The Truth About Sherman
[This article originally appeared in the Austrian Economics Newsletter, Summer 1991, pp. 1-6. You read can also read it in .PDF.]
Today regulation is generally recognized as a mechanism by which special interests lobby the government to create barriers to entry or other special privileges. Research has shown, for example, that the Civil Aeronautics Board cartelized the airline industry, the Interstate Commerce Commission helped monopolize the railroad and the trucking industries, the Federal Deposit Insurance Corporation sharply limited entry into the banking business, and occupational licensing created entry barriers into hundreds of occupations. Much of the history of regulation chronicles monopoly privileges procured through the auspices of the state, as Adam Smith pointed out more than 200 years ago in The Wealth of Nations.
Oddly, antitrust regulation is still widely viewed as government's benevolent response to the "failures" and "imperfections" of the marketplace. Even economists who are usually skeptical of regulations enacted in the name of the public interest seem to lose their perspective when it comes to antitrust. George Stigler, for example, has stated: "So far as I can tell, [the Sherman Act] is a public-interest law ... in the same sense in which I think having private property, enforcement of contracts, and suppression of crime are public-interest phenomena.... I like the Sherman Act." [Quoted in Thomas Hazlett, "Interview with George Stigler," Reason, January 1984: 46).]
A 1984 survey of professional economists revealed that 83 percent of the respondents believed that "antitrust laws should be used vigorously to reduce monopoly power from its current level." [Bruno Frey, et al, "Consensus and Dissension Among Economists," American Economic Review (May 1984): 986-84.] His opinion is widespread despite common knowledge among antitrust scholars that in practice the antitrust laws restrain output and the growth of productivity have contributed to a deterioration of the competitive position of U.S. industry, and are routinely used to subvert competition.
Why then do the antitrust laws continue to command such powerful support among economists and legal scholars when the pervasive failures are so well known? There are several possible explanations. Antitrust consultants and expert witnesses often stand to make a good deal of money, so financial self-interest may preclude criticism of antitrust. Many economists are also unable to voice informed opinions on antitrust. If it is not their area of expertise, they may not have kept up with research over the past 30 years, or excessive concentration on mathematical models may have left some economists somewhat detached from economic reality. Finally, it is widely believed that there was once a "golden age of antitrust" during which the public was protected from rapacious monopolists by benevolent public servants. According to this perspective, although mistakes have been made, more knowledgeable and public-spirited regulators can successfully reform antitrust. Once reformed, antitrust policy can then perform its original purpose and defend competition and free enterprise.
Unfortunately, the Sherman Act was never intended to protect competition. It was a blatantly protectionist act designed to shield smaller and less efficient businesses from their larger competitors. There never was a golden age of antitrust. The standard account of the origins of antitrust is a myth.
Interest Group Politics and the Sherman Act
In the late 1880s, widespread economic change produced myriad pleas from relatively small–but politically active–farmers who sought protection from larger, corporate competitors. Historian Sanford Gordon offered an example: "Perhaps the most violent reaction [against industrial combinations] of any single special interest group came from farmers.... They singled out the jute bagging and alleged binder twine trust, and sent petitions to both their state legislators and to Congress demanding some relief. Cotton was suggested as a good substitute for jute to cover their cotton bales. In Georgia, Mississippi, and Tennessee the [farmers'] alliances passed resolutions condemning the jute bagging trust and recommended the use of cotton cloth. [Sanford Gordon, "Attitude Toward the Trusts Prior to the Sherman Act," Southern Economic Journal (July 1963): 158.
Southern farmers were annoyed that consumers increasingly preferred jute to the cotton cloth they produced, and they sought antitrust legislation that would dissolve their competition. Such special-interest behavior was characteristic of the farm lobby. During the 51st Congress, Gordon notes that "64 petitions and memorials were recorded in the Congressional Record, all calling for action against combinations. These were almost exclusively from farm groups.... The greatest vehemence was expressed by representatives from the Midwest." (P. 162)
Farmers complained to their national representatives that the products they bought from the trust were increasingly expensive relative to the prices of farm products, but the facts do not support this contention. From 1865 to 1900 farm prices were falling, but at a slower rate than the general price level. This produced real income gains for farmers. In addition, the rapidly increasing quality of manufactured goods further improved the farmers' standard of living. The volatility of farm prices caused the farmers to be politically active.
Many other groups joined the antitrust coalition: small business organizations, academics (though not economists), and journalists. They argued that the "giant monopolies" were creating a "dangerous concentration of wealth" among the capitalists of the day. Although the conspicuous wealth of entrepreneurs such as Rockefeller, Vanderbilt, Mellon, and Morgan added fuel to this charge, it does not appear to be true. In fact, economic historians have concluded that from 1840 to 1900, the division of national income between labor and property owners (capital and natural resource suppliers) remained in a 70 to 30 ratio." [R. Gray, and J. Peterson, Economic Development in the United States (New York: Irwin, 1965).] Over the same time span, both capital and developed natural resources increased faster than the labor force. This means that labor income per unit of labor rose compared with profit and interest per unit of property input.
Although there was no significant redistribution of wealth from labor to capital owners in the aggregate, competitive markets always alter the distribution of income in ways that some do not like. There was no "dangerous concentration of wealth," but many supporters of antitrust legislation found that their own income had fallen (or not increased rapidly enough). The push for antitrust legislation was an attempt to use the powers of the government to improve their economic status.
Economic conditions were changing rapidly in the latter part of the nineteenth century. Expansion of the railroad and inland shipping industries greatly reduced the cost of transportation. Technological developments led to large-scale (and lower-cost) production of steel, cement, and other goods. Communications technology rapidly expanded, especially the use of the telegraph. And the capital markets became more sophisticated. The United States also underwent a rapid transition from a predominantly agrarian to an industrial society. In 1810 the ration of farm to non-farm labor was approximately 4.0. This ratio fell to 1.6 by 1840, and by 1880 the labor force was about equally divided between farm and non-farm endeavors. Meanwhile, individuals and groups uncomfortable with rapid change were becoming increasingly adept at using the regulatory powers of the state. In this increasingly mercantilist atmosphere, the Sherman Act was passed in 1890.
Were the Trusts Monopolistic?
In introducing federal "antitrust" legislation, Sen. Sherman and his congressional allies claimed that combinations or trusts tended to restrict output and thus drive up prices. If Sherman's claims were true, then there should be evidence that those industries allegedly being monopolized by the trusts had restricted output. By contrast, if the trust movement was part of the evolutionary process of competitive markets responding to technological change, one would expect an expansion of trade or output. In fact, there is no evidence that trusts in the 1880s were restricting output or artificially increasing prices.
The Congressional Record of the 51st Congress provides a list of industries that were supposedly being monopolized by the trusts. Those industries for which data are available are salt, petroleum, zinc, steel, bituminous coal, steel rails, sugar, lead, liquor, twine, iron nuts and washers, jute, castor oil, cotton seed oil, leather, linseed oil, and matches. The available data are incomplete, but in all but two of the 17 industries, output increased-not only from 1880 to 1890, but also to the turn of the century. [The following discussion is based on Thomas J. DiLorenzo, "The Origins of Antitrust: An Interest-Group Perspective," International Review of Law and Economics (June 1985).] Matches and castor oil, the only exceptions to the general rule, hardly seem to be items that would cause a national furor, even if they were monopolized.
As a general rule, output in these industries expanded more rapidly than GNP during the 10 years preceding the Sherman Act. In the nine industries for which nominal output data are available, output increased on average by 62 percent; nominal GNP increased by 16 percent over the same period. Several of the industries expanded output by more than 10 times the increase in nominal GNP. Among the more rapidly expanding industries were cottonseed oil (151 percent), leather goods (133 percent), cordage and twine (166 percent), and jute (57 percent).
Real GNP increased by approximately 24 percent from 1880 to 1890. Meanwhile, the allegedly monopolized industries for which a measure of real output is available grew on average by 175 percent. The more rapidly expanding industries in real terms included steel (258 percent), zinc (156 percent), coal (153 percent), steel rails (142 percent), petroleum (79 percent), and sugar (75 percent).
These trends continued from 1890 to 1900 as output expanded in every industry but one for which we have data. (Castor oil was the exception.) On average, the allegedly monopolized industries continued to expand faster than the rest of the economy. Those industries for which nominal data are available expanded output by 99 percent, while nominal GNP increased by 43 percent. The industries for which we have data increased real output by 76 percent compared with a 46 percent increase in real GNP from 1890 to 1900.
As with measures of output, not all of the relevant price data are available, but the information that is at hand indicates that falling prices accompanied the rapid expansion of output in the "monopolized" industries. In addition, although the consumer price index fell by 7 percent from 1880 to 1890, prices in many of the suspect industries were falling even faster.
The average price of steel rails, for example, fell by 53 percent from $68 per ton in 1880 to $32 per ton in 1890. The price of refined sugar fell from 9 cents per pound in 1880, to 7 cents in 1890, to 4.5 cents in 1900. The price of lead dropped 12 percent, from $5.04 per pound in 1880 to $4.41 in 1890. The price of zinc declined by 20 percent, from $5.51 to $4.40 per pound from 1880 to 1890.
The sugar and petroleum trusts were among the most widely attacked, but there is evidence that these trusts actually reduced prices from what they otherwise would have been. Congress clearly recognized this. During the House debates over the Sherman Act, Congressman William Mason stated, "Trusts have made products cheaper, have reduced prices; but if the price of oil, for instance, were reduced to one cent a barrel, it would not right the wrong done to the people of this country by the 'trusts' which have destroyed legitimate competition and driven honest men from legitimate business enterprises." [Congressional Record, 51st Congress, House, 1st Session (June 20, 1890), p. 4100.] Sen. Edwards, who played a key role in the debate, added, "Although for the time being the sugar trust has perhaps reduced the price of sugar, and the oil trust certainly has reduced the price of oil immensely, that does not alter the wrong of the principle of any trust." [Ibid., p. 2558.] Perhaps it would be more accurate to describe the Sherman Act as an anti-price-cutting law.
One final argument could be made that the trusts were practicing predatory pricing, that is, that they were pricing below their costs to drive out competitors. But in more than a century of looking for a proven real-world monopoly actually created by predatory pricing, an example has yet to be found. Moreover, prices charged by the nineteenth-century trusts continued to fall for more than a decade. What rational businessman would continue to price below cost for more than ten years?
In sum, the nineteenth-century trusts were not guilty of the charge levied against them by Sen. Sherman. There is no consistent evidence that they restricted output to raise prices.
Government: The True Source of Monopoly
It appears that one function of the Sherman Act was to divert public attention from a more certain source of monopoly-government. In the late nineteenth century, tariffs were a major source of trade restraints, but the Sherman Act made no provision for attacking tariffs or any other government-created barriers to competitive entry. In fact, evidence exists that a major political function of the Sherman Act was to serve as a smoke screen behind which politicians could grant tariff protection to their big business constituents while assuring the public that something was being done about the monopoly problem.
In a particularly revealing statement during the debates over the antitrust act, Sen. Sherman attacked the trusts on the ground that they "subverted the tariff system; they undermined the policy of government to protect ... American industries by levying duties on imported goods." (Page 4100). This is certainly an odd statement from the author of the "Magna Carta of free enterprise." But increased output and reduced prices in these increasingly efficient industries apparently dissipated the monopoly profits previously generated by the tariffs. This worked against the objectives of the protected industries and their legislative champions, including Sen. Sherman.
Even more damning is the fact that just three months after the Sherman Act was passed, Sen. Sherman, as chairman of the Senate Finance Committee, sponsored legislation popularly known as the "Campaign Contributors' Tariff Bill" that sharply raised tariff rates. On October 1, 1890, the New York Times reported: "The Campaign Contributors' Tariff Bill now goes to the president for his signature, which will speedily be affixed to it, and the favored manufacturers, many of whom ... proposed and made the [tariff] rates which affect their products, will begin to enjoy the profits of this legislation."
The New York Times further reported that "the speech of Mr. Sherman on Monday [September 29, 1890] should not be overlooked, for it was one of confession." Apparently, Sen. Sherman withdrew his speech from the Congressional Record for "revision," but a reporter obtained an unabridged copy of the original. The New York Times reported: "We direct attention to those passages [of Sherman's speech] relating to combinations of protected manufacturers designed to take full advantage of high tariff duties by exacting from consumers prices fixed by agreement after competition has been suppressed.... Mr. Sherman closed his speech with some words of warning and advice to the beneficiaries of the new tariff. He was earnest enough in his manner to indicate that he is not at all confident as to the outcome of the law. The great thing that stood in the way of the success of the bill, he said, was whether or not the manufacturers of this country would permit free competition in the American market. The danger was that the beneficiaries of the bill would combine and cheat the people out of the benefits of the law. They were now given reasonable and ample protection, and if they would resist the temptation attaching to great aggregations of capital to combine and advance prices, they might hope for a season of great prosperity... He did hope, the Senator concluded, that the manufacturers would open the doors to fair competition and give its benefits to the people.... He hoped the manufacturers would agree to compete one with another and would refuse to take the high prices that are so easily obtained."
It was absurd, of course, for Sen. Sherman to say that a protective tariff would actually help consumers if only manufacturers could be trusted to refrain from raising prices. The whole purpose of tariff protection is to allow domestic manufacturers to raise prices, or at least to avoid reducing them. Such hypocrisy led the New York Times to withdraw its support of antitrust legislation. The Times concluded: "That so-called Anti-Trust law was passed to deceive the people and to clear the way for the enactment of this ... law relating to the tariff. It was projected in order that the party organs might say to the opponents of tariff extortion and protected combinations, 'Behold! We have attacked the Trusts. The Republican party is the enemy of all such rings.' And now the author of it can only 'hope' that the rings will dissolve of their own accord." Thus, the Sherman Act seems to have been passed to help draw public attention away from the process of monopolization through tariff protection.
The Sherman Act won legislators votes and campaign contributions from farmers and small businessmen who thought antitrust regulation would protect them from their more efficient competitors, and the tariff bill was supported by all U.S. manufacturers, both large and small. In a political sense, then, the Sherman Act was very efficient. Congress itself seems to have been one of the principal special-interest groups to benefit from antitrust legislation.
Economists and the Emergence of Antitrust
Although most economists today favor stricter antitrust regulation, from the 1880s until the 1920s the economics profession expressed nearly unanimous opposition to antitrust. When Sanford Gordon surveyed professional journals in the social sciences and articles and books written by economists before 1890, he found, "A big majority of the economists conceded that the combination movement was to be expected, that high fixed costs made large-scale enterprises economical, that competition under these new circumstances frequently resulted in cut-throat competition, that agreements among producers was a natural consequence, and the stability of prices usually brought more benefit than harm to the society. They seemed to reject the idea that competition was declining, or showed no fear of decline." [Sanford Gordon, "Attitudes Toward the Trusts," p. 158.]
George Stigler has also noted economists' initial disapproval of antitrust: "For much too long a time students of the history of antitrust policy have been at least mildly perplexed by the coolness with which American economists greeted the Sherman Act. Was not the nineteenth century the period in which the benevolent effects of competition were most widely extolled? Should not a profession praise a Congress which seeks to legislate its textbook assumptions into practice?" [George Stigler, "The Economists and the Problem of Monopoly," American Economic Review (May 1982): 1] Stigler offered three possible explanations. First, economists did not appreciate the importance of tacit collusion. Second, they had too much confidence in other forms of regulation as a means of dealing with monopoly. Third, they underestimated the income they would receive as antitrust consultants.
These explanations are plausible, but there may be an even more important reason for the transformation of economists'attitudes toward antitrust. In the late nineteenth century most economists viewed competition as a dynamic, rivalrous process, similar to the theory of competition embodied in the work of Adam Smith and today's Austrian economists. Consequently, they tended to regard mergers as a natural consequence of the competitive struggle and not something that should be interfered with by antitrust legislation. [The following discussion is based on T.J. DiLorenzo and Jack C. High, "Antitrust and Competition," Historically Considered," Economic Inquiry (Summer 1988). Although some industries were becoming more concentrated in the late nineteenth century, rivalry was still as strong as ever, as the rapid expansion of output and the decline in prices attest. Thus, the economists of the time saw no reason to interfere in market processes with antitrust regulation.
Beginning in the 1920s, mathematical economists developed the so-called perfect competition model, and it replaced the older theory. To economists competition no longer meant rivalry and enterprise. Instead, it meant the equation of price and marginal cost. Most important, it meant that there must be "many" firms in "unconcentrated" industries. Once economists began to define competition in terms of market structure, they became more and more enamored with antitrust regulation as a way of forcing the business world to conform to their admittedly unrealistic theory of competition.
Economist Paul McNulty has noted: "The two concepts [of competition] are not only different; they are fundamentally incompatible. Competition came to mean, with the mathematical economists, a hypothetically realized situation in which business rivalry ... was ruled out by definition." [Paul McNulty, "A Note on the History of Perfect Competition," Journal of Political Economy (August 1967): 398]. F. A. Hayek has made an even stronger statement: "What the theory of perfect competition discusses has little claim to be called competition at all and ... its conclusions are of little use as guides to policy." [F.A. Hayek, "The Meaning of Competition," in F.A. Hayek, Individualism and Economic Order (Chicago: University of Chicago Press, 1948), p. 92.] Moreover, wrote Hayek, "If the state of affairs assumed by the theory of perfect competition ever existed, it would not only deprive of their scope all the activities which the verb 'to compete' describes but would make them virtually impossible." Advertising, product differentiation, and price undercutting, for example, are all excluded by definition from a state of "perfect" competition which, according to Hayek, "means indeed the absence of all competitive activities."
Those economists who use market structure to measure competition are likely to have a favorable attitude toward antitrust regulation. Stigler asserted more than 30 years ago, "One of the assumptions of perfect competition is the existence of a Sherman Act." [George Stigler, "Perfect Competition, Historically Contemplated," Journal of Political Economy (February 1957): 1.] To the nineteenth-century economists, however, an antitrust law was incompatible with rivalry and free enterprise. The perfect competition model and its corollary, the structure-conduct-performance paradigm of industrial organization theory, have seriously misled the economics profession, at least as far as antitrust policy is concerned.
The two principal reasons for the "antitrust economists' paradox," then, are the lack of historical knowledge–particularly about actual economic events in the late nineteenth century–and the failure to appreciate that competition is best viewed as a dynamic discovery procedure, as Hayek contends. Economists who believe that there was once a "golden age of antitrust" have never produced any evidence of such an age. As this paper has shown, the Sherman Act was a tool used to regulate some of the most competitive industries in America, which were rapidly expanding their output and reducing their prices, much to the dismay of their less efficient (but politically influential) competitors. The Sherman Act, moreover, was used as a political fig leaf to shield the real cause of monopoly in the late 1880s- protectionism. The chief sponsor of the 1890 tariff bill, passed just three months after the Sherman Act, was none other than Sen. Sherman himself.
In the late nineteenth century most economists viewed competition as a dynamic, rivalrous process, much like the contemporary Austrian theory. Accordingly, they nearly unanimously opposed antitrust on the grounds that such a law would be inherently incompatible with rivalry, Once the economics profession embraced the "perfect" competition theory which, as Hayek has said, means "the absence of all competitive activities," it also embraced antitrust regulation. For once competition came to mean "many" firms and the equation of price to marginal costs, rather than dynamic rivalry, most economists became convinced that antitrust laws were needed to force markets in the direction of their idealized model of "perfect" competition. Consequently, antitrust has for over a century been a tremendous drag on competition, rendering American industry less productive and less competitive in world markets. Robert Bork might not have been exaggerating when, writing in his book, The Antitrust Paradox, he remarked that if government were to somehow force the economy into "competitive equilibrium," it would have approximately the same effect on personal wealth as several strategically placed nuclear explosions.