Todo: Fix this Job Training Meta-Analysis

This article argues that a good paper would be to augment a particular job training meta-analysis.

Ehrenberg and Smith (2016) is a highly successful (in terms of citation count) and current text on labor economics and public policy.

In a section called “Is Public Sector Training a Good Social Investment?” it is effectively demonstrated that the answer is no, but in doing so reference is made to Greenberg (2003).

Greenberg (2003) argues that has a few issues:

  1. No correction for opportunity cost (Ehrenberg and Smith try to resolve that).
  2. Misleading results statement where they take the highest-gaining subsample (women) and overstate their gains (“less than 2000/year”).
  3. Does not calculate ROI, correct for inflation, compare to alternatives
  4. Does not account for crowding out of private job training
  5. Does not compare to private job training outcomes (an alternative policy is to require private job training instead of requiring public job training).
  6. Does not correct for Ashenfelter’s dip, or the outcomes of similar individuals who did not engage in training.
  7. An augmented paper would further explore the apparently negative coefficient on the program cost variable. If it is significantly negative then demonstrates spending is already beyond the bliss point.


Ehrenberg, Ronald G., and Robert S. Smith. Modern labor economics: Theory and public policy. Routledge, 2016.

Greenberg, David H., Charles Michalopoulos, and Philip K. Robins. “A meta-analysis of government-sponsored training programs.” Industrial & Labor Relations Review 57.1 (2003): 31-53.


Leave a Comment