5. The “Bargaining Power” Concept

MYTheory of Collective Bargaining contains a history and critical discussion of the notion of “labor’s disadvantage in bargaining,” an idea which, first introduced in Adam Smith’s Wealth of Nations, remains a powerful influence in popular opinion and has not disappeared from the field of labor economics. In Alchian and Allen’s excellent textbook, the concept of “bargaining power” is described categorically as “a vacuous concept of little analytical substance.”¹ I hope in this chapter to show that this description is justified. But the term is still widely used. It is treated with such respect by labor economists of high apparent authority that it is at least expedient in a work of this kind to examine the notion.

In a masterly work, Fritz Machlup has collected and classified a wide range of attempts at giving meaning to notions of “inequality of bargaining power.”² His rigorous examination of these ideas has exposed very effectively the intellectual muddle they have caused. But since his book appeared, the terms and phrases he criticized have continued to be used as though his contribution had never been made.³ In the following discussion, which is indebted to Machlup’s, I shall repeat the challenge in a different approach.

Since I treated the subject in 1930, my interpretation of Adam Smith’s famous passage about “labor’s disadvantage” has changed slightly. When he referred to “labor’s disadvantage in the dispute,” I now think he was not asserting that labor was at a disadvantage in the free market. He visualized two different things.

Firstly, he seems to have had in mind an attempt to force higher wage rates through an actual strike; and he apparently felt that, in the circumstances of his day, efforts to achieve sectional gains in that manner were more likely to fail than succeed (despite the violence which he believed often accompanied strikes). To the extent to which this was his point, there need be no argument. The size of a strike fund can, beyond doubt, influence the chances of the strikers getting better terms for those of their members who retain employment⁴ at the higher labor costs. And it is true that the individual worker can seldom effectively disrupt the production process,⁵ while a few workers among many may have hardly any ability to do so.

Secondly, Adam Smith believed that the masters possessed a tacit monopoly in purchasing labor—what we now call “oligopsony.” Now if workers who do not act collusively face a monopsonist or effective oligopsony (see Chapters 8 and 9), they may be said to be at a disadvantage. Again, there need be no argument on that point. But economists who followed Adam Smith began to use the idea of “labor’s disadvantage” in contexts in which, by implication, neither a strike situation nor the presence of monopsony is assumed.

The idea which this century has inherited is that combinations among the workers are needed to offset “labor’s disadvantage in bargaining” or “labor’s inferior bargaining power,” because the free market value of labor is depressed through some disadvantage other than monopsonistic exploitation. At times, however, monopsony is assumed, while at other times the “bargaining disadvantage” implied is the inability of a group of workers to exploit either the other cooperant parties to production or consumers, in the absence of concerted action. Sometimes one or more of these ideas seems to be woven into an argument expressed in terms of a “balance of power” between “employers” and “unions,” which collective bargaining somehow brings about. Usually the assumption is that the meaning of all these terms is self-evident. In fact, their use has seriously confused thinking about the wage-determining process.

We can begin by considering the phrase “balance of power.” If all this phrase is intended to mean is a situation in which no one can say with certainty which party will succeed in a wage dispute (whatever “success” can imply here),⁶ it has at least a modicum of meaning. It suggests that neither strikes nor management-ordered work stoppages will be attempted because, in the presence of this “balance,” the outcome is likely to be indecisive and hence resort to aggression too costly (see pp. 54-55). Such a connotation centers attention upon a threatened work stoppage and borrows a perfectly clear notion from political science. War between two powers is, according to the “balance-of-power” theory, unlikely if the armed strength of the two parties is about equal. In those conditions, according to the theory, neither side is certain that it will be victorious, while the costs of war are high. Similarly, the high costs of work stoppages help deter attempts to determine labor cost through the imposition of nonmarket values. Hence a “balance of power” will cause the maintenance of the status quo. This is obvious, of course. The burden of any work stoppage on both parties will always be weighed against the prospective gains from it. If both parties think that possible winnings do not justify the stakes, there will be no aggression. “Industrial peace” will prevail.

This is, however, seldom the sort of consideration which is in the minds of economists and others who use terms like “bargaining power,” “labor’s disadvantage in bargaining,” and so forth. To get near to what may be meant when monopoly-monopsony is not ruled out, we can, I think, simplify the question by the following approach. In every case the phrases we are examining suggest that, in the conclusion of any wage contract, the party with “strong bargaining power” will get better terms while the party with “weak bargaining power” will get worse ones. Let us suppose that what is envisaged has reference to whether the wage rates or prices determined are above or below what the free market value would be. If this is indeed what is meant, then the possibilities can be expressed with fair conceptual clarity in terms of monopolistic or monopsonistic influences. A simple diagram can represent the possibilities by representing the wage rates which would be determined under different assumptions.

OY₅ Monopolistic Labor against Competing Investors

OY₄ Monopolistic Labor against Monopsonistic Investors

OY₃ Competitive Labor against Competitive Investors

OY₂ Monopolistic Labor against Monopsonistic Investors

OY₁ Competitive Labor against Monopsonistic Investors

On the above diagram, wage-rate possibilities are represented on the vertical axis and numbers employed on the horizontal axis. OY₅ represents the wage-rate that a union would think to be to its members’ advantage to enforce (if it could) against uncombined, competing firms. It is described as “monopolistic labor against competing investors.” The numbers employed will be OX₁. OY₁ represents the opposite. It is the wage rate which managements (on behalf of investors) possessing the power to purchase labor monopsonistically might think it profitable to enforce (if they could) where the workers were wholly unorganized. It is described as “competitive labor against monopsonistic investors.” The numbers employed will be OX₁. OY₃ represents what the price of labor would be in a free market. It is described as “competitive labor against competing investors.” The numbers employed will be OX₅. OY₄ and OY₂. represent what we might assume about the extreme possibilities under what used to be called “bilateral monopoly,” i.e., with “monopolistic labor against monopsonistic investors.” The numbers employed will be OX₃ and OX₄ respectively, or somewhere between.

If by convention we always described wage rates higher than the level OY₃ as due to the “superior bargaining power” of labor (or the “inferior bargaining power” of investors) and those below the level OY₃ as due to the “superior bargaining power” of managements on behalf of investors (or the “inferior bargaining power” of labor), we should at least have intelligible notions. And I think we do find that all attempts at rigorous definition on this topic are groping toward a definition of the degree to which the price of labor diverges from the competitive level, OY₃. Thus, the term “bargaining power” could be used to refer to the ability of one or other party (a) to force a price or wage rate above or below the free market level, or (b) to neutralize, in whole or in part, an opposed monopsony or monopoly, thereby forcing a price or wage rate toward or across OY₃, in other words, toward or across what the free market level would have been.

Often, however, terms like “bargaining strength” or “labor’s disadvantage” are used in a manner which does not enable us to relate them simply to this conceptual framework. For instance, under monopsony, if A has higher-paid alternative employments than B, he may be paid more than B for an identical kind of work. B is then said to have “weaker bargaining power.” But A’s services are more valuable to society in other uses than B’s (although not in the particular job to which A is attracted, where they contribute no more to the value of the product than do B’s). If we describe A’s greater “opportunity value” as his “greater bargaining power,” we must bear in mind the full implications of the facts, firstly, that under competition for labor, no individual worker would have any advantage or disadvantage in relation to his competing comrades; and secondly, that even a monopsonist will employ no A’s for the job until there are no additional B’s available.

Actually, within any competitive labor market, where each worker is free to move to where he believes he can earn most, his “bargaining power” will rise as his services become more valuable through training or experience. The term is then simply a synonym for value! In every case the crucial consideration is the worker’s alternatives. Where competitive conditions rule, the refusal of any offer is simply a means of saying, “I have (or I believe I have) a more favorable alternative.” Such a communication may result in an improved offer. Where a worker (or his union on his behalf) can say truthfully to management, “I should like to work (or continue to work) for your undertaking but I think I can do better elsewhere,” that is, where he refuses an offer because he knows of or expects higher remuneration or better prospects in another firm, we all know what is meant when it is said that his “bargaining power” is strong.⁷ But it could then be argued that there is hardly any point in using that term. It only means that his free market value is as high as any other prospective offer. Of course, a person might successfully represent that the market value of his services is higher than it really is. We can then (if we wish) describe his bargaining bluff as “bargaining power.” It is not a very helpful usage.

Even in a noncompetitive market, we can still say that a worker’s “bargaining power” is represented by his “opportunity value,” which means the value of his alternatives; but we then mean that this power will be weak (the “opportunity value” low) when alternative employments can be somehow withheld from him.⁸ In Chapter 7, I shall explain that exploitation (whether monopolistic or monopsonistic in origin) occurs through the process of shutting off alternatives for those exploited. The important point to remember is that the “weak bargaining power” notion which I have here suggested might be used with meaning and consistency has reference to the individual worker. The notion cannot be simply transferred to labor in general. We cannot talk of “labor’s disadvantage in bargaining,” although we can discuss the individual’s. The remedy for the individual’s “bargaining weakness” is to raise the value of his work. His “bargaining power” depends (a) on his having scarce and valuable powers, which simply means that he can provide goods and services which consumers need, and (b) on his effective right to use those powers.

All the workers in an occupation may be exploited under monopsony, however, if managements can tie them under contracts which are not for their benefit, or shut them in in other ways. The remedy in that case is action to remove the barriers to other employments.

But monopoly or monopsony may influence the value of complementary services in the production process without exploitation, that is, without the monopolist or monopsonist shutting out any alternatives from any participant. As will become more clear in Chapter 8, discrimination is exploitative only when the monopsonist himself has in some way held off competing opportunities. Because the worker who has only lowly-paid alternatives has “weak bargaining power,” this does not mean that he is more exploitable (under the definition of “exploitation” we are using) than the worker who has relatively well-remunerated alternatives. Thus, a monopsonist may pay the workers he recruits (who have well-paid alternatives) a wage rate just sufficiently above their free market value in their former jobs to compensate for mobility costs, while those he retains (who have poorly-paid alternatives) may be paid just sufficiently to make it not worth their incurring the mobility costs of accepting other alternatives in the free market. As we are about to see, the lower wage rate needed to hold the latter does not mean that they are exploited under the definition we are using (see p. 3).

An often-used illustration, assuming (usually tacitly) monopsony, compares the worker who has “reserves” and can “hold out” with the case of the worker who is prepared to accept a lower wage rate than the average because he has no “reserves.” The typical example takes an untypical case, namely, that of an unemployed person. The argument is that if he has a large fund of savings, he can refuse what he thinks is a poor offer and for the time being purchase leisure, or finance his prospecting for a better opportunity, out of his capital. His less-thrifty colleague, on the other hand, who may have spent his last dollar, will be desperately in need of an immediate material income, so that a monopsonist will be able to discriminate against him.⁹

Now although such discrimination can occur if the monopsonist knows the worker’s situation, it is not exploitation unless the monopsonist himself has been able in some way to withhold employment opportunities from the worker in question. Let us consider a monopsonist who has closed no doors to any other employments, offering annual contracts of employment to individual employees through advertisements, and again assume that he recruits two classes of workers, those we can call As, who have well remunerated alternative employments and those we can call Bs, who have relatively poorly remunerated alternatives. Let us imagine also that the wage rates he offers initially are on the low side, in order to test the market. The first workers to accept the terms offered and to tie themselves for, say, a year (the Bs) will be those who can better their condition most thereby (because they have but poor alternatives, including poor “reserves”). If the monopsonist has been unable to attract all the labor he needs by the first offer (through which he recruits Bs), we can imagine that he will be able to attract some As, by offering better terms to them. It is important to notice that the condition of the Bs is improved, not depressed, in relation to their initial condition. The contracts under which the Bs earn less than the As are accepted by the former because they open the way to more highly paid alternatives than society is offering in other ways.

It can indeed be to the Bs’ advantage that they shall be discriminated against. The principle is particularly clear when economies of scale effect “natural monopsony” (either as the inevitable concomitant of “natural monopoly” or independently.¹⁰ For it is conceivable that only under the economies achievable through discrimination against the Bs will it be possible for the new, naturally monopsonistic venture that raises their earning power to be established at all. The general principle here (to be explained in Chapters 8 and 12) is that discrimination is justifiable (in the interests of the “optimization” of the community’s “welfare”) when the parties discriminated against are nevertheless the beneficiaries of that discrimination.

Hence phrases like “the workers’ bargaining strength,” in referring to each individual’s alternatives, may envisage either (a) alternatives determined in the free market or (b) alternatives influenced by a monopolist or a monopsonist. In the former case, no exploitation is involved. In the latter case, exploitation may be a factor. But labor’s “bargaining power” in a different sense will be influenced by the elasticity of demand for the end product (which will determine the exploitability of consumers) and by the elasticity of supply of complementary factors (which will determine the exploitability of suppliers of raw materials and/or investors in fixed capital). “Exploitative power” would be more apt.

When “bargaining strength” is sought by the unions by way of the strike threat, that itself involves what in practice is the most important shutting off from alternatives for the less-fortunate workers. Any duress-imposed wage rate in excess of the free market level denies access for some to jobs of higher productivity and remuneration. It may be held therefore to reduce their “bargaining power.” At the same time, it means that the “bargaining power” of a privileged group is strengthened. The principle is, then, that any individual denied access to any bargaining table is usually left with curtailed “bargaining power” in the employment outlets for which he is allowed to bargain. The depleted earnings he must accept in order to get immediate or early alternatives to the job he has lost, and his weakened security of employment, are consequences of restraints enforced through the strike threat.

In a competitive labor market, a temporarily displaced worker is confronted with a wide range of “take-it-or-leave-it” choices, just as is a shopper regarding a range of competing commodities and competing shops. Under such conditions, managements buy labor through wage-rate offers and shops sell goods through price offers which those who want employment (or better paid employment) or goods accept. It is said, however, that the worker is “at a disadvantage” compared to the shopper because he has no “waiting power.” But as we have seen, the lack of “reserves” is no disadvantage in a competitive market. Shoppers equally have no “waiting power” regarding certain of their purchases. They must buy food for the immediate future or starve.

The word “monopoly” is conventionally used to describe what Ludwig von Mises has argued cogently ought to be called the union’s “supply restriction.” Mises’s case is that “monopolistic action” is advantageous to a monopolist only if total proceeds at a monopoly price exceed total net proceeds at the potential competitive price,¹¹ That is, the monopolist must allow for the loss in respect of the capacity he withholds; whereas the organizers of supply restrictions, like unions, “are not concerned with what may happen to the part of the supply they bar from access to the market. The fate of the people who own this part does not matter to them.”¹² The important but generally overlooked difference in principle to which Mises is here referring is one which I tried to get round in 1935 by introducing the term “contrived scarcity” (contrasted with “natural scarcity”).¹³ in Chapter 7, I shall use this and related concepts in an attempt to clarify the issue further. But when I use the term “monopolistic restriction” or “monopsonistic restriction” I shall have in mind the shutting out or shutting in of those who believe they could otherwise improve their earning power or profits by moving their labor or their capital to another occupation, industry, firm, or area. And every such “restriction” (or “shutting out” or “shutting in”) involves “exploitation.”

I propose now to draw the reader’s attention to some typical uses of the term “bargaining power” in contexts in which the market alternatives available to the bargainer are obviously not what the writer has in mind. Suppose we consider the situation which arises when unionized workers, acting in concert, are in a position to say to management, in the friendliest way, “We have the power to ruin your stockholders. We don’t want to do that. A fight will involve some costs to us; but we shall feel forced to use this power unless you agree to our very reasonable terms. If you refuse to make the consumer pay a modest extra sum on our behalf, you will have to transfer some of your investors’ income to us. What we ask won’t actually ruin your undertaking. We know what your profits are. If you want what is good for your stockholders, you will agree to our terms.”

This coercive power may, if we wish, be termed “bargaining power.” But it is then an exact parallel to the “bargaining power” of the salesman of jukeboxes, gambling machines, and vending machines, who (in the United States) sometimes tells his prospective customers, with the greatest politeness, “If you know what is good for you, you will sign this agreement for the installation of one of these machines.” Every customer then knows that failure to agree will bring costly damage to his shop or personal violence against himself.

“Bargaining power” in that sense bears no resemblance to the ability to command a certain wage rate because of the existence of actual or potential competing offers. Hence although we might say that that party’s bargaining power is greatest which can convince the opposing party that it is in a position to do that party the greater harm, ought we not to convey our meaning by a simple descriptive term and say that the “stronger” party is that which has the greater “coercive power”? If I am stopped by a hijacker who has a gun while I have only fists, isn’t it just a misleading euphemism to refer to his superior “bargaining power”?

However, if the term is defined to mean “power to disrupt and hence to coerce,” it is both realistic and understandable, but it cannot then be used in the other connotations! Presumably managements rely upon a similar power if they themselves order a work stoppage with coercive intentions. But the literature of the labor movement leaves the impression that the actual use of a parallel power by management is virtually unknown and the threat almost as rare. Mere refusal to accept union conditions is not a work stoppage. As we have seen, there is a continuous offer of employment from management’s side, at stipulated wage rates, just as there is a continuous offer of goods, at stipulated prices, in a shop.¹⁴

Moreover, the downward adjustment of wage-rate offers when a wage contract is renewed, where necessary to price some current output within reach of the community’s income (that is, following reduced entrepreneurial bidding for the available labor supply), is not coercive. It is a means of avoiding (a) the displacement of labor and (b) (from society’s angle) the depletion of the wages flow. It is no more disruptive than is inflation which, since the Keynesian era, has been a crude alternative way of reducing real wage rates (as a means of mitigating the depletion of the wages flow caused through strike-threat action). The step taken by managements is not evidence of their superior “bargaining strength” but a consequence of their subordination to consumers’ sovereignty, as representatives of the residual claimants on the value of the product.

I know of no recorded case in which general wage-rate reduction has been used as a threat by managements, or punitively. But the workers in an undertaking have often been warned about the impossibility of continuing the employment of former numbers in the event of a forced wage-rate increase. It is easy enough to describe such warnings as “threats”; but they are no more coercive than warnings that smoking is liable to cause lung cancer. Admittedly, in the course of negotiations conducted in the shadow of the strike threat, managements are likely to paint the consequences of enhanced labor costs in unduly somber colors. Hence there may sometimes be a “threat” element present.¹⁵ In fact, I believe, the threat of disruption as a “bargaining weapon” (in the sense of a means of coercion) is used almost entirely by the unions. But no matter what party may actually be guilty of using it, either the threat of or the use of private coercion in the “bargaining process” is intolerable.

N. W. Chamberlain brings in such factors as “the pressure of immigration, the cityward movement of farm population, the speed of mechanization and mass production techniques” among the circumstances which create “labor’s disadvantage,” because they “place employment at a premium.”¹⁶ But all this means is that one’s “bargaining power” is raised or lowered by anything which causes demand for what one has to sell to rise or fall or its competing supply to fall or rise! Chamberlain may possibly intend, however, that these are circumstances which limit the exploitative power of the strike threat. He regards labor’s “bargaining weakness” as overcome when “by common action workers could prevent themselves from being played off one against the other.”¹⁷ But the “playing-off” of one would-be seller or buyer against another is the only way in which any party entering into a contract can protect himself against exploitation; and the collusive action which is here represented as rectifying the “bargaining disadvantage” turns out merely to be a particular method of contriving a scarcity. In other words, “labor is at a disadvantage in bargaining” comes to mean, “labor finds it difficult to exploit displaced or excluded workers, consumers, risk-takers and the providers of complementary assets in production.”

Most discussions of this issue view the parties to negotiations over a wage contract, under the strike-threat shadow, in a most unrealistic way. They portray the workers (or the union) on the one side and the “employer” (presumably management representing investors) on the other side. But the reality that successful wage negotiations require the expression of agreements or settlements in a form which maintains the prestige (and hence maintenance in office) of the union officials is seldom mentioned (see above, p. 47).

In practice this is often (if not most often) the vital consideration. As skilled negotiators, managements understand the expediency of allowing the union officials to receive the whole credit for every improvement in wage contracts. When managements make concessions, they must do so in such a way that the union rulers will be able to show a capitulation on the part of the enemy—“the employer.” Indeed, managements normally permit every intermittent restoration, full or partial, of real remuneration in the course of an inflation to be claimed as a union triumph. And more generally, when the maximum concessions managements can contemplate are small, it is at times good strategy for them to stage a fight against certain fringe benefits demanded, which cost little. After thus making an effective show of opposition, their (the managements’) eventual retreat will preserve the illusion of successful strike-threat pressures, and hence make it possible or expedient for the leaders of the unions to recommend acceptance of the terms. Occasionally, by such tactics, managements can get away with relatively small concessions. In some cases indeed, the prestige of the union officials can be protected by the acceptance of conditions of which the burden may fall wholly on the union membership itself.¹⁸

Moreover, what actually happens in almost every case when both parties have become accustomed to, and experts in, the “bargaining process” is that managements begin negotiations with a fairly clear idea of the limits to which they are prepared to go in concessions (in order to avoid an actual stoppage). They will declare at the outset their determination not to concede more than a very small part of what they expect to concede; while unions will begin by asking for more than they believe managements will concede (with or without actual recourse to the strike). For instance, managements which, in a free labor market, would have found it profitable to offer, say, $400 per month in order to retain (from alternative employments available) the most profitable number of workers may think it good tactics to begin by offering, say, $350. They know that, in the market circumstances ruling, such remuneration offered would be disadvantageous (or even ruinous) to them because it would mean their losing too many of their most valued personnel; but they regard their initial offer, maintained for a time, as serving the practical purpose of enabling an apparent capitulation; for such a capitulation is above all essential in order to satisfy the union leaders.¹⁹ If we are realistic, then, we must perceive that there are three parties involved in “collective bargaining” over wage contracts. It is not the conflicting interests of the workers in relation to the interests of the investors which constitute the really crucial issue in the struggle, but the interests of the elected rulers of the union.

In thus insisting that the essential condition is the satisfaction of the union rulers, I am envisaging as realistically as I can the internal democracy which is characteristic of the majority (but by no means all) unions. In general it is true that, in B. C. Roberts’s words, “while there is little direct check upon leaders during the course of negotiations, they must be able to . . . secure the votes of members in referendums and elections for office.”²⁰ But it is just this concern of the leaders with reelection that is the critical issue with which managements are confronted during wage negotiations. Union officials are faced with the dilemma that they do not wish to kill the goose; and sometimes they clearly do not wish to harm the general economic conditions which provide sustenance for the goose (see pp. 140-142). Yet most union members expect immediate results, while rivals for power are prepared to claim that they could win more. Indeed, there will often be competitors for office who are as unscrupulous and irresponsible in promise-making as the typical politician during election campaigns!

We sometimes hear it said that it is the duty of unions and of managements to bargain in good faith, not merely to demand. Does this mean that it is their duty not to dig their heels in and refuse to make concessions? If so, it implies that it is the duty of both sides to begin by asking more favorable terms than they are prepared eventually to accept. What else can it mean? Because the strike-threat system exists, it may well be good tactics—even inevitable tactics—for managements to carry on more or less continuous discussions with union officials. But recognition of this practical reality must not blind us to the basic principle that it is managements’ task to offer such wage rates as they predict will justify the purchase of labor services for investment into work in progress; and this must be seen as a facet of the continuous process of investment of stockholders’ capital into the retention, replacement, or accumulation of inventories of materials and fixed capital.

This does not mean that under nonstrike negotiations the wage rates set can be said to have been determined “unilaterally,” “without negotiation,” “dictatorially,” and therefore unjustly. The word “unilateral” which has been used in this connection is a red herring. Collective bargaining is not an alternative to unilateral decision-making. Every offer accepted must be bilateral. Prices are marked for goods offered in shops but this does not prevent every purchase from being a bilateral transaction. Would a system of haggling over every retail transaction result in greater consumer freedom, security, or justice? Similarly, no employee is forced to retain or accept any employment at the remuneration set by any management in a free labor market. On the termination of any wage contract he may leave, as he will, if he can get better terms or prospects elsewhere. It is that which, as we have seen, constitutes his “bargaining power” (in the least unsatisfactory use of that term). And a management may be just as helpless in respect of the wage rate at which it forecasts that the most profitable number of workers can be retained by or attracted to the enterprise as is a shopkeeper in fixing prices which he predicts will retain or attract the most profitable number of customers (when he is not in a position to act monopolistically). (See p. 14.)

Where no unions exist, one management makes offers for different types and grades of workers on behalf of many stockholders, while an individual decision to accept an offer must occur before any person becomes an employee or continues as an employee. On the other hand, at the bargaining table, a union may accept or reject offers not specifically on behalf of individuals but on behalf of large groups of those it represents. That is not a necessary arrangement, however, whereas it is impossible for each individual stockholder to make a separate contract with every worker; for the plant and the firm constitute a unity. The only parallel to this unity on the labor side is when several persons form a partnership, like a cooperative theatrical group, and offer their services as a whole to the highest bidder. But then they have usually already arranged among themselves for their relative individual remuneration. And dealing in that case with the executive of the group, or with the union executive, does not cause the decision-making to be more (or less) bilateral.

Ideally, collective bargaining involves the union officials negotiating, so to speak, bulk contracts for groups of individuals. In some circumstances, we can regard all the workers employed as having agreed, in advance, to accept whatever terms the officials accept on their behalf, subject to their retaining employment under those terms. But if union officials are seen as servants and not the masters of union members, each member should be free, if he should so wish, to make an independent contract with management, especially those who might otherwise be laid off.

As every worker knows if he participates in sports, human abilities in different directions vary considerably. Hence bargaining freedom requires that those who contribute differentially to the satisfaction of consumers’ demands should be allowed to make available their services for remuneration in proportion to what they believe the value of their contribution to be. And that value may be higher than or (when the alternative is displacement) less than the wage rate which a union might have negotiated. “Bargaining justice” requires therefore that while a union should retain the right to advise an individual against what is thought to be his wrong judgment of the value of his services, that word advise should not be allowed to be interpreted in the way in which the words “persuade” or “induce” have come to be so often interpreted, namely, as synonyms for “intimidate” or “coerce.” The individual’s “bargaining power” means, indeed, his “bargaining freedom.”

Let us consider in this context the case of a worker whose way of life faces the prospect of disruption through his threatened displacement following a change in consumer preference. The right of such a person to offer his services at a reduced wage rate in order to avoid displacement (which consumers would otherwise enforce) ought to be recognized as the source of a basic security—a fundamental freedom. Even if he is confronted with the dilemma of accepting or rejecting a wage rate set below what he believes will ultimately be his free market value, he can still retain an income by temporarily accepting a wage cut, and live on the curtailed income while he (or his union) is looking for an opening remunerated at what would perhaps be the present value of his services in a better-coordinated society.

When industry-wide bargaining is enforced (as it is for instance in the steel, coal, automobile, tire, trucking and other industries in the United States) a monopsonistic structure representing the managers of the producing units is almost necessarily created. But in the absence of agreements not to “poach” labor (and assuming that antitrust is effective) individual managements will be competing for such labor supply as the standard rate leaves profitably employable by the industry as a whole. The real danger in these circumstances is not of exploitation of labor but of the public; for with the support of unions, corporations with which the wage contract is made have the power to create “joint monopolies” with labor,²¹ or to encourage unions to bring under their wing the employees of concerns (perhaps of a different type) which, by reason of some greater technical efficiency or superior price policy, appear to be taking business from them.²²

Unions are fully conscious of their exemption from antitrust or similar penalties, and whenever competition from substitutes appears, they may recognize a solidarity of interest with stockholders. What, then, becomes of the notion of labor’s supposedly inferior “bargaining power?” A divergence of interest about the division of the spoils of joint monopoly will remain, and the “just” sharing of gains at consumers’ expense will be as indeterminate as the “just” sharing of stolen booty always is. This is one of the cases in which the “bargaining power” notion seems to be concerned with who will get the biggest share of ill-gotten gains; that is, the case in which the superior “bargaining power” will reside in the party which is apparently in a position to harm the most. When unions and managements resort to such collusion, unions will, naturally enough, express concern for justice toward their competitors in the expanding enterprises which are offering substitutes that are better value for money. They will want the interlopers to be brought into one “bargaining unit” for the protection, they will say, of their competitors as well as of themselves.²³ The merging of the steel and the aluminum workers in the United States, as technical progress caused formerly complementary products to become competing products, is a case in point. But the “protection” achieved is essentially of the power to exploit.

Let us now return to the question of the actual determination of a wage agreement when the strike threat is an influence. If an offer by the management of a corporation is at first refused by a union, there is very little that the economist can usefully say about the terms which are likely to be eventually accepted. That is, the determinants of the wage agreement (wage rates plus fringe benefits for work during a certain period) can then be subjected to purposeful economic analysis only in the sense that the use of resources in the course of warfare can be studied.²⁴

Now in warfare, the ultimate and overriding objective, namely, victory, is a product to one party and a negative product (namely, a deprivation) to the other, and there seems to be little purpose in showing that, if we have sufficient data, we may be able to forecast the probable result of any resort to the strike threat (or to an actual strike). The result is certainly not relevant to the vital question that concerns society today: Can we tolerate economic warfare unless wehave satisfactory grounds for believing (a) that it is inevitable, or (b) that it is likely to result in victory for the good and defeat for the wicked? Obviously, analysis of strikes and the strike threat can throw no light on the relevance of their outcome to the socially desirable division of the value of output. Yet it is just this relevance with which many writings on collective bargaining have ostensibly been chiefly concerned.

Attempts have been made nevertheless to study rigorously the factors which are likely to lead one or the other party to victory in the clash between the unions and the “employers.” Studies in this field may be said to have begun with J. R. Hicks’s Theory of Wages (London, 1932), although Edgeworth, Marshall, Bowley and others had previously discussed the indeterminateness of value under bilateral monopoly.²⁵ But in my judgment, Hicks’ discussion and all subsequent explorations in this field, down to recent contributions by N. W. Chamberlain,²⁶ C. Stevens,²⁷ B. D. Mabry²⁸ and others have done little more than elaborate the truism that, in warfare of the kind analyzed, and looking at the issue from the standpoint of one of the parties, (a) the greater the resources in supplies and weapons of aggression possessed by that party, and (b) the less onerous the terms of surrender it offers to the other side, the more likely it will be that that party will be able to force the capitulation of its opponent, with or without an actual strike or management-imposed work stoppage; and that, in the event of a party meeting more stubborn resistance than it had expected, it may have an incentive to soften the terms of surrender it demands.

I conclude that “labor’s inferior bargaining power,” which is said to be strengthened by agreements in concert to refuse a particular employment at less than a certain wage rate, refers most often to monopolistic power to exploit three groups, namely, and in order of importance: firstly, displaced or excluded comrades; secondly, consumers; and thirdly, the suppliers of complementary productive services, including those investors in fixed resources who have not adequately allowed for the risks arising from the strike threat when they have invested. Only when monopsonistic power can be wielded by managements to maintain or reduce wage rates below the free market level, by somehow excluding access to alternative employments, can it be claimed that phrases like “labor’s disadvantage in bargaining” are anything more than unintended euphemisms.

NOTES

¹ Armen A. Alchian and William R. Allen, University Economics (2nd ed.: Belmont, Calif.: Wadsworth Publishing Company, 1967), p. 420.

² Fritz Machlup, The Political Economy of Monopoly (Baltimore: Johns Hopkins Press, 1952), pp. 333-58.

³ To the best of my knowledge there has been no attempt at refutation of the chapter in question.

⁴ The terms of a so-called wage contract, “agreed to” under coercion, seldom binds any firm to provide so much employment at the stipulated wage rate. The firm merely binds itself not to employ any person who might find that he can better his condition by working for less than the “agreed” figure.

⁵ I say “seldom” because, as we have seen, a single key worker may, at a crucial point of time, be in a position to disrupt operations by the withdrawal of his services.

⁶ The use of the strike threat or the strike may be said to be “successful” from the standpoint of those who retain employment at a higher real wage rate. Those who are laid-off in consequence, or whose prospects are damaged, will (if they perceive how they are affected) regard it differently.

⁷“What bargaining power may mean . . . is simply the highest salary one can get from other jobs” (Alchian and Allen, op. cit., p. 402).

⁸ His remuneration could then be represented as OY₁, in the diagram on p. 63.

⁹ Actually, as Alchian and Allen point out, “the employee does not lose his entire source of income; he loses the premium he was getting in his former job over the next best alternative adjusted for moving and job-exploration costs,” (p. 402).

¹⁰ The adjective “natural” describes monopolies or monopsonies which emerge without collusion. (See pp. 103, et seq.)

¹¹ Ludwig von Mises, Human Action (New Haven: Yale University Press, 1949), p. 374.

¹² Ibid., p. 373.

¹³ W. H. Hutt, “Natural and Contrived Scarcities,” South African Journal of Economics, 1935.

¹⁴ I return shortly to the suggestion that for managements simply to make wage offers would be for them to determine labor’s remuneration “unilaterally,” that is, without bargaining (see p. 71).

¹⁵ The real danger today, however, is that managers who explain objectively the market factors through which social discipline could replace economic warfare, and who, in so doing, point to the probable immediate labor displacement consequences when labor costs are determined under the strike threat, may be accused and even convicted, in the United States, of “unfair labor practices.” I have been assured that, if a negotiator points out that the number of workers employable in an undertaking will increase more slowly, cease to grow, or even decline as a result of consumers being forced to pay more for output, his assertion might be ruled by the NLRB as in the nature of a threat to rob the workers of their livelihood!

¹⁶ N. W. Chamberlain, Collective Bargaining, 2nd ed. (New York: McGraw-Hill, 1965), p. 123.

¹⁷ Ibid., p. 124.

¹⁸As one shrewd commentator has put it, “the novel feature in the compulsory (pension) plans promoted by most union leaderships does not lie in what the unions take from the companies but rather from what they take from their own members, namely, the power to decide freely on how to dispose of each member’s income.” Philip D. Bradley, Involuntary Participation in Unionism, in Labor Unions and Public Policy (Washington: American Enterprise Association, 1958), pp. 57-58.

¹⁹ Hence it is theoretically possible that, through bargaining tactics, the settlement will determine wage rates no higher than the free market would have guaranteed. Through such a stratagem, managements may be said to have performed their duty to the community as well as to stockholders.

²⁰ B. C. Roberts, in John Dunlop, Theory of Wage Determination (London: Macmillan, Ltd., 1957), p. 109.

²¹ See the next paragraph and pp. 50, 122, 128.

²² For example, some of the self-service stores in the United States, confronted with the more effective economies of the discount houses, seem once to have followed this (probably short-sighted) policy.

²³ The reader should be reminded that the expanding enterprises of which the progress is thus curbed had been attracting labor from less productive to more productive and more highly remunerated work.

²⁴ The concept of “economic” is not irrelevant in war operations. In rationally conducted warfare, although forecasts in detail, and hence proximate objectives, are in process of constant revision, or even fundamental change (as the pattern of events confirms or negates forecasts), each proximate objective can still be pursued in a manner calculated to minimize detriment to other proximate objectives.

²⁵ If there is a corporation with monopsonistic powers facing a labor union with monopolistic powers there will be, for a certain range, nothing resembling a market price to determine the wage rate (e.g., on the diagram on p. 63, the range will be between OY₂ and OY₄). The position will be similar to that which exists under “pure barter.” Marshall illustrated the situation by the exchange of nuts and apples between two persons in isolation. (See above pp. 62-63.)

²⁶ N. W. Chamberlain, Collective Bargaining, 2nd ed. (New York: McGraw-Hill, 1965).

²⁷ C. Stevens, “On the Theory of Negotiation,” Quarterly Journal of Economics, February 1958; Strategy and Collective Bargaining Negotiations (New York: McGraw-Hill, 1963).

²⁸ B. D. Mabry, Labor Relations and Collective Bargaining (New York: Ronald Press, 1966). Mabry’s book contains a short bibliography covering this topic (pp. 239-240).