The Subversion of Collective Bargaining
Unhappy is a society that has run out of words to describe what is going on. So Thurman Arnold observed in connection with the language of private property—the myths and folklore of capitalism—which even thirty years ago was hopelessly out of date. How to find real words to describe the recent strike in the steel industry, or the consequences of the wage-price negotiations of the past decade? Two parties are locked in struggle, each seeking to articulate its claims over the other, while from the sidelines arise the moralistic alarms of spectators worrying about damage to the innocent public. But the desiccated language of collective bargaining is a trap; its syntax too constricting, its images too mechanical. The complex fact? The combat is a mimetic one, painfully real in the sense that emotions are aroused, but unreal because no economic loss can occur; in fact, each party, knowing in advance the price it will have to pay, pretty much gets what it sets out to get, and both end up with a profit—the corporation, usually, the greater gainer.
I am not suggesting that all this, like wrestling, is “fixed.” Far from it. The antagonism between the contenders is quite genuine. But a highly intricate mechanism is at work in the game, and by now each side knows the unwritten rules. Sociologists have a phrase, “the unanticipated consequences of purposive actions,” to suggest that things don’t always work out as planned; and this is often the way of the world. As far as the corporations are concerned, the pattern of collective bargaining in this country follows an opposite principle which can be called “the utilized consequences of non-purposive action,” meaning that even if you didn’t plan it that way, you can turn it to your advantage.
Over the past decade, the corporations have precisely learned how to turn the collective bargaining process—and the strikes—to their advantage. The powerful unions gain impressive wage increases; the powerful corporations gain an excuse for impressive price increases—which, in the case of steel and auto, have in almost every instance been more than proportional to the jump in wages. Who loses? Unorganized workers (e.g. textiles), workers in marginal industries, rentiers, pensioners, and the like. Is this just? It is hard to define an equitable standard. A deflationary situation would benefit the rentiers and pensioners. But why should this group gain rather than another? The present situation reflects existing market power, which in turn shapes the rules of the game. The first thing to be determined is not who wins and who loses, but the nature of the game itself, and whether it ought to be revised. The following, therefore, tries to sketch some basic characteristics of the current wage-price situation, to puncture some myths, to delineate some consequences, and to present some alternatives.
The single most important fact about contemporary corporate capitalism is that expansion comes about through “self-financing,” through retained earnings derived from high, protected prices. In formal theory—the mythical language of private property—a firm went to the capital market for financing. It floated stock, people “risked” their money and got a share of the enterprise as their equity. Later, corporations went in for institutional borrowing; insurance companies or banks would lend large sums of money to a firm, taking debentures or preferred stock in return. In either case, some outside control theoretically existed; a legal equity was always exchanged for the money raised. The actual situation is vastly different. Few firms today, except for utilities, go to the capital market for funds. Tax laws make it costly to distribute all retained earnings as dividends and then “recoup” the capital, by the investors’ fiat, through new stock. The managerial decision to utilize retained earnings for expansion allows the managers to reinforce their social power and give them independence from outside control; their ideas become the decisive factor in determining the social use of capital surplus.
But it is equally important to understand that when expansion is financed through high protected prices, it is the consumer who does the financing, and he neither receives equity in the firm (not even the promise of future price reductions) nor has any say about how his money should be used. In effect, the whole process depends upon what I shall call a “hidden tax mechanism” through which corporations can raise huge sums of money without giving away anything in return. Public taxation is openly and hotly debated in Congress and in the legislatures; bills involving the raising or spending of money are subject to all kinds of pressure and become the cause of great political divisions. Yet under the banner of “free enterprise,” a corporation can, through a protected price policy, “tax” consumers for its own purposes and do whatever it wishes with the money. In consequence of the recent expansion of steel capacity, for example, the industry—at present demand under protected prices—can supply all the steel the country uses in nine months rather than twelve. This means that the industry can “take” a three-month strike almost without reducing the average profit it would have made had no strike been provoked. In short, the strike is—that is to say, was—financed by the consumers.
The key term in the above argument, of course, is the concept of “protected price.” A firm is interested primarily in its profit margins. If sales fall, the firm cuts production—and employment—rather than price. Firms with some degree of market control can do this. The chief complaint of the farmers is that, being unorganized—other than through government crop reduction programs—they cannot adjust production to demand, but have to let prices fall.
To anyone who has read with care the Kefauver Committee reports of 1958 on “administered prices” in the steel and auto industries, it will be clear that these industries exercise an extraordinary degree of market control and thus have been able to place themselves in a protected position. We can best understand how such control is exercised by looking at the “standard volume” system for setting prices used by the auto industry. This system, which was developed by Donaldson Brown for General Motors in 1924, is based on an equation of three variables—price, estimated average rate of plant operation calculated in terms of a percentage of total annual capacity, and net return on investment. The price set for a single car is thus a function of the other two variables. But how are these variables determined? Net return on investment is simple: General Motors has decided that it must get roughly 20 per cent a year after taxes every year. “Estimated average rate of plant operation” is more complicated, however. The company figures in its best year on reaching only 80 per cent of its theoretical maximum operating capacity because of seasonal and other fluctuations in sales; in an average year, it figures on reaching 80 per cent of the production that can be achieved in its best year: thus it figures, theoretically, on using 64 per cent of its capacity in any normal year. But in actual practice, the “standard volume” has generally been calculated on a 55 per cent capacity. That is, General Motors so sets its prices as to plan for a return of about 20 per cent after taxes on the assumption that its plants will operate through the year for a total of only 180 days, or 36 weeks. (General Motors could “take” a four-month strike and still come out at its predetermined margin by operating for the rest of the year at full capacity.)
The long-range target of “standard volume” is to make it possible for General Motors to recoup its net investment in five years, but this goal has been surpassed by a phenomenal margin. In 1955, for example, net earnings (after interest and income taxes) were sufficient for the company to recoup its entire net plant investment in only two years. The American Institute of Management, which made this calculation, pointed out that such a record was not exceptional for General Motors, but was, “in fact, a continuing characteristic of the enterprise being equalled or bettered in 12 of the preceding 20 years.”
From 1950 to 1957, for every year except the last, General Motors’ actual sales were, on the average, about 30 per cent higher than the “standard volume” on which the company set its prices.1 When asked whether some of these gains from the large volume should not have been passed along to consumers through lower prices, Harlow Curtice, then head of General Motors, told the Kefauver Committee: “[Our prices] are as low as they can be and still produce the indicated return on the net worth at the standard volume.” Even, it would seem, when actual output was 50 per cent greater than the “standard volume.” As the Kefauver Committee said rather stodgily in its report: “It is clear that the use of standard volume as the basis of pro-rating expected costs and the desired aggregate profit in order to establish prices adds a considerable element of rigidity to these prices.”
What this has meant in terms of the price power of General Motors can be grasped by tracing the company’s net worth. In 1947, General Motors had a net worth of $1,428,-000,000—on which it made a return, before taxes, of 38.8 per cent. A decade later, General Motors’ net worth was $4,582,000,000—on which it made a return of 35.6 per cent before taxes. Of this increase in net worth of more than 3 billion dollars (or 221 per cent), all but $395 million came from profits which were ploughed back into the company. In short, the increase came from consumers who were making an “involuntary investment” in General Motors.
One important clue to the efficiency of a company and its ability to reduce prices is “the break-even point”—a measure that is based on the relationship between costs (divided into fixed and variable) and sales—and that gives us the figure at which the company begins turning a profit. Computations by Mr. Fred Gardner, a prominent management consultant, indicated that, including a high allowance for depreciation, General Motors’ “break-even point” in 1956 was 48.8 per cent of sales. Sales, of course, do not represent full capacity; if one took full capacity into account, General Motors’ “break-even point” would probably come to somewhere between 40 and 45 per cent of capacity.
These figures become even more significant in the light of the fact that Ford’s “break-even point” in 1956 was 64.7 per cent and Chrysler’s 87.4. Clearly General Motors has little to fear in the way of real price competition from the other automobile companies. In a serious price war, General Motors, with its superior efficiency, could probably run Ford and Chrysler into the ground. It doesn’t do so, first, because a position as a single auto monopolist would simply invite public regulation, and second, because Chrysler and Ford, as the marginal firms, hold up a neat “price umbrella” for General Motors.
Since the products of the steel industry are more diversified than those of the auto companies, the steel companies do not use any such simple measure as “standard volume” for setting prices. In general, the industry figures on making a 15 per cent net return on investment when operating at 100 per cent of capacity. U. S. Steel argues that every dollar’s worth of increase in employment costs will lead to an increase of more than $2 in total costs of production, and sets its prices accordingly. How does this claim square with U. S. Steel’s “breakeven point”?
In its presentation to a government commission investigating monopoly in 1937 (the Temporary National Economic Commission), U. S. Steel estimated that its “breakeven point” then was 63.3 per cent of sales. In 1956, an analysis by management consultant Gardner for the Kefauver Committee showed the “break-even point” to be 44.3 per cent of sales. After the price increases in 1957—which had followed a large union wage increase—the “break-even point” went down to 38.6 per cent of sales, or (since sales are lower than capacity) about 32 per cent of capacity. In other words, by working less than two full days a week, U. S. Steel could move out of the “red” and make money.
How low should a “break-even point” be? The average “break-even point” of all U. S. industry is roughly 50 per cent of capacity. A “break-even point” between 50 and 60 per cent of capacity is considered “sound” since it gives most companies a margin wide enough to cover their fixed costs if sales fall sharply. In twenty-five years of computing “break-even points” for 1500 companies, Mr. Gardner testified, the lowest he ever encountered was that of U. S. Steel. The Corporation, he said, could cut its prices by 10 per cent, and still end up with a “break-even point” of 50 per cent of capacity.
The nub of the analysis, as applied specifically to the wage-price situation of 1957 (the last major wage-price increase in steel), is that when wages went up, prices—and profits—went up even higher. After the new contract with the union had been put into practice, a correlational technique showed a higher rate of profit at a lower rate of operating costs than in the previous year, and a decline in the “break-even point” as well. In brief, it was quite clear that the steel companies had used the negotiations as an excuse for boosting prices, in order to jack up their profit margins. As the Kefauver Committee concluded, “U. S. Steel [can] cover its costs at an operating rate below 40 per cent of capacity and make very satisfactory profits while a substantial part of its capacity lies idle.”
The point of all this (to return to the role of collective bargaining) is that the net effect of union pressure—apart from the gains which have been won for the small group of highly organized workers—has been to help install a mechanism whereby the large corporation is able to strengthen its price position in the market. In the past, price protection was achieved by “basing point” systems (now outlawed), price umbrellas (in which U. S. Steel set the lead), or informal collusion. Today the union serves as the vehicle. (According to Walter Reuther, for every dollar of increased labor costs since 1947, General Motors by 1956 imposed about $3.75 in cumulative price increases on the American car buyer. In effect, the United Auto Workers, taking a small share of the increased profits, had become, albeit unwillingly, the “junior partner” of General Motors.) The companies can truthfully say that they do not like the union negotiations, since other than wage demands are often involved (work rules, fringe benefits, etc). And the companies are usually inclined to resist the union’s demands strenuously. But it invariably turns out that the union negotiation offers a lovely opportunity to increase prices—and, with exquisite irony, to blame the union for inflation.
Are the unions responsible for inflation? Industry’s argument is that they are, because by raising costs, they set off a wage-price spiral. But simple economic logic exposes the patent falsity of this charge. To determine the true effect of union wage pressure, one has first to make a distinction between the structure of wages (i.e., the relative spread between industries—say, steel and textiles), and the level of wages, which is the total wage bill in relation to other economic factors. What union pressure may do is to affect the structure of wages: that is, it can increase the gap between one group of workers (who have a strong union) and another (which does not). It is quite possible that wage and subsequent price increases in one area of the economy may have a linked effect on others—though with all the propaganda about the wage-price spiral, the actual spread of this effect has never been traced, and even so eminent a conservative economist as Milton Friedman of the University of Chicago doubts that it can go very far.2
For actually the degree of impact of a wage increase in one area on the rest of the economy depends, simply, upon the stock of money in circulation. If this stock were held constant, then an increase in wages and prices in one sector could only cause a shift in the share of money to that particular sector—provided it were strong enough to impose its increases (i.e., provided that people needed the products of that sector more than they needed other products; or, in technical terms, the demand was relatively inelastic). Thus, there would be a change in the structure, but no effect upon the general level of wages and prices.
In practice, however, money supply is not held constant, but goes up (theoretically about 4 per cent a year, or slightly ahead of the growth pace of the economy). This increase in the supply of money, which is the result of political decisions by the monetary controllers, has a far greater effect on the general price level than wage pressure could ever possibly have. The current inflationary situation is due in large measure to the $13 billion budgetary deficit that the government ran in meeting the 1957—1958 recession. Although the administration refused to use direct government spending to counter the recession, it achieved the same effect by indirect methods (accelerated spending on committed programs, lower tax receipts, and the like). If any single factor can be held responsible for the inflation-deflation seesaw of recent years, it is the erratic timing of the Federal Reserve Board, which has either stepped a little too hard on the gas or jerked the brake a little too abruptly (as it is doing now). While the business community contradicts the basic precepts of economic theory in ascribing inflation to union wage pressure, conservative economists who know better have kept shamefully quiet.
But there is another argument which holds that union wage pressure can put a particular firm or a particular industry in a difficult competitive position vis-àvis other firms or products—it can “price them out of the market.” This argument is plausible in theory, except that if it were true, production would fall and unemployment would mount—which does not seem to be the case in the relevant sectors of the economy. Actually, the one element in the whole wage-price picture which has been almost completely ignored—the emergence of a large class of non-production workers within the manufacturing firms, and the consequent rise in salary costs—is a more likely candidate, if any single one is to be cast, for the role of “villain” than the unions.
By now it is commonplace that in the last decade the white-collar force has been expanding rapidly while the blue-collar force has remained virtually stable. This increase in white-collar force, however, has taken place not only in the so-called “tertiary” area (insurance, banks, real estate, services, education, recreation, and the like), but—through the proliferation of administrative services (personnel, marketing, merchandising, etc.), of research, and of automation—within the area of manufacturing itself. From 1947 to 1957, the number of non-production workers in manufacturing increased by 60 per cent (from 2,400,000 to 3,900,000) while the blue-collar force remained almost stationary (a little under 13,000,000). In 1947, salaries (the mode of payment to white-collar workers) were one-fourth of the labor costs in manufacturing; by 1957 they had gone up to one-third.
There are two important consequences to this change in the composition of the work force. One relates to productivity, the other to unit labor costs. In all their propaganda on the effect of wage increases, the corporations have talked of the rise in employment costs—but this is never broken down into unit wage costs (usually the unionized sector) and unit salary costs. Given the nature of industrial organization, direct production costs (wage costs) are more immediately subject to control than white-collar costs (salary costs). In other words, if productivity is broken down on a man-hour per production worker basis, and on a man-hour per salaried worker basis, the corporation can recoup its costs more easily in the first sector, where it can achieve economies and technological savings by substituting machines or tightening production schedules. The rise in the proportion of the salaried worker has acted as a drag on productivity, and on unit costs.
In sum, the argument I am making is that a significant share of the rise in manufacturing costs in the last decade has been due not to direct wage costs, but to an extraordinarily large increase in salary costs, which usually become an added fixed cost.
The Federal Reserve data available before the recession of 1957—58 shows this shift in cost burdens quite clearly. Between 1947 and 1957, unit payroll costs (total wages and salaries) rose 26 per cent, while unit wage costs increased by only 16 per cent.3 Much of the payroll rise was a consequence of the rise in unit salary costs which in 1957 were almost 30 per cent higher than in 1953.
This burden becomes even greater during a recession, for when production falls the large corporations cut down their blue-collar force while the white-collar force is maintained whole.
But even more instructive when we consider the effect on costs of this new balance between production and non-production workers is a comparison of what happens to each class during a recession.
|Skilled workers and foremen||175,000||155,000||11%|
|Laborers, helpers, misc.||120,000||90,000||25%|
|Clerical and sales||75,000||75,000||None|
|Professional and technical||30,000||30,000||None|
Two other items from the Kefauver data round out the picture. When the steel-workers (or other such unions) win wage increases, the companies usually give “tandem” increases to the unorganized nonunion workers. When U. S. Steel submitted cost data to the Kefauver committee on the effects of the 1957 negotiations, it indicated that its employment costs had gone up 21 cents an hour, against a union claim that wage costs had only increased 16.4 cents. “Supplementary information provided by the corporation disclosed that the 21-cent figure is a weighted average of benefits extended to 161,500 members of the United Steelworkers, estimated at 19.4 cents per hour, and simultaneous increases granted to 47,600 other employees, estimated at 26.6 cents per hour.” The union in its data had estimated that non-union employees (principally white collar) would receive the same cents-per-hour adjustment as union members. Instead, non-members received increases which on the average were 37 per cent higher than the increase called for in the union contract. As the Kefauver committee said primly: “This may be excellent personnel policy, but there is some question as to the propriety of charging the cost of such a policy to the union agreement.”
It should also be observed that in calculating the “break-even point” for the steel industry, the Kefauver committee analyst used the generous concept of “standby cost” rather than “fixed cost.” Fixed cost includes the conventional items of overhead, interest payments, depreciation, and the like. Standby cost covers all these plus management salaries, payments to supervisory and maintenance employees and to the sales and office personnel who are usually retained even though sales and production may fall. This is obviously a more realistic concept than fixed cost since such salaries do become a “fixed” charge for the corporation. Nevertheless it is a fact that the “break-even point” kept falling steadily after every union agreement even though standby costs went up—which indicates how little the corporation actually suffered from these agreements.
What then, does the argument add up to? Because the steel companies refuse to give breakdowns on unit costs, one cannot fix the relative weights accurately, but the inference is warranted that a large portion of their increased employment costs—and those of other major manufacturing enterprises—is the consequence not of union pressure but of the rise of a salary sector which has become an added fixed expense.
This would seem to give the unions a powerful bargaining point. They could say, in effect, that the production workers, by raising productivity, are pulling their weight in a situation of increased costs, while the white-collar workers are not. But the unions can’t. For they are desperately trying to organize the growing white-collar sector within the manufacturing industries, which makes it impossible for them to “blame” these groups for benefiting unduly from wage increases and pushing up employment costs.
The recent steel strike will not change the basic pattern I have been sketching here. If anything it will tend to reinforce the power of the corporations. Despite the “shotgun settlement,” the terms show clearly that the industry was the economic victor, though the union may have gained a certain “symbolic satisfaction” in having avoided an even greater disaster.
From a rational point of view, there was little ground for a strike. The union knew that this was not “its year.” It had no new radical demands to make. It also knew that its wage levels, because of previous victories, were higher than those of auto, with which it has been linked in historic tandem. Privately it was willing to settle for modest wage increases. It certainly was not going to challenge the industry on prices. Financially, the steel companies could easily afford a wage increase without raising their prices. The profit figures for the first half of 1959 were spectacular and even embarrassing. Profit as per cent of stockholders’ equity, after taxes, was 14.2 per cent, the highest in twelve years; profit as per cent of sales, after taxes, was 7.6 per cent, the highest since 1950. That some of this was due to stepped-up production because of inventory buying by customers who were anticipating a strike merely proves that the strike was not caused by economics. Everybody knew, in fact, that the companies had decided to force a strike no matter what the union did.
For years the industry has been spoiling for a strike, and this time the situation was favorable; the companies recognized that they could afford a long one. Technological development has been so great that at present levels of demand the industry can fill virtually all the orders it gets by operating for only nine months of the year. (In 1958, the industry operated at 60.6 per cent of capacity; in the fiscal year ending June 1959, it was at 77.5 per cent of capacity.) The rest of the time can be absorbed in short work weeks, lay-offs, or a strike. The effects are the same, and the economic loss minimal.5 And the industry wanted a strike.
For years the industry has smarted at the union’s power. During World War II, the union had forced through a wage rationalization program to even out rates between jobs and end the discriminatory practices by which supervisors could practice favoritism. In the Inland Steel case, in 1949, which went to the Supreme Court, the union won a break-through on pensions. And fortuitous timing on contract expirations had allowed the steel union to come in on the upswing of the business cycles and come out with larger wage gains than other unions. But even though the corporations had learned how to take advantage of the wage pattern for price increases, there were important psychological and symbolic issues at stake. This was particularly true at U. S. Steel, where in 1958 a new management team, composed of Roger Blough as chairman of the Board and Conrad Cooper as industrial relations vice-president, had taken over. Blough is a lawyer with no experience in production; his forte is finance. Cooper is an engineer, with little feel for the human give-and-take of a bargaining situation. These two had replaced the old team of Ben Fairless, a production man who had come up through the mills, and John Stephens, whose background was in personnel. Fairless and Stephens had been pragmatic operators, tough-minded but not rigid, inclined to make a “deal” whenever it seemed necessary. Recognizing the vanity and weakness of Dave McDonald, the union chief, they had adopted the tactic of “sweet-talking” him, boosting his ego, making him feel like an equal, arranging for joint trips to the steel plants, where Fairless and McDonald—the two symbols of Management and Labor—would stride the floor together. And McDonald, pipe in mouth, chest puffed out, and distinguishing himself from that red-haired, radical fellow Walter Reuther, would talk in orotund tones of the “mutual trusteeship” concept of management and labor, the equal responsibility of union and industry to free enterprise.
But Blough and Cooper would have none of this charade. Both were men of principle, ideologists, and their main principle was that labor should be put in its proper place. In an age of growing managerial power, when the corporate manager was taking credit for the remarkable performance of American industry, when managing itself was deemed a great new complex skill involving the administration of salesmen, production men, finance men, merchandising men, public relations men, engineers, and personnel men, there was no reason to assume that union leaders should be treated as equals—for labor, after all, was only one of a large number of “coordinates of production.” Labor was to be reduced to its proper dimensions as a small problem, to be handled by the labor relations department.
Equally, the legal mind and the engineering brain wanted to tie up the ragged ends of the work rules issue. The steel companies have always had the right to introduce new machines or any other technological changes without interference from the union. But existing practices were another matter. In many plants, informal work rules or traditional ways of doing things had become the norm, arising, as is common in any human situation, out of custom and habit. To the engineer, such practices may be “irrational.” To the worker it is his “way.” The two are bound to clash. In 1945, when the wage rationalization analysis of U. S. Steel was completed, Cooper, who had been brought into the company to implement the program, insisted that the new contract contain a pledge by the workers to do a “fair day’s work.” Like the verbiage of loyalty oaths, this phrase is mere mumbo-jumbo; everyone accepts the idea of “a fair day’s work for a fair day’s pay”—but how is a fair day’s work to be measured? To the surprise of all parties, Mr. Cooper had a measure. A “fair day’s work”—and this was written into the contract negotiated on May 8, 1946—is “that amount of work that can be produced by a qualified employee when working at a normal pace. . . . a normal pace is equivalent to a man walking, without load, on smooth level ground at a rate of three miles per hour.” Presumably, therefore, the energy that would be used to walk 24 miles a day should go into a fair day’s work. Cooper began applying this viscerotonic definition to every job in U. S. Steel. The following year Phil Murray gave the corporation an ultimatum. The definition might remain, but all existing work rules were to be preserved or the union would refuse to sign any contract—and the companies capitulated. Thus the so-called 2-B clause was frozen into the contract. U. S. Steel has never forgiven that ultimatum. They regarded it as a violation of the previous contract. Technically it was, since Lee Pressman, then the union counsel, had, without realizing the consequences, accepted the company’s authority to change work rules, and the following year Murray had repudiated the agreement negotiated by Pressman. A more flexible management would have understood that the program was inherently unworkable anyway. Men are not automatons, and despite time-study rules and all the paraphernalia of “scientific” work analysis, they will go on their own, stubborn ways. But U. S. Steel has rarely been known for its flexibility.
The fact is that the 2-B rule was primarily a symbolic test of authority. There was no vital economic issue at stake. The corporations talked of the losses in efficiency, but when pressed by George Taylor (the head of the government fact-finding board) to document their case, they could only come up with one example—the “man in the air-conditioned crane cab,” which was entirely irrelevant. And the same symbolic issues of authority and power were the crucial factors in the recent steel strike; real economic questions were again non-existent.
In the end, the industry lost the symbolic fight. It was quite clear that the work rules issue could not be translated into any terms capable of enlisting public support. Moreover, management’s stubborn insistence on this point served to solidify the union ranks. At the start of the strike, the company had made some headway among the steel workers by raising the spectre of inflation. And a strike conducted solely to increase wages would not have spurred enthusiasm among the workers. But the work rules issue gave the union a rallying cry meaningful to every steel worker in the form of a threatened speed-up, or a cut in piece-rate for certain specific jobs.
And yet, in economic terms, the steel companies did uncommonly well in the negotiations. For one thing, they have virtually succeeded in knocking the cost-of-living clause out of the contract, thus putting a fixed ceiling on wage increases. Under the new arrangement the steel workers can get a maximum of six more cents an hour over a thirty-month period, but at the same time the companies can deduct rising insurance costs from the cost-of-living increment—which will in effect wipe the latter out altogether as an expense. Furthermore, since the contract is not retroactive, the steel companies also get a two-months’ free ride on wages. In straight wage terms, the steel workers will receive an increase of 8.2 cents an hour in December 1960, more than a year after the strike, and a second, 7.6 cents, increase in October 1961. In straight cost terms, the increase will come to 3.75 per cent an hour, as against gains by the steel union of 4.5 per cent an hour in the can industry, and of 5 per cent an hour in aluminum.
Meanwhile, prices are already beginning to shoot up, and to paraphrase an old remark, “before they hurt.” A story in the Wall Street Journal recently carried the headline: “Steel Pact Triggers Factory Price Boosts.” The lead paragraph quoted a furniture manufacturer as saying: “Two weeks ago, I was considering raising prices a little in March or April but since the steel settlement I’ve decided to increase them at least 3 per cent and do it as soon as possible.” One retailer remarked: “This type of settlement makes it easier . . . to bring my markup back up. Most of my customers work at Crucible Steel Co., and they know that the more they make the more they will have to pay for most products, including furniture. I started buying more furniture here as soon as I heard about the steel settlement.” The fact that furniture has little to do with steel, and that furniture costs have not yet really gone up is, of course, irrelevant in such a conditioned atmosphere. The price spiral is already under way.
And at some point during the year, when much of the publicity has died down, the steel industry will quietly raise its prices. Where then, does the whole situation leave the union, and the country?
No dogmatic or simple answers are possible. The Joint Economic Committee of the Congress, headed by Senator Douglas, in warning of the power of the large corporations “to raise the price of their goods or services in the absence of excess demand pressure,” suggests the need for “government participation in the price-wage setting process . . . [at least] for a fact-finding procedure . . . on the justification and desirability of such proposed increases.” The economist Abba Lerner has proposed that where capacity lies idle, corporations be forbidden to raise prices, and when pockets of unemployment in an industry persist, unions be barred from asking for wage increases. Such yardsticks are important to have, though it is hard to see how Lerner’s proposal could be carried out administratively without becoming overly bureaucratic and cumbersome.
But surely more is involved than the question of price increases. What is really at stake is the question of the “legitimacy of power” of the managerial groups. Who gives the manager his mandate? The traditional theory of private property has little legal or social validity in the age of the large corporation. A more plausible justification of managerial power is the argument that it allows for multiple, decentralized decision-centers to counter the dangers of arbitrary bureaucratic planning. But what checks exist on the enormous market power of the large corporation itself? Certainly not the market. As I have already tried to show, the corporations have been able to create a “hidden tax mechanism” which allows them to manipulate the market and to raise large sums of money for private expansion. Is such a thing socially desirable? For the situation amounts to this: in response to their own drives for status and power, the large automobile and oil companies have created a huge productive capacity, which in turn forces them to wage large coercive campaigns in order to stimulate consumption of their products—not only through advertising, but through political lobbying as well. One consequence is that Congress can pass a 12 billion dollar road-building program more easily than it can appropriate a billion dollars for schools. And we have the ludicrous spectacle, in New York and Los Angeles, of fantastic sums being levied for expressways—with houses torn down, views blocked, and open spaces cluttered—while public transport, which is faster and more economical, goes hang. Except for a few books by people like A. A. Berle, there has been little critical study in recent years of the social power of the corporations, and the questions of limits to that power.
As for collective bargaining, the other term in the general situation, we can say with some degree of truth that it has almost reached the end of its long career as an instrumentality for economic and social justice. The fact that in the major industries the big corporations have been able to subvert negotiations with unions by utilizing them as a device for masking a protected price policy calls the social utility of collective bargaining into deep question.
Collective bargaining has always been regarded as the chief means of achieving the traditional goals of unionism. These goals can be listed as follows:
- to raise sub-standard wages;
- to eliminate wages as a lever for comparative advantage between firms;
- to eliminate discrimination and favoritism in the treatment of workers and to establish the worker’s conception of equitable standards: e.g. the principle of seniority in lay-offs and promotions;
- to provide a juridical mechanism for grievances outside the arbitrary decisions of management;
- to provide basic security and welfare for the individual worker through “fringe” benefits like medical care, pensions, and supplementary unemployment benefits;
- to obtain a “fair share” of the profits of a firm;
- to redistribute income in favor of the lower class groups;
- to maintain consumer purchasing power, particularly during recessions.
Considering that the modern American trade union movement is only twenty-five years old, the unions have been remarkably successful in achieving most of their goals. But where does the labor movement go from here? The answer depends upon one’s conception of the social role of the trade union. If a union’s aim is simply to get a higher wage for its own members—the attitude of the building trades union, par excellence—it can then only become a partner in a collusive enterprise which strong-arms the rest of the community. This is what has happened—albeit unwillingly—to the auto and steel unions. But if the union has a wider view of its role in society—and seeks to enlist liberal and intellectual support for its claims—then it may have to begin reorienting itself and to think of collective bargaining in a new and different light.
Of the eight objectives outlined above, the unions have been able to achieve the first four, and most of the fifth. But the three strictly economic aims, which form the heart of present-day collective bargaining, have gone by the board. There has been little redistribution of shares between profits and wages, either as proportion of the national income or within firms. Nor has collective bargaining been the agency for maintaining purchasing power. The chief result of bargaining has been to favor strong unions at the expense of weak ones, to strengthen the monopoly positions of highly organized industries, and in consequence, to affect the structure of wages, but not the level (i.e. comparative shares). And the added fact that wage increases now run close to the ceiling levels of increases in productivity sets strong outer limits on the ability of unions to have any salutary effect on the economy through bargaining.
A simpler mechanism than collective bargaining for raising the standard of living of low-income groups, or maintaining purchasing power during recessions, or creating relative equity between different groups of wage workers, is government fiscal policy. The unions could use their influence to win a tax reduction for the lower-income classes of the country; this would be more equitable than pressing for the advantage of a particular group of workers, for it would be “across the board.” Another thing the unions might do is exert pressure on the corporations to reduce prices, which would provide for a more equitable distribution of savings in productivity. Admittedly this is difficult. A union, its leaders say, exists to serve its own members; and the best way to do this is to fight for wage increases. As for other workers, let them go and do the same. But if union leaders adopt such a completely parochial view, they then forfeit the claims unionism has to the sympathies and allegiances of the liberal middle class and intellectuals. To help other workers—especially during recessions—it may be necessary to forgo direct wage increases and rely on government tax policy as the economic gyroscope. To engage in such action, however, the labor movement would have to become more political and begin thinking in broader social terms than it has grown accustomed to doing.
But is there no further innovating role for collective bargaining? I think that there is, that one last historic step remains to be taken—a true annual wage. Sociologically, this is the most revolutionary step the unions can take, but they will have to take it if they wish to consummate their long effort to give workers a legitimate place in society.
Historically, the worker has been treated as a commodity, to be paid by the piece or by the hour for his labor. However much one may declare (formally, as in the Clayton Act, or piously in Labor Day addresses) that labor is not a commodity, the existing system of wage payment shows that that is exactly how the worker is regarded. General Motors still pays its blue-collar force on the basis of every tenth-of-an-hour worked, and despite some union-imposed restrictions, such as “call-in” pay (which guarantees a man at least four hours pay if he is called to work that day), wages are still determined by time or piece, as with any other commodity. The most bitter complaint of auto workers is that they have no way of knowing, from one week to the next, how many hours they will work in any given week; through the year, a man may get as many as twenty “short work weeks.”
All this emphasizes the distinction between the production worker (who is regarded simply as “labor”) and the salaried worker who is paid by the week, month, or the year. Salaried workers (usually of the white-collar class) are laid off less often (they are carried as part of standby, or fixed, costs), they are entitled to sick leave, excused for jury duty, and given a whole host of amenities often denied to the production worker. Why? Such practices are in part a carryover from the old notion of the production worker as an “interchangeable hand,” and in part simply a status distinction enforced by traditional cultural attitudes toward manual labor. But is there any reason of an economic, sociological, or moral character for this “double standard” to continue? Increased costs, says management. Yet what of the gain in status that would accrue to the worker—the gain in psychological as well as economic security—if the double standard were abolished.6
In March 1958, the International Business Machines Corporation made the unprecedented move of placing its 20,000 regular production workers on a weekly salary basis—a move which, surprisingly, received little public attention. Like most blue-collar workers, the IBM production men had been paid on an hourly rate. As salaried employees they became entitled to full pay during absences due to illness or accident, as well as to paid time off for authorized personal reasons (jury duty, death in the family, etc.). IBM is not, of course, in a “seasonal” industry, and has therefore been able to take this step with comparative ease; but few industries in the U. S. today are seasonal—even auto is not wholly so—and those who are can use counter-seasonal pricing devices to even out demand.
It is unlikely that American industry will eliminate piece work and hourly rates voluntarily; union pressure is needed through collective bargaining. But it is also unlikely that the unions, psychologically dispirited or with aging fat-cat leaders, will launch the necessary campaign in the near future. Nevertheless, such a move would be the most important means the unions could find for reducing the “status barrier” between blue-collar and white-collar work—the very barrier in the way of organizing the white-collar workers. And without organizing the white-collar worker, American unionism, in the long run, cannot survive.
1 Thus, in 1950, General Motors estimated its “standard volume” at 2,250,000 units, in order to give it a 20 per cent net return, and sold 3,812,000 units, or a 69 per cent margin of safety. In 1955, “standard volume” was 3,000,000 units, and factory sales were 4,368,000, or 54 per cent above target. In 1957, when “standard volume” was estimated at 3,470,000, sales were a shade under (3,418,000)—the only year of the eight in which the target was not exceeded.
2 The steel companies themselves were in a wonderfully quixotic position. On the one hand they claimed that steel wages, by rising faster than productivity, were inflationary; on the other hand, they protested vehemently that the effect of the increase of steel prices—which should reflect the steel wage inflationary pressure—was “negligible” on the cost of living.
3 In the period from 1953 to 1957, when the greatest increase in non-production workers took place, salary payments rose by 37 per cent while wage payments to the blue-collar force rose by only 7 per cent.
5 The steel industry, then, has to some extent begun to approach the situation that existed in coal when John L. Lewis would order staggered strikes that created a three-day work week in the industry. Newspaper editorialists screamed about the loss of production, but as a detailed analysis later showed, there were, given the level of demand, roughly 165 workdays in the industry anyway, so that it made no difference whatever whether the miners struck or were laid off. Lewis had called the strikes in part as a service to the industry since the anti-trust laws forbade collusion of companies to limit production and maintain prices.
6 Increased costs has been the cry of employers against every innovating device from shorter hours to pensions. Fifty years ago, corporations resisted workmen’s compensation for accidents and the installation of safety devices on the ground of increased costs. Yet today, in the changed climate of public opinion, what corporation would object to installing safety devices on the grounds of cost? On the grounds of mental health, one can justify the increased costs of reducing the pace of work, or the extreme division of labor. On the grounds of justice, one can argue for the elimination of the treatment of labor as a commodity.