United States. Congress. Senate. Committee on the.

The Industrial reorganization act. Hearings, Ninety-third Congress, first session [-Ninety-fourth Congress, first session], on S. 1167 (Volume pt. 7) online

. (page 25 of 140)
Online LibraryUnited States. Congress. Senate. Committee on theThe Industrial reorganization act. Hearings, Ninety-third Congress, first session [-Ninety-fourth Congress, first session], on S. 1167 (Volume pt. 7) → online text (page 25 of 140)
Font size
QR-code for this ebook


at least two and maybe more, in addition to IBM, will find a ix>sition in the
field. The chances are that they will come from the top of the above list.

It is not easy to advise a client to withdraw from a market and forego what
might later prove to be a substantial profit opportunity. Our conclusion as
to IBM's monopoly power was questioned by many. Attachment 3 is a copy
of an editorial carried in one of our industry's foremost trade publications
shortly after our report was submitted to Bendix's top management. In essence,
the writer concluded that free enterprise was alive and well and that innovation
and superior price/product performance would tell. Of the authors 9 examples,
6 have been forced to withdraw from the market. Perhaps of more importance,
of the 16 companies that we had identified as serious entrants into the computer
market in the late IQoO's, ten (asterisked in the preceding list) have since
failed or withdrawn from the market, and of those remaining, none has achieved
more than a 10% share of the market.

It took almost two years and a total investment in excess of $40 million before
Bendix was able to extricate itself from the industry. They were lucky— they
sold the division to Control Data Corporation for $10 million in CDC stock (see
Attachment 4, and in turn sold their stock of CDC's all time high. When it was
all said and done, Bendix proved to be one of the few "winners" in the computer
industry.
Weakness of Competitors An Indicia Of Monopoly Power

Judge Sherman A. Christensen, in rendering his decision in the Telex vs. IBM
case [72-C-18, United States District Court for the Northern District of
Oklahoma— Sept. 17, 1^73] said :



5092

"The strength of competitors is relevant to an assessment of market power.
Monopoly power presupposes the power to control what happens in a relevant
market. Ease of entry may be an indication of lack of market power on the part
of an alleged monopolist. Difficulty in entering, weakness of competing companies
and dependence of competitors upon dominant forces in the market are among
indicia of market control on the part of an alleged monopolist. ... If the per-
centage of a relevant market controlled by an alleged monopolist is high an in-
ference of market power may be drawn. Where its control is moderate no inference
of market control may be permissible. In case of a medium range, it may be
impossible to infer or to rule out monopoly so that factors other than market
percentage must be looked to primarily. Where there is direct credible evidence
of market domination or predatory practices which are productive of control in a
particular relevant market, inferences need not be depended upon but this more
direct evidence may be determinative. Other factors to be considered are any
necessity on the part of an alleged monopolist to meet competition in technology
and pricing, the equality of performance in the industry and its comparative
youth, growth and dynamics or change."

Judge Christensen went on to say, "Claimed necessity of responding to com-
petitive influences beyond the control of the alleged monopolist may be only its
excuse for anticompetitive conduct for the purpose of maintaining or extending
monopoly power or to surmount threatened competition, and monopoly is possible
in a young, dynamic and complex industry, as well as in an old or static one, and
may be even more feasible in special cases through masking of selective market
strategies in the overall technological developments. Sophistication of users or
competitors may discourage monopoly but equal or greater sophistication on the
part of an alleged monoplist may be a counter-balancing factor, and industry
dynamics may continue in evidence through technological momentum beyond the
inception of monopoly." (Emphasis added)

Although many companies (less than two dozen by my count — more than 100
by IBM's count) have sought to carve out a profitable niche in the general pur-
pose computer systems market, few have succeeded. One must ask. how is this
possible? It is almost inconceivable that the vast majority of these companies
were poorly managed, under-financed, or incapable of satisfying a valid customer
need. Those who have sought to penetrate this market have lost in total literally
billions of dollars of their shareholders' money — yet their corporate power and
ability ranks them among America's most successful corporations — Ford, RCA,
Litton, GE, Bendix, Philco, North American Aviation, and Xerox to name only
a few — have become household words, yet none has been able to reach break-
even in the computer business. One can only conclude that a formidable giant
must guard the gate to the computer market place and to the land of profits.

Innovation

Fortunately, the attempts of other companies to enter the computer market
did not represent a compete loss. The industry and our nation benefited greatly
from these companies' transitory participation. In their efforts to penetrate the
market, they became the principal sources of innovation during the 1950-1970
time-frame. Though the record shows that IBM has obtained more than 10.000
patents, a substantial majority of the improvements in computer performance
have been first brought to the market by others. In almost every instance, IBM
has lagged behind rather than led the technology in the market place. An indica-
tion of their failure to innovate, is vividly shown in Attachment 5 and 6. Attach-
ment 5 was taken from the Minutes of the IBM Management Committee. Attach-
ment 6 is the summary portion of an internal IBM appraisal of their products
ver.sus those of their competitors. Both documents were entered into evidence in
the Telex vs. IBM case.

This is not to say that they have not been inventive — undoubtedly they rank
among the top two or three companies in the world in research and development.
However, they have seldom brought the fruits of their labors to the market placp
until forced to do so by competitive i)ressures.

For example, IBM points to the continuing improvement in the price/perform-
ance ratio of computers as an indicia of the competitiveness of the computer
market. The facts, however, show that costs and prices have largely been lowered
by conversion from vacuum tubes to solid state devices. The latter were based
on research by Bell Laboratories. Similar situations may be noted in the use
and application of virtual memory, time-sharing, remote terminals, and most
other important advances in the state of the art.



5093

The reason IBM invents— but doesn't innovate is obvious. If you rent your
product, and recover your investment in X months, the bulk of the revenues
derived beyond this point are profit. The longer the product can be kept in place,
the more profitable it becomes. Under these circumstances the motivation is to
maintain the status quo as long as possible— bringing a new product to market
onlv when competitive pressures make it mandatory.

If innovation has not been the source of IBM's dominance of the computer
industry, what has?

MONOPOLY POWER IX A HIGH TECHNOLOGY INDUSTRY

Judge Christensen answered this question in part when he said in his
conclusions :

C18 Correspondingly, the court concluded that . . . maintenance of IBM's
monopoly power in the relevant product market for plug compatible pe-
ripheral" products was not the result of IBM's superior skill, foresight, or
industry and was not the result of superior products, business acumen
or historic accident. To an extent, it was its failure, as IBM itself recognized,
to develop new technologj' and superior performing products as rapidly and
effectively as it had hoped, and the capability of plug compatible manu-
facturers" to keep abreast of, and in limited instances surpass, some of the
technological developments and jeopardized its monopoly position in the
relevant i)roduct market, and that motivated it to undertake predatory
pricing and long term leasing to stem the growth of its plug compatible
competitors.

Judge Christensen also noted :

C15 From its found predominant market shares, the court infers and
concludes that IBM had and exercised monopoly power in the relevant
market and submarkets here defined. Circumstantial evidence apart from
that relating to market share is indicative of IBM's market power in the
relevant market and submarkets as they have l)een defined. Its own
strategy, investigations, and planning were premised to an important degree
upon tiie assumption that it had such power. The very predatory intent
with which, as already has been found, its strategies were planned, as well
as the nature and direction of its competitive responses, strongly suggest
a consciousness of market power and a determination to utilize it to the
extent that it was considered this could be done without a breach of its
confidential plans or its becoming involved in legal difl5culties. This is
not to say that there was any ruthless or nakedly aggresive programs
contemplated or carried out : anything that was done by way of strategy
was sophisticated, refined, highly organized, and methodically processed and
considered. But in this day and age .such conduct is hardly less acceptable
than the naked aggressions of yesterday's industrial powers if unlawfully
directed against competition. The organized, selective, subtle and sophisti-
cated approach, indeed, may pose more danger under modern conditions
than instantly more obvious strategies.

IBM V.P. Confirms Judge Christciiscn's Findings

Although sequestered in the Telex vs. IBM case, a document was entered into
the public record in the U.S. vs. IBM case that bears upon IBM's use of a com-
bination of sophisticated and subtle techniques in its programs to maintain
absolute market control. This memo, written l)y IBM's Director of Business
Practices, and discovered in files of IBMs Director of Marketing, is shown as
Attachment 7 and a facsimile is reproduced below :

THOUGHTS FOR CONSIDERATION

The liability of IBMs risk lease is dependent on price leadership and price
control.

By means of price leadership. IBM has established the value of data process-
ing usage.

IBM then maintains or controls that value bv various means: (timing of new
technology insertinm ; functional pricing : coordinated management of delivery :
support services and inventory : refusal to market surplus used equipment :
refusal to discount for age or for quantity ; strategic location of function in
l»oxes : "solution selling" rather than hardware selling; refusal to support sub-
sequent use hardware, etc.



5094

Unbumdling has created a new threat to IBM's price control.

By eliminating fixed price solution selling of hardware eventually leading to
increased price competition.

Functional pricing already under pressure because of OEM activity. Unbun-
dling will increase that pressure.

Equalization of subsequent use adds value to existing third party inventories.

Legal problems are emerging as a result of certain practices which are key
underpinnings to price control.

Refusal to market used machines and parts.

Refusal to sell bills of material.

Maintenance parts prices.

The key underpinnings to our control of price are interrelated and inter-
dependent. One cannot be changed without impacting others.

These interrelationships are not well or widely understood by IBM Manage-
ment. Our price control has been sufficiently absolute to render umiecessary direct
management involvement in the means, [the next five or so lines were deleted
from document].

The D.J. [Department of Justice] Complaint specifically covers varying profit
margins and an intensive investigation of this issue would reveal the extent of
our price control and its supporting practices. Such a revelation would not be
helpful to our monopoly defense.

If IBM's price control is seriously threatened, either from the market or
(because of unbundling) or from legal exposure (because of the supporting prac-
tices) or from the D.J. (because of demands for remedy) it is necessary that
IBM Management fully understand its import in order to decide.

Negotiations" strategy with D.J. good vs. l»ad practices remedies good vs. bad
structural remedies (or) decision to litigate.

Pricing Approach to New Systems.

Reduction of Legal Risk relating to practices.
Recommenda Hon

Assemble a small, knowledgeable, secure group to think through these issues,
particularly in their interrelationships — define the emerging environment and
the various new forces which indicate significant new change — and map the
rudimentary elements of strategy or alternative strategies for consideration by
Management.

■■If if !i: if ^ * :ii

The effectiveness of these and other market control techniques has been obvi-
ous to those of us who have worked in the computer industry.

After more than two decades, IBM still holds almost 65% of the worldwide
installed base of general purpose EDP systems. Domestically IBM's nearest
competitor has a 9.4% share of the market and even this was achieved in part by
acquiring GE's installed base. The third ranking competitor, Univac. with an
8% share acquired RCA's installed base when they closed the doors on their
computer operation.

Tables 1 and 2 show the relative sizes of the principal domestic computer sys-
tems supplier. Although the revenues of the leaders are impressive by normal
standards they are, within the context of the computer industry and IBM's
dominance, relatively weak competitors with little influence on what happens.
Like Chrysler and American Motors, they exist at the sufferance of the domi-
nant force in the industry and provide the illusion of competition.

If IBM's control of this important industry is to be reduced, it is necessary
to understand how it was obtained in the first place and how it is perpetuated
today.

"Otir Price Control Has Been Sufficiently Ahsolntc . . ."

IBM policies and pricing in the area of computer systems is consistent wirn
a long corporate history of exploitation of customers and the exclusion of com-
petitors. In the 1920's the firm dominated the market for office tabulating equip-
ment to much the same extent it now reigns over the computer industry. Then,
as now, IBM wished to extract maximum profits from customers ; the device
it settled upon was to effectively charge high-demand users a higher price for
their machinery than low-demand users by requiring all customers to purchase
tabulating puncli-cards from IBM itself~at inflated prices. In this fashion high-
use buyers contributed proportionately more to IBM profits than customers with
less need for the tabulating services. "The courts eventually declared this practice
illegal under Section 1 of the Sherman Antitrust Act.



5095

As IBM emerged as the dominant ;nipplier of computer equipment it again
devised ingenious metliods for profiting from and holding onto its monopoly.
Though the technique of tying computing supplies to computer equipment was
foreclosed by IBM's earlier confrontation with the Antitrust Division, the same
ends were accomplished through slightly different means. The approach taken
by management was to refuse to sell either computer mainframes or periplieral
equipment outright, instead offering only month-to-month leases. Maintenance
had to be purchased from IBM, with the minimum maintenance fee geared to
a particular customer's usage.

U.S. MANUFACTURER MARKET SHARES-INSTALLED BASE (GENERAL PURPOSE EDP SYSTEMS, 1973)

Dec. 31, 1973, installed base

United States Foreign Worldwide

Value Percent Value Percent Value Percent

(millions) total (millions) total (millions) total

IBM - $17,406 63.8 $11,232 65.4 $28,638 64.4

PCM'" - - 1,325 4.9 295 1.7 1,620 3.6

Honeyweir" 2,578 9.4 2,036 11.9 4,614 10.4

Univac 2,205 8.1 1,378 8.0 3,583 8.1

Burroughs 1,421 5.2 823 4.8 2,244 5.1

CDC 973 3.6 740 4.3 1,713 3.9

NCR - 737 2.7 468 2.7 1,205 2.7

DEC "" - 134 .5 53 .3 187 .4

Xerox"" - 390 1.4 81 .5 471 1.1

Others.. 132 .4 68 .4 200 .3

Total __ 27,301 100.0 17,174 100.0 44,475 100.0

Note: Totals may not add due to rounding.

Source: International Data Corp. (conference report Mar. 5, 1974, pp. 36.1-36.3).

1973 MARKET SHARES

ELECTRONIC DATA PROCESSING
MAJOR U.S. MANUFACTURERS



$949 - 7. 0%



Control Data
$573 - 4. 2%



National Cash Register

.fzas - 2. 1%

Digital Equipment Corp
$265 - 2.0%



>-)t



Note: TotaLs jT.a/ not add due to




Sourco: .Standard £: Poors except:

IBM - CIA adjustment of Martin Simpson, Standard k Po

Honeywell - 1973 Annual Report



5096

Tying Arrcmgements and Discrimination

In two ways this lease-only, required-maintenance marketing structure en-
abled IBM to effectively charge higher prices for computing services to high-
demand customers. First, customers with a greater need for computers would
tend to want more peripheral equipment and, with a monopoly of both main-
frames and peripherals, IBM could (and did) charge inflated peripheral prices.
In this way the leasing of peripherals served to meter customer demand for
data processing, and just as in the case of punch-cards and tabulating machinery
25 years earlier, higher profits were extracted from high-demand users.

The second method of discriminating in price between customers lay in terms
of the maintenance contract required of all lessees. More intensive users had to
pay "extra-shift differentials" in addition to the minimum maintenance fee on
the pretext that extra use resulted in added wear and tear on the machines. In
point of fact, this provision simply forced the customer who found computers
most useful to contribute a disproportionate share of IBM's earnings. The 1952
government antitrust suit sought to end this practice.

The Justice Department signed a Consent Decree with IBM in 1956, supposedly
curbing IBM's abuse of its market power ; the lease-only element of its marketing
policies was eliminated, however, maintenance continued to be tied.

As in 1932, Sherman-Clayton had served to pry loose 1 of the 10 fingers closed
around the customer's thoat. The free enterprise system was somehow supposed
to loosen the others. Once again, the tenacity and creativity of IBM's manage-
ment prevailed. Additional strategies were developed and implemented in order
to freeze out potential competition.

The most significant anti-competitive policies through the 1960's were the
maintenance lock and "software bundling" (the provision of operating and
applications software at no extra charge to lessees). Both practices tended to
erect insurmountable barriers to new company entry.

The Cost of Competing Has Been Kept High

No new firm could be expected to absorb the enormous start-up costs of dupli-
cating either the IBM software library or its nationwide consulting and mainte-
nance network in order to provide services to tlie scattered IBM customers
choosing to buy rather than lease. Therefore any new competition would have
to come from a total systems company which would have to risk the massive
capital outlay required to design, manufacture, market, maintain, and provide
software for a new line of systems. The potential rewards tempted RCA. GE,
Philco, Bendix and others ; their attempts resulted in some of the more renowned
financial fiascos in American corporate history.

Until the late 1960's IBM's competitive stance was basically passive; the
barriers to entry erected by long-term strategies liad been successful in protecting
its monopoly, and the firm was rarely required to take visible offensive action
against competitors. But after the introduction of IBM's "System/360," a crack
in the IBM armor was opened by one of the very practices which had earlier
enabled the company to price discriminate among customers— separate pricing
of mainframes and peripherals.

Predatory Pricing

In 1967 the Telex Corporation and other small electronics manufacturers
began producing peripheral equipment designed to replace IBM tape drives, disc
drives, and printers. Their products simply plugged into the IBM mainframe—
hence the terms "plug compatible manufacturers" or "PCM's"— and were tech-
nologically superior yet less expensive than the IBM equipment they replaced.
By 1970, business for PCM's was booming.

PCM's had captured roughly 10% of the plug-compatible market, leaving, of
course, the remaining 90% to IBM. The IBM response was quick. A series of
price cuts was initiated in 1970 which, according to internal IBM documents
subpoenaed by Telex in its successful antitrust suit against IBM, were calculated
to kill off the plug compatible competition. These were followed in 1971 and 1972
with long-term leasing plans (prohibited for 10 years by the 1956 Decree) designed
to further erode the viability of the plug-compatible competitors. To offset
planned revenue decreases in the peripherals area, mainframe prices were raised
to maintain IBM's average profit at 32%. (See Attachments 9, 10, 11.)

Judge Sherman Christensen in his September 1973 decision in Telex t\ IBM
characterized IBM tactics as ". . . unlawful predatory conduct . . . intended to . . .
maintain its monopoly position." The effect on the PCM's was devastating. Telex,
for example, lost over half its sales from 1971 to 1972. and has yet to operate
profitably since the IBM attacks.



5097

The long IBM history of employing marketing strategies which exclude com-
petitors with a view towards maintaining its monopoly power and profits is a
textbook Ciise of the abuses the authors of the antitrust statutes intended to
prevent. When unchecked, IBM complacently reaped the fruits of monopoly ; when
Challenged, it responded with vigor to bring competitors — no matter how small —
to financial ruin.

Traditional antiti-ust enforcement has focused largely on eliminating blatantly
illegal practices such as collusion in restraint of trade, reciprocity, below cost
selling, price fixing, and other such obviously unethical practices.

Perhaps it has been the failure to understand that the modern-day monopolist
employs a variety of far more subtle techniques to maintain his monopoly power,
that has caused the Department of Justice to miss the target on two prior occa-
sions. Examples of some of the more subtle techniques now in use are provided
below.

Computer Systems and Black Boxes

As Mr. Faw has so helpfully pointed out in his "Thoughts for Consideration,"
". . . solution selling rather than hardware selling . . ." is a key underpinning
of IBM's monopoly power.

At the very beginning of the design cycle, the architecture of an IBM system,
including the interconnection and interactions between each of its separate
functional parts, is structured to achieve efficient data processing and to lock the
customer into IBM products and services.

This technique is employed in the design of the logical, mechanical, and elec-
trical interfaces between the central processing unit (mainframe) and the sub-
systems such as disc drives, tape drives, memories, and printers. It is also used
in the design of the software that operate the system. Several examples serve to
illustrate the interaction between architecture and monopoly power.

Hanlivarc Interfaces

With the introduction of the IBM 370 family of computers, IBM standardized —
internally — on the interfaces between its central processing units and its periph-
eral devices. This move briefly increased the computer user's flexibility by allowing
him to replace various devices as his needs changed.

In 1969, however, IBM realized that many customers were interconnecting
peripheral devices produced by other suppliers — devices that were superior in
performance and/or lower in price than the IBM equivalent — and took steps to
remedy the situation. As one of several strategies aimed at stemming the com-
petitive tide, IBM moved the electi'onic controller out of the peripheral device
and into the computer mainframe. This forced competition to redesign its prod-
ucts and also restricted their opportunities to innovate in the electro-mechanical
design area.

Because IBM hardware interfaces are kept secret until first product shipment,
a user or a competitor wishing to interconnect must obtain the new product,
reverse engineer the interface, and then design their device so that it works
properly with the host IBM computer, obviously lo.sing valuable time. It is this
time that IBM's marketing force uses to sign up customers on one- or two-year
contracts. When the competitor finally gets to the market, he finds that it is fore-
closed.

If GE generated electric power and made electric appliances, an analogous



Online LibraryUnited States. Congress. Senate. Committee on theThe Industrial reorganization act. Hearings, Ninety-third Congress, first session [-Ninety-fourth Congress, first session], on S. 1167 (Volume pt. 7) → online text (page 25 of 140)