Control Data Australia Memories compiled by Brian Membrey

“The U.S. Justice Department’s Vietnam”


The Justice Department's interest in IBM dated back to January 21, 1952 when the U.S. Government brought the corporation before a court in a hearing that spent around four years investigating the company's conduct in the market for punched card and tabulating equipment, later extended to include "electronic data processing machines".

The 1956 consent decree [1], brought down by Judge David N. Edelstein, Chief Judge of the Federal Court for the Southern District of New York required the Armonk, New York-based IBM to sell its machines as well as lease them as had been their exclusive previous practice, to provide the same services to purchasers of equipment with charges at the same rates as leased machines, and also to provide service and sell parts and documentation for IBM computers after they were no longer owned by IBM. The latter provision ensured a market would be created in used equipment that competed with IBM's new machines and limited its monopoly power in the computer market.      

[1]  A “consent decree” is an agreement or settlement that resolves a dispute between two parties without admission of guilt (in a criminal case) or liability (in a civil case), and most often refers to such a type of settlement in the United States. Such agreements are often brokered by a judge, but without involving a legal ruling. The 1956 consent related only to punched card equipment where IBM held an estimated 90% of the market, but was later interpreted as also including electronic computers.

Control Data Corporation at this time was, of course, just a figment of William C. Norris’ imagination, but the judgement had a significant impact in the later settlement of the CDC - IBM suit :

“Section VIII (a) - IBM is hereby ordered and directed to transfer, within one year after the date of the entry of this Final Judgment, all its contracts for service bureau business to a corporation (hereinafter called the Service Bureau Corporation), which may be wholly owned by IBM, and IBM shall thereafter be enjoined and restrained from engaging in the service bureau business except on a nondiscriminatory basis for the Service Bureau Corporation and for service bureaus operated by other persons”.

(IBM was later forced to divest all interest in SBC and restrain from competing in the data services industry for six years as part of the settlement of the CDC - IBM suit).

Additional restrictions to ensure an “arm’s length” operation  prevented the Service Bureau Corporation from using any corporate name containing the words “International Business Machines” or “IBM”, from utilising the service of any person also employed by IBM, or for employee of SBC to solicit on behalf of IBM any order for the sale or lease of any IBM equipment. The decree also restricted SBC, after a period of three years’ grace, from subleasing more than 20% of its service locations from IBM.    Read the full judgement

(Somewhat beyond the scope of this coverage, but IBM was still attempting to overturn some of the conditions of the 1956 consent decree as late as 1996 when it was entangled in legal disputes with Microsoft. A settlement came in June of that year with the approval of U.S. District Judge Allen Schwartz in Manhattan to phase out over five years the remaining provisions of the landmark 1956 consent decree, most of which related to mid- and mainframes, especially the AS/400 series; he had already approved in January the elimination of certain provisions relating to PC-based products where IBM faced considerable opposition  A spokesman for IBM suggested had cost it more than $100 million annually by restricting how the company sold and serviced computers).


Although the 1956 decree was effective in the short-term, considerable disquiet had arisen by the mid-1960’s over IBM's dominance in the rapidly emerging new "computer" industry and many of the practices the corporation was adopting, including the announcement of a number of models directly in opposition to rivals, but which in many cases were nothing more than "vaporware" intended to convince potential customers to hold off purchasing equipment from  rival vendors until a "real" IBM alternative was available.  

On 17 January, 1969, around five weeks after Control Data filed its suit (based on much the same "vaporware" concept) and following several years of investigation, U.S. Attorney General Ramsey Clark instructed the Department of Justice to initiate an anti-trust suit against IBM.

The Anti-Trust case had a whiff of political intrigue from day one – the 17th of January was the last working day of the Lyndon B. Johnson administration and many observers believed Clark took the action to ensure the efforts of the incumbent government had not gone to waste, rather than leave any decision to the new Richard Nixon-appointed Attorney General.

The Government's 1969 case does not appear to have sought financial penalties – in reality, the U.S. Government suffered no financial disadvantage form IBM's activities – instead it sought structural reorganisation based on IBM illegally acquiring and maintaining a monopoly of general purpose digital computers through "exclusionary and predatory conduct going beyond the 1956 Decree”.

(A letter issued by Control Data chairman CEO, Bill Norris after the corporation's out-of-court settlement with IBM suggested that as a minimum this would be have IBM split to mainframe, plug compatibles and software companies, but he conceded that while technically feasible, it would have been "politically unpopular" and unlikely to happen – the connotations of the term "politically unpopular' remain a fascinating unknown!)

The Government contended that from 1961 to 1969, IBM engaged in anti-competitive practices in "for the purpose or with the effect of restraining or attempting to restrain actual or potential competitors from entering the relevant markets"; specifically,  price discrimination such as giving away software services for the purpose of enabling IBM to maintain or increase its market share; that IBM's bundling of software with related computer hardware equipment for a single price was anti-competitive; and that IBM predatorily priced and pre-announced specific "fighting machines" knowing they had unusually low or no expectation of returning a profit, but were simply designed to discourage customers from acquiring competing products before IBM could produce a “real” machine..

In fact, little more was heard of the action until immediately after the settlement of CDC's case early in 1972 – the Justice Department openly admitted in pre-trial hearings that it lacked both the technical expertise and knowledge of the computer market to move its case forward, and that it was leaning heavily on Control Data's expertise in these matters.  CDC's support appears to have been withdrawn after the settlement, although it is not clear whether this constituted part of the agreement or just a fact of life after the corporation disentangled itself from the legal dispute

IBM was still negotiating out-of-court settlements with several smaller claimants, but shortly after the conclusion of the CDC suit and the failed attempt to have the controversial evidence database restored, Judge Edelstein assigned the Justice Department case to himself, although the trial itself did not commence until May 19, 1975.

From the outset, the Government, perhaps again exposing its lack of knowledge of the computer industry estimated that the presentation of its case would last 60 days. Instead, it took six years and with no discernible result either way!


For all practical purposes, the U.S. Justice Department Anti-Trust case against IBM ended in a New York City Federal courtroom on 1 June, 1981.  

Judge Edelstein was still presiding over the Federal Court case and it was suggested that final summations were still at least seven months away, when Thomas D. Barr, IBM's lead attorney in the six-year-old trial asked for a conference with Edelstein and William F. Baxter, head of the Justice Department's Antitrust Division.

Barr suggested two other failed anti-trust decisions against giant corporations - Eastman Kodak Co. and American Telephone & Telegraph - had crippled the government's case against IBM, and this, "together with the vast changes in the industry since 1969 [2], compel us to find a way to stop this unprecedented waste of public and private resources".  

  [2] In terms of technological advances, it could be said that there had been two "generational changes" in the intervening twelve years  and although not expressly referenced in the final judgement, many observers believed that the emergence of PC products which were relatively inexpensive to manufacture had exposed to more competition than it had ever contended with previously

Barr's contention was that even if Judge Edelstein ruled that IBM had a monopoly over the market in question (a point which of course he did not concede), the government could no longer hope to prove that IBM behaved illegally because of the recent rulings.

Baxter agreed to the postponement, but said he would not discuss the case or settlement options until he had finished studying the case and offered no prediction when he would complete his investigation

On 8 January, 1982, the Anti-Trust case was formally abandoned when Baxter declared it was "without merit".

If there was ever an gold medal for concise language, Baxter surely would have won it by the proverbially "country mile" – although there are some minor variations in the figures, one account suggests that the court had heard testimony from 81 live witnesses, heard 928 depositions from other witnesses, 11,644 documents had been presented, at one stage IBM had 200 attorneys working on the case and the transcript of the case amounted to over 100,000 pages – obviously all "without merit"!

In a later press conference, Baxter concluded ...

"the record is very, very weak on the existence of any such unpermitted or unlawful practices ... IBM continues have a very substantial market share, depending on how one defines the market precisely".

Perhaps more significantly, he added :-

"it would not be impossible to say that IBM has a monopoly position within the meaning of Section 2 doctrine, although that is a debatable question. But it is not a violation of Section 2 to have a monopoly position; one must have either attained it unlawfully or maintained it by unlawful acts, and it is perfectly clear from the record that IBM obtained its very large market share in an entirely legal way".

The main thrust of the Government's case was that IBM was able use to its massive wealth to absorb a temporary loss of revenue when selling products at low prices that denied the competing firms a sufficient profit to survive.

Newspaper reports at the time put IBM's worldwide revenues in 1981 at a lazy $72 million PER DAY, and it had maintained a position as the second-most profitable corporation in the world (behind Exxon Corporation) since the anti-trust case began; it was however noted that its share of the mainframe market had dropped from 70 to 62 percent since the case commenced in 1969.

The Government case had relied heavily on IBM memos which appeared to show top company officials following a price-cutting strategy.

One such document (which IBM claimed was “privileged information” but which came to light during the Justice Department case) quoted was from a meeting of IBM top executives in 1965, when the company's competition with Control Data Corporation was discussed, the author, IBM attorney, H. Barrow Farr, noting that Control Data in its Annual Report had publicly blamed its declining earnings on IBM "and its frequent model and price changes. There was some sentiment that the charges were true".

CDC Says …

Very little was reported of the testimonies presented, but one report that appeared in Computerworld in conjunction with a product announcement (18 July, 1977) was from Gordon Brown, CDC's vice-president of marketing and planning (right).  

Rather than any direct reference to IBM, Brown primarily addressed the existing marketing environment with generalised references to the latest draft of Control Data’'s long-range plan “which understandably CDC wanted to keep confidential”.

He suggested the the corporation had reduced its expected growth in OEM products to 1980 from 18 percent down to 15 percent per annum, although he added he "personally favoured" 12 percent.

Claiming the OEM market was more sensitive to slowdowns in the general economy, Brown's reasons were given based on an expectation of two more recessions over the next five years, but he believed that on the other hand during depressed conditions, users were more likely to invest in peripheral products to improve performance of their mainframes, and thus CDC were anticipating substantial growth in peripheral systems (including IBM-compatibles) which he set at 22 percent.

As a final prediction, Brown suggested a consistent 10-percent growth in business products (which unlike our operation in Australia, also included printed business forms), referring at one point to the product line as “a cash-cow”..

Apparently questioned directly on the corporation’s competition with IBM, Brown stated that IBM’s activities did impact CDC’s short and long-term planning and the corporation was highly aware of the need to evaluate where its products stood in relation to those of IBM ...

"the business risks are probably most critical in the plug-compatible business ... these risks can include unforeseen product announcements, price cuts and the potential for for subsystems to be rendered inoperable and/or incompatible by IBM actions in microcode encryption techniques".

(The Charles Babbage Institute archives include an oral interview with Gordon Brown in which he may have elaborated more, but unfortunately, it is one of perhaps a half dozens such interviews with CDC executives that is not available on-line).

David Norton Edelstein (1910-2000)

was born in New York City. Edelstein received a B.S. and an M.A. from Fordham University, and an LL.B. from Fordham University School of Law.  After working for several years in private practice, he joined the U.S. Justice Department in 1944 and was appointed a Federal Judge in November, 1951 by President Harry S. Truman.

Edlestein served as the chief judge of the Southern District of New York from 1971 to 1980 and for years was the longest-sitting active judge in the country.  He was still serving when died in New York on 19 August, 2000 at 90 years of age - obituaries noted his 48 years on the bench where he presided over two of the Justice Department's most historic undertakings -- the antitrust action against IBM and his long ongoing effort to rid the Teamsters union of corruption.




Telex v IBM : the winner is ... Curiouser and Curiouser 1973 : The I.B.M. settlement The New York Times says ... The New York Times says ...