Producer or predator
The computer industry is younger than most of the people who are reading this article, yet it already rivals the automobile industry for first place in affecting our lives. IBM (International Business Machines) dominates the industry and is one of the five largest corporations in the United States. It employs more than a quarter of a million people, has a gross income of $10 billion a year, and, in 1971, paid $1.1 billion in income tax.
Since 1968 a dozen antitrust suits have been filed against IBM. The most significant of these are the suits filed by the Telex Corporation, Control Data Corporation (CDC)—one of IBM's main competitors in the manufacture of computers—and the Department of Justice. The Telex suit went to trial in 1973 and was lost by IBM; the decision is still under appeal. The CDC suit was settled out of court in January 1973. The government suit is still pending and could result in the break-up of the giant company.
The purpose of this article is to provide some background and analysis of these suits and IBM's position in the computer industry.
We may divide the computer industry into several segments, the first consisting of main-frame manufacturers; that is, manufacturers of processing units and main memories. There are only about half a dozen major companies in this area, which is not surprising, since a very large amount of time and capital is required to develop, produce, and market a computer system.
Another segment of the industry consists of manufacturers of peripheral equipment such as disk and tape drives, typewriter and graphics terminals, and other input/output devices. The main-frame companies also make peripherals for their own machines. Several firms, including Telex, make plug-compatible peripherals; that is, peripherals specifically designed for computers made by other firms.
Software and service firms make up a third section of the industry. Included here are companies that do contract programming, consulting and systems analysis, lease computers, rent computer time and supply educational services. There are over 1000 firms in these areas, many of which are small. Barriers to entry are low and competition is intense.
IBM is the major main-frame manufacturer and it dominates most of the other sectors as well. Sixty years ago it was an uninspiring, unprofitable collection of companies called the Computing-Tabulating-Recording Company which produced scales, tabulating machines, and time recorders. In 1913, when Thomas J. Watson was given the job of running CTR, its gross income was a little over $4 million. Ten years later, when Watson changed the name to International Business Machines, income had more than tripled. By 1954, shortly after IBM sold its first computer, income was $460 million. On the average, gross revenue has more than doubled every six years and after tax net income has consistently been above 10 percent!
The government's pending antitrust suit is the third one it has filed against IBM. The earlier suits (1932 and 1952) concerned the company's activities in the tabulating machine industry, but the 1956 consent decree (which settled the 1952 suit) placed significant restrictions on IBM's computer business , and is responsible for the very existence of some segments of the computer industry. It required that IBM sell its machines as well as lease them—there are now over a hundred firms whose sole business is leasing IBM computers! They buy from IBM, depreciate the machines over a longer lifetime, and charge lower rental fees. (Recently some of these leasing companies have expressed their desire that IBM be forced out of the leasing business altogether.)
Another provision of the decree aided the growth of the computer services business. It required that IBM operate its services business as an independent (wholly owned) subsidiary, Service Bureau Corporation (SBC), and that any equipment and services provided for SBC be on the same terms as for any other company. Ironically, as part of the settlement of the CDC suit last year, CDC bought Service Bureau Corporation from IBM, and since CDC is not under any antitrust restrictions, it expects SBC to be much more profitable! 
The 1956 decree also required IBM to train personnel and provide technical manuals for anyone who owns, repairs, maintains, or distributes IBM electronic data processing equipment. The decree also ordered IBM to grant unrestricted licenses to companies desiring to use its patents, and enjoined IBM from instituting any suits for patent infringement which occurred prior to the date of the decree.
The aim of these provisions was to cut IBM's share of the market and to "enhance" competition. IBM adjusted its activities to conform with the decree, and then, seemingly treating the new regulations as minor annoyances, continued to grow and prosper. In 1968 the new attack began.
THE RECENT SUITS
The Justice Department and the numerous companies that have filed antitrust suits since 1968 charge IBM with monopolizing and intending to monopolize the computer industry. The central theme underlying these suits is that IBM's size, structure, experience, and salesmanship make competition difficult and therefore are unfair.
The Justice Department and CDC suits begin by citing statistics and characteristics of the computer industry which illustrate IBM's dominance. IBM's share of the main-frame computer market is roughly 70 percent (and has been as high as 90 percent). IBM spends more than $200 million a year on research and development, more than the total value of computers shipped by any other company. IBM has obtained thousands of patents which it uses to "entrench its monopoly power." IBM dominates the education market, training people on IBM machines, and a substantial number of personnel in other firms were formerly employed by IBM. Thus, there is a large pool of computer users who are predisposed to use IBM equipment. IBM has large quantities of accumulated software and related services to offer its customers. IBM has spent many millions of dollars advertising ("touting," to use CDC's word) its extensive installations in all industries, its software libraries, and its massive force of systems analysts, etc.
Certainly all of these factors make the job of IBM's competitors very hard, but they are simply evidence that IBM does a good job and has been productive in the computer field for a long time!
Most of the actual "offenses" listed in the suits concern IBM policies allegedly designed primarily to take business away from competitors. A survey of the major complaints follows.
IBM anticipates and usually gets very high profits (20-30 percent before taxes) on its products. One might think that the suits would criticize these profit rates. However, IBM's "price umbrella," as it is called, is responsible for the existence of many firms such as Telex which sell and lease at lower prices. Telex complained in its suit that IBM deprived it of the right to compete—by reducing prices to meet Telex's competition. Similarly, the Justice Department suit complains that IBM introduces low-profit machines in markets where competitors have been very successful. The government has asked that IBM be prohibited from producing equipment on which it is not likely to make profits comparable to those it expects on its other products. Apparently some firms want to compete with IBM but don't want IBM to compete with them!
Until shortly after the government suit was filed, IBM priced a computer to include software and related support (for example, maintenance, and training of personnel) thus making it more difficult for firms that specialize in software and support services to compete. This practice, called "bundling," often required buyers to pay for a lot of services they did not want at all or could have obtained more cheaply elsewhere, but they wanted IBM equipment enough to accept the package deal. Several of the suits list bundling as one of IBM's offenses, claiming that the firm's dominance of the industry makes it an anticompetitive tactic.
In June 1969, IBM announced an extensive plan to unbundle. The plan included a three percent decrease in computer prices coupled with separate charges for training customer personnel, for some software, and for other services. The unbundling plan was hailed by many software firms as a great boon to competition and, in particular, to their business. Others complained that it did not go far enough and have continued to press for further unbundling. Some firms were very critical of IBM's decision to consider a large number of its programs in the public domain and hence available for free, since naturally no one would pay a software company to write the same or only slightly better programs. 
The suits complain that IBM practices price discrimination (by outright discounts and other means) favoring customers in areas where competition is intense or where IBM stands to lose a customer or gain other benefits such as prestige. Here again, in the name of fair competition, IBM is not supposed to compete.
The government suit and several firms specifically criticized IBM's extremely large discounts to universities, the reasoning being that students who learn how to program on IBM machines will be predisposed toward them later and that some of the computer science research done at the universities may be mainly relevant to IBM machines. That the university, as well as IBM, benefited from the discount was ignored. In 1969 IBM modified its discount policy for universities (much to their distress) and now gives the same discounts as other manufacturers.
IBM is also accused of discriminating against companies that buy rather than lease equipment, by providing poorer and slower service. Such tactics were forbidden by the 1956 consent decree.
During the late 1960's there was a trend of IBM main-frame users toward buying peripherals from other companies because of the poorer quality, higher prices, and remote delivery dates of some comparable IBM products. (It should not pass unnoticed that IBM loses customers when they can get better or cheaper service elsewhere.) According to testimony in the Telex trial, in 1970 IBM formed a (secret) task force to find ways to reverse the trend. This task force recommended long-term lease plans, very large price cuts (disguised by giving existing models face-lifts and new model numbers), and strategic timing of product announcements to confuse the marketing plans of other companies and keep computer users always expecting something new from IBM. IBM allegedly would plan several improvements and new models for a product in advance, then announce them one at a time until they regained the sales they had been losing. All of these policies drew heavy criticism from Telex in its suit.
Aside from strategic timing of announcements for products it really produced and sold, IBM is accused of announcing machines with specifications it could not and never intended to meet in order to cause delay or cancellation of orders to competitors. A full page of the CDC suit, entitled "Paper Machines and Phantom Computers," is devoted to this charge. CDC believes it was victimized twice by IBM's announcements. In the early 60's when CDC decided to enter the scientific computer market (where IBM was not active), IBM announced a computer for the same market. The specifications were spectacular, but they were not met. After at least four delays in the delivery date and modifications of the design cutting the power in half, less than a dozen were ever delivered. In the mid 60's, when CDC introduced its model 6600, IBM, admitting that the 6600 was the fastest computer on the market, announced that it was developing one that would be "more powerful." Again there were delays, and IBM never really sold many of its high speed models. In both situations, however, its announcements caused serious losses for CDC.
Although it is not identified as such by IBM's opponents, the issue underlying CDC's criticisms of IBM announcements is whether such activities are fraudulent, or reflect poor planning and the uncertainties of designing new computer systems in a rapidly changing technological environment. To what extent can or should a company be held legally responsible for the claims it makes in its product announcements? Certainly, the tactic of announcing "phantom computers" is unethical, and won't work indefinitely.
Charges of IBM coercion, intimidation, and misrepresentation are frequently made, but details are rarely included. The CDC suit devotes a section to these charges and gives as an example of coercion the setting of quotas for salesmen. Some users have complained that IBM threatens to stop maintenance service or cancel leases if the user attaches equipment made by a competitor to an IBM main-frame. One writer, very critical of IBM, cites the statements "Of course, it's your neck that's going to be on the chopping block if you select non-IBM equipment and it fails," and "You could be making a serious mistake by not maintaining contact with IBM technology," as typical examples of threats against data processing managers who make acquisition decisions. The weight of these "threats," he says, derives from the fact that computer vendors, particularly IBM, serve as an informal employment service for data processing managers. The same writer complains that IBM circulates "inflammatory literature" to discredit its competitors. His examples include articles from the WALL STREET JOURNAL and other non-IBM sources. (He proposed that IBM "be restrained from the use of criticism of their competition, whether stated verbally or written by an IBM employee or other individual." Apparently he objects to freedom of speech!)
Much of what IBM's opponents call coercion and intimidation is high pressure salesmanship. If IBM uses its considerable influence in the computer industry against individuals who have crossed the firm, it is guilty of a breach of ethics, but even IBM's critics do not accuse it of criminal activity in this area.
There is no doubt that IBM does many of the things it is accused of, but, except for those which may involve force or fraud which should be resolved in a criminal court with IBM's size and position in the computer industry being irrelevant, its actions are illegal only if the firm is engaged in "unfair competition." Thus, the legal cases depend on whether IBM is simply trying to protect its business and profits or is trying to "hurt" its competitors. The distinction is more than a little obscure.
In the Justice Department case, when IBM asked for a separate trial to define exactly which market it supposedly monopolizes, the government vigorously opposed the request. The 70 percent figure often quoted as IBM's share of the market refers only to the main-frame market, but it is quite clear that IBM's opponents want to curtail its operations in all areas of the computer industry. The government, CDC, Telex and others have asked that IBM be split into several smaller firms: the points they debate are the number and type of pieces, and how to ensure that none of the pieces become strong.
The Justice Department has admitted several times that it knows little about IBM and the industry. In order to provide "industry input" into the case, a group of very vocal IBM opponents (executives from about eight competitors, mostly peripherals makers, including Telex) formed the Computer Industry Association (CIA). The CIA has diligently drawn up plans for splitting IBM, submitted a proposal to the Justice Department for "early relief," held numerous meetings, and issued myriad statements encouraging the government to take action against IBM. The CIA's activity has not been in vain. Several of the proposals in its early relief plan, submitted to the Justice Department in February 1973, found their way into the decision in the Telex trial. These proposals and the provisions of the decision indicate that the aim of IBM's opponents is "a piece of the action," and that they are likely to get it via antitrust.
One of the major complaints of the CIA is that IBM does not release the information needed to design plug-compatible peripherals. According to the president of the association, IBM destroys "their competition by announcing new industry standards which, because of their prior knowledge, give IBM an unfair competitive advantage." By "new industry standards" he means simply new design of IBM products. The statement continues, "Anything IBM does becomes a de facto industry standard, and if your product isn't compatible with IBM nobody will buy it" (emphasis added). Thus he is complaining about "standards" which are enforced not by IBM but by computer customers! (Such distinctions would undoubtedly be lost on the gentleman.) The CIA proposed that IBM be required to disclose complete specifications of the interface between all its products. The judge in the Telex case ordered IBM to disclose "information describing the design of the electronic interface" to each peripheral.
The CIA proposed that IBM be forced to release customers from penalties for lease cancellation with 30 days' notice, thus voiding provisions of contracts that have already been signed. It also asked for restrictions on penalty clauses in future contracts. Both proposals (with slight modification) were part of the Telex decision.
The CIA proposed that IBM be enjoined from pricing at "a low-profit or no-profit level with the intent or effect of eliminating competition." Telex charged IBM with predatory pricing during the competition mentioned earlier over the peripherals market. The judge found IBM guilty and enjoined the company from predatory pricing—even though its profit margin on the products in question was at least 20 percent. Observe that IBM's 30 percent profit margins are commonly considered prima facie evidence of its monopoly power, while its 20 percent profit margins are considered evidence of predatory pricing! (It should be emphasized that the CIA offered its early relief plan as suggestions for immediate injunctive action. Their long range goal is dissolution of IBM.)
Telex won the case and was awarded triple damages amounting to $350 million, which is more than five times its annual gross revenue. (The judge later reduced the fine to $259 million.) There was one offense brought out in the trial that involved violation of a company's rights: Telex was found guilty of infringing IBM patents and stealing IBM trade secrets!
The Telex case inspired several small companies to get in on the pickin's. One company which had filed a suit in 1971 announced after the Telex decision that it was considering filing another. A month later it dropped the earlier suit and the plans for a new one—in exchange for a $3.5 million contract to do product development work for IBM. The president of the company stated, "We're cancelling our previous suit in return for our contract."
A few more examples should suffice to illustrate the goals of IBM's opponents.
CDC's suit complained of IBM's "anticompetitive" and illegal monopolistic practices, but CDC settled out of court for some attractive business concessions. It acquired Service Bureau Corporation, the free use of IBM equipment at SBC (for six months), about $30 million in research and development contracts from IBM, an agreement that IBM would stay out of the data services business for six years, and other benefits worth millions of dollars.
An executive of Memorex, claiming that financing, not technology, is the major problem facing every company in the industry except IBM, has proposed that IBM's "excess capital" be made available to users to finance computer purchases from other manufacturers. He didn't seem concerned with why IBM had this capital, nor did he explain why it was "excess." (Memorex is in serious financial condition as a result of its unsuccessful attempt to enter the main-frame business.)
Information about how IBM operates is a valuable commodity. The firm has had to turn over to the government, and to companies that have filed suit, more than 40 million pages of documents. Many individuals, organizations, and firms in the industry have hailed the release of once confidential IBM papers, claiming that the disclosure of such information is of "major benefit to the industry and the nation." The most notorious public release consisted of some 40,000 pages of IBM documents that went into the trial record in the Telex trial. The court prepared a copy of the documents for the Computer Industry Association, which sold copies at $5000 each to other firms. The documents include confidential minutes of top level management meetings, discussion of an entirely new computer system that would require fundamental changes throughout the industry, detailed reports on profits and costs for individual products, details of research and development expenditures, pricing and profit policy, plans for gathering information on competitors, and perhaps of most interest to computer users, IBM's own "consumer reports"—results of its very detailed critical analysis of its computers which rated several models of its System 360 inferior to other machines on the market.
Clearly these documents will be valuable reading for IBM's competitors and for anyone interested in the computer industry. Just as clearly, I think, these eager readers have no more right to them than they have to read anyone's personal papers without permission.
THE USERS' VIEWS
The antitrust suits have been filed by the government and by competitors of IBM. The public statements have been issued mostly by competitors and politicians. What do the customers think?
DATAMATION, a monthly magazine for computer users, carried out a formal survey of 400 data processing managers aimed at finding out what users think of the eight major main-frame manufacturers. In the areas of "after-sales service," "product reliability," "support," and "live up to claims," IBM ranked highest, being far ahead of the second ranked company in the first three of these categories. IBM ranked lowest in "product performance per dollar." Finally, IBM was a close second in the "most ethical" category. The discrepancy between the user viewpoint of IBM's ethics and that presented in the antitrust suits and trade journal articles implies that either the charges are without substance or customers find other companies worse than IBM.
DATAMATION asked its panel to rank the factors they considered when making acquisition decisions. Most of the respondents placed product performance and vendor support ahead of price. However, those who use non-IBM main-frames usually ranked price ahead of product performance. There isn't much difference of opinion about exactly what IBM's good points and bad points are.
Many critics claim that other companies produce better computers and incorporate more of the latest technology into their machines. Such claims are often true, but to many customers they are beside the point. The critics overlook the importance to the computer user of reliability, program support, and services.
I have mentioned the complaint by the president of the Computer Industry Association that IBM sets the standards in the industry. DATAMATION tested the accuracy of the claim that "IBM controls the release of new technologies" because "people are reluctant to buy something new until IBM gives its blessing." Eighty-three percent of the respondents said it's not true; they trust their own staff's ability to judge new technologies. (The DATAMATION editor was skeptical.)
Sixteen percent of the panelists who replaced IBM memories or peripherals with other manufacturers' equipment said that IBM maintenance service (on their remaining IBM equipment) declined. Apparently IBM sometimes gets nasty if you deal with its competitors.
The rankings of RCA in the survey were enlightening. The fact that RCA, a large, well-financed company, withdrew from the computer business is used by IBM opponents (the Computer Industry Association and the government) as evidence that IBM "destroys competition" and that immediate government action is needed to protect it. RCA ranked lowest in 4 of DATAMATION's 6 categories; it received extremely poor ratings in the ethics, lives up to claims, after-sales service, and support categories. It looks like it was competition, not the lack of it, that drove RCA out!
This survey of the charges in the antitrust suits and the motives of IBM's most vehement critics and opponents shows that the suits are really attacks on success.
What has been responsible for IBM's fantastic success and growth? Although IBM didn't produce the first electronic computer, it has been the major impetus in the creation and promotion of the whole field of information processing by computer. It did not simply come along and take over an existing market; it created the market. IBM's sales campaigns have convinced companies to acquire computers to do jobs that were previously done without them. This policy of seeking out new uses for its machines has always been an integral part of its marketing policy. A study of the young computer industry written in 1955 stated that "IBM's outstanding long term growth record reflects excellent management, a superior research organization, aggressive sales policies and a capable service department. Large expenditures for research and engineering have kept the Company in a strong competitive position."  This statement would be an accurate description of IBM at any time since shortly after Watson took over. The four areas listed—management, research, sales, and service—are all areas where IBM excels, according to its strongest critics as well as its friends.
The government suit is scheduled to go to trial late this year. Some observers see the great complexity of the case and the importance of IBM in the industry as reason to believe that the government won't take strong action against the company. The result could be just the opposite. Recognizing the extraordinary amount of time and money needed to gather and organize evidence for such large antitrust cases, Senator Philip Hart has introduced legislation that would replace the expensive, time consuming step of proving guilt with statistical criteria such as profit rates and share of the market.  He cited the IBM case in support of his claim that the government should be permitted to break up big companies without having to prove any wrong doing. This attitude is an all too common one which will continue to exert a strong influence on the IBM case. The Justice Department, the Computer Industry Association, and some IBM competitors and critics agree with Hart; IBM's size and success are crimes enough.
Sara Baase teaches computer science and mathematics at San Diego State University. She received her Ph.D. in mathematics from the University of California at Berkeley. In addition to her teaching career, she is chairman of the Libertarian Alternative in San Diego.
NOTES AND REFERENCES
 IBM ANNUAL REPORT (1972).
 The complete texts of the CDC and Justice Department suits appear as appendixes in Bruce Gilchrist & Milton Wessel, GOVERNMENT REGULATION OF THE COMPUTER INDUSTRY (AFIPS Press 1972).
 William Rodgers, THINK: A BIOGRAPHY OF THE WATSONS AND IBM (Stein & Day 1969), p. 81; Thomas G. Belden & Marva R. Belden, THE LENGTHENING SHADOW: THE LIFE OF THOMAS J. WATSON (Little, Brown, & Co. 1962), p. 123; and Arthur D. Little, Inc., THE ELECTRONIC DATA PROCESSING INDUSTRY (White, Weld, & Co. 1956), p. 68.
 The text of the consent decree appears in Gilchrist & Wessel, op. cit.
 Ibid., p. 167.
 "IBM's 'Unbundling' Move Makes It Target of Growing Criticism by Computer Makers," WALL STREET JOURNAL, July 24, 1969, p. 4.
 "Control Data Settlement Seen as Setback By Others Suing IBM," WALL STREET JOURNAL, January 22, 1973.
 The Telex suit charges IBM with monopolizing the market for plug-compatible peripherals for IBM main-frames.
 "IBM Discloses Plan for Separating Its Computer and Services Prices," WALL STREET JOURNAL, June 24, 1969, p. 38. Prior to the unbundling, one software company actually tried to obtain an injunction prohibiting IBM from providing for its customers a free sorting program similar to one the plaintiff rented for $200 per month. The judge denied the injunction, but not because IBM had a right to give away its programs, nor because the injunction would force users to pay for something they could have had for free. His reason was that the evidence did not show that IBM provided the program mainly to take business away from the other firm, or that the firm was damaged irreparably. ("Bid by Programmatics to Restrict IBM in Distributing Computer Program Denied," WALL STREET JOURNAL, June 8, 1969, p. 8).
 COMPUTERWORLD May 2, 1973, numerous articles.
 "Datamation in Business and Science," DATAMATION February 1961 and January 1962 for rescheduling and reduction of power. See Rodgers, op. cit., pp. 254, 258 for CDC's attitude.
 The CDC 6600 was developed on contract from the AEC, as were several later CDC "number crunchers" (big, fast, scientific computers). CDC provides the federal government, the computer industry's largest customer, with 65 percent of its computers priced over $1,500,000. (Only 21 percent of all the federal government's computers were made by IBM.) INVENTORY AND SUMMARY OF FEDERAL ADP ACTIVITIES FOR FISCAL YEAR 1972 (General Services Administration).
 "IBM Says It Will Develop Giant Computer Twice as Speedy as Fastest Current Model," WALL STREET JOURNAL August 19, 1964, p. 6, and Rodgers, op. cit., p. 288, pp. 304-06.
 When IBM began breaking promises about delivery of some System 360 models in 1965, other companies ran advertisements reminding users that they, unlike IBM, did not repeatedly promise what they could not produce. Rodgers, op. cit., p. 288. CDC has also missed deadlines for several of its scientific computers, most recently its STAR computer.
 Anonymous contribution to "The Forum," DATAMATION October 1973, pp. 183-86.
 For the CIA's early relief plan see "CIA Gives Justice Early Relief Plan," COMPUTERWORLD, February 7, 1973, p. 25. Details of the Telex suit and decision are in WALL STREET JOURNAL articles "Telex Asks $877 Million In Damages From IBM for Alleged Monopoly," January 25, 1972, p. 21, "Telex-IBM Trial Begins," April 13, 1973, p. 9, and "IBM Loses Antitrust Action for First Time," September 18, 1973, p. 3; and DATAMATION articles "Telex vs. IBM" July 1973, pp. 65-68 and "Decision in Tulsa," November 1973, pp. 134-35.
 Dan McGurk, quoted in "Newsdata," COMPUTER DECISIONS, August 1972, p. 4.
 "Potter Instrument Decides Against Suing IBM After Winning Work From Big Firm," WALL STREET JOURNAL, October 15, 1973, p. 6.
 IBM ANNUAL REPORT (1972) and DATAMATION, February 1973), p. 98.
 "Guzy Plan: IBM Should Spread Its Wealth Around to Users," COMPUTERWORLD, January 10, 1973, p. 23.
 "Inside IBM's Management," BUSINESS WEEK, July 14, 1973, p. 46.
 Richard A. McLaughlin, "Monopoly Is Not a Game," DATAMATION September 1973, pp. 73-77.
 Arthur D. Little, Inc., op. cit., p. 67.
 Industrial Reorganization Act, S. 1167, CONGRESSIONAL RECORD, Vol. 119, No. 38, pp. S4361-S4366. Section-by-Section Analysis available from Senator Hart. See also DATAMATION May 1973.