes increased women’s schooling by more than a year.
In the summer of 1997, the Ministry of National Education (MONE) increased compulsory education from five to eight years by combining the first two tiers.
Children born in or after January 1987 were affected by the policy. They use Regression Discontinuity design and compare women born months apart who are affected and not affected by the policy.
Increased compulsory schooling years reduce the probability of teenage marriage by age 16 and first-births by age 17 substantially. However, these effects are short-lived; they disappear after age 17 for marriage and after age 18 for first-births because the policy increases the marriage hazard rate at age 17, shortly after these women leave school, and there is no policy effect on the time to first-birth after marriage.

21. 12/27/2017
Questions for Thought
1. Suppose the net present value of an educational investment is highly positive. What can you infer about the investment’s internal rate of return relative to interest cost of borrowing?
2. Comment on the following statements:
(a) Older workers are less geographically mobile than younger workers.
(b) An economic recession tends to stimulate college enrollments.
(c) The age-earnings profiles shown earlier, clearly indicate that people with more education earn more than people with less education; therefore, personal spending on education is always a good investment.
4-21

22. 3. Human Capital Investment and the Distribution of Earnings
12/27/2017
3. Human Capital Investment and the Distribution of Earnings
4-22

23. Diminishing Rate of Return
12/27/2017
Diminishing Rate of Return
The marginal rate of return to education declines as additional schooling is acquired.
Rate of Return
Investment in education is subject to the law of diminishing returns. The increases in knowledge decline with each additional year of schooling. r
The return also falls because the explicit cost and opportunity cost of education rises with additional schooling.
Years of Schooling
4-23

24. Demand for Human Capital
12/27/2017
Demand for Human Capital
Since individuals should increase schooling so that the marginal rate of return of schooling (r) is equal to the interest rate (i).
r, i 1
Using the r = i rule, at interest rate i2, the optimal level of schooling is e2. i1 S1 2 i2 S2
At i1 the optimal level is e1. 3 i3 S3
At i3 the optimal level is e3. r
, DHC
Each equilibrium point (1,2,3) indicates the price” and quantity demanded of human capital. In other words, the demand for human capital. e1 e2 e3
Years of Schooling
4-24

25. 12/27/2017
Ability Differences
Alfonse is low-ability person. He has low mental/physical talents and/or low motivation and self-discipline. His demand for schooling is DA.
r, i A B
Bob is a high-ability person. He has a greater demand for schooling at DB because he can better translate schooling into higher productivity and earnings. i S DB DA
For a given interest rate, Bob will obtain more schooling which will compound the earnings differential between low and high ability persons. eA eB
Years of Schooling
4-25

26. 12/27/2017
Discrimination
Albert is black and is discriminated against in the labor market. His demand for schooling is DA since he has low ability to convert additional schooling into higher earnings.
r, i A B i S
Brett is white and has a greater demand for schooling at DB as he can reap the benefits of additional schooling. DB DA
For a given interest rate, Bob will obtain more schooling which will compound the earnings differential between whites and blacks. eA eB
Years of Schooling
4-26

27. 12/27/2017
Cost of Funds
r, i
Ann is from a wealthy family and faces a low cost of borrowing funds (iA). Her optimal level of schooling is eA. B iB SB A iA SA
Betty is from a poor family and faces a high cost of borrowing funds (iB). Her optimal level of schooling is eB.
DA= DB eB eA
Years of Schooling
4-27

28. Interactions
12/27/2017
The ability, discrimination, and cost of funds factors that affect schooling levels may interact to cause even larger earnings inequality.
If a person faces labor market discrimination, lenders may charge a higher interest rate since they are less certain of getting repaid.
Discrimination will reduce both the supply and demand for schooling.
Anti-discrimination policies may reduce earnings inequality as a result.
4-28

29. Capital Market Imperfections
12/27/2017
The capital market is biased in favor of physical rather than human capital.
Lenders can’t repossess human capital.
Young people, who are most likely to invest in human capital, don’t have established credit ratings.
The government may have to intervene by subsidizing human capital loans in order to make the returns on physical and human capital equal.
4-29

30. 12/27/2017
Question for Thought
1. Describe the expected effects that college scholarships based on (a) student ability and (b) student need are likely to have on the distribution of earnings.
4-30

31. 12/27/2017
4. On-the-Job Training
4-31

32. 12/27/2017
Costs and Benefits
Firms will invest in on-the-job training if the present value of the benefits of the training exceeds the present value of the costs.
The costs to the firm include:
Direct costs such as classroom instruction and greater worker supervision.
Indirect costs such as reduced worker output during training.
The benefit is greater worker productivity.
4-32

33. General and Specific Training
12/27/2017
General and Specific Training
General training is training that is usable at all firms and industries.
Word processing skills or accounting skills
Specific training is training that is usable at only at the firm that provides the training.
Assembly procedure unique to a firm’s product
Most training is a mixture of general and specific training.
4-33

34. General Training Wage & Marginal Revenue Product Wp= MRPp Wu= MRPu
12/27/2017
Wu and MRPu are the wages and marginal revenue product for an untrained worker. Marginal revenue product is the increase in total revenue associated with the employment of an additional worker.
Wp= MRPp
Wage & Marginal Revenue Product
Since general training is usable at other firms, workers must pay for the entire cost of the training. They receive a lower wage (Wu> Wt) that is equal to their diminished productivity (MRPt).
Wu= MRPu
Wt= MRPt
After the training period, workers receive a higher wage Wp that is equal to their new higher level of productivity MRPp.
Post Training
Training
4-34

35. 12/27/2017
Specific Training
Since specific training is not transferable to other firms, the employer must pay for the training.
During training, the employer pays a wage greater than the worker’s productivity (Wu> MRPt).
Wage & Marginal Revenue Product
MRPp
After training, the employer gets a return on her training investment by paying a wage less than the worker’s productivity (Wu < MRPt).
Wp’
Wu= MRPu
MRPt
The employer may pay a higher wage to decrease worker turnover and thus protect her training investment (Wp’).
Post Training
Training
4-35

36. Modifications
12/27/2017
Faced with a minimum wage, some firms may pay for general training.
The firms recoup their expenses by paying workers less than their MRP after the training is completed.
This is possible because workers are not perfectly mobile across jobs—there are costs to switching jobs.
Workers with the most formal education also receive more on-the-job training.
They have shown they can receive training more readily and thus less cost.
4-36

37. Empirical Evidence
Union workers are more likely to receive training than nonunion workers. (Dustmann, Schonberg 2009)
Positive productivity effects have been found for general training but not specific training.(Barret, O’Connel, 2001)
Accumulation of OJT raises a worker’s real wages. Each year of training with the current employer, among younger workers raises earnings by 40 percent. (Parent, 1999)
A 1 percent increase in training is associated with an increase in productivity of about 0.6 percent and an increase in hourly wages of about 0.3 percent. (Dearden et al. 2006)
Formal training lengthens employment durations. (Grossberg, 2000)
Gender difference in the amount of OJT play only a small role in the gender wage gap. (Sicilian, Grossberg 2001)

38. 12/27/2017
Question for Thought
1. Suppose that after graduation you take a job with an employer that offers to pay full tuition for employees wishing to return to school to get an MBA degree during non-work hours. You are not required to continue working for the firm after getting your MBA. What type of training is this, and who you think actually pays for it?
4-38

39. 5. Criticisms of Human Capital Theory
12/27/2017
5. Criticisms of Human Capital Theory
4-39

40. Investment or Consumption?
12/27/2017
Investment or Consumption?
Not all education expenditures are an investment because some part is a consumption expenditure.
Courses such as music appreciation yield consumption benefits rather than investment benefits.
By ignoring the consumption benefits of education, researchers overstate the investment costs of education and understate the rate of return.
4-40

41. Non-Wage Benefits
12/27/2017
Studies that only examine the earnings of high school and college graduates understate the rate of return for two reasons.
College graduates have greater fringe benefits as a percent of pay than high school graduates.
College graduates tend to work in more pleasant surroundings and have more interesting jobs than high school graduates.
4-41

42. Ability Problem
12/27/2017
Those with more ability (i.e., intelligence, motivation, and self-discipline) are more likely to go to college.
Even without a college degree, they would have earned more than those who decided not to go to college.
To the extent that the higher earnings of college graduates reflects their greater ability rather than schooling, the rate of return is overstated.
Omission of ability biases the rate of return estimates by a small amount.
4-42

43. Econometrics Recap When is there a causal interpretation?
Conditional mean independence assumption
Example: wage equation
The explanatory variable must not
contain information about the mean
of the unobserved factors
e.g. intelligence …
The conditional mean independence assumption is unlikely to hold because
individuals with more education will also be more intelligent on average.

44. Econometrics Recap
Standard assumptions for the linear regression model
Assumption SLR.1 (Linear in parameters)
Assumption SLR.2 (Random sampling)
In the population, the relationship
between y and x is linear
The data is a random sample
drawn from the population
Each data point therefore follows
the population equation

45. The Simple RegresEconometrics Recap sion Model
Assumptions for the linear regression model (cont.)
Assumption SLR.3 (Sample variation in explanatory variable)
Assumption SLR.4 (Zero conditional mean)
The values of the explanatory variables are not all
the same (otherwise it would be impossible to stu-
dy how different values of the explanatory variable
lead to different values of the dependent variable)
The value of the explanatory variable must contain no information about the mean of the unobserved factors

46. Econometrics Recap Theorem 2.1 (Unbiasedness of OLS)
Interpretation of unbiasedness
The estimated coefficients may be smaller or larger, depending on the sample that is the result of a random draw
However, on average, they will be equal to the values that charac-terize the true relationship between y and x in the population
„On average“ means if sampling was repeated, i.e. if drawing the random sample und doing the estimation was repeated many times
In a given sample, estimates may differ considerably from true values

47. Screening Hypothesis
12/27/2017
The screening hypothesis argues that schooling increases earnings not by increasing productivity but providing a way to identify high-quality workers.
Screening does not affect the private rate of return, since college graduates still have higher earnings.
The social rate of return is overstated, as screening does not increase productivity.
The empirical evidence shows screening appears to play a small role.
Same amount of education between salaried vs. self employed workers
4-47