12   TWO DECADES OF ANTITRUST SUITS, 1960s–1980s

In effect, the government complaint against IBM was that because it was so big it must be an antitrust violator even though no illegal behavior could be shown. In other words, the company was guilty of making a good product and pricing it right. Horrors.

—THE MEMPHIS PRESS-SCIMITAR, 19821

FRANK CARY SPENT much of the 1970s in court. There were many days when his driver showed up at his house early in the morning to take IBM’s CEO either to an office filled with lawyers and IBMers working on lawsuits at 1133 Westchester Avenue—DPD HQ—in White Plains, New York, or to the federal courthouse at Foley Square in Manhattan, where he was grilled by hostile Department of Justice (DoJ) attorneys or by IBM’s own. Some court sessions lasted many hours. Either way, Cary had to be attentive, careful in how he spoke, and always stick to a rehearsed approach. He then attended post-testifying debriefings back at Armonk before his driver took him home, pouring out the exhausted Cary at his front door. The CEO had attended thousands of meetings sprinkled around the world, often dealing with contentious issues, so he understood how to behave in front of the government’s lawyers. In this instance, if he and his company failed to defend their behavior, IBM could be split into multiple companies or pay a frightfully large financial penalty. There were days and weeks when he spent up to 80 percent of his time focused on U.S. v. IBM (1969–1982), the longest antitrust case in U.S. history.

Besides the disruption to IBM’s affairs, Cary must have been irritated by the fact that the lawsuit concerned the company’s activities with the S/360. In the 1960s, both he and John Opel had expressed skepticism about the strategy of building systems in modular fashion that could be tailored to a customer’s needs. Their experience with the S/360 led him to join a generation of senior executives who never again wanted to engage in such a massive project. IBM created new classes of competitors, notably plug-compatible mainframe and peripheral vendors, as well as 250 rival leasing companies, so despite the enormous increase in revenue and the placement of over 18,000 S/360s by the end of the 1960s in the United States alone, generating $16 billion in revenue and $6 billion in profits,2 he occasionally wondered if it was worth it. Now he had to deal with a raft of lawsuits. For most of his tenure as the head of IBM, the DoJ was hot on IBM’s heels.

IBM’s experience was not unique in American industry. As new technology firms emerged in the late nineteenth century, early entrants such as AT&T, DuPont, and US Steel gained great market dominance, which was protested by regulators and then government lawyers, who wanted them to scale back to encourage competition and innovation. While Americans loved their litigation, other nations had other methods of dealing with such issues: the Japanese used MITI to govern industrial rivalries, innovation, and imports, while European regulators altered or prohibited specific practices. European and Asian governments invested in local vendors and created favorable regulations to support them. The U.S. government did, too, while simultaneously trying to contain the competitive surge that accrued to companies like IBM. That dual approach was unique to the United States, and such behavior continued right through the twentieth century. Economists have studied the implications of these kinds of litigation, and we encounter their work in this chapter. Business historians, however, have a great deal of catching up to do, so this chapter should provide insight about the effects of litigation on their favorite topics: strategy, organization, R&D, corporate cultures, and information ecosystems. The problem is part of the larger one labeled by Matthias Kipping and Behlül Üsdiken as being “stuck elsewhere: business history between history and economics.”3 While interest in business-government relations has enjoyed a recent boom, antitrust and other iterations of lawsuits remain outliers in that broader discussion. Litigation needs to be further integrated into the histories of companies because, as Matthias Kipping put it, “governments have played an important role in business matters.”4 My point is less of a criticism than a recognition that litigation is profoundly influential on a corporation, and IBM is a case in point.

On May 18, 1998, the DoJ filed an antitrust case against Microsoft Corporation, accusing it of behaving as a monopoly in violation of U.S. antitrust law. Twenty U.S. state attorneys general joined the suit. The particulars of the case were less important than that government lawyers focused on the issue that had concerned them about IBM: practices that reduced the ability of competitors to sell to Microsoft’s customers and that blocked entry of new rivals. Microsoft and the government went to trial and on November 5, 1999, the company was found guilty of maintaining a monopoly over its operating system for PCs and of constraining rivals, such as Apple, Lotus (owned by IBM), Netscape, and others. The following June, the court ordered Microsoft to split into two firms. Microsoft’s lawyers appealed, and the company and the Justice Department negotiated a settlement. The trial judge approved the deal in June 2004.

So IBM was not alone in experiencing the U.S. government’s love/hate relationship with big business. It did not end with Microsoft. In April 2012, it became Apple’s turn, when it was accused of raising the prices of e-books. Trade publishers and high-profile authors joined the case, filing their own suits. These mimicked earlier ones: appeals, negotiated settlements, and court approval of parts of agreements, although the U.S. Supreme Court let stand a fine of $450 million in a ruling handed down in March 2016.

After being relatively quiet on matters concerning information technology companies in the 1990s, in the 2010s European regulators probed Apple, Facebook, Google, and Amazon, as well as IBM, over a variety of issues related to their market dominance and behavior. So the issues involving IBM were part of a much larger topic involving how high-tech companies gained control and operated in information technology markets. These usually concerned data privacy and “foreign” dominance of national markets. Could regulators and lawyers keep up with markets and technologies that continued to change faster than the wheels of justice turned? This chapter adds evidence that a disconnect existed between changing technologies and markets on the one hand and how quickly the legal system could respond, a phenomenon that IBM’s experience demonstrated.

The more important story, however, is not a legal one but rather how litigation affected the daily operations of a firm. The task before us is to reconstruct IBM’s perspective on litigation and how its employees were affected, because for decades afterward their behavior was conditioned by the legal events of the 1950s to 1980s. Our purpose is less to add new insights regarding antitrust law and more to appreciate what occurred inside one corporation. To do that, the story is organized into three discussions. First, I briefly describe the legal landscape in which IBM operated in midcentury; second, I discuss the private antitrust lawsuits and the federal case; and third, I examine the consequences of the federal case. For students of IBM’s history, the events of the private and federal cases will be familiar, as they are routinely discussed in company histories, although here points of emphasis differ, as we take advantage of the passage of time, providing perspective and insights on internal operations. The consequences of the federal case described here represent new material, with the notable exception of historian Richard S. Tedlow’s observations about how IBMers talked about their business, reported on in this chapter. Central in much of the discussion of the federal case has been the economic assumptions underpinning the government’s argument against IBM, also reviewed in this chapter, challenging the effectiveness of government economic assumptions and emphasizing the cost to all parties involved in both private and public cases.5

With the federal case, the broader issue of how to apply century-old antitrust laws in an economy that appeared so different by the 1950s and 1960s, when technological and economic changes came more rapidly than the legal system could keep up with—a point discussed here but only recently taken up by legal scholars—is informed by the role of information as an economic influence on events.6 Not discussed in this book are patents and copyrights, but they are discussed by others.7 IBM’s experience with antitrust litigation is not well understood, hence our focus on it here.8 Historians have paid more attention to antitrust activities of the 1950s than to the more important ones of the 1970s. The earlier event takes on importance, historians say, because it resulted in Thomas Watson Sr. turning over control of IBM to his son, Tom Jr.9 Historian Robert Sobel was one of the few to point out that antitrust activity at IBM should be seen as a whole, not as bits and pieces scattered in histories of IBM.10 I agree with him.

Finally, we correct the chronology of this story. This case did not begin in 1969, as historians have asserted. It started years earlier, when the DoJ began exploring the possibility of a suit against IBM. In fact, it is not completely clear when that process began. Why is this distinction about dates important? When IBM lawyers and senior management learned that the DoJ was entertaining antitrust questions, they responded to this turn of events with concern. The DoJ’s activities affected IBM’s strategy, the terms and conditions of its offerings (such as unbundling), and the behavior it expected of its employees. More than any other company in the information technology world, then or now, it had experience in such matters. In less than a decade, IBM management knew it faced an existential threat to the firm’s survival.11

THE LEGAL LANDSCAPE, 1950s–1970s

Large U.S. corporations had fretted over antitrust issues since 1890, when the Sherman Antitrust Act became law, followed by subsequent legislation in the Clayton Act of 1914 and other patent, trademark, and copyright laws. As time passed, firms in an increasing number of industries took into account essentially three antitrust issues when developing their products and marketing strategies. First, the law forbade companies from exercising monopolistic control over their markets, and they could be found guilty of violating the law even if they were perceived to have the power to do so. As the largest companies gained such power, it proved impossible for them not to exercise it. IBM was no exception. Second, government lawyers and judges sought to stop “tie-ins,” practices that forced customers to buy one or more goods or services when acquiring a specific product, such as when IBM required a tabulator customer to use IBM cards. Third, as companies developed patents they were allowed to convert into commercial products, officials wanted them to share (lease out) rights to competitors, such as when the DoJ forced AT&T to do so in the 1950s after it invented the transistor.

By the mid-1950s, enough cases had gone through the courts to make all three considerations relevant for senior executives to keep in mind when formulating product development strategies and in shaping the structure of their organizations. IBM shared these same concerns with other U.S. enterprises, notably Alcoa, AT&T, and United Shoe Machinery Company, firms that tangled with antitrust litigation. IBM’s experiences in 1936 and 1956, in which its activities were restrained by consent decrees, became high-profile examples of the federal legal community working to nurture competition within the U.S. economy.

Section 2 of the Sherman Act was the main event. It forbade companies from stifling competition, which could be caused by becoming large or dominant in a market. That notion was at the center of the DoJ’s case. The Clayton Act of 1914, in the words of a student of that law, “brought greater specificity to antitrust by enumerating the forms of corporate activity that were subject to prosecution,” such as outlawing discriminatory pricing, “exclusive dealing,” and “tying contracts,” among other actions.12 In 1914, the U.S. Congress established the Federal Trade Commission (FTC), giving it broad authority to advise and regulate on antitrust matters. The economic issues stimulated considerable discussion among economists in support of either IBM or the DoJ, providing a detailed plot to the case.13

Competitors who sued IBM focused additionally on Section 3 of the Sherman Antitrust Act, which extended the law to cover actions in the District of Columbia and in all U.S. territories. In 1977, Congress passed the Foreign Corrupt Practices Act, which essentially said that one could not do overseas what laws prohibited a company from doing in the United States. When an Argentine IBM executive was charged with bribing a local government official a few years later, it was as much a criminal offense for his U.S. manager as it was for him in Argentina. Regardless of whether another country chose to implement similar antitrust laws, management in New York had to take U.S. practices into account when operating World Trade or in implementing anticorruption practices around the world. The “Business Conduct Guidelines” that governed the behavior of all IBM employees worldwide were a direct outgrowth of U.S. anticorruption laws. In practice, the two fused together within IBM.

It made sense that this would happen. Recall that Thomas Watson Sr. had been convicted of violating the Sherman Antitrust Act largely because of business practices at NCR that in future decades might have been viewed also as acts of corruption. The case against IBM in the 1930s did not have the tinge of bribery or corruption but certainly had the feel of constraining competition and the look of tie-ins to control markets. The case leading to IBM’s consent decree in 1956 had a patina of illegality in addition to the more formal concerns about antitrust behavior. In the 1950s, one response by IBM was to make sure all employees understood the terms of its consent decree with the DoJ and to prescribe the kind of behavior employees could not engage in, integrating into its guidelines responses to concerns over antitrust, corruption, and business ethics, making them part of the “IBM Way.” In time, these became the Marketing Practices Guidelines that every employee has been required to adhere to ever since.

American legal battles of the twentieth century were often dramatic and the source of much corporate rhetoric, posturing, and press coverage, but they were usually resolved within a few years. The cases IBM faced in the 1960s and 1970s were of long duration and of high cost to the federal government and IBM. In the federal antitrust case, taxpayers footed the bill. IBM spent in excess of a quarter of a billion dollars. We may never know for sure what this case cost the nation.

Government policies remained remarkably consistent throughout the twentieth century. The desire was to break up links in commercial behavior that at best were, to use Usselman’s word, “tenuous,” always to ensure competitors’ survival.14 Their survival rested on the belief that competitive behavior led to technological innovation, competitive prices, and quality products. Practices such as bundling goods and services were forbidden; unbundling became a solution. No company seemed to have done more to inject the word “unbundling” into the American business lexicon than IBM. By 1970, most data processing managers around the world understood the concept. Not until IBM introduced the term “Personal Computer,” in 1981, did so many people become exposed to an IT phrase emanating from one company.15 To avoid confusion, however, there is a difference between policies and practices. The U.S. government went through periods when it enforced antitrust laws more firmly than at other times, such as during the Progressive Era and again right after World War II, but less so during the 1920s and 1980s. But when it did, what it enforced and how it interpreted legislation remained relatively consistent.16

The legal environment enveloping the cases discussed in this chapter can be summarized quickly. On the one hand, federal prosecutors, regulators, and courts spent enormous energy to ensure the economy was vigorous in its ability to sustain competitive activities. The United States did this more than any other nation. To put it in blunt terms, if a country wanted to know how best to manage national antitrust laws, regulate patents and trademarks, and control corporate corruption, they turned to the United States. On the other hand, no nation did as much to encourage development of new sciences, technologies, and products, funding them with hundreds of billions of dollars buttressed by laws to protect, yet regulate, these innovations. Such investments led to the creation of the computer industry in the United States, and its growth was faster, more vigorous, and occurred earlier than anywhere else. One could argue the same about telecommunications, medicine, air travel, and weapons systems in the post–World War II period. In summary, corporations like IBM were encouraged to do well, and when they did, they often had their market power challenged. Corporations and government officials learned how to navigate this seemingly contradictory behavior. IBM became a master player.

CDC, TELEX, AND ALL THOSE PRIVATE LAWSUITS

Recall Control Data Corporation (CDC), one of the post–World War II start-up computer companies that began by specializing in large scientific computing, successfully producing large-end computers that competed against IBM’s largest computers. It ran head-on against a problem when IBM discussed producing larger S/360s that CDC management concluded were designed to dislodge CDC’s products. In 1968, CDC’s management decided to open a legal front against Big Blue, accusing IBM of announcing higher-end models of the S/360 aimed at CDC’s 6600.17 CDC’s management had cause to be suspicious about IBM because the CDC 6600 did fine until IBM told potential customers for CDC’s system that it was about to bring out an S/360 Model 90 that would outperform CDC’s machine.

In fairness to IBM, it did in fact work on such a machine and in fact built and delivered 13 of them, but it is also true that these machines did not perform as promised by IBM and experienced technical difficulties during development, all of which led IBM to lease them for less than originally planned. IBM lost about $100 million on that effort. IBM documents revealed during legal proceedings demonstrated that its executives were upset with the project and with the Product Test Department (responsible for quality control), which refused to sign off on a machine built largely in response to competitive pressures, notably the Models 90 and 91.18 But IBM’s announcement of the forthcoming Model 90 had its effect. Customers waited for IBM’s computer, so CDC’s sales of the 6600 slowed. In its annual report in 1965, CDC blasted IBM’s “frequent announcement of changing characteristics and new models, at reduced prices, of large computers reported to be under development.”19

CDC’s president, William Norris (1911–2006), referred to IBM’s S/360s as “phantom machines” because they did not exist at the time of their announcement. CDC’s lawyers accumulated case studies of IBMers selling this product, documenting what IBM told customers about how effective it would be. This had three consequences. First, other firms embraced the idea of fighting IBM’s market dominance in court, so over the next few years they launched their own suits, listed in table 12.1. Evidence and arguments from one case resurfaced in others. Second, the DoJ, barely ten years after settling with IBM in 1956 to stop market-dominating practices, again began exploring what IBM was doing with the S/360s. That investigation led to its antitrust suit, filed in 1969.

Table 12.1

Chronology of antitrust lawsuits filed against IBM, 1969–1975

Case filing date

Case name

1969

U.S. v. IBM

1969

Greyhound v. IBM

1972

Telex v. IBM

1973

CalComp v. IBM

1973

Hudson v. IBM

1973

Marshall v. IBM

1973

Transamerica v. IBM

1974

Forro v. IBM

1974

Memory Tech v. IBM

1975

Sanders v. IBM

Source: List modified from Franklin M. Fisher, James W. McKie, and Richard B. Mancke, IBM and the U.S. Data Processing Industry: An Economic History (New York: Praeger, 1983), 379–380.

Third, IBM agreed to a rapid settlement of the CDC case. It sold its Service Bureau Corporation (SBC) to CDC for a fraction of its value, which added to CDC’s own service bureau business, and paid CDC an additional $101 million in compensation in cash and contracts. In exchange, CDC turned over its case studies to IBM, evidence that other litigants would have loved to use.20 In his memoirs, Tom Watson Jr. admitted being persuaded by his lawyers to destroy the file before it could become evidence. “I had no choice,” he wrote, “ ‘Burn it,’ I said.”21 It was done that night. In hindsight, he noted, “Even though the best lawyers money could buy assured me I wasn’t breaking any law, it made me feel uneasy.”22 He had denied the DoJ and IBM’s rivals a major body of evidence.

The Telex case was another matter. Formed in 1936 to sell hearing aids, in the 1960s it entered the computer peripheral equipment business. In 1972, it filed an antitrust suit against IBM along the lines CDC had four years earlier, and three years later, the U.S. government filed its own suit. Unlike CDC’s case, this one came down through history as a sleazy attempt at judicial redress by a vendor unable to sell competitive products.23 It asked that IBM be broken up because Big Blue was selling and leasing machines at prices Telex could not match. Its complaints piggybacked off the federal case. Telex won its case in the initial trial, being awarded a sum five times its gross revenue, but which the court later lowered. Meanwhile, in a countersuit, Telex was found guilty of infringing IBM patents and stealing trade secrets. In 1975, a federal court overturned the conviction against IBM. IBM agreed not to collect the penalty levied against Telex, so the legal fight ended. The case took three years and seven months to settle.24

The industry trade press covered it blow by blow. Computerworld, the premier trade newspaper in the industry, covered every twist and turn.25 The other leading publication, Datamation, did, too. Industry watchers provided running commentary. These included Gideon Gartner, who made a living interpreting IBM’s actions for the brokerage industry. Everyone in the computer industry paid attention to these cases because, as Gartner put it, IBM was the largest vendor: “Big Blue ruled the info-tech industry and vendors and users alike danced around IBM like courtiers at Louis XIV’s Versailles. When IBM made a product announcement, everyone listened.”26 Fanning interest during the Telex trial, a judge, over IBM’s objections, ordered all documents in the case made public, resulting in 40,000 pages of memoranda, minutes of IBM meetings, product plans, notes, and other IBM ephemera becoming available to reporters, industry watchers, prosecutors at the DoJ, and the other firms suing IBM. The Computer and Communications Industry Association (CCIA) came into existence to supply the anti-IBM community with information. It sold many copies of these documents to IBM’s rivals for $5,000 a set.27 These documents damaged IBM’s cause.

The Telex drama was not the only legal activity under way. In 1972, CEOs at CDC, Univac, NCR, and Honeywell called on the DoJ to get on with its case against IBM, offering to assist to move it forward; that is, paralyze IBM’s ability to compete against them. Reporters favorable to IBM tended the minority view: “The most recent case [IBM vs. Telex] is a classic illustration of the slimy function and goal of antitrust: the sacrifice of ingenuity and ability and success to mediocrity and incompetence and failure. In ruling against IBM, Judge A. Sherman Christensen has sanctioned the sacrifice of IBM’s ingenuity and ability and success to the mediocrity, incompetence and failure of a virtually bankrupt little fifth-rate company, Telex.” This reporter’s argument was that, rather than attacking the firm, IBM’s rivals should thank it for inventing the computer industry and thus, “Consequently, it is IBM that made and makes it possible for them to succeed.” The writer blamed Telex for “poor management and a colossal failure to produce a product that very many people care to buy.”28 Another observer commented that the “Telex case inspired several small companies to get in on the pickin’s.”29 In other words, some saw litigation as an opportunity to reach a cash settlement with IBM based on the Telex documentation.

Table 12.1 shows IBM’s competitors joining a growing list of vendors complaining about Big Blue’s actions, relying on the expanding body of available trial evidence. While the majority of these cases had ended by 1980, during the 1970s they added to the cloud of IBM’s legal problems. IBM survived them all. For example, a U.S. district court dismissed the Greyhound v. IBM case, one of the first filed against the company, in 1972. On appeal, a federal appeals court called for a new trial. To cut back on the anticipated expenses of such a trial, IBM’s lawyers negotiated a settlement, paying out $17.7 million. Both parties agreed that IBM would admit no wrong.

Other cases were resolved similarly. The CalComp case went to trial in 1976, but the court dismissed it in 1977 because of a lack of evidence of wrongdoing, even before IBM’s lawyers could mount a defense. Forro went to trial in 1977 and, a month later, the judge threw out that case. The following year, IBM’s lawyers were back in court, this time with Memorex, with the judge handing down a verdict for IBM. In 1979, the same happened in the Transamerica trial. Smaller cases involving Marshall, Memory Tech, and Sanders were settled out of court. No private case had established that IBM monopolized any market for mainframes or peripherals or had illegally crushed leasing companies.

IBM’s lawyers had private U.S. cases to deal with every year from the late 1960s until the early 1980s. Duane W. Krohnke, a young lawyer at IBM’s antitrust legal firm (Cravath, Swaine & Moore, better known as CS&M inside IBM), recalled what it was like to work on these cases, citing the example of the Greyhound litigation. As a young father of two, he found IBM’s cases grueling. He left home very early in the morning and did not return until after his children were asleep. Working weekends became routine. One lawyer supposedly reported 27 hours in one day as a result of working across three time zones!

Greyhound wanted to question IBM officials under oath. IBM spent large sums on lawyer fees to look at vast quantities of its own documents to prepare for that questioning. As Krohnke described,

One of the IBM officials to be deposed in the Greyhound case was its President, Thomas J. Watson, Jr., and I was put in charge of assisting Tom Barr in preparing Mr. Watson for his deposition. This was a daunting challenge. It meant collecting and analyzing as many IBM and public documents as possible that were potentially relevant to the Greyhound and other cases, figuring out the possible questions that might be asked of Mr. Watson by opposing counsel and then meeting with him and IBM’s General Counsel, Nicholas Katzenbach, to go over these documents and questions, all in a relatively short time period.30

To get Watson Jr. prepared, he and Tom Barr (the young lawyer’s boss) had to meet with him on IBM’s jet flying back from California to New York, after the two lawyers flew out to California so that they could fly back with Watson. Inside IBM, such cases always involved a similar tedious process of research, endless rounds of meetings, and years of depositions.

That IBM had been aggressive, clever, and determined in competing was never in question. The plaintiffs and IBM argued over the sizes of the markets, its rivals wanting a definition that made IBM look like it dominated, while IBM contended that, by its definition of market size, its share was smaller.31 The lack of common definitions caused much confusion in each case and led to hung juries and to judges tossing out cases. The key point IBM made to its customers, the public, and its employees is that the firm was never found legally guilty after all the verdicts were handed down and appeals ended. As irritating and time consuming as they were to IBM, all those cases proved to be the legal sideshow of the 1970s. The main event was the federal case.

THE FEDERAL ANTITRUST CASE: U.S. v. IBM, 69 CIV. 200 (S.D.N.Y. 1969)

The federal case ran so long that when it ended in January 1982, half of IBM’s employees had not been at the company when it began. They never knew a time when the case was not hanging over them. European and Asian IBMers were not immune either, because the effects seeped out worldwide. It was a case that had its fits and starts, stops and progress, quiet periods and frenetic moments, its ups and downs, but remarkably the key players in the case remained involved for over a decade. Admirers of federal judge David N. Edelstein (1910–2000) considered him a kind and fair man, but IBM’s legal team saw him as being prejudiced against the company and big business in general. Many of his rulings gave them much to worry about. He adjudicated the government’s case for ten years. IBM’s lead lawyer was Thomas D. Barr (1931–2008) of Cravath, Swaine & Moore, who in time directed the work of some 200 people on the case. His counterpart at IBM, Nicholas Katzenbach (1922–2012), a senior vice president, became involved after joining the company in 1969.

Then there was the cast of over 1,000 other participants. The government brought 118 people to the witness stand. IBM brought a similar number, plus there were 97 IBM depositions of government officials. All these people needed to be prepared for their roles and played their parts, often more than once, as did Frank Cary. The list of IBMers deposed or brought into court read like a Who’s Who of IBM: John F. Akers (vice president and group executive), George B. Beitzel (general manager of the DP Group), Frederick P. Brooks Jr. (formally of IBM and a software manager on the S/360), Frank Cary (then chairman of IBM), Bob O. Evans (vice president of engineering programming), Hillary Faw (who had advised senior management about pricing strategies), Ralph E. Gomory (director of research), Gilbert E. Jones (chairman of the board of World Trade), T. V. Learson (retired chairman of IBM), John R. Opel (senior vice president), Ralph A. Pfeiffer (corporate vice president and president of the DP Division), Paul Rizzo (vice president, Finance and Planning), “Buck” Rodgers (vice president, Marketing), Tom Watson Jr. (retired chairman), and so many lesser figures. In other words, an entire generation of IBM executives participated.

Figure 12.1

Thomas Barr led IBM’s defense against the federal antitrust suit in the 1970s and early 1980s. Photo courtesy of IBM Corporate Archives.

But it did not end there. Other luminaries from the world of computers were drawn in as well. These included J. Presper Eckert of Univac, developer of the ENIAC and UNIVAC I; Edward Feigenbaum, an academic and rising star in the emerging world of artificial intelligence; one-time IBM employee Gideon I. Gartner, then Wall Street’s leading authority on IBM; John W. Mauchly, cofounder with Eckert of their own computer company (later called Sperry Rand); William C. Norris, chairman of the board and CEO of CDC; Frederic G. Withington of Arthur D. Little and, like Gartner, an industry expert; and an assortment of executives from IBM’s rivals, plus the 97 officials deposed for the case. All were cast for the longest antitrust case in U.S. history.

When it started with IBM’s rivals urging the Justice Department to open an inquiry into the company’s behavior, Attorney General Ramsey Clark (b. 1927), a progressive New Dealer in President Lyndon B. Johnson’s cabinet, became interested. In 1967, his department began exploring the computer industry and soon realized IBM’s central role. IBM’s lawyers felt it would become obvious that the computer industry was competitive and therefore IBM had not violated any antitrust laws. Discussions with IBM began almost immediately, with the company’s lawyers cooperating in 1967 and 1968. It was their belief that the industry was open to competition and could be accessed by many new rivals, contradicting the DoJ’s concern that IBM was constraining trade. Management wanted to quickly close the case. IBM presented the DoJ with 15,000 pages of supporting evidence. Federal lawyers chose initially to define the market in narrow terms (mainframes and associated peripherals), while IBM advocated for a broader definition. By the DoJ’s definition, IBM had few competitors. To use one expert’s description of the DoJ’s thinking, “Market concentration and the existence of substantial barriers to entry carried with them the presumption of illegality.”32

By the end of 1968, it was clear that the DoJ would probably file an antitrust case against IBM. In early January 1969, the government’s lawyers met in Clark’s office to weigh the issues. The Johnson administration was to end in a couple of weeks, so if after two years of work they were going to file a case, they had to move quickly. They knew the next administration, a Republican one, would undoubtedly not press charges, so they acted. On their very last day in office, January 17, 1969, a Friday, Clark’s DoJ filed suit against IBM and then everyone went home for the weekend, leaving it to the next administration to deal with it.

How did the DoJ inform IBM, and how did its employees learn about it? On that last day of the Johnson administration, Edwin M. Zimmerman, assistant attorney general in the Antitrust Division, called IBM’s vice president and general counsel, Burke Marshall, just as the parking lot at Armonk was emptying, with everyone going home. Zimmerman informed him that the case was being filed in New York City. Marshall then called various senior executives to share the news. Almost instantly, the news media began broadcasting the announcement; IBMers driving home up and down the east coast of the United States began hearing the news on their car radios. IBM quickly sent a notice to every IBM facility in the United States through its internal network and then around the world. Within minutes, IBM’s response was also being broadcast, over “the wires” of United Press International, Associated Press, and Dow Jones.

When the wider IBM community learned about the filing the following week, employees suspected something was not right; the announcement did not look good. Just the fact that the DoJ filed its case on the very last day of an administration did not pass the sniff test, tainting the suit right from the beginning and leaving a lingering odor of suspicion for the life of the case. By early February, IBM was explaining to employees the details of the case, the government’s position, and IBM’s, all in a matter-of-fact manner.33 The actions by the presiding judge added to the widely held belief within IBM that the case was bogus, but IBM’s lawyers quickly realized it would be the most important one of their careers, although none thought it could go on for more than three to four years. On the hopeful side, a Republican administration was about to take office the following week.

What was the complaint? The DoJ argued that IBM had violated Section 2 of the Sherman Act, which stated that no firm can “monopolize or attempt to monopolize … trade or commerce.” It summarized its case in 12 double-spaced typed pages. That was it. As the case lumbered along over the next decade, those 12 pages launched what U.S. solicitor general Robert Bork famously called “the Antitrust Division’s Vietnam.” The DoJ dived into the never-ending case, which consumed vast resources, filing the case in Manhattan in the U.S. District Court for the Southern District of New York, then let it sit there for nearly three years.

IBM’s lawyers wanted to get going on tussling with the DoJ. Tom Barr, IBM’s lead lawyer, already working on other antitrust cases for IBM and now responsible for this one as well, recalled that, “The antitrust division did nothing, and I mean absolutely nothing. They wouldn’t do anything to prosecute the case, and they wouldn’t talk to us about settlement.”34 Actually, some activity occurred, because in 1970 IBM began providing the DoJ with pretrial information, as is customary before a lawsuit, where both sides collect—“discover”—evidence needed for a trial. Soon, however, the DoJ’s lawyers stopped the process, deciding simply to use the information gathered for the CDC case against IBM, though it dated back to 1968 and that case had already been settled out of court. In January 1972, Judge Edelstein, chief justice of this court, assumed responsibility for presiding over the case.

Progress slowed again. IBM wanted a different census of the industry’s activities used, one that had already been developed by the federal judge involved in the CDC case. The New York court upheld the DoJ’s objection to IBM’s request. A new census was created that involved interviewing—deposing—800 people to create a file describing the computer market. The one not used included a survey of 1,700 companies. The cost to all parties to create this new census ran into the millions of dollars. The DoJ now submitted a broad definition that, as the trial unfolded, became too unclear for the case to progress in its favor, but the judge allowed it, while denying IBM’s request for a more precise definition of the market. Next, the DoJ asked the judge to order IBM to produce thousands of pages of material from the CDC case, even though the government’s lawyers had agreed earlier not to do that. Judge Edelstein agreed to the DoJ’s request. The U.S. Supreme Court sided with the judge, even though IBM claimed many of the papers involved privileged communications between clients and lawyers.

In October 1972, IBM asked that a trial be held quickly to define the market, and then hold the main trial for the antitrust charges on the basis of that first trial’s result. The judge refused. In 1974, the DoJ broadened the charges to include evidence that came out of the Telex suit. The DoJ wanted to add these to its case, since IBM was initially found guilty. Edelstein approved the DoJ’s request. Soon after, the 10th Circuit Court of Appeals in Denver threw out IBM’s conviction in the Telex case, but it was too late; the DoJ could now use the massive file. That summer, the DoJ secured permission to depose more witnesses; IBM was prohibited from doing the same.

Accounts of the suit routinely ignore one important constituency: everyone’s customers. What did they think? Their actions spoke loudly in the late 1960s and 1970s. They acquired computing products and services from every vendor in ever-larger quantities, but were they agnostic about the performance of one firm over another? In 1973, Datamation set out to provide insights by surveying 400 data processing managers concerning the eight largest computer manufacturers. The magazine focused largely on “after-sales service,” “product reliability,” “support,” and whether the company could “live up to claims.” IBM ranked number one and far ahead of all the others across the first three categories. In a separate question about cost versus performance, it came in lowest. In other words, customers thought IBM’s products and services were the most expensive. It asked respondents to rank a company’s ethics, and IBM came in a close second. Since the DoJ and other firms were hammering IBM on ethics, why were customers happier with IBM’s performance?35 The survey never resolved that mystery, but two conclusions were either they did not think the charges against IBM had merit or everyone else was worse than IBM.36

On the charge that IBM controlled the market, respondents disagreed. The survey asked them to vote to what degree they believed the statement “People are reluctant to buy something new until IBM gives its blessing.” Eighty-three percent of the respondents disagreed with the statement, because they believed they were qualified to judge the value of a new technology for themselves. Datamation’s writer expressed skepticism over their responses, but it was quite an emphatic result, and the respondents were knowledgeable about computing. They were the people most affected by market realities, by whether IBM had a stranglehold on the market’s behavior.37 Finally, since litigants accused IBM of driving rivals out of business, often citing the highly publicized case of RCA, an iconic U.S. corporation that had entered the computer business but without deep knowledge about how to compete in the world of computers, what did respondents say about RCA? They ranked RCA lowest in four out of six categories of performance. They hammered RCA over its ethics, inability to honor its claims, and the quality of its after-sales and support services.38 Their responses begged the question: Did RCA just not compete as well as other firms, including IBM?

Finally, on May 19, 1975, the trial began in the Manhattan courtroom. All parties assembled at 10:30 A.M., to begin what it would take the DoJ three years to do—present its case.39 The DoJ brought in 52 witnesses, 32,000 exhibits, and 72,000 pages of transcripts. File cabinets lined the walls of the courtroom. IBM’s motions for mistrial, submitted as the case shifted, were denied, so the trial continued. On April 26, 1978, the DoJ completed its presentations, and IBM began its defense that afternoon.

Meanwhile, the DoJ had swapped out its prosecutors, so it wanted to reinterview IBM’s witnesses. Edelstein approved their request. IBM had by now delivered to the court 300,000 pages of materials, so many in fact that to manage this huge file it created an online database for use by its lawyers. Later, it sold this software as a product to law firms. What the DoJ had now asked for would have entailed creating five billion pages of text drawn from the files of some 2,000 locations worldwide, which IBM estimated would cost it $1 billion to gather. The judge agreed to a compromise: Cary would testify about pre-1974 activities at IBM and about evidence from his prior depositions. Nevertheless, his driver had to bring him down to the courthouse for 10 days just for this phase of the trial. By now, IBM’s lawyers were fed up with the judge, and in July 1979 it asked him to recuse himself because of “bias and prejudice” against Big Blue. He declined to do so and was upheld on appeal. However, even the appellate court observed that perhaps a settlement would be in order. Nothing came of the recommendation, despite a few meetings.

Meanwhile, in January 1981, the Democratic administration of President Jimmy Carter ended, and a new one, led by President Ronald Reagan, a Republican, came into office. That meant new DoJ leadership and different priorities at the White House. Both sides completed presenting their cases in June. All through the trial years, IBM maintained its innocence. Cary stated, “I want all IBMers to understand that our competitive practices are fair and ethical. That is the way we have always conducted our business and the way we always will.”40 He never missed an opportunity to repeat the message. Each year, IBM included a one or two sentence update in its annual report to stockholders. It was almost unnecessary, as the business trade press routinely covered every step.

Regardless of IBM’s position, the bigger drama unfolded outside the courtroom. The case lasted so long that most of the circumstances on which it was based—actions of the 1960s—no longer existed. The industry had fundamentally changed. S/360s and S/370s were history. The industry now had a vibrant software market, another for midrange systems, yet another for desktop computing (largely PCs, worth billions of dollars), and a rapidly expanding IT services industry. Most of the competitors affected in the 1960s and early 1970s were gone. The major players in the early 1980s did not exist when the case began: Amdahl, Apple, and Data General, among others. Japanese competitors were vigorous and active in the U.S. market. The press was already describing the case using terms such as “anachronistic” and “irrelevant.” It was costing the DoJ millions of dollars and becoming a source of embarrassment to the White House.

Then, just after 4 P.M. on January 8, 1982, on another Friday afternoon, a deputy assistant attorney general, Abbott Lipsky, raised his tall, thin frame before the now aging Judge Edelstein in that familiar courtroom number 1105. Lawyers from both sides of the case, newspaper reporters, and industry watchers packed the room. The buzz for months had been that maybe the case would end soon. Lipsky uttered the words IBM’s defense team had wanted to hear for so many years, that “the case is without merit and should be dismissed.” A week earlier, on Wednesday, IBM’s legal team heard that the DoJ was about to bring the case to a close, so conversations began. On Thursday evening, Assistant Attorney General William F. Baxter, the chief of the Antitrust Division, signed off on shutting down the case. That fateful Friday morning, Tom Barr added his signature to the “stipulation of dismissal.” Baxter and his staff had spent a couple of months reviewing the mountain of evidence from the case and could not see sufficient reason for continuing. Barr felt vindicated, exclaiming later that day that, “Sixteen Federal judges have decided in our favor.” He added, “These claims are without merit. IBM has been completely vindicated.”41

Four hours before Lipsky stood up in that courtroom in New York instead of Baxter, who normally would have uttered those fateful words, the latter was in Washington, D.C., where he announced the conclusion of an out-of-court settlement of the government’s antitrust suit against AT&T, the world’s largest company. That suit had gone on for seven years, and its settlement resulted in the company’s breakup into 32 operating companies, forever changing how telecommunications evolved. The Reagan administration had decided to clean house of old antitrust cases and open a new chapter in the federal government’s relations with large corporations. Baxter found time to comment about the IBM case, saying simply that dropping it was “the only sensible thing to do,” since it had been based on “flimsy” evidence.42 He acknowledged that in addition to being a poor case, it had cost the U.S. government between $1 and $2 million per year for 13 years. His cost estimates seem low; nonetheless, it proved expensive.

IBM CEO John Opel was thrilled, calling the decision “wonderful news,” a “burdensome” case for IBM now over. Everyone at IBM Corporate seemed relieved. IBM’s media machine got out all the usual press announcements, arranged for interviews, and every IBM bulletin board around the world carried the news by Monday morning. Years earlier, Cary had called it the “Methuselah of antitrust cases,” and Baxter, the new U.S. attorney general, agreed. The case unfolded across four presidential administrations, the tenures of nine attorneys general, and three IBM chairmen (Watson Jr., Cary, and Opel). The New York Times called the conclusion of the AT&T and IBM cases “the end of an era in antitrust law and very likely the beginning of an era in information and telecommunications technology.” Business Week wrote that “Justice [Department] deserves the thanks of the country for blowing the whistle on a case that was weak to start with and became weaker with the passage of time.”43

But not everyone was happy. Richard T. DeLamarter, an economist who worked for the Justice Department on the case, published a book-length retrial of it, savaging IBM’s activities of the 1960s. DeLamarter wrote that “IBM is in fact a major problem for practically every country in which it does business, even the United States. By maintaining everywhere a self-perpetuating, lopsided market structure in which it towers over all competitors, IBM has amassed entirely too much power to be considered ‘good’ for any nation as a whole.”44 Five economists who worked on IBM’s side of the case defining the size of the computer market took the opposite view in two fat books. Three of those economists were blunt about the matter: “The government’s argument was based on erroneous conceptions of competition and a grossly inadequate understanding of the facts, particularly in respect to the nature and definition of the market in which IBM operated, the experience of other firms in that market, and the significance of IBM’s introduction and pricing of new products in the context of market developments.”45 Among the beneficiaries were business historians, who ended up with access to hundreds of thousands of pages of information about the computer industry.46

No event of this magnitude just ends on a Friday afternoon. Consequences for IBM unfolded over the next quarter century, casting a shadow over the company. But first, what happened to the key players? Attorney General Ramsey Clark, under whose watch it all began, went on to support various liberal political causes and to provide legal defense services to Saddam Hussein, among others. Born in 1927, as of this writing (2018), he was in his nineties. Judge David N. Edelstein retired from the bench in 1994. When the IBM case ended, he was three weeks shy of his seventy-second birthday. He died on August 19, 2000. William F. Baxter (1929–1998), who had his moment of fame with this case in 1981–1982, was applauded by economists and criticized by some in political life and law for not taking on large, vertically organized corporations. He spent most of his life as a law professor and in legal practice and died at the age of 69. IBM’s lead attorney, Thomas D. Barr, had joined Cravath, Swaine & Moore, IBM’s law firm, in 1958 and remained with the practice for 40 years. He died in 2008, one day after turning 77. IBM’s senior legal official, Nick Katzenbach, had been a prisoner of war during World War II, a Rhodes Scholar, a lawyer after the war, assistant attorney general in the Kennedy administration, deputy attorney general from 1962 to 1965, and then attorney general. IBM hired him as its general counsel in 1969, and he retired from IBM in 1986, after which he turned to private practice. Like so many of the players in the antitrust story, he, too, led a long life, passing away at age 90.

Cary and Opel rarely spoke in public about their time at IBM, and never about the antitrust suit. The only IBM CEO who did was Tom Jr.:

Looking back I see a lot of sad irony in the whole affair. I think a lot of people would agree that at the outset the Justice Department’s complaint had merit. IBM was clearly in a commanding position in the market, and some of our tactics had been harsh. We eliminated many of these practices ourselves, and our overall record during the case was pretty clean. But I’ve always thought that if Judge Edelstein had carried the suit along rapidly, we’d likely have ended up with a consent decree in which we might have formally agreed to hold back from announcing our machines until we were a little further along with their development, to loosen our grip on the educational market, and so on. Instead, the case stretched on unresolved for so long that before it was over history showed my arguments to Ramsey Clark to have been right. IBM kept growing, but the computer industry grew even more, and the natural forces of technological change etched away whatever monopoly we may have had.47

WHAT THE ANTITRUST CASE DID TO IBM, THE COMPUTER MARKET, AND OTHER CORPORATIONS

Inside IBM, accountants and executives were tabulating the costs of all these federal and private lawsuits. The U.S. government was doing the same. At IBM, the cost of a constantly expanding legal staff in the 1960s and 1970s became a permanent feature of IBM, with a population of dozens of attorneys expanding to hundreds. Historian Richard S. Tedlow estimated IBM’s costs just for the litigation of the 1960s–1970s in the tens of millions of dollars.48 A more accurate cost would easily be ten times that amount.49 Various observers spoke of the damage done to IBM’s confidence, its way of acting, and weakened élan vital. While the antitrust case was still ongoing, Watson Jr. had this to say about the rise of the lawyers: “It depressed me to see IBM back in the lawyers’ hands. The antitrust case began to color everything we did. For years every executive decision, even ones that were fairly routine, had to be made with one eye on how it might affect the lawsuit.”50

Tedlow cleverly picked up on a subtle change in how IBMers spoke and wrote since the end of the case. Certain phrases could not be used, such as “market share” and “crush our competition,” yielding to a new corporate language filled with code words. Slide decks carried less text, bulleted lists instead of sentences.51 Tedlow stated that “the antitrust suit traumatized IBM for more than a decade. The special ‘code’ being developed to deal with normal business situations introduced an element of unreality into the company.”52 While all of this was true, as we will see later when discussing the history of the Personal Computer, executive management also proved eager to reenergize the company in the 1980s. With new leadership coming during the 1990s, the effects of the lawsuits began to wane.

Step out of corporate headquarters and one could see others fretting over the rising power of lawyers at IBM. After all, it was the job of lawyers to keep IBMers out of jail and the company out of court. Sales thought the legal community was overly cautious. Every executive had assigned to them a lawyer they could or should go to for every significant decision, even to review comments to be made outside the firm. David Mercer, a British IBM manager, wrote in 1987 about doing business in World Trade: “One department that is uniquely powerful in IBM, where it probably barely exists in most other companies, is Legal. IBM has a long history of almost continuous legal battles with the U.S. government, typified by the decades long anti-trust suits. As a result it is pathologically nervous of the legal implications of almost any decisions or activities; details that other companies would not even consider contentious. That is reflected in a unique, and important, offshoot: the Business Practices department.” While the Business Practices Department was normally seen as a constraining force, Mercer admitted that the resulting “Business Conduct Guidelines” used worldwide made it easier for IBMers to know what they should and could not do.53 He also reminded us that the rise of the lawyers and their influence dated to before the antitrust suit, to the time of the consent decree of 1956.

Lost to history was another antitrust case that should be acknowledged. For reasons similar to those the U.S. Department of Justice had used, the European Union also explored the possibility that IBM had violated its laws. The charges were similar. Its litigation process took a couple of years and, in August 1984, IBM and the European Commission settled the case, with IBM agreeing to end some of its practices. The European Union started a new investigation of IBM’s mainframe maintenance practices in 2010. The regulatory probe ended in late 2011 when IBM agreed to modify how it provided information to its competitors regarding mainframes. The Europeans had learned from the U.S. cases. Joaquin Almunia, European commissioner for competition, explained: “Timely interventions are crucial in fast-moving technology markets.”54

Historian Steven Usselman made the point that the combative relations between the DoJ and IBM “never devolved into a simple doctrinaire feud between habitual adversaries.” Their disputes “involved complex questions regarding the ways in which market structure and firm organization shaped incentives and promoted or impeded technological change.” These U.S. suits involving IBM “essentially provided a forum in which lawyers, economists, and other experts actively worked through the issues” concerning how a modern corporation could function and the parameters set by the law.55 It is a discussion still under way, occurring with the EU in its dealings with Amazon and Facebook in the 2010s. Usselman was not alone in pondering the significance of the earlier cases. Lawyers and economists did, too, publishing scores of lengthy articles about the effects. A general cooling of antitrust prosecutions permeated the work of the Antitrust Division at the DoJ for the next quarter century.56 Economists debated the effects of emerging technologies on national economics and corporate behavior.57 Commentators combined the cases of IBM and AT&T, the first as a winner, the second often as a loser, in the high-stakes game of antitrust litigation.

But Watson Jr. may have come closer to the ultimate truth of the matter, that however more cautious IBMers became at all levels of the corporation, the markets kept growing. Opportunities kept expanding to such an extent that the company could increase revenues, expand profits, and grow its global footprint in the 1980s while still carrying the psychological burdens imposed by caution, slowed innovation, and increased bureaucracy. Tom Jr. hinted at the hit to IBM’s confidence and bravado that occurred. IBMers living through the period felt chastened, and when the government’s case ended, they were more relieved than excited. For most, it meant the company would not be broken into two or more pieces, as was happening to AT&T. Did the case affect other large corporations, particularly American ones? Any major U.S. corporation that was doing well in the United States was vulnerable to litigation either from its less successful adversaries or by the DoJ. Nobody could ignore the risks, especially when the DoJ experienced success over the course of the second half of the century.

Was the exercise of so many lawsuits worth it? The U.S. Department of Justice spent a fortune, yet the IBM case was essentially a loss. However, it got what it wanted out of the AT&T case. While the IBM case certainly diminished its appetite for antitrust suits, if it intended to foster more competition, it succeeded, as the 1980s and 1990s turned into a Golden Age of innovation in information technologies, although much of it would have come regardless of what happened to IBM. We know that because the company was not broken up or severely hampered, while new technologies came along, evolving rapidly into viable commercial forms. They surfaced from many sources, across numerous industries and countries. The most spectacular one, of course, was the explosive use of the Internet, which we now realize was one of the major events of human history.

Breaking up AT&T released pent-up demand for digital telecommunications waiting in the wings in Canada and Europe, which gushed forth in digital telephonic switches, cell phones, and mobile computing. Would IBM have unbundled its software if there were not concerns over potential antitrust moves by the DoJ? The evidence suggests potential government litigation pushed IBM to take that action. The antitrust cases facilitated the emergence of a large commercial software industry that might otherwise have remained far smaller for a very long time.

IBMers in the 1980s widely believed that the cost to IBM’s spirit and energy unquestionably had been severe. The dollars involved proved ultimately less of a factor than how the company operated. IBM could afford to carve out blocks of Cary’s time for the case because customers kept acquiring IBM products for all the reasons they had always done so. Stockholders remained supportive as well. Nevertheless, it would be difficult to find an IBMer who lived through the period who thought it was a positive experience. IBM’s U.S. employees remained critical of the U.S. Department of Justice for the rest of their careers and normally did not discuss it, even among themselves.

The Antitrust Division and the Department of Justice both changed over the course of the lawsuit. The IBM case was widely described as the DoJ’s “Vietnam,” the idea being that it lost its war against IBM and that the case lasted as long as the U.S. war in Southeast Asia, at enormous cost to the nation. However, Marc Allen Eisner, in his study of the DoJ’s antitrust activities, argued that it practiced poor economics in the 1960s but did far better by 1982. The Reagan administration limited the antitrust activism of the department that had so marked its activities from the 1950s through the 1970s,58 although the DoJ never gave up its valuation of the importance of market power in determining whether to pursue antitrust litigation.59

Historical accounts of IBM move from the antitrust cases to narratives of IBM’s experience with the Personal Computer. However, one large, unfilled part of the history of the company ignored or glossed over by journalists, economists, historians, and IBMers was IBM’s interactions with Communist Europe. If someone thought IBM World Trade lay slightly outside Domestic—the core of IBM—then what would they think of the Cold War world that existed on the other side of the Berlin Wall? The company’s experience there did not parallel what the rest of IBM underwent in Western Europe or the United States, and it is to that story we must turn next.

Notes