All Commentary
Thursday, July 1, 1965

Regarding the Minimum Wage

Dr. Blair is a research chemist in New Jersey.


Government regulation of wages is, of course, an old prac­tice. In Western Europe there ex­isted a wide range of wage regu­lations prior to the rise of classi­cal liberalism in the nineteenth century. The North American continent was largely free from government manipulation of wages during the period which saw wages here become the highest in the world; however, from 1912 to 1923 a humanitarian concern for the poor resulted in the estab­lishment of minimum wage laws in fifteen states, Washington, D. C., and Puerto Rico. These laws, when tested in the courts, were declared unconstitutional as violations of the Fourteenth Amend­ment.

The idea of a minimum wage law was revived with the Fair Labor Standards Act of 1938, based on the Federal government’s power to regulate interstate com­merce. The legal minimum wage was set at 25¢ an hour in October, 1938. This was raised to 30¢ in October, 1939 and to 40¢ in 1945. The rate went from 40¢ to 75¢ on January 25, 1950, then to $1.00 on March 1, 1956, and was raised to $1.15 on September 3, 1961 at the same time extending coverage to additional persons at $1.00 an hour. In September, 1963 the minimum wage was increased to $1.25 in all jobs covered prior to 1961 and to $1.15 in jobs added in 1961. New York City unsuccess­fully attempted to establish a $1.50 minimum wage law in 1964, and “progressives” today are ad­vocating a Federal minimum of $2.00 an hour. Indeed, if it is possible to raise wages to any desired level by governmental de­cree, one wonders why large seg­ments of the population, espe­cially those in lower paid jobs, are usually excluded from mini­mum wage coverage, and also why the level is held down to what a typical low income worker is thought to need.


Before presenting the results of empirical studies made on the effect of minimum wage laws, it is desirable to review the results predicted by the classical theory of economics, to know what to look for in the mass of data pub­lished by the Department of La­bor’s Bureau of Labor Statistics. Let me hasten to assure the reader at this point that it is scientifi­cally sound to use theory as a guide in the interpretation of data. Indeed, I hope to show that failure to consider data deemed impor­tant by theory has led many peo­ple to overlook some of the harm­ful consequences of our present minimum wage law.

Simple application of the law of supply and demand suggests that employers forced to pay higher wages will employ fewer workers. This indicates that the industries affected will respond to minimum wage increases by either laying off existing help, or hiring fewer new workers than they otherwise would have done.

As a consequence of reduced employment opportunity in indus­tries “protected” by the minimum wage coverage, one would expect an influx of workers into indus­tries not covered by the law—workers who would “normally” have been employed in the pro­tected industries. Thus, theory predicts that unprotected indus­tries should show increases in employment, or unemployment, or both, depending on the particular industry’s ability to absorb the new workers as the minimum wage is raised. As a result of the increased competition for jobs one would also expect that wages in the nonprotected industries would either fall, or else rise more slowly than normal, when the Federal minimum wage is in­creased.

As the legal minimum is ex­tended to more workers, or is raised higher above the market value of the worker as determined by his productivity, the nonpro­tected industries will be less able to absorb the workers precluded from employment in the protected industries. Hence, one would ex­pect (all else being constant) an increase in the number of persons structurally unemployed. This should be greatest among persons with little skill, or those who are for one reason or another likely to seek employment in the low-pay jobs most affected by the law.

In the normal operation of the market economy, if unemployment develops in a given location (due, for example, to decreases in the demand for a product produced in that area), wages paid in that region tend to be reduced. The lower wages serve as an induce­ment for industry to move into the area, particularly industry such as textile and light manu­facturing plants which do not re­quire highly specialized skills in their workers. Insofar as mini­mum wage laws tend to reduce the wage differential between “de­pressed areas” and areas of nor­mal employment, they would be expected to retard the movement of industry into depressed areas.

The “Ricardo Effect”

Another consequence of mini­mum wage laws (discussed by Ludwig von Mises in Human Ac­tion, pp. 767-769 of the 1949 Yale edition) is the “Ricardo effect,” i.e., a high minimum wage causes employers to substitute machin­ery for labor because of the in-creased cost of labor. It has been suggested by some that the Ri­cardo effect is desirable because it promotes automation. But this neglects the fact that it is usually lack of capital which checks a businessman’s endeavor to im­prove the equipment of his firm. Since the minimum wage law does not create additional capital, the forcing of more capital expendi­tures in one industry leaves less for other industries where it would have been employed more efficiently, i.e., would have yielded a higher return on investment. Thus, the economy as a whole does not benefit from the Ricardo effect. And while the worker in the pro­tected industry who has higher pay benefits from the law, the worker who is laid off or replaced by a machine may see things in a different light.

The Consequences of Intervention

To summarize, classical eco­nomics predicts reduced employ­ment opportunity in protected in­dustries, lower wages and in­creased employment in nonpro­tected industries, and more un­employment in both types of in­dustries than would have other­wise been the case. In addition, a shift of capital expenditures from the rest of the economy into some protected industries would be ex­pected. If the unprotected industries could not absorb the influx of workers precluded from the protected industries, the decreased employment opportunity in the latter would cause an increase in structural unemployment contrib­uting to the development of “de­pressed areas.” The classical the­ory does not claim that no worker will benefit, or even that wages in the protected industries cannot be raised for those workers fortunate enough to remain employed.

Proponents of minimum wage laws, or at least all of them that I have read, base their support on the assumption that the classi­cal theory is invalid (if indeed they indicate having thought about it at all) and that employment opportunity will not be affected. In addition, they like to stress the humanitarian purpose of the law. On this point I offer two observations: (1) Since a law is not animate, we should rather talk about the purpose of the legislators who supported it. But this is impossible to determine without telepathy or a truth se­rum. Perhaps a congressman voted for it because he thought it would help the poor, or because it would aid in his re-election, or because he wants to reduce the likelihood of industry moving into depressed areas instead of into his state or district. (2) The “purpose” is not relevant to the actual effect.

The Data to Prove It

When the Federal minimum wage law was passed in 1938, there were no data available, from this country at least, on the effects of such a law. One could claim that he “knew in his heart” the classical theory is wrong. Now, however, there are both theory and data.

It is sometimes suggested in jest (and even in earnest) that since average wages in this coun­try have increased, and the legal minimum wage rate has increased, the latter caused the former. This argument does not even qualify as post hoc, ergo propter hoc since the increases in the legal minimum in each case followed the average wage increase. I, for one, find it difficult to believe that the national average wage rose past the 75t per hour mark in the 1940′s due to the minimum wage boost to 75t per hour in January of 1950. Studies on min­imum wage law impact have to be done a bit more carefully than this.

For one thing, since most work­ers are not directly affected by any given boost in the legal mini­mum, either because they already earn more than that level or be­cause they are excluded from coverage, the effect can be seen only by studying those industries or geographic areas where a relatively large portion of workers receive low wages. An intelligent study must consider employment as well as wages, and must study the effect on industries excluded from coverage of the law as well as those included.

Support for the law comes from studies printed in the Department of Labor’s Bureau of Labor Sta­tistics’ publication, Monthly Labor Review, so let us consider these carefully. The May, 1960, Monthly Labor Review (v. 83, no. 5, pp. 472-83) contains the most recent such study entitled “Effects of the $1.00 Minimum Wage in Six Areas 1956-59.” The six areas were selected in low-wage regions of the South where the law has a measurable effect. The survey re­ports average hourly wages in the areas before and after the legal minimum was raised to $1.00 on March 1, 1956, and shows that in the industries covered by the law, average hourly wages jumped by around 10 per cent in most of the six regions. But no data are given on employment and unem­ployment figures in these regions at this time, and nothing is said about possible reductions in the work force. Indeed, the average wage in an industry can be raised by simply firing the lowest-paid employees. Hence, this Bureau of Labor Statistics’ study is almost completely useless as an attempt to test the predictions of classical economic theory.

The study does contain some interesting figures, however. Wages in the industries in these six areas which were not covered by the minimum wage law showed an average reduction in one of the areas (Dothan, Alabama), and they either stayed the same or increased by only a per cent or two in the other five areas. This was during an upswing in the economy as a whole, when wages would normally be expected to rise. Thus, the theoretical predic­tion that wages in unprotected industries will either fall or rise more slowly than usual appears to be supported. It should also be noted that this study shows that in all six areas wages in covered industries were already higher than in uncovered industries be­fore the $1.00 legal minimum went into effect. Thus the law produced an even greater differ­ence in wages between the “high” and “low” pay jobs.

Higher Wages—Fewer Jobs

As a consequence of studies such as the above, Monthly Labor Review in an article on the “Re­sults of U. S. Minimum Wage Laws” (March, 1960, pp. 238-42) concludes that we know from ex­perience that it is possible to raise the average pay for workers in low-paying industries by mini­mum wage laws. Classical eco­nomic theory doesn’t dispute that.

America magazine (April 4, 1959, p. 8) at least deals with an actual prediction of the theory when the editors observed: “In the halting progress of the legal minimum wage from 40¢ an hour in 1939 (sic) to $1.00 today, none of these dire predictions has been fulfilled. There has been no erosion of jobs….” It should be noted in connection with this observa­tion, made with no evidence cited to support it, that since World War II, with the Federal minimum wage law in operation, successive business cycles have each left an increase in unemployment. This pattern has been interrupted only recently, probably by the tax re­duction. We have also witnessed a “depressed area problem.” Both are in accord with the theoretical predictions of the effect of mini­mum wage laws, but since there are many factors working in the economy we cannot conclude with­out detailed studies that these problems are caused by the law. However, detailed studies of the law have been made and I will present some of the results.

Effect on Low-Wage Industries

The effect of raises in the legal minimum on employment in var­ious low-wage industries covered by the law is summarized in Man­power, Productivity, and Costs by Professor Yale Brozen of the University of Chicago. In the two years following the establishment of the 25¢ per hour minimum wage rate in October, 1938, 14 per cent of the workers in seamless hosiery plants lost their jobs. Likewise, when the rate was raised to 75¢ an hour employment in southern pine saw mills dropped by 17 per cent. Similar employment drops occurred in the cigar, fertilizer, shirt, footwear, and canning industries. The Bureau of Labor Statistics found an 8 per cent decline in total employment dur­ing the year following the in­crease to $1.00 in the five low-wage industries it chose for de­tailed examination. The applica­tion of the $1.00 minimum wage in 1961 to a certain sector of retail trade brought an 11 per cent de­cline in employment to that part of retail trade, while retail trade employment in the other sectors and in the nation rose. In each of these cases cited above, while em­ployment in the protected low-wage industries dropped, sales, production, and employment were rising in the United States as a whole, because these figures were compiled during a cyclical up­swing.

Another study of the economics of the minimum wage law is the Ph.D. thesis of David E. Kaun, Stanford University (1964), which is summarized in Dissertation Ab­stracts, 25, no. 2, p. 881. Kaun studied fourteen low-wage indus­tries, with large segments located in the South (where the direct effects of the minimum wage are greatest). He considered the be­havior of wage distributions, em­ployment, and labor force compo­sition, among other things. His list of findings include “relative adverse employment effects occur­ring where the impact of the min­imum wage is greatest,” and “in­creases in the minimum wage ap­pear to have adversely affected em­ployment opportunities for cer­tain classes of labor, namely, Negroes, females, younger work­ers, and workers living in rural farm areas.” He concludes that his analysis “results in conclusions generally in agreement with the implications derived from the com­petitive hypothesis,” i.e., classical theory.

A CornellUniversity study of the $1.00 minimum wage law on New York retail trade, some of the results of which are given in Monthly Labor Review, March, 1960, pp. 238-42, found that the law resulted in

·         lower profits to stores

·         reduced hours for part-time help

·         the laying off of workers, es­pecially “inefficient” ones, which, the study explains, means elderly, handicapped, and part-time help

·         reduced store hours, and

·         “more careful recruitment of employees,” which is ex­plained to mean exclusion of the elderly, Negroes, and other “less acceptable” em­ployees.

The Ph.D. thesis of M. A. Malik, University of Michigan (1963), summarized in Dissertation Ab­stracts, 25, no. 3, p. 1616, reports that of twelve low-wage industries studied in the United States, eleven experienced employment declines in the immediate period of two or three months after the establishment of the $1.00 mini­mum (remember this was during a general economic upswing). Of these, ten continued to show em­ployment declines a year later. Since there are many other con­stantly changing factors which in­fluence the employment situation in any given industry, Malik tried to find alternate explanations for the employment reductions in these industries. But in at least five of the industries he could find no other reasonable explanation—the employment decline must be due to the minimum wage law. As expected, the industries where the law had the greatest impact registered the largest declines in em­ployment.

The final study I shall cite is the effect of minimum wage law increases on a noncovered indus­try, household workers, by Yale Brozen in the Journal of Law and Economics, 5, pp. 103-109, Octo­ber, 1962. Studying the period from 1950 to 1962, Professor Bro­zen’s figures, from the Depart­ment of Labor and Bureau of the Census, show that in each instance when the minimum wage rate rose, the number of persons em­ployed as household workers rose. The rise was not the result of un­employed household workers find­ing jobs, since there was also a rise in the percentage of house­hold workers unemployed in each instance (except 1961-62, when the decline in unemployment percent­age accounts for only 15 per cent of the rise). This increase in both employment and unemployment in the noncovered industry with raises in the legal minimum wage is exactly as predicted by classical economics, and indicates that workers driven or precluded from jobs in covered industries by the law must seek work in noncovered industries (like household work). Figures given on wage rates in household employment indicate that the wages are lower than they would have been without the Fair Labor Standards Act.

Other Reasons Offered, But They Are Invalid

As evidence of curtailment of employment in low wage indus­tries resulting from the minimum wage law has mounted, some pro­ponents of the law have adopted a new rationale for their posi­tion; they say the law is good be­cause it helps to eliminate “sweat­shops.” Since some industries are covered and some exempt from coverage by the Fair Labor Stand­ards Act, if some “sweatshops” have been eliminated, it has caused people employed in them to find jobs in others, generally at even lower wages. If the law covered everyone in the economy, (includ­ing babysitters and the like) those who were “saved from sweat­shops” would have nowhere to go to find jobs. It is all very well for the “liberal” theorist to claim that a man is better off unemployed than working in a “sweatshop,” but shouldn’t the decision rest with the man in question?

One additional observation on this point: often a low-paying job gives a person the chance to learn the business or demonstrate his ability, and can lead to a higher-paying position. Consider the number of company presidents and high officials who started their careers in low-paying jobs, and imagine where they might be to­day if they had been “protected” against being offered their first job by a minimum wage law.

Thus, we see that the minimum wage law can raise average wages in an industry by reducing the employment of low-wage help. In some respects the effects are like that of a tariff—it is easy to recognize those who benefit from the law, but harder to determine those who suffer from it. We can see the worker who is given araise because of the increased minimum, but the worker who is laid off when he otherwise would not have been, and the man who is not hired who otherwise would have been, are harder to identify. But while the harmful effects of the tariff are spread over the whole economy, those harmed by the minimum wage law are most­ly the very poor, the unemployed, the elderly, and the unskilled.

This article previously appeared in the January 1965 issue of Insight and Outlook, a conservative journal published by students at the University of Wisconsin.



Grossly Underpaid

An investigator for the Anti-Poverty Commission was recently asked to check on reports that a farmer was pay­ing his help below-standard wages. He went out to the farm and was introduced to all of the hired hands.

“This here is Gordon,” said the farmer. “He milks the cows and works in the fields and he gets $45 a week.

“This is Billy Joe, the other hired man. He works in the fields and tends the stock and he gets $40 a week.

“And this young lady is Sue Ann. She cooks and keeps house and she gets $30 a week, room and board.”

“Fair enough so far,” said the inspector. “Is there any­one else?”

“Only the half-wit,” answered the farmer. “He gets $10 a week, tobacco, room and board.”

“Aha,” said the inspector. “I’d like to speak to him.” “You’re talk in’ to him right now,” replied the farmer.