Chapter 10: The Fallacy of Providing Jobs

SINCE ANCIENT TIMES IT HAS BEEN ASSUMED NOT ONLY THAT THE government has a duty to do something for the poor, but that one of the things it can and should do is to "provide jobs." A declared objective, in fact, even of the Elizabethan Poor Law was the "setting of the poor to work." Today many are insisting that the government has both the ability and the duty to become "the employer of last resort," or to "guarantee everybody a job."

These views rest on some serious misconceptions.

In a hypothetical evenly rotating economy, with free and fluid competition, each worker would try to find work wherever his pay was highest, which means wherever his marginal productivity was highest; and each employer would likewise try to find the worker whose productivity was highest for the pay and in the job he had to offer. Therefore in such an economy workers would be allocated at their highest individual productivity among the tens of thousands of different Occupations, and there would tend to be full employment at maximum overall productivity. There would be no unemployment for the State to try to eliminate.

In any actual dynamic economy, of course, there is always a certain amount of "normal" unemployment. This is too often regarded as an unmitigated evil. It is misleadingly called "frictional" unemployment. Yet it is mainly the result of necessary and desirable economic adjustments that ordinarily take time, and for the most part ought to be allowed to take time.

In a healthy, flexible economy these adjustments are always taking place. Some industries are expanding while others are shrinking, either absolutely or comparatively. It is necessary that workers and capital transfer from the shrinking to the expanding industries. Some workers are forced to do this because they are laid off. Others quit voluntarily, draw unemployment benefits, live on their savings, give themselves vacations, rest between jobs, or spend time "looking around" and deciding what to do next. They are trying to decide where they can be most profitably or satisfactorily employed; and time for comparisons is necessary to ensure a good choice.

A man is only unemployed, as the economist A. C. Pigou once put it, "when he is both not employed and also desires to be employed." It is this subjective element that statisticians can not measure. So both those who are voluntarily and those who are involuntarily between jobs are lumped together in the same unemployment statistics.

Many people are unduly distressed by these statistics not only because they fail to make this distinction but because they picture the unemployed as a permanent army hopelessly tramping around in search of work. But the make-up of the unemployed is constantly changing. In mid-1972, to take a fairly typical example, the average duration of unemployment was eleven weeks. Only about a fifth of the unemployed were out of work for fifteen weeks or more.

It is when there is abnormal or mass unemployment that the demand is loudest that the government itself should provide jobs. What is first of all wrong with this demand is that it ignores the cause of the existing unemployment. In most cases this cause will be found to be some policy or situation for which the government itself is mainly responsible.

If the government has imposed a minimum wage law, for example, it has in effect condemned to unemployment all the workers incapable of earning that minimum. If, then, the government itself "provides work" for them, it is at best doing merely what it has prevented private employers from doing. If it pays these workers the legal minimum, as it probably would, it is employing them at an economic loss made up by the tax payers, for such workers are almost certainly producing less value than the amount of their pay.

Private employers would at least have employed them (if there had been no minimum wage law) where their productive value was highest. As the government will be under no such necessity, it will not try to do this. So on the average it will put them to even less productive work than private industry would have done.

If, again, the government tries to put the unemployed to work in lines in which there are already strong unions, it will run into opposition from these unions for increasing the competition against their members. This will further restrict the productive possibilities of whatever work the government offers.

The same kind of problems will arise no matter what the primary cause of the existing unemployment. That cause may be, as it often is, excessive wage rates brought about by labor union pressure -- by strikes or strike threats, or by concessions otherwise wrung from employers because of legal compulsions imposed on them to "bargain collectively" with specified unions. Excessive wage rates always lead to unemployment. When the unemployed compete for private jobs, this tends, in a free market, to bring average wage rates back to levels at which unemployment will disappear. But if all those who are thrown out of work by excessive wage rates are then immediately employed by the government, the normal economic pressures are removed to get wages down to a working market level. The necessary adjustments are not made. The workers employed by the government are employed at a loss borne by the taxpayers. The whole economic community, because its workers are less efficiently employed, is made poorer than it other wise would have been.

The same reasoning applies if the cause of the unemployment is a typical depression, brought about, for example, because a currency or credit deflation has led to a drop in consumer demand and commodity prices while wage rates and other "sticky" costs stay up. In a free market, this situation would eventually be cured by a downward adjustment of wage rates to the new lower demand level and price level. But if the government immediately offers make-work "jobs" to everybody dropped from employment by private industry, the downward adjustment of wage rates will never take place. The burden on the taxpayers will soon become unbearable; and the only way out will seem to be budget deficits and an inflation of the currency. Further inflation, in fact, will soon be considered the "normal" solution for all unemployment problems -- until this has led to the inevitable crisis.

Step One: Re-examine Existing Policies

The first step necessary, therefore, when there is an abnormal amount of unemployment, is for the government in power to re-examine existing economic policies and discontinue all those that have been causing the unemployment.

A certain percentage of the unemployed are unemployables. These include the physically incapacitated -- the aged, weak, disabled, or blind. They include the feeble-minded, and the people so backward that they cannot be taught elementary skills, and require more supervision than it is practicable to provide. They include, finally, the chronic loafers -- the Rip van Winkles born with "an insuperable aversion to all kinds of profitable labor" -- and the hostile, who refuse to accept any kind of discipline, who through malice or indifference do more damage than useful work. It is frustrating for any government to try to "provide useful work" for such people.

In times of mass unemployment there is nearly always a loud demand that the government should provide jobs rather than merely put people on relief, but the effort, if undertaken, soon begins to prove prohibitively expensive. In the depression of the 1930s, for instance, the Roosevelt regime set up the Works Progress Administration (WPA) to provide jobs. But as the workers needed raw materials, tools, machinery, and equipment, this employment proved extremely expensive per job provided. When this was discovered, the officials in charge tried to think of projects -- and were praised for thinking of projects -- that provided the maximum number of jobs per dollar expended; in other words, jobs requiring the least raw materials, machinery, and equipment. But what was over looked was that such jobs -- requiring the maximum hand labor in relation to capital equipment -- were precisely the least efficient and least productive jobs that could have been thought of. (It is estimated that private industry today in the United States has invested about $30,000 per production worker.)

All this points to the folly of the proposal that the government can or should "guarantee everybody a job." With the fear of dismissal completely removed, such a guarantee would demoralize even workers who might be passably industrious and efficient in unguaranteed jobs. They would have no obligation to please the boss or anybody else. Suppose they started to arrive one or two hours late? Or quit two hours early? Or chronically reported sick? Or failed to show up at all except to collect their pay? Or broke more dishes than they washed -- botched every job they were assigned to? Or refused to accept orders or any direction? Or stole? Or committed deliberate vandalism? Or beat up the boss? Their jobs would be guaranteed, wouldn't they?

Putting aside all questions of worker morale, how would the government decide where and on what to put people to work, and how many on this job and how many on that? Could it put everybody to work on his previous skill, if any, regardless of whether there was still any demand for the product of his services? Could it put men to producing goods for which there was no market?

And what pay scale would it offer? Would it pay high enough wages to prevent workers from being attracted back to private industry? Or even high enough to attract workers already employed in private industry? Would it, on the other hand, pay lower than the minimum in private industry? Would such low wages prove politically tenable?

We are forced to the conclusion that it is impossible for the government to provide useful and profitable work (apart from necessary governmental services themselves) outside of what is or would be provided by an unhampered private enterprise.

We come now to a contrasting proposal -- that the government should deny relief to anybody who refuses to take a job offered to him. This is a proposal that has been constantly put forward in the history of relief but has as constantly run into difficulties. The first objection commonly raised is that the specific job offered may not be "suitable." The man to whom it is offered may complain that the wage offered is too low, or the job too disagreeable, or too "menial," or beneath his skill, or even beyond his strength. There will always be those who will denounce the work requirement as a form of "involuntary servitude."

Such a work requirement is often written into relief laws, but the officials in charge of relief usually lack the courage to enforce it. They are afraid of being accused of "letting people starve," so the requirement quickly becomes merely perfunctory.

Practically all forced work is economically unprofitable. Toward the end of the eighteenth century, in fact, parish authorities in England were reduced to such expedients as making men stand in the parish pound for so many hours, or obliging them to attend a roll call several times a day, or making them dig holes and fill them up again.

But there was one work requirement that did have a plausible basis. It was the keystone, in fact, of the Poor Law reform of 1834. This was that no able-bodied person would be given relief unless he was willing to live in a workhouse and perform the usually monotonous and uninteresting labor there assigned to him. The Commissioners who proposed that requirement did not assume that the work so performed would be economically very useful. On the other hand, it was not imposed as a mere punishment. It was imposed primarily as a test, a test that hopefully would separate the pretenders from those in dire need. If a man was really in danger of starving, it was argued, he would accept the workhouse; if he refused it, his condition could not be too bad.

In England this system lasted, with general public acquiescence, for some three quarters of a century. Yet even in the Victorian Age protests against it became increasingly insistent. It is doubtful that it will ever again gain public acceptance.

We seem driven, then, to the conclusion that the government can neither guarantee useful and profitable work, nor directly provide it, nor compel it.

Is there any escape from this conclusion? Can the State devise -- and have the courage consistently to adhere to some work test, or work-acceptance test, that will enable it to separate the deserving poor from the shirkers and fakers? There is no more stubborn social problem than this: How can the State adequately relieve those truly in need without undermining their incentives to effort and without imposing on workers and producers an insupportable burden of relief?

This is a problem that no nation has yet satisfactorily solved.

But we have learned this: The chief thing that the State can do to reduce the problems of poverty and unemployment to minor dimensions is to permit and encourage the free market system to function.

This e-text is made available by the The Henry Hazlitt Foundation in cooperation with The Foundation for Economic Education. The Hazlitt Foundation is a member-supported 501(c)(3) non-profit corporation whose mission is to make the ideas of freedom more accessible. Please visit our flagship Internet service .

© 1973 Henry Hazlitt. For permissions information, contact The Foundation for Economic Education, 30 South Broadway, Irvington-on-Hudson, NY 10533.