Bryan Caplan  

The Unfair Labor Training Act

Does Market Power Affect Optim... The Rule of Law in the Regulat...
John Bishop's "What We Know About Employer-Provided Training" alerted me to the existence of some obscure but plausibly very harmful labor regulations.  Suppose your employer provides free job training every Saturday.  Compared to regular schooling, this is a great deal.  The firm gives you a "full scholarship"; you pay only the opportunity cost of your time.  But this great deal is normally forbidden.  Bishop:
Labor Department regulations implementing the Fair Labor Standards Act make it very difficult for firms to ask workers to contribute towards the cost of their training by taking training classes during unpaid time. They specify that:
Sec. 785.27 --- Attendance at ... training programs ... need not be counted as working hours if the following four conditions are met:
a) Attendance is outside the employee's regular working hours;
b) Attendance is in fact voluntary;
c) The course, lecture or meeting is not directly related to the employee's job; and
d) The employee does not perform any productive work during such attendance....

Sec. 785.29 --- The training is directly related to the employee's job if it is designed to make the employee handle his job more effectively as distinguished from training him for another job, or to a new or additional skill.

Sec. 785.30--...if an employee on his own initiative attends an independent school, college or trade school after hours, the time is not hours worked for his employers even if the courses are related to his job. (Bureau of National Affairs, Wages and Hours, p. 97:3208).
The piece is from 1996, but as far as I can tell these regulations remain the law of the land.  Bishop's analysis:
These regulations present employers with the following dilemma: either (a) don't provide training in general skills like reading or word processing that will improve a worker's productivity in their current job or (b) provide such training and pay all of its costs-instructional costs and trainee time costs. Thus, no matter how general the skill nor how voluntary the decision to take training, if it raises productivity in one's current job and is provided by the employer, the worker must be paid while engaged in training. Workers and employers are prohibited from cutting the following deal--the company will provide instructors, classrooms and certification, while the worker will commit uncompensated time to learning general skills that enhance productivity (Bureau of National Affairs 1993, 97:3208). Schools can offer such a deal, employers cannot.These regulations have three pernicious effects: they discourage employers from organizing formal training in general skills, they induce them to ration the training programs that they do offer and they force workers who seek such training to get it at a school. Since school based training of adults not paid for by their employer has no effect on wage rates, the regulations effectively push many workers into a type of training that does not appear to raise their wages. This constraint on how workers and employers share the costs of general training provided at the workplace is probably a very important source of market failure in the training market.
If Bishop's right, this would of course be a prime case not of "market failure" but of government failure: Markets are perfectly able to train workers once government gets out of the way.  But why can't markets circumnavigate the law by hiring workers-in-search-of-training at lower hourly wages?
The other way workers can share the costs of general training is by accepting lower wage rates during the training period. Under current regulations this is not really possible for short intermittent spells of training voluntarily undertaken by individual workers. For new hires, however, flexibility is greater because wages customarily rise with tenure on the job. In some low wage jobs, however, the minimum wage constraint is binding. Young inexperienced workers are, in effect prevented from bidding for training opportunities by offering to work for a low wage.
The minimum wage aside, another snag that Bishop doesn't consider is that most low-skilled workers are too short-sighted, impatient, or distrustful to accept lower pay in exchange for training.  As a result, these regulations may, like a banana subsidy, do great harm that would be child's play for thoroughly rational agents to avoid.

COMMENTS (10 to date)
david condon writes:

Tattoo parlors still use the older apprenticeship model Bishop proposed. I can't think of any other exceptions. Tattooing has the benefit that skills are very easy to demonstrate with a portfolio. In most other industries, employees need a sheet of paper to say they're qualified. Most employers aren't large enough to emit a strong signal in the job market so when the employees try to get a new job; their training becomes worthless in the market. Training without signaling doesn't sound like a good deal to me.

Damian writes:

My sister worked at a kitchen here in Berkeley earning $1 per hour in house currency. She also got a nice meal and cooking training. Their lawyers had them revamp the program so that they would not be liable for back wages (min wage violations). The result: a formal apprentice program that costs a couple thousand but has more formal/intense training, potentially eliminating those like my sister unable to pay. They also eliminated their volunteer positions for folks interested in simply earning the $1.

I think the minimum wage is a pernicious beast and the inability to gain experience while earning a lower wage is quite costly to certain groups who could really use that training.

RPLong writes:

Prof. Caplan, isn't this only government failure if we assume that the human capital model of education is correct, at least for "skills that would raise productivity?"

If the signalling model of education is correct, then one way the regulations could be circumnavigated is if the employer offers employees a voluntary sign-up sheet for a non-existent class and then gives a raise to all those who sign up.

Another way to do it is to offer raises and promotions only to those employees who put in a lot of voluntary overtime, which is exactly how it has worked in all the companies for which I've worked.

ThomasH writes:

More evidence that writers and commentators on Econlog get so upset over the harm minimum wage laws do to low-income workers that they cannot even consider alternatives to the minimum wage that would not harm low income workers or worry about any other distortion in the whole economy that might harm low income workers (the 15% wage tax, for example).

This seems to be an argument for eliminating the minimum wage to avoid the demonstrated harm but without considering the unintended consequences, the same kind of policy advocacy error that "Liberals" are accused of.

Don Boudreaux writes:

ThomasH: I do not understand your comment. The argument against the minimum wage is that it makes many low-skilled workers worse off than they would otherwise be.

Pointing this fact out does not require that the person who points it out also offer an alternative to the minimum wage that better 'solves' the problem that minimum-wage proponents believe mistakenly is solved by the minimum wage. If this person moves on to offer other policy alternatives, that's fine. But a physician who advises a smoker who wishes to avoid lung cancer to stop smoking cigarettes does a worthwhile service to that patient even if the physician does not then go on to also prescribe a healthy diet and exercise regimen for that patient. That the latter sort of advice might be excellent does not render the advice to stop smoking, standing alone, a mistake or unworthwhiile

Because many people wrongly believe that the minimum wage is an effective way to improve the economic lot of poor people, pointing out the many reasons why this widely held belief is mistaken is itself valuable.

As for your last paragraph, you there unintentionally sneak in the assumption that the minimum wage in fact might on net be a good policy for poor workers. You do this by asserting the existence of alleged harmful unintended consequences of eliminating (or reducing) the minimum wage. This move on your part unintentionally presumes that the argument against the minimum wage is itself mistaken. Only if the minimum wage is itself a worthwhile policy for poor workers is there reason to worry that eliminating or reducing the minimum wage will unleash harmful unintended consequences. But if those of us who argue against the minimum wage are correct - a proposition that you seem to accept in your first paragraph - then there are no net harmful unintended consequences that will arise if the minimum wage is reduced or, better, eliminated.

R Richard Schweitzer writes:


"The piece is from 1996, but as far as I can tell these regulations remain the law of the land."
emphasis added.

Those words capture this entire issue and much else in the descent of enterprise.

_NL writes:

These rules make sense if you assume nearly all workers want substantially the same compensation model and nearly all employers are prone to cheat and exploit any alternative compensation model. From that perspective, eliminating apprenticeship, training, internship, and low wages is just closing loopholes by which employers are able to coerce employees into working for free. Sure, the regulators have to basically discount the idea of individual choice and agency, but maybe they've already gotten to that opinion.

I think one of the worst things about the regulatory state is a certain widespread bias: thinking that everybody has the same goals, preferences and personalities. Therefore, the reason some people are living lives inconsistent with those preferences must be that some force is thwarting them. If we assume that choice and bargaining are negligible even for middle-class educated workers, then removing the artifice of choice will result in workers living consistent with the preferences regulators expect them to all share.

JK Brown writes:
_NL, "and nearly all employers are prone to cheat and exploit any alternative compensation model."

I can't speak for nearly all employers, but I can assert that the federal (and perhaps lower government levels as well) bureaucracy is prone to cheat and exploit alternative compensation models. During the Clinton administration, it became public that a recently deceased 20+ year employee of the National Monuments Commission (??) had been for his entire career a temp employee. This caused a lot of uproar as new rules came down severely curtailing the use of temp workers or the recurrent re-hiring of the same person as a temp worker.

You find a misuse of an employment program and you'll be able to find a government agency misusing that program to get around FTE and/or budget limits.

Private employers have competitors, government employers have mandarins seeking to circumvent Congressional constraints. Or to be fair, managers who don't review how they are treating employees by default until a crisis.

JK Brown writes:
Viewed in a commercial sense, as an exchange of consideration or values, apprenticeship can be regarded like other engagements ; yet, what an apprentice gives as well as what he receives are alike too conditional and indefinite to be estimated by ordinary standards. An apprentice exchanges unskilled or inferior labour for technical knowledge, or for the privilege and means of acquiring such knowledge. The master is presumed to impart a kind of special knowledge, collected by him at great expense and pains, in return for the gain derived from the unskilled labour of the learner.


All these things constitute technical knowledge, and the privilege of their acquirement is an element of value. The common view taken of the matter, however, is that it costs nothing for a master to afford these privileges the work must at any rate be carried on, and is not retarded by being watched and learned by apprentices. Viewed from any point, the privileges of engineering establishments have to be considered as an element of value, to be bought at a price, just as a ton of iron or a certain amount of labour is; and in a commercial sense, as an exchangeable equivalent for labour, material, or money. In return a master receives the unskilled labour or service of the learner; this service is presumed to be given at a reduced rate, or sometimes without compensation, for the privileges of the works and the instruction received.
--'The Economy of Workshop Manipulation', J. Richards (1875)

After reading the chapter on apprenticeships the quotes above came from, I did a quick search for apprenticeship regulations. U.S., the UK, Australia, all came back but with regulations that actually required higher than minimum wages for those in apprenticeship programs. Rather foolish in the light that an apprentice would necessarily be providing less return on his wages, as well as imposing a cost on the employer, due to the training, even if it is on-the-job.

Historically, an apprentice "owed" a number of years of service to the master for his training, with the only compensation being room and board.


The following is an anecdote from the Memoirs of General William T. Sherman (Vol I) where he was invited to give his opinion of the South on the eve of the Civil War at a dinner party at the Louisiana governor's mansion, while he was employed as superintendent of what is now LSU.

And, again, I would advise the repeal of the statute which enacted a severe penalty for even the owner to teach his slave to read and write, because that actually qualified properly and took away a part of its value; illustrating the assertion by the case of Henry Sampson, who had been the slave of Colonel Chambers, of Rapides Parish, who had gone to California as the servant of an officer of the army, and who was afterward employed by me in the bank at San Francisco. At first he could not write or read, and I could only afford to pay him one hundred dollars a month; but he was taught to read and write by Reilley, our bank-teller, when his services became worth two hundred and fifty dollars a month, which enabled him to buy his own freedom and that of his brother and his family."

That training was not strictly provided by the employer, although under current labor laws it might be construed as being, but it does demonstrate the productivity value of general education an employer might wish to provide, even as it improves the overall condition of the employee. Even today, an employer-provided literacy program, perhaps after-hours on-site, could do near as much as for many low-paid employees today as it did for Henry Sampson.

ThomasH writes:


As a matter of fact I DO think that a minimum wage not too much above what would prevail without it can do more good than harm, depending on the elasticity of the demand for labor. If the elasticity is small then only a few (but the worst off!) low income workers will loose or not get jobs but many more will gain a lot of extra income.

That is the reason I do not think knowing that the elasticity is not zero is enough to a minimum wage must be a bad policy for transferring income to low-wage workers as a group even if some are harmed.

As with all real life policy decisions there are trade-offs. [I'm sure you know the joke about two political scientists walking down the street. One suggests what he says is a Pareto optimal policy and the other say it can't be. If it were, it would already have been put into practice.]

In the case of a minimum wage increase, an objection that "there will be some unemployment" sounds to the proposer, if she has already made a cost benefit analysis, as if the objector thinks she does not know that there will be unemployment or is himself uninformed about the magnitude of income that will be gained by those who remain unemployed or even that he objects to transferring income consumers and owners of firms to low income workers.

It would seem more persuasive to say, "Your proposal will cause a loss of X thousand jobs for an increase in income of only Y million in increased wages. Do you really think it's worth it?" And it would be even more persuasive to say "An increase in the EITC with a a dead-weight loss to taxpayers of Z millions could raise the incomes of low income workers without causing increased unemployment. Isn't that a better deal?

[And I also agree that it would be more persuasive if proponents of a higher minimum wage would say, "I know that my proposal causes a loss of X thousand jobs, but it leads to an increase of Y million in increased wages." And it would be even more persuasive if she also said, "Of course an increase in the EITC with a a dead weight loss to taxpayers of just Z millions could raise the incomes of low income workers without causing increased unemployment, but I don't think that's politically feasible at the moment."]

Comments for this entry have been closed
Return to top