Human capital v sorting

1
Human capital v sorting

• Remember
• Why do individuals invest in education?
• To acquire knowledge and skills that tend to
  raise the productivity and therefore earnings
  potential
• known as the human capital perspective.
• However, an alternative view to human capital
  theory
• The screening or signalling perspective
• suggests that schooling is a device that allows
  employers to identify individual ability.

2
Human capital v sorting
Weiss (1195) uses the term sorting
device. Signalling models have the more informed
workers moving first, where in screening models
the first move is made by the firms. Students
choose the length of schooling to signal their
ability to employers. Employers demand a minimum
level of schooling from their applicants to
screen their workers. Both signalling and
screening serve to sort workers according to
their unobserved characteristics.
3
Human capital v sorting
For the individual decision, the exact mechanism
of earnings enhancement, production of human
capital or sorting, is indifferent. Either way,
individuals will be induced to invest in
education based on the comparison between returns
and costs. The identification of the model in
place is hard to achieve using labour market
data. How have people tested it?
4
Human capital v sorting

• Estimates of effects of specific courses on
  earnings holding constant the number or years of
  schooling.
• Main message from studies on course work, test
  scores and grades is that learning in high school
  does not seem to be a significant factor in
  explaining the correlation between secondary
  schooling and wages.
• Effect of birthdate on returns to schooling
• Use of twins studies
• Riley includes only observable variables in wage
  equations, and finds that wage equations fit
  better for people in occupations in which
  screening is important. Returns to education are
  higher for unscreened occupations.

5
Human capital v sorting

• Evidence about job tenure and wages. Senior
  workers seem to suffer the most with layoffs or
  plant closures.
• Evidence on wages in jobs with probationary
  periods.
• Interesting application
• Brown, Sarah and John Sessions, 2006, Evidence on
  the relationship between firm-based screening and
  the returns to education, Economics of Education
  Review, 25, 498-509

6
Human capital v sorting
By combining information on firms that use
personality/attitude or performance/competency
tests during the hiring process with data on
employee level characteristics, They are able to
explore how the use of hiring tests by employers
impacts on the signalling role of education. To
summarise, their analysis suggests that
educational screening plays an important role in
the labour market, especially where firms do not
conduct their own hiring tests. Such tests are
found to raise an individuals earnings ceteris
paribus, and to reduce the impact of educational
certificates on such earnings. This would suggest
that neglecting the distinction between direct
and indirect screening would bias any attempt to
estimate the returns to education. It is
possible that the allocation of workers to a
testing or non-testing firms is not random,
and also that the factors affecting the
probability of assignment may also impact on an
individuals wage within a particular sector. For
example, it may be the case that high ability
individuals are attracted to firms conducting
hiring tests.
7
Human capital v sorting
Table 6 reports selectivity-corrected earnings
equations for individuals employed by firms that
do and do not conduct performance/competency and
attitude/personality tests for the individuals
particular occupation grouping.
8
Human capital v sorting
Ignoring the selectivity issue would imply a
positive (negative) bias in the estimated
earnings of individuals employed within firms
that (do not) perform tests. The magnitude of the
estimated coefficients on all of the educational
variables is greater for the indirectly screened
samplei.e. those individuals employed by firms
who do not conduct performance/competency tests
at the hiring process suggesting that
educational certificates may play a more
pronounced screening role in this particular
case. It is evident that the use of an
attitude/personality test is associated with a
relatively lower return to education for all
education levels whereas the use of a
performance/competency test is only associated
with lower returns to education for university
level qualifications. This would seem to be
consistent with a screening model where education
is signalling motivation and personality rather
than particular cognitive skills.
9
Human capital v sorting
Why do people object to sorting models? Sorting
is grouped with credentialism, i.e., wage
differences are independent of productivity (or
education has no effect on productivity) Sorting
models are Pareto inefficient People want to know
the social rate of return to education Sorting
may mean that the impact of education is
overstated, therefore making a weaker case for
spending in education. However there is a strong
case for the importance of primary education
education surely improves productivity at certain
jobs and there is more to education than labour
market outcomes.