Making on-the-job training pay

It is well established that the best “job training” takes place on the job. But it is also well known that young, low-skilled workers are highly mobile from firm to firm, giving their current employers next to no incentive to contribute to those workers’ non-firm-specific human capital (which is economese for teaching them anything that would increase their value to their next emploers).

Why not figure out a way to estimate the human capital of each worker entering the workforce or re-entering it after some moderately long period (say, six months) and then give the employer a financial stake in improving that stock by paying the first employer a share of the worker’s payroll taxes over the next several years (maybe a declining share over time) no matter where that employee was then working? This would encourage hiring entering or re-entering workers, encourage giving them jobs in which they learn useful skills, and encourage employers to take the process of giving job references for current and former employees seriously.

Obvious problems:

1. It costs money. But given the public stake (thorough taxes and transfers) in increasing the human capital of low-skilled workers, and especially young ones, a well-designed program ought to be a net benefit even looking at the public fisc alone.

2. The market will take care of it; workers will accept lower wages for firms that invest in their non-firm-specific human capital, so there’s no market failure to fix. That assumes more knowledge and foresight than many young workers may have, and assumes away all the rigidities in the labor market that prevent such bargains from being struck.

3. It will disadvantage existing labor-force participants and those with high measured human capital compared to new entrants and re-entrants with low measured human capital. True enough, though the size of the effect is open to question. This problem could be eased some by phasing out payments gradually rather than sharply as previous labor force participation and measured human capital rise.

4. (The flip side of #2): In the past, subsidy programs for hard-to-place workers have proven to do more harm than good because program eligibility became a stigmatizing factor. Less of a problem if the program applies to lots of workers.

5. Many employers may ignore the payments in making management decisions, thus wasting the entire subsidy paid to them.

6. Whatever the system to measure previous labor force participation and human capital, some companies will try to game the system by taking workers whose actual job prospects are better than they look.

7. Another form of gaming would involve maintaining heavy turnover while providing next to no training, in hopes of collecting the subsidy from those workers who, on their own, later did well in the job market. A minimum tenure requirement at the first employer in order to trigger the payments, and a formula that considered the performance of all eligible employees, with deductions for companies that produced large numbers of “duds,” would help avoid that at the cost of increased complexity and perhaps some addition incentive to “cherry-pick.”

8. By reducing the cost to the firm of employee turnover, the program would diminish firms’ incentives to treat low-skilled new labor market entrants well once they were hired. Again, the subsidy formula might be jiggered to reduce that effect.

I don’t assert that any given subsidy program would be the optimal one. This idea needs a significant investment in design and testing.

What almost can’t be right is to continue to leave the incentive for firms to invest in the non-firm-specific human capital of low-capital new or returning labor force participants at zero.

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: