Empirical Evidence

1
Topic 3 Part V
Empirical Evidence
2
Education and Earnings Empirical Evidence

• Benjamin, Gunderson and Ridell use data from
  Statistics Canada (Individual Public Use
  Microdata Files, 1996 Census of Population) to
  evaluate the relationship between education and
  lifetime earnings
• Benjamin, Gunderson and Riddell, Labour Market
  Economics, 2002

3
Education and Earnings Empirical Evidence

• They found a strong relationship between
  education and lifetime earnings
• The income streams (or age-earnings profiles) of
  those with more education lie above the streams
  of those with less education

4
Human Capital Model Empirical Evidence

• Vaillancourt and Bourdeau-Primeau use 1990 and
  1995 Canada data on earnings and education to
  calculate the rate of return on investment in
  education
• Vaillancourt and Bourdeau-Primeau, The Returns
  to University Education in Canada 1990 and
  1995. In Renovating the Ivory Tower Canadian
  Universities and the Knowledge Economy, D.
  Laidler ed., Policy Study 37, C. D. Howe
  Institute, 2001

5
Human Capital Model Empirical Evidence

• Estimates of the rate of return to education are
  obtained by comparing different individuals (with
  different levels of education) at a point in
  time, rather than following the same individuals
  overtime
• After controlling for other observed factors that
  influence earnings (using multivariate regression
  analysis), they estimate the rates of return to
  the individual

6
Human Capital Model Empirical Evidence

• This is accomplished by estimating the human
  capital earnings function. In its simplest form,
  this function is a least-squares regression of
  earnings on education, with controls for other
  factors believed to affect earnings (like work
  experience)
• The coefficient on years of education represents
  the return to schooling

7
Human Capital Model Empirical Evidence

• Their findings (for the 1995 data)
• The rates of return are highest for the
  bachelors level (as would be expected if there
  is diminishing returns to education)
• There are large differences in rates of return
  for the bachelors level by type of education
  obtained

8
Human Capital Model Empirical Evidence

• Those obtaining degrees in health sciences (as
  medicine, dentistry) earn the highest returns
• Those graduating in humanities earn the lowest
  returns
• Commerce and economics degrees have lower rates
  than health sciences but much higher than
  humanities

9
Human Capital Model Empirical Evidence

• The interpretation of the coefficient on
  education as the return to investment on
  education due to human capital enhancement is
  challenged by the signaling approach
• If more educated people are also more able people
  (innate ability), the coefficient on education
  will represent the returns not only to higher
  productivity due to education but also the
  returns to higher innate ability (signaled
  through education)

10
Education as Signaling Empirical Evidence

• Empirical tests of the signaling hypothesis have
  not been conclusive
• Riley, Testing the Educational Screening
  Hypothesis, Journal of Political Economy, 1979
  Lang and Kropp, Human Capital Versus Sorting
  The Effects of Compulsory Attendance Laws,
  Quarterly Journal of Economics,1986

11
Education as Signaling Empirical Evidence

• Most would agree that the pure signaling model in
  which education has no impact on productivity
  does not appear capable of explaining observed
  behavior
• (Rosen, Human Capital A Survey of Empirical
  Research. In, Research in Labor Economics, ed.
  Ehrenberg, 1977 Weiss, Human Capital versus
  Signaling Explanations of Wages, Journal of
  Economic Perspectives, 1995)

12
Education as Signaling Empirical Evidence

• Professional programs such those in medicine, law
  and engineering clearly are more than elaborate
  signaling devices

13
Education as Signaling Empirical Evidence

• However, as Weiss emphasizes, there is a
  considerable body of evidence that suggests that
  education acts as a filter to some degree
• Weiss, Human Capital versus Signaling
  Explanations of Wages, Journal of Economic
  Perspectives, 1995

14
Education as Signaling Empirical Evidence

• Altonjis study support the signaling hypothesis
• Altonji, The Effects of High School Curriculum
  on Education and Labor Market Outcomes, Journal
  of Human Resources, 1995
• Most of the empirical evidence on estimated
  returns to schooling focuses on the return to a
  year of education, without any regard for the
  actual courses taken over that year

15
Education as Signaling Empirical Evidence

• His study provides one of the first estimates of
  the effect of high school course selection on
  future labor market outcomes
• Using the National Longitudinal Survey for the
  US, he observes the wages in 1985 of a large
  sample of high school graduates from the class of
  1972

16
Education as Signaling Empirical Evidence

• Controlling for a variety of family background
  variables, as well as characteristics of the high
  schools, Altonji finds almost no economic payoff
  for students who took more academically oriented
  high school courses (like sciences, math)
• This evidence is more consistent with the
  signaling view of education than the pure human
  capital model. Why?

17
Education as Signaling Empirical Evidence

• Remember that the premise of human capital theory
  is that it is the purely productivity-enhancing
  features of education that employers are paying
  for
• If this is the case, course content should matter
  more than just the number of years of education

18
Education as Signaling Empirical Evidence

• On the other hand, the signaling model emphasizes
  the fact that employers infer innate
  characteristics from the level of education (such
  as perseverance and good work habits related to
  high productivity)
• In that case, employers will care more about
  whether the person finished the high school than
  whether or not she took this or that course

19
Final Remarks

• The fact is that schooling investments offer
  individual monetary rates of return that are
  comparable to those received from other forms of
  investment
• For individuals to recoup their human capital
  investment costs requires willingness on the part
  of employers to pay higher wages to people with
  more schooling
• For employers to be willing to do this, schools
  must be providing a service that they could not
  perform more cheaply themselves

20
Final Remarks

• The fact that employers continue to emphasize
  (and pay for) educational requirements in the
  establishment of hiring standards suggests one of
  two things
• Either more education does enhance worker
  productivity. Or,
• It is a less expensive screening tool than any
  other that firms could use to detect the
  unobserved productivity type of applicants

21
Final Remarks

• In either case, the fact that employers are
  willing to pay a high price for an educated
  workforce seems to suggest that education
  produces both private and social benefits