In the middle of the twentieth century, the emphasis of US antitrust policy shifted from protecting small producers to protecting consumers. As we saw in the previous chapter, in the first half of the twentieth century antitrust was only one of a number of methods of artificial political protection of small producers, many of them located in the rural South and Midwest and threatened by national competitors in the Northeast. Along with antitrust, the arsenal of small-producer protectionism included unit banking laws, which privileged local banks; anti–chain store laws, which protected owner-operated general stores from competition with national chain stores; and resale price maintenance legislation, which limited the ability of large retailers to offer discounts. Unit banking laws lasted until the late twentieth century, but the rest of the producer republican agenda was moribund by World War II. Its champions, among them Wright Patman and Louis Brandeis, seemed like throwbacks to an earlier era of agrarian populism in the increasingly industrialized and urban America that developed after 1945. Already in the late 1930s the Antitrust Division of the Department of Justice under Robert Jackson and Thurman Arnold was changing the focus of federal antitrust policy away from small-producer protectionism to protecting consumers and rival businesses from the alleged dangers of excessive market power and market concentration. Arnold declared that the sole test of antitrust policy should be, “Does it increase the efficiency of production or distribution and pass the savings on to consumers?”1
This new strain in the antimonopoly tradition displaced the older producer republican strain. For a generation after World War II, US antitrust policy was dominated by the Harvard school of antitrust, also known as the “structure-conduct-performance” (S-C-P) school, which ignored the benefits of scale in many industries and indiscriminately treated large firms and high market shares as bad. For them, markets in which a single firm or a few firms controlled more than a certain percentage of production were automatically suspect, not so much because small firms got hurt as because such markets were alleged to lead to allocative inefficiency, consumer harm (e.g., higher prices), and increased inequality. As a result, their goal, as David Hart has written, was to “establish market structures for ideal performance.”2
In the last third of the twentieth century, the simplicities of the Harvard school came under attack from two rival schools. One was the Chicago school of economics and its more mainstream variants, both of which were even more extreme than the Harvard school in their religious faith in markets. However, the most accurate and impressive criticism of postwar antitrust policy came not from the Chicago school but from the branch of institutional economics known as evolutionary economics or innovation economics. Economists in this tradition—Schumpeter, Galbraith, Baumol—argued that imperfect competition among large oligopolistic firms in such sectors as manufacturing was a positive good that promoted technological progress, not a dangerous deviation from free market purity. And in a world economy dominated by global oligopolies based in particular nations and often helped by home governments, for one country to pulverize its leading firms inadvertently could lead to the capture of both home markets and foreign markets by the national champions of rival countries. As we will see below, this happened in the United States owing to overly aggressive antitrust policy.
These overlapping critiques of mid-century antitrust policy persuaded administrations of both parties, beginning with the Reagan administration in the 1980s, to adopt more nuanced and balanced approaches to antitrust policy that continue to this day, focusing less on market structure and more on business conduct and looking beyond short-term effects on prices and profits to longer-term effects on efficiency and even occasionally innovation. But as we note in the next chapter, a small but vocal school of neo-Brandeisians now seeks to turn back the clock in US competition policy.
As the administration of Franklin Roosevelt came to the end of its first term and faced setbacks from Congress and the Supreme Court, the critics of big business within the administration started to assert themselves. The antimonopoly tradition found a champion inside the Roosevelt administration in Robert H. Jackson, who soon headed the Justice Department’s Antitrust Division in 1938–1939. Jackson was born on a farm in rural Pennsylvania, which he later described nostalgically as “socially classless” and “the nearest to Paradise that most of us ever know.”3
His successor, Thurman Arnold, born in rural Wyoming, sometimes also sounded like a producer republican from the agrarian periphery. In 1961, in his seventies, he told a correspondent, “This process [of consolidation] repeated in industry after industry during the period between the first World War and the depression created a system of absentee ownership of local industries which made industrial colonies out of the West and South, prevented the accumulation of local capital and siphoned the consumers’ dollars to a few industrial centers like New York and Chicago.”4 But in his 1937 book, The Folklore of Capitalism, Arnold mocked trust-busting campaigns as “entirely futile but enormously picturesque.”5 According to Arnold, the problem was not the size of large corporations but their behavior: “There can be no greater nonsense than the idea that a mechanized age can get along without big business.”6
Partly in response to the threat of populists such as Louisiana governor and senator Huey P. Long, the Roosevelt administration tilted toward the antimonopolists during the second New Deal of 1935–1936. The federal government broke up public utility holding companies and in 1936 passed an undistributed profits tax inspired by the mistaken theory that large corporations were prolonging the Great Depression by hoarding cash instead of giving it to workers in the form of wages or to shareholders as dividends.
In 1937 the United States experienced a sharp recession. Most economists view the “Roosevelt recession” as an avoidable mistake, the inadvertent result of federal policies with contractionary effects: an increase in bank reserve requirements, which caused banks to reduce lending; the first collection of Social Security payroll taxes, which reduced take-home pay; and the end of a bonus for World War I veterans. Reducing the federal deficit by 2.5 percent, these measures caused a downturn in an economy that still had not fully recovered.7
Unfortunately, this accurate Keynesian explanation of the 1937 recession was rejected by Roosevelt. In his memoirs, Robert Jackson wrote that Roosevelt “knew that there were evils in the suppression of competition and that there were evils in competition itself, and where the greater evils were he never fully decided.”8 Having sided with supporters of big business and public-private collaboration in his first term, Roosevelt now refused to accept partial responsibility for the economic downturn and ignored the small group of Keynesians who understood the dynamics of aggregate demand. Instead he listened to Jackson, Benjamin Cohen, Arnold, and other “antitrusters” who had opposed the government-industry-labor partnership of the first New Deal. This group favored the erroneous theory that the Great Depression had been caused not by a collapse of aggregate demand triggered by a financial crisis but rather by monopolies and oligopolies hoarding money and creating artificial “bottlenecks.” Arnold shared this view, mocking the Keynesian theory of inadequate aggregate demand: “The cause of spending to prime the pump is the destruction of a free market, and at the same time it is the reason why such spending never does prime the pump according to expectations.” The problem, according to Arnold, was not a lack of demand but “economic toll bridges” caused by inadequate price competition among firms.9 More competition, the theory went, would lead firms to reduce prices, and spur demand, which would in turn spur hiring.
During its brief ascendancy in the late 1930s, the antimonopoly school won a few symbolic victories, including Roosevelt’s endorsement of its mistaken theory in his April 29, 1938, message to Congress on curbing monopolies: “Managed industrial prices mean fewer jobs. It is no accident that in industries, like cement and steel, where prices have remained firm in the face of falling demand, payrolls have shrunk as much as 40 and 50 percent in recent months.” Unwilling to concede that the problem was inadequate aggregate demand rather than the perfidy of big business, Roosevelt asked Congress to give more resources to the Antitrust Division and also to fund a study of the effect of concentration on the economy.10
Ironically, one of the few people ever to read the resulting 1941 congressionally mandated Report on the Concentration of Economic Power was the hero of the free market libertarian right, Friedrich von Hayek. In his 1944 manifesto The Road to Serfdom, Hayek quoted the report, which argued that “the superior efficiency of large establishments has not been established.” Hayek, like today’s anti–big business libertarians, denied that there was any tendency toward oligopoly in increasing-returns industries and blamed the idea that there was on “the influence of German socialist theoreticians.”11
Under Arnold, the Justice Department focused on preserving small competitors and insulating them from “coercion” by their suppliers and from competition from larger firms. One area of special emphasis was distributional restraints. In an era when larger chain stores were emerging, Arnold believed big retailers were undermining the freedom of action of independent dealers. A majority in Congress held this view as well, as evidenced by the passage of the Robinson-Patman Act of 1936, which attempted to protect small businesses from cost-based pricing by their suppliers. The big retailers’ price reductions were not the kind the New Dealers wanted to get the economy moving.
His mission, as Arnold saw it, was to break up large firms or discipline them with consent degrees under the theory that somehow this would be the shock therapy needed to get the economy out of the Great Depression. Under his leadership the Antitrust Division grew twenty-seven times, from eighteen lawyers to 500. As Senator Elizabeth Warren (D-MA) has admiringly noted, “In Arnold’s five years running the Division, those lawyers brought almost as many cases as there had been in the previous thirty-five years. Antitrust law was real—and American corporations knew it.”12
With the approach of World War II, however, trust busting fell out of favor, given the need of the federal government to mobilize large industrial corporations to be what FDR termed the “arsenal of democracy.” FDR removed Thurman Arnold from the Antitrust Division by the expedient of appointing him to the US Court of Appeals for the District of Columbia Circuit, a post from which he resigned after only two years to become a lawyer and lobbyist.
After World War II, large national, and in many cases, multinational corporations emerged in many industries and often gained considerable market share. There was a growing concern among followers of the Brandeisian tradition that firms in some industries had become too powerful and too concentrated. These large firms came to be seen as retarding entry and innovation rather than driving growth.. The focus of antitrust policy shifted from protecting small firms to policing “oligopolies.”
With the establishment of the Small Business Administration (SBA) in 1953, subsidies replaced protectionist measures such as fair trade laws and anti–chain store laws as the preferred method for American politicians to bestow favors on small businesses.
In addition to multiplying subsidies and tax breaks for small firms, Congress expressed concern about increasing market concentration in the two decades after 1945. Adolf Berle and Gardiner Means predicted in their 1932 book, The Modern Corporation and Private Property, that, based on the growth in large corporations from the first three decades of the 1900s, the largest 200 firms would hold all corporate wealth by 1970. In fact, as Elaine Tan writes, “The largest 200 and 500 corporations held 38.85% and 50.08% in 1946 respectively, and kept their share in tandem until a peak in 1968, when the top 200 share dropped steadily, losing more than a third of their share by 1997. In contrast, the biggest 500 corporations saw only a gentle decline from 54.64% in 1968 to 50.14% in 1997.”13
To check the feared growth of big business, the Celler-Kefauver Act of 1950 sought to strengthen the Clayton Antitrust Act of 1914, which itself amended the 1890 Sherman Antitrust Act. The Celler-Kefauver Act empowered the federal government to thwart competition-limiting vertical mergers, in addition to the horizontal mergers that had been the focus of previous antitrust laws. The high-water mark of political concern about concentration was the 1968 submission of the Neal Report, a task force report commissioned by President Lyndon Johnson.14 It recommended enactment of a “concentrated industries act” and a “merger act” that would mandate deconcentration of any “oligopoly industry” and limit conglomerate mergers.
Much of this policy action was inspired by the prevailing school of antitrust policy in the postwar era. This was a time in which highly mathematical, neoclassical economics displaced other, more pragmatic and empirical approaches in academic economics departments. The dominant paradigm in antitrust policy became the structure-conduct-performance paradigm devised at Harvard University by Edward S. Mason in the 1930s and 1940s and developed by his student Joe S. Bain Jr. and others. In the highly deterministic theory of the Harvard school, market structures in radically different industries could be modeled by the same methods and using the same variables. This view, best characterized by Carl Kaysen and Donald F. Turner, was that market power per se is harmful and therefore should be illegal.15 The focus of analysis was on market structure rather than on business conduct as the source of adverse economic performance. Because they argue that market forces, including technological innovation, are insufficient to challenge the entrenched power of a dominant firm, the S-C-P school emphasized structuralist solutions, such as aggressive merger enforcement and the breakup of large firms. Even George Stigler, who subsequently abandoned this school, was sympathetic when he wrote, “An industry which does not have a competitive structure will not have competitive behavior.”16
The S-C-P approach gave a veneer of economic scientism to Brandeisian prejudice against big business. Something like Brandeis’s view of “the curse of bigness,” detached from the earlier small-producer protectionist agenda that included unit banking and anti–chain store laws, became, if not dominant, at least highly influential in the period from the 1930s to the 1970s. In the 1948 case of United States v. Columbia Steel Co., 334 US 495, Justice Douglas wrote, “No monopoly in private industry in America has yet been attained by efficiency alone. No business has been so superior to its competitors in the processes of manufacture or of distribution as to enable it to control the market solely by reason of its superiority.”
The Harvard school influenced federal case law in this era. The Supreme Court in United States v. Von’s Grocery Co. in 1966 rejected a merger that would have produced a firm with just 7.5 percent of the relevant market because of “threatening trends toward concentration.” A few years earlier, in Brown Shoe Co., Inc. v. United States (1962) the court declared that Congress intended the Clayton act “to promote competition through the protection of viable, small, locally owned businesses.”17 In 1963, Adolf Berle, the veteran New Dealer and rival of Brandeis and Arnold, criticized the result:
The court found that the four largest shoe manufacturers, including Brown, produced 23% of the nation’s shoes, and Brown was third largest, and also that there was a “trend toward concentration.” Brown and Kinney together would control 7% of retail shoe stores, and 2.3% of all retail shoe outlets of all kinds. Monopoly thus was not even remotely involved.18
From the era of Thurman Arnold to the 1970s, US antitrust enforcers took a very Brandeisian view of the economy, prosecuting a wide array of firms for having supposed market power. But in many cases this aggressive antitrust enforcement was blind to either important technological innovations that were reshaping market conditions or rising international competition. Indeed, in their zeal to limit market power, US antitrust enforcers imposed real damage on a number of important firms and industries, and in so doing seriously set back the US economy, the effects of which continue to be felt to this day. As historian John Steele Gordon writes:
Those with the hammer of antitrust in their hands have had a record of doing at least as much harm as good. Often their timing has been nearly surreal. Standard Oil was broken up just as Royal Dutch Shell was beginning to provide true competition. In 1948, the very year that television really took off in this country with the debut of the ‘Texaco Star Theater,’ with Milton Berle, Hollywood studios were forced to change several practices. The purpose was to lessen their stranglehold on popular visual entertainment; the result was to move power in Hollywood from the Samuel Goldwyns to the Barbra Streisands. I am not sure that is progress.19
Over the course of several decades, aggressive antitrust enforcement significantly weakened, and in some cases helped kill, a number of leading American technology companies. As Lynn and Longman write approvingly, “Antitrust enforcers weren’t content simply to prevent giant firms from closing off markets. In dozens of cases between 1945 and 1981, antitrust officials forced large companies like AT&T, RCA, IBM, GE, and Xerox to make available, for free, the technologies they had developed in-house or gathered through acquisition.”20 They praise such actions as opening up patents to other companies: “Over the thirty-seven years this policy was in place, American entrepreneurs gained access to tens of thousands of ideas—some patented, some not—including the technologies at the heart of the semiconductor.”21 Lynn elsewhere notes that “a study done in 1961 counted 107 judgments just between 1941 and 1959, which resulted in the compulsory licensing of 40,000 to 50,000 patents.”22
In some cases compulsory sharing of trade secrets—mostly ones, it should be noted, developed fairly through hard work and investment—no doubt helped spur innovation, at least in the short term. But this overlooks two serious problems. The first is the absurdity of having a nation’s industrial policy carried out not by the Commerce Department or another agency tasked to promote long-term national productivity growth and export success but by the Justice Department, in adversarial settings dominated by lawyers and academic economists, on the basis of government litigation or threats of government litigation.
Praise for the Justice Department as the promoter of innovation through compulsory sharing of intellectual property also overlooks the very severe damage that this policy did to leading US technology companies, along with the benefits to foreign companies at enormous cost to American economic development, innovation, and job creation.
The AT&T case is a case in point. After inventing the transistor at its Bell Labs facility, the company faced pressure from antitrust regulators to license that technology. And so in 1952 AT&T licensed the technology for a small fee to thirty-five companies. At one level this spurred innovation, as it helped emerging companies, such as Texas Instruments and the predecessor of Intel, Fairchild. But because of government pressures AT&T also licensed this technology to foreign companies, including Sony, which was the core advantage Sony needed to propel itself to global leadership, in the process taking market share from the leading US consumer electronics firms of the time. At the time, no one in the US government could conceive that a company like Sony could pose a competitive threat to US companies.
RCA is another case of unintended damage done by antitrust policy. As Gordon writes, “Perhaps the best example of the harm antitrust has sometimes done to our economy is RCA.”23 Because RCA had a dominant share in the emerging color television industry, achieved by its own internal research and development, the Department of Justice required RCA to share its patents with US companies for free, stating, “By this criminal indictment, we seek to restore competition in this significant industry so that all competitors of RCA can compete with it at every level from the research laboratory to the sale of end products.”24 As an article in Time magazine noted, “In what the department considers ‘a stroke of industrial statesmanship,’ an agreement was reached on a color TV patent pool.”25
The Justice Department required RCA to provide its valuable patent portfolio to US competitors at no cost. However, RCA was allowed to license the patents to foreign companies for the usual royalty arrangement. Because RCA had long relied on licensing revenue, it now was essentially forced to license its technology to foreign firms, in this case predominantly Japanese firms who were seeking, with little success to break into the color TV market. As James Abegglen, a leading technology historian, has written, “Unwittingly, RCA actually assisted the Japanese by selling them whatever technology licenses they required. It was a highly profitable exercise. … Clearly … Japan was dependent on foreign sources for virtually all of the technology employed even to the stage of color television. … RCA licenses made Japanese color television possible.”26 But without the criminal indictment by the US Department of Justice, RCA would in all likelihood not have licensed its crown jewels to foreign companies, and very well could have survived to this day as a global, leading TV producer.
Armed with this valuable technology, produced through years of research and engineering costing RCA billions of dollars, the Japanese TV manufacturers, which were protected from foreign competition by the Japanese government, soon took over the US market, and an industry invented in America was destroyed. The Japanese government understood the remarkable gift RCA, under pressure from the US government, gave Japanese TV makers. Indeed, when RCA CEO David Sarnoff visited Japan in 1960, he was awarded the Order of the Rising Sun for his contributions to the Japanese electronics industry.
What was the real cost to consumers of this RCA “monopoly”? One study found that it raised the price of televisions by just 2.26 percent.27 This was despite the fact that most of the product and process innovations in the TV industry came from RCA because RCA had the scale and scope to be able to invest in innovation. As one study of radio producers at the time, including RCA, found, “Firms that were larger and had prior radio manufacturing competence innovated far more than other firms, and pursued more challenging innovations including more mechanization innovations, confirming cost-spreading models of innovation incentives.”28 Indeed, the two leading producers of radios, RCA and Philco, produced more process innovations (e.g., innovations related to how to produce TVs) than any other firm in the industry. As Margaret Graham notes in her history of RCA, “If RCA was no longer entitled to claim licensing revenues for maintaining the whole state of radio-related research, the obvious question for the company was which kind of research should it continue to support, and at what level?”29 The answer was there would be less research funding and the research would be much less risky. In summary, as Gordon writes, “To protect an American industry from the dominance of one company, antitrust had killed off the entire industry. That’s a bit like using a guillotine to cure a headache.”30
Oblivious to the possibility that its actions might end up driving out of business a US technology leader, the government went after other leaders, including AT&T, Xerox, Kodak, and IBM. Indeed, nearly a hundred of America’s most innovative companies were forced to give away their patents, over 50,000 of them, by 1960.31 A 1954 consent decree put Eastman Kodak on notice that its attempt to protect its film-processing technology would be heavily constrained. One effect of the FTC’s intervention was to allow Japan’s Fujifilm to enter the US market for film essentially unimpeded by a competitive response.
A decade later, in 1969, the Antitrust Division sued IBM. As Gordon writes, “With 65 percent of the market at that time, IBM was the eight-hundred-pound gorilla of the computer industry. But by the time the case was finally abandoned as unwinnable, in 1982, the next oversize anthropoid of computing, Microsoft, was already shipping its software and IBM was headed into the worst decade of its existence.”32 A few years later the Federal Trade Commission filed suit against Xerox, accusing it of monopolizing the office copier business, with the head of FTC’s Bureau of Competition stating that he would be “dissatisfied if Xerox’s market share isn’t significantly diminished in several years.”33
And indeed, Xerox soon did lose half its market share, but mostly to Japanese firms, in large part because Xerox was forced to provide its Japanese competitors with “written know-how, including drawings, specifications and blueprints for existing and subsequent machines. It made an estimated 1,700 patents available to its competitors.”34 As former Boston Consulting Group consultants Mark Blaxill and Ralph Eckardt write, “Practically speaking, they forced Xerox to license their patents to the world. The company agreed to license any three of its patents for free, the next three for a maximum royalty of 1.5% and then the entire remainder of its portfolio for nothing.”35 However, the unintended consequence of the FTC’s compulsory license was to donate Rochester’s technology to the Japanese, who were able to take decades of American investment and innovation and deploy it in their own products for free. Moreover, because Xerox was so afraid of increasing its market share owing to challenges from US antitrust authorities, it did not respond to emerging Japanese competition by lowering its prices. Within a few short years after the consent decree, Xerox’s market leadership had withered as Japanese competitors such as Canon, Toshiba, Sharp, Panasonic, Konica, and Minolta each claimed a significant share of the US market.36
This aggressive competition policy enforcement, which blithely ignored the threat of global competition to the US economy and jobs in traded-sector firms, had one other pernicious effect. Because companies were so restricted from merging to gain scale and domestic market share, companies turned instead to horizontal mergers in completely unrelated industries. As former deputy assistant attorney general in the Antitrust Division William Kolasky has noted, “In the 1960s the United States experienced a wave of conglomerate mergers, driven in part by overly restrictive antitrust policies toward horizontal and vertical mergers.”37 In Lords of Strategy, a history of the business consulting industry, Walter Kiechel writes, “Antitrust law ruled out acquisitions in your own industry. … So to plow the proceeds back into your company and to keep getting bigger, you often seemed to have only one choice: buy something in an area unrelated to those you were already in.”38 The problem with these kinds of mergers is that they generated little added value through scale economies or synergies, and were largely undone twenty years later. But the damage had been done: companies not able to gain the scale they needed to effectively compete with rising international competitors, often backed by their governments with subsidies and trade protection, instead spent their valuable managerial time and effort on largely worthless mergers.
Unfortunately, this kind of national industry policy in reverse in the name of antitrust continues in the United States. In 2016 the Federal Trade Commission required that the semiconductor maker NXP divest its RF power business as a condition for its $11.8 billion acquisition of US-based Freescale Semiconductor Ltd. While this was done with a focus on the consumer, it opened up the business for acquisition by the Chinese Jianguang Asset Management Co. Ltd. (a company with the financial backing of the Chinese government). Just like that, critical US technology capabilities went to China, thanks to an action undertaken by the US government. This give-away to foreign state capitalism was anything but pro-competition and reflected a lack of understanding of the new nature of global competition in the technology industry, where the Chinese government has a strategy to enable its firms to acquire foreign technology assets to ultimately displace the US technology leaders in the marketplace.39
In the last third of the twentieth century, the S-C-P school with its hostility to any and all market power came under attack from two different schools. In part because of the recognition of the damage done to the US economy by the S-C-P school and changing economic conditions, especially increased global competition, the Chicago school was able to gain followers, particularly during the Reagan administration. Adherents of the Chicago school argued that markets were much more contestable and disciplining than the postwar intellectual heirs of Brandeis and Arnold believed, and that government attempts to intervene vis-à-vis antitrust legislation caused more harm than good.
In addition, the Chicago school gave more weight to efficiency than did the populists, who focused more on distributional questions. The Chicago school argued, for example, that if a merger led to increased market power and prices (which reduces allocation efficiency), it still could lead to overall societal welfare if the gains from productivity increased more than the losses from allocation inefficiency. As Robert Bork, a founder of the school, describes it, “The whole effort of Chicago was to improve allocative efficiency without impairing productive efficiency so greatly as to produce either no gain or a net loss in consumer welfare.”40
Another assault on the structuralist orthodoxy of the Harvard School came from growth-oriented institutional economics, known also as “evolutionary economics” or “innovation economics.”41 Adherents of the innovation doctrine argue that antitrust policy, and merger policy specifically, need to incorporate analysis of longer-term dynamic effects. Joseph Schumpeter explained dynamic efficiency as “competition from the new commodity, the new technology, the new source of supply, the new organization … competition which commands a decisive cost or quality advantage and which strikes not at the margins of the profits and the outputs of the existing firms but at their foundations and their very lives.”42 In this view, a merger might be justified, even if it results in greater market power, if the increased profits are invested in research and new product development.
Like Schumpeter, John Kenneth Galbraith argued that large firms are essential for modern technological progress: “[A] benign Providence who, so far, has loved us for our worries, has made the modern industry of a few large firms an excellent instrument for inducing technical change.” The very market power enjoyed by large firms in markets with limited competition, according to Galbraith, ensured that they could recoup the costs of developing innovations, before the innovations were adopted by competitors. “The net of all this is that there must be some element of monopoly in an industry if it is to be progressive.”43
To prove his point by counterexample, Galbraith pointed to American agriculture, dominated by family farmers in competitive markets who did not enjoy sufficient profits to allow them to engage in research and development (R&D): “There would be little technical development and not much progress in agriculture were it not for government-supported research supplemented by the research and development work of the corporations which devise and sell products to the farmer. The latter, typically, are in industries characterized by oligopoly. The individual farmer cannot afford a staff of chemists to develop an animal protein factor which makes different proteins interchangeable as feeds.”44 Galbraith noted that as long as large firms were rare, one “could demand prosecution of the offending monopoly under the Sherman Anti-Trust Act with a view to its dismemberment, or, if this latter were impractical as in the case of the utilities, he could advocate public regulation or public ownership.”45 But that was not possible in an economy dominated by large industrial firms: “It is possible to prosecute a few evil-doers; it is evidently not so practical to indict a whole economy.” He mocked the Harvard school approach to antitrust: “To suppose that there are grounds for antitrust prosecution wherever three, four or a half dozen firms dominate a market is to suppose that the very fabric of American capitalism is illegal.”46
In terms of challenging the dogmas of the Harvard school, the innovation school had far less influence on US antitrust policy than the Chicago school of economics, just as it had and continues to have less influence on US economic policy overall. Both the Harvard and the Chicago schools worked within the unrealistic, mathematically modeled universe of neoclassical academic economics. But adherents of the Chicago school, including legal scholar and federal judge Robert Bork, favored a more hands-off approach to market concentration for several reasons, including a belief that any temporary monopolies will soon be corrected by the entry of new competitors, the illogic of punishing firms for success, and a general suspicion of government power and wisdom. The influence of the Chicago school produced a more lenient approach to mergers and concentration under Republican administrations, beginning with the Reagan administration.
When the history of antitrust is reviewed from today’s perspective, the most striking thing about it is how ineffectual it has been in achieving either of its two somewhat conflicting goals of producer republicanism and market fundamentalism. More than a century of federal antitrust policy has failed to preserve a society of small producers; at less than 10 percent, the number of self-employed Americans is about the same is it is in other advanced industrial economies. And in many industries, the market fundamentalist objective of preventing one or a few large firms from controlling large market shares has also failed—fortunately so, in many cases, from the perspective of the innovation school. As the historian Philip Cullis notes, “Antitrust might have prevented American industry from becoming dominated by monopolies, but it did little to stem the rise of oligopolies.”47
The failure of US antitrust policy to eliminate most oligopolies in increasing-returns industries (it did eliminate some, including the entire US TV manufacturing industry) should be a cause for gratitude, not lamentation. Baumol has warned that because of the contributions to economic progress of innovative oligopolistic firms, they should not be
targets for antitrust prosecution simply because their prices are discriminatory or are not close to marginal costs. … [T]hey should not be deemed vulnerable to prosecution simply on the claim that their pricing patterns show them to be the possessors of monopoly power. Such a course can easily constitute a major handicap to the steadily growing expenditure on innovation by private industry, which is arguably a mainstay of the U.S. economy’s unprecedented growth record.48
Adolf Berle’s words from 1963 are just as relevant today:
Large corporations are in existence; not many will shrink. Modern conditions being what they are, the largest will tend to prevail in the competitive market. … Meanwhile, the small units will clamor to the political state for help or protection, or both. The result is likely to be not nineteenth-century competition, but twentieth-century oligopoly. Atomization as a remedy will not do. I doubt if we can bring back the England of the late eighteenth or nineteenth centuries, or that we would tolerate it if it returned.49
The innovation doctrine makes it clear that the focus of antitrust thinking should be on the long-term trajectory of product value and price, not just current consumer welfare measured by short-run prices. As Harvard’s Michael Porter rightly argues, “Since the role of competition is to increase a nation’s standard of living and long-term consumer welfare via rising productivity growth, the new standard for antitrust should be productivity growth, rather than price/cost margins or profitability.”50
However, as we have seen, the neo-Brandeisians are back. As we saw earlier, the major heirs of Brandeis following World War II abandoned his support for measures such as anti–chain store laws that protected small producers from competition but retained his conviction that big firms and concentrated markets are social and economic evils. This view informed the Harvard school of antitrust policy, which focused on the structure of industry, with rigid views regarding market share.
Nevertheless, the neo-Brandeisian approach fell out of fashion beginning in the 1970s, largely as a result of the emergence of the conservative Chicago school of antitrust policy, which focused much more than the Harvard school had on efficiency concerns. This view—whether in the conservative Chicago version or the liberal neoclassical economics version—dominated US antitrust thinking until recently. (In contrast, the Schumpeterian school that we favor has had little influence on antitrust policy.)
But now a small but intelligent and articulate school of neo-Brandeisians seeks to turn back the clock, if not to the era of anti–chain store laws and unit banking laws, at least to the heyday of the populist S-C-P era of the 1950s and 1960s, which treated even minor levels of concentration in markets as per se illegitimate and dangerous. Today’s new Brandeisianism has found a home in the twenty-first century Democratic Party, which seeks to demonstrate its solidarity with the working class and consumers through its desire to break up big companies.
For many of these neo-Brandeisians, the defeat of the Harvard team by the Chicago team was the result of a conspiracy by corporations and their academic and political puppets to eviscerate antitrust laws. Even worse, relaxed attitudes toward antitrust enforcement are alleged to have caused or contributed to many of the economic and social ills of the present, from higher inequality to slower wage growth and even the global financial crisis of the global recession of 2008–2009.
According to the Yale law school student Lina Khan, writing in the progressive Democracy Journal: “America’s monopoly problem today largely results from a successful campaign in the late 1970s and early 1980s to change the framework of antimonopoly law.”51 Derek Thompson writes that “antitrust law shifted over the course of the 20th century from principally protecting competition to principally protecting consumers.”52 Today Brandeisianism has returned, and “many reformers are calling for the pendulum to swing back.”53 Khan writes, “Some policymakers and politicians today are starting to realize that America once again has a monopoly problem.”54 Khan and Sandeep Vaheesen say that “policymakers and the public ought to recognize antitrust as another tool for achieving a more progressive distribution of income and closing the staggering economic disparity we see today.”55 Barry Lynn and Philip Longman write, “What would a True Populist do today? Immediately restore America’s traditional anti-monopoly philosophy.”56 Nell Abernathy and coworkers of the Roosevelt Institute want to “tame the corporate sector” by reviving “an open markets agenda for the 21st century.”57
For neo-Brandeisians industry concentration has developed into crisis proportions and breaking up big companies should be the animating goal not just of antitrust policy but of US economic policy generally. In short, when all you have is a hammer, everything looks like a nail. There is almost no economic problem that cannot be laid at the feet of large corporations and a supposed increase in firm size and concentration and cured by the panacea (literally, “cure-all”) of antitrust.
Neo-Brandeisians blame income inequality on concentration and claim that breaking up big corporations is the key to boosting incomes for average Americans. The liberal Center for American Progress writes: “Income inequality is rising, middle-class incomes are stagnant, and much of the current economic policy debate is centered on finding ways to counter these trends. A renewed focus on antitrust enforcement could make a significant contribution toward accomplishing this goal.”58 Robert Reich writes that increased concentration has “resulted in higher corporate profits, higher returns for shareholders, and higher pay for top corporate executives and Wall Street bankers—and lower pay and higher prices for most other Americans. They amount to a giant pre-distribution upward to the rich.”59
Lina Kahn blames monopolization for the “fact” that “the vast majority of American workers have seen their hourly wages flatten or decline since 1979.”60 Yet, as we discussed in chapter 4, recent evidence suggests that most of the growth of the “one percenters” is not in corporate managers and CEOs but among the professions: dentists, doctors, lawyers, financers, and the like, most of whom work for small companies or are self-employed. Moreover, according to the US Bureau of Labor Statistics, not only did establishments with more than 500 workers pay their workers 77 percent more than establishments with fewer than fifty workers, but from 2004 to 2016, real, inflation-adjusted compensation for their workers (based on cost per hour worked) grew by almost $4 per hour ($3.88) compared to just $1.45 for establishments with fewer than fifty workers.61
Neo-Brandeisians also argue that the modestly declining share of income going to labor is evidence of rent seeking from industry concentration. But as former Obama administration officials Jason Furman and Peter Orszag note, “The decline in the labor share of income is not due to an increase in the share of income going to productive capital—which has largely been stable—but instead is due to the increased share of income going to housing capital.”62 In other words, more of output of society is going to land and building owners. This is a zoning issue, not an antitrust issue. And a war on the landlords would seek guidance from the ghost of single-taxer Henry George, not the ghost of Louis Brandeis.
If our goal is to increase after-tax wages, there are only two ways to do so: increase productivity or change the distribution of income, through government policies of redistribution or by increasing the bargaining power of workers in wage negotiations with employers. Neo-Brandeisians are decidedly ambivalent about the former, productivity growth, because it sometimes leads to job losses. As Lina Khan writes, “One result of consolidation is fewer jobs, as companies routinely lay off thousands of workers after merging.”63 Barry Lynn complains that increased concentration leads to “fewer and lesser jobs.”64 But when companies lay off workers after merging this is almost always because the combined firm needs fewer workers to do the work. Keeping the workers employed by prohibiting the merger would reduce productivity and per-capita income growth since the workers would be doing work that could be done more productively with a merger. Most, if not all, laid-off workers find new jobs and produce additional output for society.
Some neo-Brandeisians concede that big firms give us higher productivity, and hence lower prices and higher real incomes, but dismiss the value of that. The Roosevelt Institute scholar Sabeel Rahman writes, “If consumer prices are our only concern, it is hard to see how Amazon, Comcast, and companies such as Uber need regulation.”65 Others, however, follow Brandeis, who went to great pains to try to paint small firms as efficient as or more efficient than large firms, writing, “A corporation may well be too large to be the most efficient instrument of production and of distribution.”66 Barry Lynn agrees, dismissing arguments and data about the superior efficiency of larger firms as “metaphysics.”67 Abernathy and coworkers from the Roosevelt Institute writes that firm size is “the product of distinct political and policy choices.”68 To admit that in most cases big businesses benefit from technology-enabled economies of scale would force the neo-Brandeisians to abandon their economic case for antitrust and base their advocacy on producer republican arguments.
Perhaps a world of small, less productive firms with higher prices is fine because we can afford to pay a premium for our craft-brewed beer, organic arugula, free range chickens, and coffee even fancier than Starbucks’.69 As Thompson writes, “The bigness of business is a result of federal policy, which, in the past three decades, has deliberately made it easier for large companies to dominate their markets, provided that they keep prices down.”70 So low prices are not a goal? Few Americans would prefer to live in India, where mom-and-pop retailers are protected by government against more efficient larger retailers and because of that are about 6 percent as productive as US retailers.
If productivity growth is not to be emphasized, this leaves a second option, redistribution. And there are only two ways to get the money for that: from individuals and from corporations. Increasing taxes on wealthy Americans and using that revenue to reduce the taxes on middle- and lower-income workers (or to expand public services they use, such as health care) could increase the after-tax incomes of the latter. But significant tax hikes on the rich would be opposed by almost all Republicans and some centrist Democrats. And from the point of view of the small-is-beautiful school, redistribution from the rich would be undesirable because it would leave the pre-tax concentration of income and wealth intact.
What about paying for higher redistribution through raising taxes on corporations rather than individuals? How much would the average American benefit if the corporate profit rate was the same as it was in the glory days of the 1950s and 1960s, when antitrust enforcement was much tougher and wage growth much higher? Returning to the profits rate of that era would, at least in the short term, actually make American workers worse off, since corporate profits as a share of GDP were higher then. From 1947 to 1968, corporate profits accounted for 10.3 percent of GDP, while from 1994 to 2015 they accounted for 8.2 percent. To be fair, one reason is because more companies have organized as partnerships and S corporations rather than C corporations to avoid paying the corporate income tax. But even with this trend, income going to “proprietors” (e.g., business owners of non-C corporations) declined from 10.4 percent of GDP to 7.2 percent over the last two decades. So did corporate profits. According to the US Bureau of Economic Analysis the average corporate profit rate from 1965 to 1966 was 13.6 percent, slightly higher than the average profits for 2015 through the third quarter of 2016 (13.3 percent).71 Yet during the 1960s median wage growth was significantly higher, even with those profit levels.
Let’s assume that neo-Brandeisians get their way and break up most large companies into medium-sized ones, and that this somehow reduces corporate profits 25 percent, to 6 percent. Further, let’s assume that the resulting decline in business investment from lower profits has no effect on growth and that all the fall in profits goes to price declines. Let’s further assume that none of the price declines benefit the top 10 percent (we will assume they pay the old higher prices) but that all the loss of profits hurts the top 10 percent (we will assume that middle-income Americans own no stock whose value would decline from lower profits). So how much would median incomes increase? A whopping—get ready for it—3.1 percent.72 Not only is this a de minimis amount, it’s a one-time effect. The bottom 90 percent would enjoy 3.1 percent higher incomes for the rest of their lives relative to the base case.
One reason for this is that there is just not that much money here. As finance professor Craig Pirrong writes, “In 2015, after-tax corporate income represented only about 10 percent of US national income. Market power rents represented only a fraction of those corporate profits. Market power rents that could be affected by more rigorous antitrust enforcement represented only a fraction–and likely a small fraction–of total corporate profits.”73
But the negative effects of this would quickly swamp this small static redistributional gain. Firms would be smaller and therefore often less productive. If we assume that the Brandeisians were successful in shrinking the size of large firms so that the United States had the same firm size structure as Canada, US per capita GDP would decrease by 3.4 percent, because on average, smaller firms are less productive than larger firms. But this too is a one-time effect. The negative dynamic effects are likely even larger. Lower profits would reduce investment in R&D and machinery and equipment; according to one study, a cut in profits by $1 would reduce investment by between 32 and 62 cents.74 This decline would in turn reduce productivity and wage growth. So, it is clear that campaign to reduce the size of large corporations would result in reduced living standards for most Americans.
As an alternative to redistribution, higher wages, whether resulting from a higher minimum wage, more unionization, or tighter labor markets, would be denounced by small businesses that are less able than big firms to respond by investing in labor-saving technology.
The ironies in the archeological excavation of the long-buried antimonopolist tradition in the early twenty-first century are striking. Today’s progressive antimonopolists claim they want a return to the golden age of high wages in the decades after World War II. But John Kenneth Galbraith, the leading liberal economist of the mid-twentieth century, thought that the cult of small business was anachronistic. And the labor unions of the New Deal era preferred to negotiate with dynamic industrial oligopolies like the Big Three automakers. Organized labor faced unremitting hostility from the kind of small business owners idealized by today’s gentry liberals of the small-is-beautiful school.
Far from being a return to New Deal liberalism, the Brandeisian revival is based on the two strands of the older antimonopoly school that were decisively marginalized by the mainstream New Deal Democrats between the 1940s and the 1970s—producer republicanism and market fundamentalism. Indeed, during the heyday of New Deal liberalism, champions of small firms and “open markets” were more likely to be found on the anti–New Deal right than on the liberal left.
Equally striking, to the eyes of the economic historian, are the parallels between the growth in influence of small-is-beautiful antitrusters on the center left in the late 1930s and in the 2010s. In Franklin Roosevelt’s second term, as in Barack Obama’s two terms, recovery from a catastrophic global economic collapse continued to be painfully slow. In both cases the incumbent Democratic administration worsened the macroeconomic situation by prematurely trying to balance the budget. In both the Roosevelt and the Obama administrations, this provided an opportunity for members of the antimonopoly tradition to argue for the same erroneous diagnosis of the slow-growth problem—excessive concentration of economic power—and the same mistaken diagnosis—much more aggressive antitrust. In both the late 1930s and the 2010s, this analysis ignored the real problem of inadequate aggregate demand, caused in both cases by a financial panic, not overconcentration (businesses did not abruptly become bigger right before 1929 or 2008).
In the Roosevelt years, wartime mobilization solved the problem of low aggregate demand. And the prosperity of the postwar years was generated in large part by the emergence of a new powerful Schumpeterian innovation wave based on electromechanical technologies, while the ability of workers to share the high profits of industrial oligopolies, with the help of unionization and tight labor markets, spurred wage growth. The explanation of the Great Depression as a result of too much concentration of industry was discredited and relegated to the footnotes of historians. If we are correct, today’s neo-Brandeisian school will share a similar fate at some point in the next generation when the next Schumpeterian innovation wave begins to kick in and higher levels of growth result.