Making learning on the job a reality for low-skill workers

The problem: getting companies to invest in the skills of their young, poorly educated workers. It’s a problem because such workers tend to be highly mobile from job to job. If you’re an employer, teaching them how to make your particular widgets (what the economists call job-specific human capital investments) will pay off; if they know how to make your widgets, they’re more productive working for you than for someone else, which means that their value to you just went up but their value on the outside job market didn’t. On the other hand, teaching them either generalizable skills, or useful behaviors such as showing up for work and not calling their foreman a dickhead, has no next to no payoff to the employer; yes, it makes the worker more valuable, but it makes him more valuable to the employer next door, so you either lose the worker or have to pay him a higher wage.

In a perfect world, employees would invest in their own non-job-specific human capital, but (1) they may have only a vague idea about how much a given job contributes to their human capital stock; (2) they can’t easily borrow against future earnings, so financing such improvements (by taking lower wages for jobs that give them better future prospects) may require forgoing more present consumption than they’re willing to forgo); and (3) they may be irrationally present-oriented, just like the rest of us.

The long-term apprentice contract in a guild system can be seen as a partial solution to this problem; the master craftsman whose apprentice was bound to him for seven years could “internalize” a good chunk of the benefit of what the apprentice learned. (Slavery and indentured servitude could also be seen as solutions, though of course slavery gave the “worker” very little incentive to learn and indentured servitude gave the master too much incentive to starve and work his servant to death.)

Okay, but we’re not doing any of those things any more — thank Heaven — so what can we put in their place? The problem is still a very significant one.

Candidate solution: direct financial incentives. Announce that the first employer of some defined group of workers with poor prospects (as defined, for example, by census tract of birth, to avoid any perverse incentive effects) would (as long as the worker stayed on the job some defined amount of time, say at least a year) receive a small percentage of that worker’s Social Security contributions over the next ten years. Administratively, it would be a piece of cake; once the eligible workers were identified, tracking their future Social Security contributions would be trivial.

It’s not clear how employers would respond to this in the real world. One can imagine some firms springing up that would in effect specialize in providing on-the-job training. (Even if some firms figured out ways of gaming the system by picking out eligibles who in fact had good prospects, that screening activity alone might be worth paying for in a world where the immediate financial rewards of making a better-than-average record at your rotten high school are close to nil.) On the other hand, previous incentives for hiring “at-risk” kids turned out to have the perverse effect of labeling eligible workers as likely losers.

Still, in principle, it’s not hard to see the market failure here, and the direction of the fix seems obvious. Worth trying out?


William Sjostrom at AtlanticBlog points out two problems and offers two solutions. Problem #1: Getting fired can be a highly educational experience. By penalizing an employer for firing a new employee, my plan would tend to reduce the number of employees who learn valuable lessons in that way. Problem #2: the next employer, on my scheme, has no incentive to continue the process. Right and right. Solution #1: Make the payoff a continuous function of length of employment, and provide a cut to second and subsequent employers. Solution #2: Make the whole subsidy arrangement more cost-effective by increasing the percentage of the marginal dollar that goes back to previous employers but starting it at some floor. Again, right and right. Both the problems and the solutions should have been obvious to me, but weren’t. Thanks for Prof. Sjostrom, and hurrah for blogging!

Henry Farrell suggests that regulation can substitute for subsidy, and claims Germany as an example. I have heard that the German system depends in part on a level of informal social control that would be hard to reproduce here: the managers of German firms that slack off on their apprenticeship-training efforts are felt by their peers to be acting badly, with consequences for their personal social standing. Mr. Farrell also suggests that providing a better social safety net might extend the planning horizons of young, low-skill workers, and thus facilitate their own human-capital investment. That’s an intriguing thought, and it’s one more reason to be sorry (alongside some reasons not to be sorry) that the U.S. seems committed to a somewhat ragged social safety net.

Author: Mark Kleiman

Professor of Public Policy at the NYU Marron Institute for Urban Management and editor of the Journal of Drug Policy Analysis. Teaches about the methods of policy analysis about drug abuse control and crime control policy, working out the implications of two principles: that swift and certain sanctions don't have to be severe to be effective, and that well-designed threats usually don't have to be carried out. Books: Drugs and Drug Policy: What Everyone Needs to Know (with Jonathan Caulkins and Angela Hawken) When Brute Force Fails: How to Have Less Crime and Less Punishment (Princeton, 2009; named one of the "books of the year" by The Economist Against Excess: Drug Policy for Results (Basic, 1993) Marijuana: Costs of Abuse, Costs of Control (Greenwood, 1989) UCLA Homepage Curriculum Vitae Contact: