Posner's Economic Analysis of Law

Economic Analysis of Law. by Richard A. Posner

Review by: Peter A. Diamond
The Bell Journal of Economics and Management Science, Vol. 5, No. 1 (Spring, 1974), pp. 294-
300
Published by: RAND Corporation
Stable URL: http://www.jstor.org/stable/3003108 .
Accessed: 28/06/2014 07:29

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.

RAND Corporation is collaborating with JSTOR to digitize, preserve and extend access to The Bell Journal of
Economics and Management Science.

http://www.jstor.org

This content downloaded from 193.142.30.98 on Sat, 28 Jun 2014 07:29:58 AM

All use subject to JSTOR Terms and Conditions
 Posner's Economic
Analysis of Law
Reviewed by

Peter A. Diamond
Professor of Economics
Massachusetts Instituteof Technology

* RichardPosnerhas writtenan exceptionallyinteresting book.'

It is a wide rangingdiscussionof the impactof the legal system
on the allocation of resourcesby the marketand of economic
aspectsof the behaviorof participantsin the legal system,written
as a text for law studentsknowinglittleor no economics by a
nonmathematicalChicagoan. (Of course, it is only the absence
of mathematicsthatI hold againstthe author.) I shall brieflylay
out the range of analysisand varietyof audiences who will find
this book interesting,some limitationsof the absence of mathe-
matics,and some examples of the tone of the book. Then, in a
more leisurelyfashion,I shall discuss contractlaw to show an
example of the typeof analysisin the book.
This is a textbook for law students,intendingto teach
economic principleswhile conductinga surveyof the rules and
institutionsof the legal system,summarizing,and extendingthe
applicationsof economicsto law. I expectthatit does its job well
and would be an interesting book to teach or studyfromin this
sortof course,forit is verywell writtenand interesting in content.
As an economist,I naturallybelievethatthe book wouldbe better
for law studentsif it taughta littlemore economics,particularly
notingwhichassumptionsplay major roles in the analysis.(I also
suspectthat some discussionof the different nature of marginal
and lumpy decisions would be valuable, i.e. the complications
arisingfromnonconvexities.)The textwould also serveverywell
for a law and economics course for economics students(either
advanced undergraduatesor graduate students), although the
more conventionaleconomics subjects, such as taxation and
regulation,would need considerableexpansion,or better,omis-
sion since theyare alreadya standardpartof economicscurricula.
While thisis basicallya text,it is also a monographin a fieldof
growinginterestto economists,and I will reviewit primarilyon
that basis.
The scope of thisbook is exceptionallywide. It encompasses
the traditionallyanalyzed areas of government intervention in the
market(antitrust,regulation,taxation), the areas of greatrecent
interestof the impact of the legal systemon privateuse of the
market (property rights,contracts,torts, crime control, anti-

1 Richard A. Posner, Economic Analysis of Law (New York: Little,

294 / PETER A. DIAMOND Brown and Co., 1973).

This content downloaded from 193.142.30.98 on Sat, 28 Jun 2014 07:29:58 AM

All use subject to JSTOR Terms and Conditions
discrimination regulations)and of theeconomicsof thefunctioning
of thelegal system(due process,civiland criminalprocedure,law
enforcement).Withthehypothesis thatthecommonlaw is seeking
economic efficiency, it also seeks to explain judicial decision-
making.2The styleis lucid but unfortunately, to my taste, non-
mathematicalexceptfor numericalexamplesand a few diagrams.
Mathematicsis both a method of expositionthat makes some
assumptions(or theirimportance)more apparentand a method
of thoughtthat makes some types of errorsmore difficult. For
example,in the discussionof expenditures on litigation,3
underthe
assumptionthateach partytakesthe litigationexpendituresof the
otheras given,it is statedthatincreasedoptimismresultsin higher
expenditures. But the perceivedprobabilityof successis a function
of expenditures.For an upward shiftin the functionto increase
expenditures, the derivativesmustchange appropriatelysince the
first-order condition involves the derivative,not the level of
probability.4
Anotherexample where mathematicsmightclarifycomes in
the discussionof freeentryand profitability. Withfreeentryand
identicalcost functionsforall actual and potentialfirms,all firms
have zero profits(i.e. just earn the normalreturn)in competitive
equilibrium.Whendifferent firmshave different cost functions,the
marginalfirm(with a continuumof firms)will have zero profits,
but otherfirmsmay be earningpositiveprofits(or rents) without
encouragingfurtherentry,since potentialfirms(which have not
yetentered) have highercosts and cannotduplicatethe profitable
positionof some of the existingfirms.A priori I do not see any
reasonto assume thatall potentialentrepreneurs would startfirms
withidenticalcost functions.Particularlyin a book where trans-
action costs play a major role, we would expect existingfirmsto
have at least an informational advantage,vis-'a-viscustomers,rela-
tive to potentialfirms.Empirically,in particularindustries,the
supplyof potentialentrepreneurs may be such that the identical
firmassumptionis a good one. I do not believeit is a good assump-
tionin all industries.This distinctionis not drawndespiterepeated
use of a zero profitassumption(and the impliedeventualbank-
ruptcyof any firmthatis not successfully profitmaximizing).At

2 Of course, this hypothesis is not universally accepted. See, e.g.,

George Fletcher, "Fairness and Utility in Tort Theory," Harvar-d Law
Review (January 1972).
3 A dose of game theory might be helpful here, particularly in the
discussion of the existence of equilibrium when each party is aware that
litigation expendituresmay induce a response.
4 It seems to me that clearly distinguishingaverage from marginal
ranks as an important economic lesson. Thus sloppiness on this point is
a weakness in the book. See, e.g. "he will stop spending [on an appli-
cation for a broadcast license] at the point where, if he prevails, the
differencebetween his expenditures and the value of the license will be
large enough to compensate him for the risk," (p. 14). For the example
in the text, expected profitof the plaintiffis fl(E)D-E, where E is expendi-
tures,D damage award if the case is won, fl(E) the probabilityof winning
if E is spent. The first-ordercondition is fl' D = 1. E will increase with
a shift in fl that increases LI' at the previous optimum (given suitable THE BELL JOURNAL
regularityconditions). rl could shift up and Hl' shift down in the relevant OF ECONOMICS AND
range. MANAGEMENT SCIENCE / 295

This content downloaded from 193.142.30.98 on Sat, 28 Jun 2014 07:29:58 AM

All use subject to JSTOR Terms and Conditions
 least in organizedcrimeone would expectan existingfirmto have
if theyare goingto
an advantageover potentialfirms,particularly
shootit out fora (temporary?)monopolyposition.(Do potential
firmsuse potentialbullets?)
Now let us turnto a few quotationswhichwill give the tone
of the book more easilythandescription:
There is abundantevidencethatlegislativeregulationof the economy
frequently, resultsthan the
perhapstypically,bringsabout less efficient
market-common law systemof resourceallocation.[Footnotecontaining
omitted.]5
references
A newer view, which draws on economicanalysis,argues that the
structureof the administrative process is designedto increase political
controlover the processof legal regulationratherthan to,increaseeffi-
ciencybut that withinthe constraints imposedby the fundamentally po-
liticalpurposeof regulationthe evidenceis consistentwiththe hypothesis
thatthe agencies,like mostotherorganizations, are rationalutilitymaxi-
mizers.6
The analysisin the precedingsectionassumesthatjudges make their
But what is the
decisionsin accordancewith the criterionof efficiency.
and the promotionof efficient
linkage betweenthe judges' self-interest
resourceuse?7
Delay is notdue to thefactthatthedemandforlitigationis highand
the amount of judge time limited. . . . An appropriatelygraduated system
would
forpeople desiringto have theircases heardpromptly
of surcharges
[clearthe market.]8
Without statutoryprotectionagainst disinheritance,women could
negotiatewith their husbandsfor contractualprotection.The statutory
costs.9
provisionminimizestransaction

As this last quotationsuggests,Posner does not examinethe

question of the effectof institutions on preferencesand inter-
personalrelations.Rather,he uses the standardconceptof Pareto
optimalityin termsof givenpreferences.10 This mode of analysis
is verywell employedand pushed intonew regionsof uncertainty
and transactioncosts. One curious aspect of the structureof the
book is the treatment of the transactioncosts of the legal system.
In much of the book, theyare taken to be zero (particularlyin
the discussionsof the correctmeasurement of damages to induce
behavior). Then, when the proceduralsystemis directly
efficient
discussed,the measureof damages whichwould be optimalfor a
legal systemis assumedto hold. (This can, of course,
frictionless
be viewedas a versionof how the legal systemis currently trying
to measure damages. Posner does discuss separately damages
wherethe probabilityof apprehensionis less than one.) Treating

5p. 329.
6P. 386.
TP. 325.
8p. 355.
9 P. 250.
10While the effectof legal rules on the incentiveto settledisputesis
examined,I was sorryto see no discussionof the veryinteresting analysis
by ArthurLeff("Injury,Ignoranceand Spite-The Dynamicsof Coercive
Collection,"Yale Law Journzal,October1970) of the dynamicsof settle-
mentand impactof legal ruleson different typesof litigants(wage earners-
corporations,high volumelitigants-infrequent litigants).
296 / PETER A. DIAMOND

This content downloaded from 193.142.30.98 on Sat, 28 Jun 2014 07:29:58 AM

All use subject to JSTOR Terms and Conditions
simultaneously the measurementof damages and the structureof
procedureswould be interesting.
To see more of the flavorof thisbook and of thisburgeoning
fieldof research,let us considerone topic in more detail. I have
selectedthe area of contracts.11Discussion of the role of contract
law is greatlycomplicatedby the presence of the contracting
parties.By hypothesistheyare (were, will be, or could easilybe)
negotiating.This preventsthe simple analysis of tortlaw where
the assumed high transactionscosts preventparties affectedby
externalitiesfromnegotiatingto consider all of the social costs
conventionally viewedby a generalequilibriummodel withexter-
nalities.Rather we mustmove into the finestructureof transac-
tions to findany role for the law at all, other than dissuading
outrightcheating (and the costs induced by feai of cheating).
There are three types of effectsof the law which seem worth
consideringseparately-adjustingallocationwherethepartiescan-
not or have not workedeverything out, savingon theirtransaction
costsin theprocessof workingthingsout,and affecting information
flows,and so, what theyare tryingto work out. Put differently,
the threetypesare wherethe partiesare indeed strangers,where
theyare fillingin detailsof an agreedbargain,and wheretheyare
determining the natureof the bargain.
In the firstcategoryPosner considersa contractwhere the
deliverydate stipulatedomitsthe month.A, new to the industry,
means next month;B, an old hand, knows and is followingthe
custom of the industrythat an omittedreferenceto the month
means the currentmonth.The court will interpretthe contract
as incorporating the customsof the industry.Posner points out
thatthisencouragesnewcomersto masterthelanguageof thetrade
promptly,which presumablyincreasesefficiency in the industry.
One should also pointout thatit also affectsthe costs of entering
a newfield,and so, bothlong-runefficiency and incomedistribution.
Imperfectcommunicationcan also take the formof faultytele-
grams and garbled messages. In addition,where communication
takes time,there are risks of changingconditionsthat must be
bornein some divisionbetweenthe parties.By decidingwho bears
the risk (of increasedcosts,say, betweentransmission and receipt
of an offer)the court can affectthe efficiency of the bearingof
this risk and the decisionsaffecting its magnitude.Since in these
situationsthe parties are in imperfectcommunication,court im-
posed rules will generallydeterminewhere risks are borne (and
so who adjuststo bear them).
An example in the second categoryconsideredby Posner is
a quantityof wool witha latentdefectthathas not been examined
by eitherthe selleror tne buyer.The buyerthenuses the wool in
manufacturing and so ruinsthe suitsmade fromit. Efficiency calls
for inspectionby the partywith lower inspectioncosts,'12or no

11 The selection reflectsmy own interests,of course, as well as the

fact that Posner's analysis here is, to my knowledge, new and very inter-
esting.
12 These costs are not simply physical inspection costs but also reflect
other transactions, for example whether self-protection against cheaters THE BELL JOURNAL
already called for some inspection by one of the parties. What is needed OF ECONOMICS AND
MANAGEMENT SCIENCE / 297

This content downloaded from 193.142.30.98 on Sat, 28 Jun 2014 07:29:58 AM

All use subject to JSTOR Terms and Conditions
 inspectionif expected damages are lower than inspectioncosts.
By tryingto place liabilityon the cheaperinspector,the courtcan
attemptto minimizethe sum of expectedinspectionand damage
costs. In the world of "transactorsknowledgeableof the law,"
misplacedliabilitymerelyleads to a correctivetransaction, shifting
the liability.But correctivetermsin the contracthave transactions
costs, so by gettingliabilityright,the court permitsa cheaper
reliance on the common law to substitutefor a more expensive
explicitcontract.13Partiallyrestatingthe categorizationof topics,
Posner describesthree economic functions,"maintenanceof ap-
propriateincentives"(i.e., puttingrisk where it is most cheaply
borne), reductionof "the complexityand hence cost of transac-
tions by supplyinga set of normalterms"'14 and, not mentioned
above, furnishing "prospectivetransacting partieswithinformation
concerningthe manycontingenciesthat may defeatan exchange,
and hence to assist them in planning their exchange sensibly.
The parties,throughtheirlawyers,are guided around the pitfalls
in the process of exchange revealed by the opinions in decided
contractcases."'15
Here we move into the thirdcategory,of makingtransactors
more knowledgeable.The more difficultversion of this is con-
sideringhow people mighthave acted had theybeen moreknowl-
edgeable and consideringhow, if at all, legal rules applyingboth
to carefullyand carelesslyworkedout bargainsshould adjust for
the careless. Let us look more closely at the role of information
exchanges.Posner considersa contractinduced by "an outright
and calculatedlie." In thiscase, he argues "the presumptionthat
a contract,if carriedout, will producea value-increasing exchange
fails . . . [and] nonenforcement may discouragesuch conductin
the future."Consideringthe case of withholdingof information
he argues,"It would be inefficient to requirethe sellerof a house
to obtain and disclose every fact that mightbe material to a
purchaser;it would be particularly to requirethe seller
inefficient
to obtain and disclose information that the buyercould obtain at
lowercost [footnoteomitted]."16 He does not considerthe tougher
case of nondisclosureof alreadyobtainedinformation of value to
the purchaser.Efficiencyat that point would call for disclosure.
This would be a disincentiveto the seller to obtain information
which he mightmake use of but which mightlater negatively
affectthe selling price. It would also be a disincentivefor the
buyer to obtain informationwhich he mightpresume the seller
has checkedfor him, althoughthis problemis easily handled by
asking.This questionis also consideredseparately(in ?3.6), where
Posner considersinformation flowsbetweenseller and consumer,
arguing that informationabout simple, frequentlypurchased,

here is consideration of the total set of costs, not just those of the single
transaction.
13 Obviously legal placement of liability independent of the contract
does not permit this shifting.
14 Posner also considers the reverse case of compelling market rather
than legal exchanges where the former are cheaper, as in determination
of price, when the parties have been vague.
15 P. 44.

298 / PETER A. DIAMOND

16Pp. 48-49.

This content downloaded from 193.142.30.98 on Sat, 28 Jun 2014 07:29:58 AM

All use subject to JSTOR Terms and Conditions
productssubjectto patentdefectsis mostcheaplygatheredby the
consumer; while informationabout complex infrequentlypur-
chased productssubject to latentdefectswill be cheaper for the
sellerto obtain. Posner also considersothertypesof information
flowsin the market,problemsof suitsfordamageswheredamages
are smalland FederalTrade Commission'sregulationof advertising.
The implicitbackgroundfor this discussionis a conventional
resourceallocationmodel wheremarketsare functioning more or
less properly,so one can focus on the singletransactionagainst
a backgroundof efficient resourceallocation.But it is naturalto
ask whetherthe same informationand transactioncosts which
create the questions at hand do not simultaneouslyimply some
systematicbiases in the entiremarketsettingwhich are relevant
for the treatmentof the singletransactionand efficiency. As an
example,let us considerPosner'sdiscussionof standardformcon-
tracts(?3.7). Many offersare made on a take-it-or-leave-it basis
by individualfirms.This patternmay well reflectsignificant cost
savingsfora firmengagedin manysmall transactions.But, he ar-
gues,competition will adjustforthis."If one selleroffersunattrac-
tivetermsto a purchaser,a competingseller,desiringto obtainthe
sale forhimself,willoffermoreattractive terms.The processshould
continueuntilthe termsare optimalfromthe purchaser'sstand-
point.Thus the purchaserwho is offereda printedcontracton a
basis does have a real choice: he can refuseto
take-it-or-leave-it
sign,knowingthat if bettertermsare possible anothersellerwill
offerthemto him.All of the firmsin theindustrymay findit eco-
nomicalto use standardcontractsand to refuseto negotiatewith
purchasers.But whatis importantis not whetherthereis haggling
in everytransactionbut whethercompetitionforcessellersto in-
corporate in their standard contractsterms that maximize the
purchaser'sbenefitsfromtransacting."17
But what has happened to all those informationlimitations
and transactioncosts that were the heart of the argument?For
termsof limitedimportanceis it worththe cost of seekinga better
contract?(Can you be sure one even exists?) Is it even worth
the cost of findingout what would be a betterterm?'8 If the
answersto thesequestionsare no, whereis the incentiveforbetter
terms?Might not all firmsfindit to theirinterestto offerbad
terms19(just as modelshave been constructed wherepricesexceed
the competitiveprice because information limitationsremovethe
advantagesofpricedecreases)? If in somecases themarketsystem-
aticallydivergesfromefficiency, is not the role of a legal system's
seekingefficiencybasicallydifferent fromthatanalyzed?20Perhaps
no rulesto cure thisproblemcan be foundwhichare betterthan

17p. 54.
18 Examples would be risks betterborne by the seller, so the consumer
benefitsfrom better terms and a higher price.
19 Free entry cannot come to the rescue here, for even with identical
firmswe merely get zero profits,not efficientterms.
20 These problems are furthercomplicated by the presence of
plain
THE BELL JOURNAL
old-fashioned careless consumers in addition to new-fangled transaction-
OF ECONOMICS AND
cost-avoiding careless consumers.
MANAGEMENT SCIENCE / 299

This content downloaded from 193.142.30.98 on Sat, 28 Jun 2014 07:29:58 AM

All use subject to JSTOR Terms and Conditions
 thosewhichignoreit. Perhapsthe legal systemcan only help effi-
ciencyforthe (excessively?)wellinformed. Apartfromtheobvious
questions,thisappears to be an extremelyimportant
distributional
efficiencyquestion,the outcomeof whichis not at all clear.
As I hope I have indicatedby thisdiscussion,Posneris examin-
and
ing an area of greatintrinsicinterestand analyticaldifficulty
makingconsiderableprogress.The book is broad as well as deep.
I foundit fascinatingto read and I look forwardto usingit as a
text.

300 / PETER A. DIAMOND

This content downloaded from 193.142.30.98 on Sat, 28 Jun 2014 07:29:58 AM

All use subject to JSTOR Terms and Conditions