This Classic Work was written by Henry Hazlitt.

Continued from Chapter 18: What Rent Control Does

We have already seen some of the harmful results of arbitrary governmental efforts to raise the price of favored commodities. The same sort of harmful results follow efforts to raise wages through minimum wage laws. This ought not to be surprising, for a wage is, in fact, a price. It is unfortunate for clarity of economic thinking that the price of labor’s services should have received an entirely different name from other prices. This has prevented most people from recognizing that the same principles govern both.

Thinking has become so emotional and so politically biased on the subject of wages that in most discussions of them the plainest principles are ignored. People who would be among the first to deny that prosperity could be brought about by artificially boosting prices, people who would be among the first to point out that minimum price laws might be most harmful to the very industries they were designed to help, will nevertheless advocate minimum wage laws, and denounce opponents of them, without misgivings.


Yet it ought to be clear that a minimum wage law is, at best, a limited weapon for combatting the evil of low wages, and that the possible good to be achieved by such a law can exceed the possible harm only in proportion as its aims are modest. The more ambitious such a law is, the larger the number of workers it attempts to cover, and the more it attempts to raise their wages, the more certain are its harmful effects to exceed any possible good effects.

The first thing that happens, for example, when a law is passed that no one shall be paid less than $106 for a forty-hour week is that no one who is not worth $106 a week to an employer will be employed at all. You cannot make a man worth a given amount by making it illegal for anyone to offer him anything less. You merely deprive him of the right to earn the amount that his abilities and situation would permit him to earn, while you deprive the community even of the moderate services that he is capable of rendering. In brief, for a low wage you substitute unemployment. You do harm all around, with no comparable compensation.

The only exception to this occurs when a group of workers is receiving a wage actually below its market worth. This is likely to happen only in rare and special circumstances or localities where competitive forces do not operate freely or adequately; but nearly all these special cases could be remedied just as effectively, more flexibly and with far less potential harm, by unionization.

It may be thought that if the law forces the payment of a higher wage in a given industry, that industry can then charge higher prices for its product, so that the burden of paying the higher wage is merely shifted to consumers. Such shifts, however, are not easily made, nor are the consequences of artificial wage-raising so easily escaped. A higher price for the product may not be possible: it may merely drive consumers to the equivalent imported products or to some substitute. Or, if consumers continue to buy the product of the industry in which wages have been raised, the higher price will cause them to buy less of it. While some workers in the industry may be benefited from the higher wage, therefore, others will be thrown out of employment altogether. On the other hand, if the price of the product is not raised, marginal producers in the industry will be driven out of business; so that reduced production and consequent unemployment will merely be brought about in another way.

When such consequences are pointed out, there are those who reply: “Very well; if it is true that the X industry cannot exist except by paying starvation wages, then it will be just as well if the minimum wage puts it out of existence altogether.” But this brave pronouncement overlooks the realities. It overlooks, first of all, that consumers will suffer the loss of that product. It forgets, in the second place, that it is merely condemning the people who worked in that industry to unemployment. And it ignores, finally, that bad as were the wages paid in the X industry, they were the best among all the alternatives that seemed open to the workers in that industry; otherwise the workers would have gone into another. If, therefore, the X industry is driven out of existence by a minimum wage law, then the workers previously employed in that industry will be forced to turn to alternative courses that seemed less attractive to them in the first place. Their competition for jobs will drive down the pay offered even in these alternative occupations. There is no escape from the conclusion that the minimum wage will increase unemployment.

A nice problem, moreover, will be raised by the relief program designed to take care of the unemployment caused by the minimum wage law. By a minimum wage of, say, $2.65 an hour, we have forbidden anyone to work forty hours in a week for less than $106.[5] Suppose, now, we offer only $70 a week on relief. This means that we have forbidden a man to be usefully employed at, say, $90 a week, in order that we may support him at $70 a week in idleness. We have deprived society of the value of his services. We have deprived the man of the independence and self-respect that come from self-support, even at a low level, and from performing wanted work, at the same time as we have lowered what the man could have received by his own efforts.

These consequences follow as long as the weekly relief payment is a penny less than $106. Yet the higher we make the relief payment, the worse we make the situation in other respects. If we offer $106 for relief, then we offer many men just as much for not working as for working. Moreover, whatever the sum we offer for relief, we create a situation in which everyone is working only for the difference between his wages and the amount of the relief. If the relief is $106 a week, for example, workers offered a wage of $2.75 an hour, or $110 a week, are in fact, as they see it, being asked to work for only $4 a week—for they can get the rest without doing anything.

It may be thought that we can escape these consequences by offering “work relief” instead of “home relief “; but we merely change the nature of the consequences. Work relief means that we are paying the beneficiaries more than the open market would pay them for their efforts. Only part of their relief-wage is for their efforts, therefore, while the rest is a disguised dole.

It remains to be pointed out that government make-work is necessarily inefficient and of questionable utility. The government has to invent projects that will employ the least skilled. It cannot start teaching people carpentry, masonry, and the like, for fear of competing with established skills and arousing the antagonism of existing unions. I am not recommending it, but it probably would be less harmful all around if the government in the first place frankly subsidized the wages of submarginal workers at the work they were already doing. Yet this would create political headaches of its own.

We need not pursue this point further, as it would carry us into problems not immediately relevant. But the difficulties and consequences of relief must be kept in mind when we consider the adoption of minimum wage laws or an increase in minimums already fixed [*]

Before we finish with the topic I should perhaps mention another argument sometimes put forward for fixing a minimum wage rate by statute. This is that in an industry in which one big company enjoys a monopoly, it need not fear competition and can offer below-market wages. This is a highly improbable situation. Such a “monopoly” company must offer high wages when it is formed, in order to attract labor from other industries. Thereafter it could theoretically fail to increase wage rates as much as other industries, and so pay “substandard” wages for that particular specialized skill. But this would be likely to happen only if that industry (or company) was sick or shrinking; if it were prosperous or expanding, it would have to continue to offer high wages to increase its labor force.

We know as a matter of experience that it is the big companies —those most often accused of being monopolies—that pay the highest wages and offer the most attractive working conditions. It is commonly the small marginal firms, perhaps suffering from excessive competition, that offer the lowest wages. But all employers must pay enough to hold workers or to attract them from each other.

All this is not to argue that there is no way of raising wages. It is merely to point out that the apparently easy method of raising them by government fiat is the wrong way and the worst way.

This is perhaps as good a place as any to point out that what distinguishes many reformers from those who cannot accept their proposals is not their greater philanthropy, but their greater impatience. The question is not whether we wish to see everybody as well off as possible. Among men of good will such an aim can be taken for granted. The real question concerns the proper means of achieving it. And in trying to answer this we must never lose sight of a few elementary truisms. We cannot distribute more wealth than is created. We cannot in the long run pay labor as a whole more than it produces.

The best way to raise wages, therefore, is to raise marginal labor productivity. This can be done by many methods: by an increase in capital accumulation — i.e., by an increase in the machines with which the workers are aided; by new inventions and improvements; by more efficient management on the part of employers; by more industriousness and efficiency on the part of workers; by better education and training. The more the individual worker produces, the more he increases the wealth of the whole community. The more he produces, the more his services are worth to consumers, and hence to employers. And the more he is worth to employers, the more he will be paid. Real wages come out of production, not out of government decrees.

So government policy should be directed, not to imposing more burdensome requirements on employers, but to following policies that encourage profits, that encourage employers to expand, to invest in newer and better machines to increase the productivity of workers — in brief, to encourage capital accumulation, instead of discouraging it—and to increase both employment and wage rates.

* In 1938, when the average hourly wage paid in all manufacturing in the United States was about 63 cents an hour, Congress set a legal minimum of only 25 cents. In 1945, when the average factory wage had risen to $1.02 an hour, Congress raised the legal minimum to 40 cents. In 1949, when the average factory wage had risen to $1.40 an hour, Congress raised the minimum again to 75 cents. In 1955, when the average had risen to $i.88, Congress boosted the minimum to $1. In 1961, with the average factory wage at about $2.30 an hour, the minimum was raised to $1.15 in 1961 and to $1.25 for 1963. To shorten the account, the minimum wage was raised to $1.40 in 1967, to $1.60 in 1968, to $2.00 in 1974, to $2.10 in 1975, and to $2.30 in 1976 (when the average wage in all private nonagricultural work was $4.87) Then in 1977, when the actual average hourly wage in nonagricultural work was $5.26, the minimum wage was raised to $2.65 an hour, with provision made for notching it up still further in each of the next three years. Thus, as the prevailing hourly wage goes higher, the minimum wage advocates decide that the legal minimum must be raised at least correspondingly. Though the legislation follows the rise of the prevailing market wage rate, the myth continues to be built up that it is the minimum wage legislation that has raised the market wage

Continue to Chapter 20: Do Unions Really Raise Wages?