It occurs to me that some of the non-lawyer members might like to
learn a bit about the ins-and-outs of how antitrust cases are put together
and presented in court. What one has to prove in order to win is of course
determined by what the Supreme Court (and the federal appellate courts) has
demanded in the prior set of cases that most closely matches the facts of
your case--the relevant "precedents." If they say you have to prove that
the moon is made of blue cheese, then that's what you have to prove--on pain
of losing your case.
The parties typically hire not just lawyers but economic "experts,"
people who specialize in the economics of monopoly and competition. (The
American Economic Association lists about 2,000 of its members who include
that field as one of their areas of special interest.) Since the U.S.
courts routinely decide antitrust cases on the basis of what they call
"economic analysis," it is generally fatal to your case to neglect to hire
such an "expert" and present him in court. He is, in effect, the High
Priest who intercedes on your behalf with the robed Deity.
You can use your antitrust economist in one of two roles. Since you
will be paying him by the hour ($250 to $500 per 60-minutes, all billed up
front), he will of course insist that it is imperative to the success of
your case for him to master the "facts" of your industry before he takes the
witness stand. He will need to visit your factory, study your files, review
the industry's history, interview your key employees and those of other
firms similarly affected, and so on--with his hourly meter humming along at
a total cost of, generally, $100,000 to $1 million.
If you are poor and/or frugal--antitrust plaintiffs, on average, are
some l/30th the size of antitrust defendants, and often bankrupt as
well--there is an alternative method of using your economic expert. Instead
of paying him (at an obscene hourly rate) to immerse himself in all the
"facts" of your case and your industry, you can simply hire him to make a
court appearance and give the judge and jury his expert opinion on the
economic implications of a set of "assumed" facts--all in response to what
the lawyers call a "hypothetical" question. In fact, it is not hypothetical
at all. To be effective, it must recite the relevant facts of the case with
meticulous accuracy, since the cross-examiner will be relentlessly asking,
"Your opinion, Professor, would be the opposite, would it not, if any ONE of
your 'assumed' facts in counsel's question should be found by the jury to be
wrong?"
Antitrust lawyers and economists deal with scores and even hundreds
of industries over the course of their careers (in my case, 39 years as an
antitrust professional, with industries from steel and cement to breakfast
cereals and supermarkets). Each has its differences but the economic
principles are, in the final analysis, always the same in them all. Adam
Smith was quite right: Monopoly is an "absurd tax" by the few on the many
and the cure is competition--preferably 20 or so vibrant firms in each
industry and market.
Every industry has millions of "facts" but not all of them have
antitrust relevance. Good antitrust lawyers and economists--particularly
the honest ones who aren't trying to maximize their own fees (an admittedly
small group)--routinely present new clients with a questionnaire and this
instruction: Have your best people put together for me the answers to these
questions, bring them back to me, and I'll then tell you whether you have an
antitrust case. The questions can be put on a couple of pages. The
answers, to say the least, are rarely so brief.
The following is one of the articles from my journal (Vol. 26, No.
4) that's included in my Web site (under the "Selected Articles" button).
Another that I especially commend to the members is entitled "Glossary of
Antitrust Terms." The public interest can only be advanced when those with
technical expertise in a monopolized industry are prepared to learn the
basics of monopoly law and policy.
Charles Mueller, Editor
ANTITRUST LAW & ECONOMICS REVIEW
http://webpages.metrolink.net/~cmueller
*******************************
ANTITRUST LAW & ECONOMICS REVIEW
Vol. 26, No. 4
ELEMENTS OF AN ANTITRUST CASE:
STRUCTURE, CONDUCT, AND
PERFORMANCE
Charles E. Mueller
The 'Industry Study'
The elements of an antitrust case are today derived largely from the
economic art form known as the "industry study," a factual description of an
industry cast in the analytical framework of the industrial-organization
economist. Economists have been studying monopoly for some 200 years
now and have worked up what might be called a laundry list of factors that
determine where an industry stands on the spectrum between competition
and monopoly. If a medical doctor is asked whether a particular patient is
"healthy," he has to conduct some tests before he can answer. How is the
patient's blood pressure? His heart beat? His blood sugar count? In other
words, he has to conduct an investigation, one designed to answer a series
of checklist questions. Once the answers to these questions are known, any
competent medical doctor can give a reasonably sound answer to the
original question about the patient's health. Similarly, there are a series˙
of
factual questions that have to be asked and answered about a particular
industry before a competent economist can say whether it is a competitive
industry, a monopolized one, or something in between those two extremes.
The Economic Expert in Court: A Hypothetical Examination
What are the questions? Suppose one of the parties in a monopoly
case puts
on the stand an expert economic witness, a properly qualified
industrial-organization economist but one who has no factual information
about the industry in question. The questioning might go something like
this:
Q. Dr. Smith, can you tell his Honor whether or not this industry is˙
a
monopoly as alleged?
A. No.
Q. Why not?
A. Because I don't know anything about it.
Q. What would you have to know in order to be able to answer the
question?
A. A lot of things.
Q. Does that mean that you can provide us with a list of factual
questions
that, if answered, would in turn permit you to answer the ultimate question
of whether the industry is competitive or not?
A. Yes.
Industry Data and Economic Analysis
Q. So if we put on the stand one of the industry executives who
knows all
there is to know about the industry, and get him to answer your list of
questions, you can then tell us whether the industry is effectively˙
competitive
or effectively monopolized?
A. Yes. Any competent industrial-organization economist can do that.˙
Q. How can you do that?
A. The competitiveness of an industry is determined by certain˙
causal
factors and evidenced by others. From the presence or absence of those
factors in a given industry, plus their magnitudes, we can determine
whether the industry fits the effectively competitive economic model, the
monopoly model, or some in-between variation.
Q. Is there any single factor that, if known, would permit you to
answer that
question?
A. No.
Q. Do you have to have the answer to every factual question on your˙
list
before you can answer that ultimate question?
A. No, not necessarily. But the more of them that are answered the˙
more
confident I can be of my diagnosis. As the answers pile up, they start to
become cumulative, simply confirming what was already pretty certain from
the facts already determined.
Two Investigative Techniques:
The 'Field' and 'Library' Alternatives
Q. Suppose you didn't have an all-knowing industry expert to provide˙
you
with the answers to your factual questions. Could you develop the answers
on your own?
A. Well, there are basically two ways to gather the kind of factual˙
data
needed to determine how competitive an industry is. One way, as you have
suggested here, is simply to go out and ask the executives of the firms in
the industry to tell you what you want to know. Most of the basic facts
about an industry can be gotten from a couple of knowledgeable--and
cooperative--officials. If I was an employee of one of the antitrust˙
agencies,
for example, I would probably be able to find someone in at least one or two
of the smaller firms who would be more than happy to fill me in on what I
needed to know. As an academic economist, I might have to work a little
harder but I could probably get the answers in due course.
An alternative investigative technique--actually, a complementary
one--is to
simply use the library. A considerable amount of information can be
developed on an industry by consulting such standard sources as the
government's statistical publications, business reference services, trade
publications, and of course the academic studies already done on the
industry, that is, the doctoral dissertations and masters theses. These˙
latter
studies, some of them running to several hundred pages of factual
information on a single industry, are sometimes complete enough to answer
virtually all of the key questions on the list. Indeed, the authors of those
studies were probably working with the same list of questions I would be
using.
'Economic' Versus 'Legal' Investigation
Q. Does that mean a skilled economic researcher can get out of a
library all
the information that would be needed to bring a monopoly lawsuit?
A. No, that's another issue entirely. The answers one can get out of˙
the
library will often tell you whether or not, as a matter of economic fact,˙
the
industry in question is effectively competitive. But whether that kind of
data is sufficient to establish the legal fact of monopolization is a˙
question I
am not qualified to answer. My own experience is that the antitrust
agencies use--or should use--the "library" investigation and the "field"
investigation sequentially. That is, one does the library investigation˙
first
and, if it suggests that the industry is in fact noncompetitive and that˙
this
condition can be remedied by an appropriate decree--and is worth the
bother--then a full field investigation is launched. The purpose of the˙
latter
investigation is two-fold: First, it confirms the findings of the library˙
study
and, second, it brings in the kind of evidence that will be legally˙
acceptable
to the courts as proof of the facts that, while probably already known from
the library investigation, might not have been provable in court through
that data alone. Data sufficient to convince a scientific investigator may˙
or
may not be admissible in a court of law. One is economic foundation, the
other is legal superstructure.
The Problem of Priorities
Q. What do you mean by "worth the bother"? Shouldn't the antitrust
agencies investigate all monopolies?
A. Not necessarily. Assuming their budgets are limited, they will be
physically unable to commence an investigation of every industry in the
economy in the first year of their antitrust program. Ideally, they would be
trying to maximize the public benefits from their work each successive year.
To do this in the first year of the program, they would have to have a
complete list of all noncompetitive industries, ranked in the order of their
monopoly overcharge figures, less the costs of making them competitive. If
there were 100 industries on the list and the two agencies only have
enough money to bring 10 cases each, they would presumably want to begin
by suing the 20 industries that showed the highest benefit-cost ratios, that
is, the highest dollar yield to the consumer from each budget dollar spent.
In succeeding years they would presumably want to continue working their
way down the list until they reached the point where it would not pay to
restore competition in the remaining ones, i.e., the point at which the
consumer losses from the remaining monopolized industries were less than
the cost of making those industries competitive. It is difficult to˙
visualize
the reaching of any such point within the foreseeable future.
Developing Antitrust 'Planning' Data
Q. Where are they going to get this list of noncompetitive
industries and the
monopoly "overcharge" figures you talk about?
A. That is where the preliminary or "library" investigation comes˙
in.
Another way to describe it is in terms of a "planning" investigation. While˙
it
would be prohibitively expensive for the antitrust agencies to conduct a˙
full
field investigation of every single industry in the United States, it would˙
be
well within their current budgetary capabilities to do a "library"
investigation across the board. On the basis of the published data assembled
by these economic researchers, it should be possible to (a) "screen out"˙
those
industries that are either effectively competitive already or that are not
"worth the bother" as discussed above and (b) estimate the probable amount
of the monopoly overcharge in the remaining industries. Perhaps 10% of
the FTC's current annual budget_that is, 10% of about $30 million_would
be enough to develop the preliminary "planning" data I am talking about
here on an economy-wide basis.
The Structure-Conduct-Performance Analytical Structure
Q. All right. Please tell his Honor what is it you would have to
know about
the industry we are concerned with here before you could tell us whether it˙
is
in fact noncompetitive in character.
A. I would need to know three things, namely, the industry's
structure, its
conduct patterns, and its several performance characteristics.
Q. What do those terms mean?
A. Each of them has several factual components. The term "structure"
refers
to, at a minimum, the industry's (1) concentration ratio, (2) the degree of
product differentiation present in it, and (3) the height of any additional
entry barriers around it. "Conduct" refers principally to the behavior˙
pattern
employed by the industry's firms in arriving at their price, product, and
output decisions, particularly whether they make those decisions
independently, collusively, or interdependently. The term "performance"
refers to the ultimate economic results produced by those structural
features and conduct patterns, that is, to its (a) efficiency, (b) price
stability,
(c) technological progressiveness, and (d) contribution to the general goal˙
of
equity in the distribution of income.
A Hypothetical Industry--Structural Elements
Q. If I gave you a hypothetical set of data on each of these
elements of this
industry's structure, conduct, and performance, could you then answer the
question as to whether or not it was effectively competitive?
A. Yes.
Q. Very well, I will ask you to assume as a fact that this industry˙
is
structured as follows:
First, concentration. There are 12 firms in the industry, with total
sales of
$10 billion. The largest of these firms has annual sales of $4.5 billion, or
45% of the industry's total sales. The second largest firm has 12% of the
market; the third, 8%; and the fourth, 6%. None of the other firms has more
than 3% nor less than 1%.
Second, product differentiation. The industry spends 10% of its
total sales or
approximately $1 billion per year on advertising, primarily TV advertising.
A new entrant would have to spend twice as much proportionally--20% of its
sales--for a period of at least 10 years in order to sell the output of a˙
single
minimum-efficient size plant at a price comparable to that of the largest
established firm. Alternatively, the new entrant would have to accept a 20%
lower price for that period of time in order to secure that minimum volume˙
of
sales.
Third, other entry barriers. The industry's product can be produced
at the
lowest possible unit cost in a single plant accounting for 5% of the˙
industry's
total sales. To create a firm of that size from scratch--including the˙
necessary
plant, selling organization, and the like--would require an investment of
$100 million. At that plant size, unit costs are no higher than in larger
plants. Below that plant size, unit costs rise rapidly: for each 1˙
percentage
point below the 5% minimum-efficient size, unit costs rise by approximately
10%.
Conduct
Turning to the industry's conduct patterns, there is no evidence of
outright
collusion on either price, product, or output volumes. Nor is there any
evidence of independence of decision here, either. Rather, there is a very˙
high
degree of interdependence in all of these areas. The largest firm is both˙
the
price leader and the style leader, with the other firms carefully following
both its prices and its product changes. Exclusive dealing is practiced in˙
the
industry, with each firm--including even the smaller ones--maintaining its
own system of "exclusive" retail distributors, those that handle no˙
competing
manufacturer's product and confine their sales efforts to a prescribed
geographical territory. There is no evidence of predatory or exclusionary
pricing in the industry, at least not in the last several decades.
Performance
In terms of the industry's performance, I will ask you to assume the
following: (1) Each of the four largest firms has had an average after-tax
return on its stockholders' equity of more than 20% over the past two
decades. The remaining eight firms have had average returns of 8% to 14%
over that period. (2) The minimum unit cost of producing the industry's
product is $100, a cost that is, as noted previously, realizable by a firm
producing 5% of the industry's output. The industry's average price last
year was 50% above that level or $150 per unit. (3) During recessions, the
industry raises its prices in order to maintain its 20% target rate of˙
return
on stockholders' equity, with output falling as low as 50% of capacity.
During the expansionary phases of the business cycle, the industry also
raises prices but at a somewhat slower rate than during the recession
periods. (4) There have been no significant technological advances in the
industry since 1910. Minor improvements made in the last two decades were
developed either by the smaller firms in the domestic industry or by even
smaller foreign competitors.. (5) 10% of the industry's sales dollar, or
approximately $15 per unit, is, as I mentioned a moment ago, spent on
advertising. There are no technical differences, however, in the products of
the 12 firms; all sell a physically identical product except for the˙
different
brand names and minor styling variations. In terms of performance, there
are no measurable differences between the 12 competing products.
Dominant-Firm Oligopoly and Consumer Prices
Now, Dr. Smith, assuming all of these facts to be true, can you tell˙
us
whether this industry is effectively competitive?
A. Yes.
Q. Is it?
A. No. You have described the classic case of an asymmetrical
dominant-firm oligopoly of the most socially-injurious kind. On the basis of
this industry's noncompetitive structure--high concentration, high degree of
product differentiation, and high entry barriers--one would expect
approximately the kind of conduct and performance your hypothetical
assumes, namely, non-independence in pricing, excessive prices,
technological stagnation, and so forth.
Q. Can you estimate the amount by which prices would fall if the
industry
should be made competitive?
A. On the basis of your assumed fact that the industry is selling a
product
for $150 that can be produced and sold for $100, including a competitive
rate of return on the capital employed, one would expect prices to fall to˙
the
latter level if the industry should be made effectively competitive. With
sales currently running at $10 billion per year, this implies an aggregate
price reduction to consumers of one-third that total, or about $3.3 billion
per year.
Restructuring and Improved Performance
Q. But can the industry be made effectively competitive without
sacrificing
some productive efficiencies?
A. Well, your hypothetical says that unit costs can be minimized by
a firm
with sales of only $500 million, or 5% of total industry-wide sales. A˙
decree
reorganizing the industry into 20 firms, each with 5% of the national
market, would permit all of those firms to operate at the most efficient
scale. And of course an industry in which the four largest firms account for
only 20% of the market--all of equal size--would be quite competitive.˙
Prices
would fall to the competitive level, as would costs and profits.˙
Technological
innovation might also reappear in the industry. And with prices less rigidly
controlled, the industry might well be unable to "beat" the business cycle˙
in
the future by cutting output in order to maintain prices during periods of
recession. Lowering prices instead of output would obviously contribute to
fuller employment and thus to the overall health of the economic system.
Q. No further questions.
Elements of the Industry Study: Price Overcharge Figure
While this hypothetical "testimony" hardly provides a comprehensive
checklist of every item of information the industrial organization economist
would like to have in evaluating the competitiveness of an industry,1 it is
sufficient to illustrate the general thrust of the art form known as the
industry study:
It should describe the key structure-conduct-performance features of
the industry.
It should give particular attention to the extent to which the
industry's actual price exceeds the price that would be expected to
prevail if the industry should be made effectively competitive.2
Without this central piece of information, the antitrust agencies are
unable to determine whether an action to make the industry
competitive would be "worth the effort" or, if so, what rank to give it˙
in
their own lists of priority industries.
It should also give special attention to the issue of scale economies.
Unless one knows how large an establishment is needed in the
industry in order to realize the lowest possible per-unit costs, there˙
is
no way to determine whether the industry can be made effectively
competitive without making the individual firms too small to be
optimally efficient in producing the product.
A recommendation should be made as to the kind of changes that
would be required in the industry's structure, conduct, or
performance patterns in order to make it effectively competitive.
_____
NOTES
1.For a fuller list of the structure-conduct-performance components, see
the editorial appendix, "A Structure-Conduct-Performance
Questionnaire," 2 Antitrust Law & Economics Review (Spring 1969),
pp. 8-12. A questionnaire used by Professor Bain in his pioneering
studies in this area is presented in an appendix to his Barriers to New
Competition (Harvard University Press, 1956), pp. 223-226. For the
analytical significance of these items of information in assessing the
effectiveness of competition, see also his Industrial Organization
(Wiley, 1968) and Charles E. Mueller, "The New Antitrust: A
'Structural' Approach," 1 Antitrust Law & Economics Review (Winter
1967), pp. 87-130.
CopyrightŠ 1997 Antitrust Law & Economics Review, inc.
*******************************
