Competition and Concentration

Author(s): GEORGE J. STIGLER

Reviewed work(s):
Source: Challenge, Vol. 12, No. 4 (JANUARY 1964), pp. 18-21
Published by: M.E. Sharpe, Inc.
Stable URL: http://www.jstor.org/stable/40718789 .
Accessed: 30/06/2012 15:49
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

M.E. Sharpe, Inc. is collaborating with JSTOR to digitize, preserve and extend access to Challenge.

http://www.jstor.org

Challenge

Interview

Competition and
Concentration
A keypublic policyquestionconstantlyfacingthe Americanpeople is how to
maintaina freecompetitivesystemwithoutundue governmentencroachmenton
privatedecisionmaking.While mostcitizensagree withthisgoal, thereis no consensus on how to determinepreciselywhen competitionhas brokendown and
it is timeforgovernmentto step in. George J. Stigler, Walgreen Professorof
AmericanInstitutions
at the University
of Chicago, has spentmanyyearsstudying
thisproblem.He recentlyexaminedthe difference
in profits
in concentratedindustriesvis-a-vishighlycompetitiveones in ''Capital and Rates of Returnin ManufacturingIndustries"writtenunder the auspices of the National Bureau of EconomicResearch.Prof.Stigleris Presidentof the AmericanEconomic Association.

^% Dr. Stigler, mostpeople believethata company

which dominatesa large share of its marketis
'^
boundto end up witha higherreturnon its investment
thana firmoperatingin a highlycompetitivemarket.
But in yourrecentbook,"Capital and Rates of Return
in Manufacturing
Industries"you indicatedthat there
was no evidenceof appreciablyhigherreturnsin concentratedindustries.What conclusionscan be drawn
fromthis regardingcompetitionin concentratedindustries?

GEORGE
J.
STIGLER

petition.For example,in 1890 almosteveryeconomist

in theUnitedStateswould
lookingat thesteelindustry
that
of
course
is
a
it
say
competitiveindustry.Any
industrywithdozensand dozensof firmsin whichthe
largestfirmhas only30 per centof theindustry's
output
was believed to be obviouslycompetitive.Nowadays
a large number of economistswould say it is an
oligopolisticindustryand thusexpectsome significant
departuresfroma perfectlycompetitivesituation.So
I would answeryourquestionby askinghow does one
prove the absence of competition?And I would say
like most economicresearch,mine thatwe haven'tdone too well in thisrespect.The basic
A Unfortunately,
led to resultswhich were ambiguous on the most measure,of course,is the share of the
output of an
interesting
questions^I did findthat in the peacetime industrythatis in a fewhands. But thatisn'tenough.
yearsbetween1938and 1957therewas less than a one
The precisescope of a particularindustrymay have
in profitratesin concentratedand
been misdefined.
And so we look forotherthings,the
per cent difference
and when certainplausible twomostcommonbeinghighratesof returnon investunconcentrated
industries,
are made in the data, even this difference ment, and price rigidity.
corrections
vanishes.But one cannot conclude that concentrated
I also foundthathigh Q Can we measure departuresfrom competitionacare fullycompetitive.
industries
industries
their
profit
preferredposition much curatelyenoughso thatpublic policieswithrespectto
keep
if
are
concentrated.
This suggeststhatcon- it can have a reasonablysound factualbasis?
longer they
centrationdoes not necessarily
allow monopolyprofits,
A We can sometimesrecognizean outrightcollusive
but thatunderfortunate
marketconditionstheycan be
structuresuch as existed in the electricalequipment
obtained for long periods.Finally,in an articlesoon
and thereis no real doubt of the need for
to appear in the Journalof Political Economy,I find industry,
certainkindsof public policyto cope withit. We can
a substantialcorrelation
betweenconcentration
in more
recognizethe so-callednatural monopolyof a public
preciselydefinedindustriesand the ratio of market utilityand considerwhat policiesare
necessaryto deal
value of stocksto book value of corporationassets.
withit. On the otherhand, the dividingline between
whatis a seriousand what is a trifling
departurefrom
by whicha modern
Q What are the major yardsticks
competitionis open to dispute.There are,forexample,
economistmeasurescompetition?
and competentacademiceconomists
manydisinterested
who
think
that
the
A It is a lot easier to describeclear, significant
decisionin the Brown Shoe case,
persistentdeparturesfromcompetitionthanit is to define which forbade a mergerthat involved triflingpercompetitionitself.Indeed, we have experiencedover centagesof the shoe industry's
output,was overzealous,
in our conceptof com- and othereconomists
timea progressive
refinement
who thinkthatit was a far-sighted
18

Challenge January,
1964

"THE ECONOMY IS STILL PREDOMINANTLY COMPETITIVE"

policies
put aside fora momentthetraditionalantitrust
and say thatso faras our otherregulatorypoliciesgo,
and by a
I thinkthatwe are in generaloverregulating,
substantialamount.I thinkwe have gottenmuch too
in the solutionof
deeplycommittedas a government
labor disputesand at a verysubstantiallong-runcostto
I thinkwe have regulatedsuch areas
the community.
How much monopoly?
as motortruckingand agriculturefartoo long and far
too much in detail. Both are capable of viable and
Q Sen.Paul H. Douglas,in a recentinterviewin Challenge ("Evaluating the EmploymentAct," October, effectiveexistencein the long run withoutdetailed
I find
1963), stated: "There is a tremendousamount of governmentalcontrol.As for underregulating,
it
to
answer
own
tastes
and
our
harder
a
since
giveyou good
my
economy."
monopoly
oligopoly throughout
run
in
the
direction
of
a
economic
Do you agree?
decentralized
clearly
systemin which thereis a minimumboth of private
A Sen. Douglas certainlyhas the right to use the monopolypowerand directpublic control.
amount,"but I don'tknowexactly
phrase"tremendous
how to interpretit. That we have a good deal of
Present antitrust policy
monopolyand oligopoly in the American economy
seemsto me beyonddispute and should be a source Q How, in general,would you evaluate our present
of seriousconcernfor public policy.There is a very policies in the antitrustarea?
commonview that Americanmanufacturingis preorganizedin oligopolisticindustries.
ponderantly
Using A On the whole, I am quite complacentabout our
the word oligopolyin its proper sense-an industry progressand satisfiedwith the effectiveness
of our
concentratedso that thereis a substantial policies.There has been a verysubstantialreorientation
sufficiently
of noncompetitive
behavior-I disagreewith of antitrustpolicysinceI servedon the AttorneyGenprobability
I would estimatethatin manufacturing eral'sAntitrust
thatstatement.
Committeein 1954-55.The controlover
our
like
one-tenth
to
two-tenths
of
has
come
to be the major part of antitrustenindustries,
mergers
something
whethermeasured by employmentor income orig- forcement
in boththeJusticeDepartmentand theFedThis, in myopinion,is a highly
inatingin them,do have a degree of concentration eral Trade Commission.
whichI hope does not becomepermanent.The restof desirabledevelopment.Of course,we cannotallow oldour manufacturing
on the otherhand, are fashionedcollusivepracticesto become respectableor
industries,
We have everyreason to widespread.While I deplorethe long-runeffecton the
deconcentrated.
sufficiently
believe that the normal formof behaviorwill be es- reputationof our economyof thingslike the General
sentiallycompetitive,and that the departuresfrom ElectricWestinghousecase of severalyearsago, I think
run
and
will
be
short
behavior
that
relatively
competitive
theyhave servedas a therapeuticreminderto the
of verylittleconsequence.As for Sen. Douglas' state- Americanbusinesscommunitythat it is engagedin a
amountof monopoly competitivesystemand it should stay that way. We
mentthatwe have a tremendous
and oligopoly,I would say that we have a serious have neverbeen veryeffective
in dealingwitha highly
amountin our economy,but that the economyitself concentrated
let us saya traditionalmonopoly
industryof thetypethatat one timewas held by theAluminum
is still predominantly
competitive.
Company of America. We have always felt-and by
Dr.
in
terms
the
Division and the federal
of
of we I mean both the Antitrust
Stigler,
general effectiveness
Q
or
unwe
are
who
render
decisionsthat one should be
the
present-day
competition,
overregulating
judges
our
in
this
or
are
antiour
loath
an
to
dismember
area,
derregulating economy
very
enterpriseand that theretrustpoliciesgenerallycorrect?
forethe remedyof dissolutionor divestiture
should be
used only in the mostextremeoccasions.This has led,
Well, one should draw a distinctionbetweentwo I think,to a greatdeal of ineffectual
pursuitof certain
kindsof regulationof our competitivebusinesssystem. highlyconcentrated
industries.This has been the longOn the one hand, thereis traditionalantitrustpolicy, standingweaknessin our antitrustpolicy.I thinkour
thepurposeof whichis to maintaina competitive
struc- policywithrespectto the mergerswill eventuallycorture and then allow the businesssystemto operate rect that. The law as now interpretedand enforced
relativelyfreely.On the otherhand, we have a set of seems to make it virtuallyimpossiblefor any major
industries
whichwe believecannoteconomicallyor for producerin anymajorindustry
to acquirea substantial
otherreasonsbe restoredto a competitivesituation. rival in that industry.Indeed, we have gone farther
than that-maybe too far.Thus it is not clear to me
Therefore,theyrequirecontinuingregulation.Let me
in a
precautionto avoid eventual overconcentration
As withmostthingsin life,we can deal
basicindustry.
on a reasonablyinformed
basiswitha substantialnumber of problems,and in a fairnumberof cases we are
just guessing.

1964
Challenge January,

19

"NOT EVERY CONCENTRATED INDUSTRY IS SERIOUSLY

whetherGeneralMotorscould successfully
mergewith
a bootblackin Detroitwithoutbringingdownantitrust
sanctions.So thatas I look at the long-runoutlook in
I feel rathercomplacent.
the area of competition,
Defining concentrated industry

Q How would you definea concentratedindustry?

A There is, of course,no unique definition,It is an
a small
in whichone, two,threeor fourfirmsindustry
number,in anycase-controla verylargefractionof the
industry'soutput. Whetherit be 100 per cent or 80
per cent or 70 per cent is not important.The conI suppose,by,say,
centratedindustry
is best illustrated,
the American automobile industry,where the Big
Three accountforwell over90 per centof our domestic
output and have accountedfor over 90 per cent for
a good many years.This is simplya matterof fact
and calls for no moral reprobationor denunciation.
The crucial question is always whetherthat concentratedindustrywill behave in a competitivefashion.
Can we quicklysay, as soon as we know that there
are two or threedominantfirms,that there is high
probabilityof importantdeparturesfromcompetition?
And here I mustsay thereis a major split in the economicsprofession.There is a substantialgroup that
embraces the doctrine of "workable competition"whichwas inventedby myformercolleagueat Columbia University,
JohnMaurice Clark. Lookingat things
broadlyand wisely-and theyhave considerableconfidencein theirabilityto do both-they are satisfied
of a greatmanyindustrieswhich
withthe performance
are relativelyconcentrated.I, myself,am much more
iconoclasticin thismatter,and I findit veryhard to
know preciselyhow, proceedingon such a basis, one
reaches a definiteconclusion. What is the precise
distinctionbetween an industrythat is behaving
and one that is not? In my view, the
satisfactorily
doctrineof workablecompetitionin effectsays that
the way in which to decide whetheran industryis
in the public interestis to have
acting satisfactorily
a youngman writea Ph.D. thesison the subject and
accepthis verdict.And while that is obviouslyan emactivityfrom the viewpoint of
ployment-increasing
I
believe
that it is a tenable basic
can't
professors,
solution. But it is also reasonablycertain that not
industryis seriouslyuncompetitive.
everyconcentrated
Q Does the rate of returnon investmentafter taxes
offera reliableclue as to the degreeof competitionin
a particularindustry?
A No. The reasonsa firmmakesa high rate of return
on an investmentare numerous.One perfectlygood
20

1964
Challenge January,

UNCOMPETITIVE"

himselfis a man of
reason is that the entrepreneur
and
so
that, even without any
prodigious ability,
recordsthatare
he
sets
profits
monopolisticadvantages,
hard to match. Mr. Sears of Sears, Roebuck made
colossalsumsof moneywithouthavingany monopolistic powersat all- havingindeed only one littlething:
genius in business organization,Henry Ford's great
successwas obviouslynot due to monopolisticpowers
but to a geniusin the productionof inexpensiveautomobiles.
chargesof theSenate
Q How valid are theoft-repeated
Antitrustand Monopoly Subcommitteeto the effect
industries
thatadministered
pricesin the concentrated
have
turn
which
in
have led to oligopolisticprofits
spurredinflation?
A I am a greatadmirerof muchof what the late Sen.
Estes Kefauverdid duringhis long yearsin Congress,
but I would say thatthiswas not amongthe mostimportantand lastingworkswhichwe shall associatewith
in
hisname.I do thinkthatthereareoligopolisticprofits
someindustries.
The real question,however,is whether
or of oligopolyhas led
the presenceof concentration
to a more rapid rise of pricesover time than would
have occurred with a more competitiveindustrial
structure.There are three broad schools here. The
traditionalschoolsaysthatmonetaryinflationhas been
the basic source of price rises, and that fiscal and
are theonlymethodsof copingwith
restraints
monetary
rise of prices.There is a second school,
the persistent
whichbelievesthat
verywidespreadin the profession,
and thatonly
factor
a
have
been
increases
major
wage
a systemof increasingsanctionsor controlsover labor
is likelyto preventthe
disputesand wage settlements
increasesby wage
of productivity
constantoutstripping
smaller
a
is
And
there
increases.
third,
group,which,
however,was dominantin the KefauverCommittee.
This last group argued that oligopolisticprofitswere
was the
risingovertimeand thedesirefortheseprofits
forcemakingfor price rises.
Administered prices

Q Can any case be made for the latterview?

A I have done a fairamountof statisticalworkin this
area, partlyas theresultof mychairmanshiptwoyears
ago of the Committeeto Review the Price Statistics
a committee
Program of the Federal Governmentcreatedbya contractbetweentheBureau of theBudget
and the National Bureau of EconomicResearch.Our
convincedme that thereis gravedoubt
investigations
that so-calledadministeredprices-by which I think
we mean primarilyprices which are unresponsiveto

"THREATS TO COMPETITION CAN COME EITHER FROM BUSINESS

OR GOVERNMENT'

thebusinessside in a desireto suppresscompetitionas

an obnoxious and troublesomephenomenon,or from
the governmental
side. On the whole,I thinkthat the
threatsusuallyoriginatefromthe businessside. They
unless theyare acceptedby
are, of course,ineffective
To give an example,I thinkit would
thegovernment.
be quite impossiblefor the domesticpetroleumindustry-andI referto the crude oil branch-effectively
to restraincompetitionwithoutgovernment
assistance.
The way in whichtheyachievetheirresultis by persuading the governmentto institutea systemof oil
Productivity gains
embargoeswith quotas on importswhich protect,at
Q Dr. Stigler,do you believethatin thefuturebusiness very substantialcost to the consumer,the domestic
shall again pass on productivity
gains to the consumer crude oil industryagainst foreigncompetition.Too
lower
rather
than
to itselfin the form manysectorsof the businesscommunityare eager to
through
prices
runto Washingtonand seekshelteragainstcompetition
of higherprofits?
whentheslightest
wave of competitive
activityappears.
A In manyinstancesbusinessis doingso now and will The currentendeavor,for
to restrictthe rate
example,
continueto do so in the future.There are other in- that is
paid by savingsand loan associationsto destances,of course,in which businessis not doing so,
positorsis an indicationof how sensitivemanypeople
and I am afraidwon't do so in the future.When a
are to pricecompetition.
new industryemergesand shiftsover fromsmall-to
side?
large-scaleproductionwith immenseincreasesin effi- Q What about the threatfromthe government
ciency,thereare,of course,declinesin prices,and of a
of the
A This usuallystemsfroma mistakenimpression
verymajor magnitude.You will recall that ball-point
of
and
of
the
for
methods
importance
monopoly,
proper
that
used
sell
for
to
are
for
now
six
$12.50
25^.
pens
with
More
and
it.
more
find
we
the
government
The colortelevisionsethas dropped50 percentin price dealing
seekingto counterwhat it thinksare unsound price
in the last fewyears.Indeed, this is the only way in
in
which new productscan make their way into new movesin the economy.Mr. Kennedy'sintervention
the
steel
in
increases
the
of
an
is
1962
price
spring
marketsand drive old productsout of existence.But
thereis a separatequestion. Puttingaside large dif- obvious example, While this particularincidentwas
highlydramatic,the attitudebehind it was certainly
ferencesin productivityincrease in one industryas
new.Mr. Eisenhower,
duringhis eightyearsin
to
about
what
the
compared another,
averageupward nothing
the
White
was
House,
persistently
strivingto persuade
driftof efficiency
the economyas a wholethroughout
and
business
labor
to
act
in
he
what
as a
characterized
a figurethat is commonlyput somewherein the two
''statesmanlike
manner"that
to
sacrifice
their
own
is,
to four per cent range. Here one almost has to get
intereststo thoseof the communityat large. I think
of an economyin
nostalgicin discussingthe possibility
whichdividendsand wagesare held stablewhile prices thatkindof exhortationis a mostundesirableone. On
of finishedproductsdriftdownwardthroughtimeand, theone hand, it exaggeratesthe powerof thesepeople
the degreeof
real incomesrise steadilyfor workers,in- to do harm,and seriouslyunderestimates
therefore,
vestorsand, indeed,everyonein thecommunity.
Unless competitionin the economy.On the otherhand, the
manin productionincreases, remedyof men actingtogetherin a statesmanlike
pricesdriftdown as efficiency
ner
or
into
it
being pushed
by stronggovernmental
however,pensionersand otherswho live on fixedinworkingof our
comes will not share the gains. But I think that pressuresleads to a seriouslyinefficient
We
are
not
to
the
economy.
going
get
right
priceswith
this argumenthas lost some merit because, after 20
the
of
the
degree
proper
flexibility,
right
places for
of
fewerand fewerpeople are
years price instability,
investmentand the right kinds of innovationif we
themselves
to
into
the
where
allowing
get
position
they
are totallydependenton a fixedincome.We all know, have to clear thesestepsthrougha CongressionalcomI believe that if
for example,that a major price level change would mitteeor througha Cabinet officer.
we
could
rekindle
the
idea
the
that
primarytaskof the
probablylead to changesin the level of benefitsunder
central
is
to
to widen
government
preservecompetition,
the social securityprogram.
economicopportunity
forall people and to protectonly
Q Whatforcesin our economydo you see as potential thosepeople who need protection,therewould be no
threatsto competition?
seriousthreatto theviabilityof thecompetitive
system.
A The threatsto competitioncan come eitherfrom Q Thank you, Dr. Stigler.
fallsin demandbut are oftenquite sensitiveto increases
in demand,so thatlike a ratchettheyworkin onlyone
direction-existon anyimportant
scale in theindustrial
sectorof the Americaneconomy.The empiricalevidenceseemsquite strongthatthisis a blind alley and
did a disservice
thattheSenateAntitrust
Subcommittee
in creatinga suspicionof oligopolyas a sourceof inflation.This certainlycontributedsomewhatto 1962's
unfortunate
steelprice embroglio.

Challenge

January,1964

21