It might be asserted that labor unions, in exacting higher wage rates on the free market, are achieving identifiable monopoly prices. For here two identifiable contrasting situations exist: (a) where individuals sell their labor themselves; and (b) where they are members of labor unions which bargain on their labor for them. Furthermore, it is clear that while cartels, to be successful, must be economically more efficient in serving the consumer, no such justification can be found for unions. Since it is always the individual laborer who works, and since efficiency in organization comes from management hired for the task, forming unions never improves the productivity of an individual’s work.

It is true that a union provides an identifiable situation. However, it is not true that a union wage rate could ever be called a monopoly price.58 For the characteristic of the monopolist is precisely that he monopolizes a factor or commodity. To obtain a monopoly price, he sells only part of his supply and withholds selling the other part, because selling a lower quantity raises the price on an inelastic demand curve. It is the unique characteristic of labor in a free society, however, that it cannot be monopolized. Each individual is a self-owner and cannot be owned by another individual or group. Therefore, in the labor field, no one man or group can own the total supply and withhold part of it from the market. Each man owns himself.

Let us call the total supply of a monopolist’s product P. When he withholds W units in order to obtain a monopoly for P – W, the increased revenue he obtains from P – W must more than compensate him for the loss of revenue he suffers from not selling W. A monopolist’s action is always limited by loss of revenue from the withheld supply. But in the case of labor unions, this limitation does not apply. Since each man owns himself, the “withheld” suppliers are different people from the ones getting the increased income. If a union, in one way or another, achieves a higher price than its members could command by individual sales, its action is not checked by the loss of revenue suffered by the “withheld” laborers. If a union achieves a higher wage, some laborers are earning a higher price, while others are excluded from the market and lose the revenue they would have obtained. Such a higher price (wage) is called a restrictionist price.

A restrictionist price, by any sensible criterion, is “worse” than a “monopoly price.” Since the restrictionist union does not have to worry about the laborers who are excluded and suffers no revenue loss from such exclusion, restrictionist action is not curbed by the elasticity of the demand curve for labor. For unions need only maximize the net income of the working members, or, indeed, of the union bureaucracy itself.59

How may a union achieve a restrictionist price? Figure 69 will illustrate. The demand curve is the demand curve for a labor factor in an industry. DD is the demand curve for the labor in the industry; SS, the supply curve. Both curves relate the number of laborers on the horizontal axis and the wage rate on the vertical. At the market equilibrium, the supply of laborers offering their work in the industry will intersect the demand for the labor, at number of laborers 0A and wage rate AB. Now, suppose that a union enters this labor market, and the union decides that its members will insist on a higher wage than AB, say 0W. What unions do, in fact, is to insist upon a certain wage rate as a minimum below which they will not work in that industry.

The effect of the union decision is to shift the supply curve of labor available to the industry to a horizontal one at the wage rate WW′, rising after it joins the SS curve at E. The minimum reserve price of labor for this industry has risen, and has risen for all laborers, so that there are no longer laborers with lower reserve prices who would be willing to work for less. With a supply curve changing to WE, the new equilibrium point will be C instead of B. The number of workers hired will be WC, and the wage rate 0W.The union has thus achieved a restrictionist wage rate. It can be achieved regardless of the shape of the demand curve, granting only that it is falling. The demand curve falls because of the diminishing DMVP of a factor and the diminishing marginal utility of the product. But a sacrifice has been made—specifically, there are now fewer workers hired, by an amount CF. What happens to them? These discharged workers are the main losers in this procedure. Since the union represents the remaining workers, it does not have to concern itself, as the monopolist would, with the fate of these workers. At best, they must shift (being a nonspecific factor, they can do so) to some other—nonunionized—industry. The trouble is, however, that the workers are less suited to the new industry. Their having been in the now unionized industry implies that their DMVP in that industry was higher than in the industry to which they must shift; consequently, their wage rate is now lower. Moreover, their entry into the other industry depresses the wage rates of the workers already there.

Consequently, at best, a union can achieve a higher, restrictionist wage rate for its members only at the expense of lowering the wage rates of all other workers in the economy. Production efforts in the economy are also distorted. But, in addition, the wider the scope of union activity and restrictionism in the economy, the more difficult it will be for workers to shift their locations and occupations to find nonunionized havens in which to work. And more and more the tendency will be for the displaced workers to remain permanently or quasi-permanently unemployed, eager to work but unable to find nonrestricted opportunities for employment. The greater the scope of unionism, the more a permanent mass of unemployment will tend to develop.

Unions try as hard as they can to plug all the “loop-holes” of nonunionism, to close all the escape hatches where the dispossessed workmen can find jobs. This is termed “ending the unfair competition of nonunion, low-wage labor.” A universal union control and restrictionism would mean permanent mass unemployment, growing ever greater in proportion to the degree that the union exacted its restrictions.

It is a common myth that only the old-style “craft” unions, which deliberately restrict their occupational group to highly skilled trades with relatively few numbers, can restrict the supply of labor. They often maintain stringent standards of membership and numerous devices to cut down the supply of labor entering the trade. This direct restriction of supply doubtless makes it easier to obtain higher wage rates for the remaining workers. But it is highly misleading to believe that the newer-style “industrial” unions do not restrict supply. The fact that they welcome as many members in an industry as possible cloaks their restrictionist policy. The crucial point is that the unions insist on a minimum wage rate higher than what would be achieved for the given labor factor without the union. By doing so, as we saw in Figure 69, they necessarily cut the number of men whom the employer can hire. Ergo, the consequence of their policy is to restrict the supply of labor, while at the same time they can piously maintain that they are inclusive and democratic, in contrast to the snobbish “aristocrats” of craft unionism.

In fact, the consequences of industrial unionism are more devastating than those of craft unionism. For the craft unions, being small in scope, displace and lower the wages of only a few workers. The industrial unions, larger and more inclusive, depress wages and displace workers on a large scale and, what is even more important, can cause permanent mass unemployment.60

There is another reason why an openly restrictionist union will cause less unemployment than a more liberal one. For the union which restricts its membership serves open warning on workers hoping to enter the industry that they are barred from joining the union. As a result, they will swiftly look elsewhere, where jobs can be found. Suppose the union is democratic, however, and open to all. Then, its activities can be described by the above figure; it has achieved a higher wage rate 0W for its working members. But such a wage rate, as can be seen on the SS curve, attracts more workers into the industry. In other words, while 0A workers were hired by the industry at the previous (nonunion) wage AB, now the union has won a wage 0W. At this wage, only WC workers can be employed in the industry. But this wage also attracts more workers than before, namely WE. As a result, instead of only CF workers becoming unemployed from the union’s restrictionist wage rate, more—CE—will be unemployed in the industry.

Thus, an open union does not have the one virtue of the closed union—rapid repulsion of the displaced workers from the unionized industry. Instead, it attracts even more workers into the industry, thus aggravating and swelling the amount of unemployment. With market signals distorted, it will take a much longer time for workers to realize that no jobs are available in the industry. The larger the scope of open unions in the economy, and the greater the differential between their restrictionist wage rates and the market wage rates, the more dangerous will the unemployment problem become.

The unemployment and the misemployment of labor, caused by restrictionist wage rates need not always be directly visible. Thus, an industry might be particularly profitable and prosperous, either as a result of a rise in consumer demand for the product or from a cost-lowering innovation in the productive process. In the absence of unions, the industry would expand and hire more workers in response to the new market conditions. But if a union imposes a restrictionist wage rate, it may not cause the unemployment of any current workers in the industry; it may, instead, simply prevent the industry from expanding in response to the requirements of consumer demand and the conditions of the market. Here, in short, the union destroys potential jobs in the making and imposes a misallocation of production by preventing expansion. It is true that, without the union, the industry will bid up wage rates in the process of expansion; but if unions impose a higher wage rate at the beginning, the expansion will not occur.61

Some opponents of unionism go to the extreme of maintaining that unions can never be free-market phenomena and are always “monopolistic” or coercive institutions. Although this might be true in actual practice, it is not necessarily true. It is very possible that labor unions might arise on the free market and even gain restrictionist wage rates.

How can unions achieve restrictionist wage rates on the free market? The answer can be found by considering the displaced workers. The key problem is: Why do the workers let themselves be displaced by the union’s WW minimum? Since they were willing to work for less before, why do they now meekly agree to being fired and looking for a poorer-paying job? Why do some remain content to continue in a quasi-permanent pocket of unemployment in an industry, waiting to be hired at the excessively high rate? The only answer, in the absence of coercion, is that they have adopted on a commandingly high place on their value scales the goal of not undercutting union wage rates. Unions, naturally, are most anxious to persuade workers, both union and nonunion, as well as the general public, to believe strongly in the sinfulness of undercutting union wage rates. This is shown most clearly in those situations where union members refuse to continue working for a firm at a wage rate below a certain minimum (or on other terms of employment). This situation is known as a strike. The most curious thing about a strike is that the unions have been able to spread the belief throughout society that the striking members are still “really” working for the company even when they are deliberately and proudly refusing to do so. The natural answer of the employer, of course, is to turn somewhere else and to hire laborers who are willing to work on the terms offered. Yet unions have been remarkably successful in spreading the idea through society that anyone who accepts such an offer—the “strikebreaker”—is the lowest form of human life.

To the extent, then, that nonunion workers feel ashamed or guilty about “strike-breaking” or other forms of undercutting union-proclaimed wage scales, the displaced or unemployed workers agree to their own fate. These workers, in effect, are being displaced to poorer and less satisfying jobs voluntarily and remain unemployed for long stretches of time voluntarily. It is voluntary because that is the consequence of their voluntary acceptance of the mystique of “not crossing the picket line” or of not being a strikebreaker.

The economist qua economist can have no quarrel with a man who voluntarily comes to the conclusion that it is more important to preserve union solidarity than to have a good job. But there is one thing an economist can do: he can point out to the worker the consequences of his voluntary decision. There are undoubtedly countless numbers of workers who do not realize that their refusal to cross a picket line, their “sticking to the union,” may result in their losing their jobs and remaining unemployed. They do not realize this because to do so requires knowledge of a chain of praxeological reasoning (such as we have been following here). The consumer who purchases directly enjoyable services does not have to be enlightened by economists; he needs no lengthy chain of reasoning to know that his clothing or car or food is enjoyable or serviceable. He can see each perform its service before his eyes. Similarly, the capitalist-entrepreneur does not need the economist to tell him what acts will be profitable or unprofitable. He can see and test them by means of his profits or losses. But for a grasp of the consequences of acts of governmental intervention in the market or of union activity, knowledge of praxeology is requisite.62

Economics cannot itself decide on ethical judgments. But in order for anyone to make ethical judgments rationally, he must know the consequences of his various alternative courses of action. In questions of government intervention or union action, economics supplies the knowledge of these consequences. Knowledge of economics is therefore necessary, though not sufficient, for making a rational ethical judgment in these fields. As for unions, the consequences of their activity, when discovered (e.g., displacement or unemployment for oneself or others), will be considered unfortunate by most people. Therefore, it is certain that when knowledge of these consequences becomes widespread, far fewer people will be “prounion” or hostile to “nonunion” competitors.63

Such conclusions will be reinforced when people learn of another consequence of trade union activity: that a restrictionist wage raises costs of production for the firms in the industry. This means that the marginal firms in the industry—the ones whose entrepreneurs earn only a bare rent—will be driven out of business, for their costs have risen above their most profitable price on the market—the price that had already been attained. Their ejection from the market and the general rise of average costs in the industry signify a general fall in productivity and output, and hence a loss to the consumers.64 Displacement and unemployment, of course, also impair the general standard of living of the consumers.

Unions have had other important economic consequences. Unions are not producing organizations; they do not work for capitalists to improve production.65 Rather they attempt to persuade workers that they can better their lot at the expense of the employer. Consequently, they invariably attempt as much as possible to establish work rules that hinder management’s directives. These work rules amount to preventing management from arranging workers and equipment as it sees fit. In other words, instead of agreeing to submit to the work orders of management in exchange for his pay, the worker now sets up not only minimum wages, but also work rules without which he refuses to work. The effect of these rules is to lower the marginal productivity of all union workers. The lowering of marginal value-product schedules has a twofold result: (1) it itself establishes a restrictionist wage scale with its various consequences, for the marginal value product has fallen while the union insists that the wage rate remain the same; (2) consumers lose by a general lowering of productivity and living standards. Restrictive work rules therefore also lower output. All this is perfectly consistent with a society of individual sovereignty, however, provided always that no force is employed by the union.

To advocate coercive abolition of these work rules would imply literal enslavement of the workers to the dictates of catallactic consumers. But, once again, it is certain that knowledge of these various consequences of union activity would greatly weaken the voluntary adherence of many workers and others to the mystique of unionism.66

Unions, therefore, are theoretically compatible with the existence of a purely free market. In actual fact, however, it is evident to any competent observer that unions acquire almost all their power through the wielding of force, specifically force against strikebreakers and against the property of employers. An implicit license to unions to commit violence against strikebreakers is practically universal. Police commonly either remain “neutral” when strikebreakers are molested or else blame the strikebreakers for “provoking” the attacks upon them. Certainly, few pretend that the institution of mass picketing by unions is simply a method of advertising the fact of a strike to anyone passing by. These matters, however, are empirical rather than theoretical questions. Theoretically, we may say that it is possible to have unions on a free market, although empirically we may question how great their scope would be.

Analytically, we can also say that when unions are permitted to resort to violence, the state or other enforcing agency has implicitly delegated this power to the unions. The unions, then, have become “private states.”67

We have, in this section, investigated the consequences of unions’ achieving restrictionist prices. This is not to imply, however, that unions always achieve such prices in collective bargaining. Indeed, because unions do not own workers and therefore do not sell their labor, the collective bargaining of unions is an artificial replacement for the smooth workings of “individual bargaining” on the labor market. Whereas wage rates on the nonunion labor market will always tend toward equilibrium in a smooth and harmonious manner, its replacement by collective bargaining leaves the negotiators with little or no rudder, with little guidance on what the proper wage rates would be. Even with both sides trying to find the market rate, neither of the parties to the bargain could be sure that a given wage agreement is too high, too low, or approximately correct. Almost invariably, furthermore, the union is not trying to discover the market rate, but to impose various arbitrary “principles” of wage determination, such as “keeping up with the cost of living,” a “living wage,” the “going rate” for comparable labor in other firms or industries, an annual average “productivity” increase, “fair differentials,” etc.68

• 58The first to point out the error in the common talk of “monopoly wage rates” of unions was Professor Mises. See his brilliant discussion in Human Action, pp. 373–74. Also see P. Ford, The Economics of Collective Bargaining (Oxford: Basil Blackwell, 1958), pp. 35–40. Ford also refutes the thesis advanced by the recent “Chicago School” that unions perform a service as sellers of labor:
  But a union does not itself produce or sell the commodity, labour, nor receive payment for it. ... It could be more fitly described as ... fixing the wages and other conditions on which its individual members are permitted to sell their services to the individual employers. (Ibid., p. 36)
• 59A restrictionist, rather than a monopoly, price can be achieved because the number of laborers is so important in relation to the possible variation in hours of work by an individual laborer that the latter can be ignored here. If, however, the total labor supply is limited originally to a few people, then an imposed higher wage rate will cut down the number of hours purchased from the workers who remain working, perhaps so much as to render a restrictionist price unprofitable to them. In such a case it would be more appropriate to speak of a monopoly price.
• 60Cf. Mises, Human Action, p. 764.
• 61See Charles E. Lindblom, Unions and Capitalism (New Haven: Yale University Press, 1949), pp. 78 ff., 92–97, 108, 121, 131–32, 150–52, 155. Also see Henry C. Simons, “Some Reflections on Syndicalism” in Economic Policy for a Free Society (Chicago: University of Chicago Press, 1948), pp. 131f., 139 ff.; Martin Bronfenbrenner, “The Incidence of Collective Bargaining,” American Economic Review, Papers and Proceedings, May, 1954, pp. 301–02; Fritz Machlup, “Monopolistic Wage Determination as a Part of the General Problem of Monopoly” in Wage Determination and the Economics of Liberalism (Washington, D.C.: Chamber of Commerce of the United States, 1947), pp. 64–65.
• 62See Murray N. Rothbard, “Mises’ Human Action: Comment,” American Economic Review, March, 1951, pp. 183–84.
• 63The same is true, to an even greater extent, of measures of governmental intervention in the market. See chapter 12 below.
• 64See James Birks, Trade Unionism in Relation to Wages (London, 1897), p. 30.
• 65See James Birks, Trades’ Unionism: A Criticism and a Warning (London, 1894), p. 22.
• 66We can deal here only with the directly catallactic consequences of labor unionism. Unionism also has other consequences which many might consider even more deplorable. Prominent is the fusing of the able and the incompetent into one group. Seniority rules, for example, are invariable favorites of unions. They set restrictively high wages for less able workers and also lower the productivity of all. But they also reduce the wages of the more able workers—those who must be chained to the stultifying march of seniority for their jobs and promotions. Seniority also decreases the mobility of workers and creates a kind of industrial serfdom by establishing vested rights in jobs according to the length of time the employees have worked. Cf. David McCord Wright, “Regulating Unions” in Bradley, Public Stake in Union Power, pp. 113–21.
• 67Students of labor unions have almost universally ignored the systematic use of violence by unions. For a welcome exception, see Sylvester Petro, Power Unlimited (New York: Ronald Press, 1959). Also cf. F.A. Hayek, “Unions, Inflation, and Profits,” p. 47.
• 68On the nature and consequences of these various criteria of wage determination, see Ford, Economics of Collective Bargaining, pp. 85–110.