Barzel - Economic Analysis of Property Rights

POLITICAL ECONOMY OF INSTITUTIONS AND DECISIONS
Editors
James E. Alt. Harvard University
Douglass C. Nonh, Washington University of St. Louis
ECONOMIC ANALYSIS
OF PROPERTY RIGHTS
SECONp EDITION
Other books in the series

Alberto Alesina and Howard Rosenthal, Partisan Politics, Divided Govemment
and the Economy . ......................:.:.::..::.:.:.:.::..::.: . :.:::':.::::::".:::.:::
.. __ ~~~ J . Alston,Thrinn~ggertsson, -andDouglass C. Nor:rh, EmfJiricalStudiesin .
",-;--'"
,.."
.InStitutionalChnge."
.
James E. Alt and Kenneth Shepsle, eds., Perspectives on Positive Political
Economy
Jeffrey S. Banks and Eric A. Hanushek, Modem Political Economy: Old Toptes,
New Directions
Robert Bates, Beyond the Miracle of the Market: The Political Economy of
.
Agrarian Development in Kenya
Peter Cowhey and Mathew McCubbins, Structure and Policy in Japan and the
.
United States
Gary W. Cox, The Efficient Secret: The Cabinet and the Development of
Political Parties in Victorian England
. Jea:nEnsminger; Making a Market!'Th-e 1nstitutional Transformation ofan
. . ..... African. Societ:)!. . .' . . ... ., ...
Murray Horn, The Poltica I Economy of Public Administration: lnstitutional
.
Choice in the Public Sector .
.
Jack Knight, Institutons and Social Conflict
Micnae1 Laver and KeI}neth Shepsle, Cabinet Ministers and Parliamentary
Govemment
Mithael Laver and Kenneth Shepsle, Making and-Breaking Govemments
Brian Levy and Pablo T. Spiller, Regulations, lnstitutions, and Commil:n1tm-1.
Leif Lewin, Ideology and Strategy: A Centuiy of Swedish Politics (English
Edition)
Gary Libecap, Contracting for Property Rights
Mathew D. McCubbins and Terry Sullivan, eds., Congress: Structure and Policy
Gary J. Miller, Managerial Dilemmas: The Political Economy of Hierarchy
. Douglass C. North, Institutions, Institutional Change, and Economic
Performance
Elinor Ostrom, Goveming the Commons: The Evolution of Institutions for
Collective Action
J. Mark Ramseyer, Odd Markets in Japanese History
J. Mark Ramseyer and Frances Rosenbluth, The Politics of Oligarchy:

Institutional Choice in Imperial ]apan
Jean-Laurent Rosenthal, The Fruits of Revolution: Property Rights, Utigation,
and French Agriculture
Charles Stewart I1I, Budget Reform Politics: The Design of the Appropriations
... "f.r.9.f.i1s.J1Jt:I?:e...fI.9.11s.e..qfR,e.presentatives, I 865- I 92I
...
George Tsebelis and Jeannette Money, Bicameralism
John Waterbury, Exposed to Innumerable Delusions: Publc Enterprise and
State Power in Egypt, India, Mexico, and Turkey
David L. Weimer, ed., The Poltical Economy of Property Rights: Institutional
Change and Credibility in the Reform of Centrally Planned Economies
CAMBRIDGE
UNIVERSITY PRESS
PUBLISHED BY THE PRESS SYNDICATE OF THE UNlVERSITY OF CAMBRIDGE
The Pitt Bulding, Trwnpington Street, Caril bridge. United Kingdom
Contents
CAMBRIDGE UNlVERSITY PRESS
The Edinburgh Building, Cambridge CB2 2RU, UK http://www.cup:cam.ac.uk
40 West 20th Street, New York, NY 10011-4211, USA http://www.cup.org
10 Stamford Road; Oakleigh, Melboume 3166, Australia
Yoram Barzel 1997
This book is in copyright. Subject to statutory exception and

.. ,'.. ,to, ,th.e. ,p.tQyisions..ohelev.ant,callective.,licensing::<l,grl!<t~l!:f:lt~, '.
"','," nOfeprodcon oi 'any~arimay 'rkep'ice without "
the written.permission of Carilbridge University Press ..
First edition published 1989
Second edition pubJished 1997
.Reprinted 1999
Printed in the United States of America
Typeset in Sabon
A catalogue record (or this book is available (TOm the British Library
, , ,LiiJrttTY. ,9f Clngr.'~s~, ,a.ta.l9g>fi"g~itt.~PubLicatjon. Data is available
ISBN 0-521-59275-5 hardback

15B1;'10-521-59713-7 paperback
pagev
Series editors' preface

Preface to the second edition .
Preface to the first edition .
ix'
xi
Introduction
1 . The property rights model
.
2
The publc domain: Rationing by waiting and price controls
3 Contract hoice: The tenancy. contract
4 Divided ownership
5 The old firm and the new organization
6 The formation of rigrus
7 Slavery
8 Wealth-maximizing constraints on property rights "
9 Property rights and non-market allocation
10 Additional property rights applications
IX
The property rights model: Recapitulation
- "I
3
"!6
33
55
65
.. 85
"''' 10 5
114"
128
139
14 8
154
References
lndex
159
Series editors" preface
The CambridgeSeries on the Political Economy of Institutions and Deci

sions is built around attempts to answer two central questions: How do
institutions evolve in response to individual incentives, strategies, ando .
choices? How do nstitutions affect the perfOl;mance of political and eco
nomic systems? The scope of the series is"comparative and historical
rather than international or specmcally American, and the focus is posi
tive rather than normative.
. ... .':nJ.le.first edition of this work has become the classic stqtement of the
propertY ~ights paradigm. In this second edition Yoram Barzel clarfies,
elaborates, and extends the argumento Clarifications consist of more
straightforward writing and making effective use of diagrams to illustraJ:e
complex theoretical points. Elaboration takes the form of greater depth
of analysis in various sections of the study; for example, the old chapter
"The Old Firm and the New Organization" has been divided into two
chapters that now include new sections on divided ownership and on
insurance. Barzel extends the argument to the role of the state in the
formation of rights, the role of government in lowering transaction costs,
and property rights to wildlfe.
The property rights model developed in this book is an important ex
tension and modification of economic theory, explaining an array of phe
nomena' that standardtheor.ycannot.successfully .~ddress.
v.ii
:>"'~.
.."';~'~''.;{''~:.~,~
...<.... ':~-"':: ,.. ::>"
.: &(;~';I,,~;':: \.~ :<".;:' '::-:
':~.\~. ><;~i.i'~! ~~I:,\;.~~;g~:i~;{;:.:~t~!:;':'/;::~;:;/I.:; 'I( :~~::
;:.
Preface to the second edition
- ~
:,-,
,':- ..
-< -
-,~-
~ ~.
..
':
-.~.
,0,'
;.;:,i
Since the publcation of the first edition of this book I have continued to
conduct research regarding economic organization and poltical econ
omy. This work is reflected primarily in Chapters 5 and 6 of the current
edition. I have aIs o separated the discussion of divided ownership from
that of the firmo The former now occupies Chapter 4, whereas in Chapter
5 I offer new thoughts on the latter. In Chapter 6 I briefly speculate on
the emergence of property rights and the state.
Teachingproperty rights courses over the last few years has led me to
discover iriihe fiisfedtiifsome rtots~iiiiihyambigities; a.n:d several ... ' .
instances that call for elabor-ation. I have attempted to correct the mis
takes, clarify the exposition, and elaborate when necessary. I have also
added a numbei of illustrations, some of which derive from new research
on property rights.
I wish to thank Dean Lueck, who read the ~n'tire manuscript, for his
helpful comments; my daughter Tamar, for enlivening the presentation
and protecting the English language; and the Earhart Foundation for its
generous financiaI support.
IX
Preface to the first edition
"
The intellec.tual content of "property rights," a term that has enchanted

and occasionally mesmerized economists, seems to lie within the jurisdic
don of the legal profession.. Consistent with their imperialist tendencies,
however, economists have also attempted to appropriate it. Both disci
plines can justify"their claims, since the term is given different meanings
on different occasions. Perhaps economists should initially have coined a
t;erm dfstinct frQmth~.n~..used for legal purposes,I;>l,l~dby now the cost
of doihg so is too high. I attempt, however, to make clear the meaning I
give to "property rights" and to demonsuai:ewhy property dghis"so"de:':
fined ar~ an appropriatesubject for economic analysis"
"Th' inaterial" ofth book is at the heart of a course I have taught in
re.ceIlt years... Undergr:.aduate students take my approach in stride. Gradu
ate srudents often vigorously resist my dissatisfaction with the z~!:2 trans
action costs modeI; converting them is, however, rewarding. This book
is influenced by the diverse classroom reactions. It is an attempt to appeal
both to those with litde training in economics and to specialists.
I am grateful to my former students Douglas Allen and Dean Lueck and
to my colleague Paul Heyne, who read early drafts of the book and forced
. me to refimulate many oEmy ideM and to clarify their presentation.
Douglass North, who supported the project from its infancy, also read
the entire manuscript and made numerous valuable suggestions. Victor
Goldberg and Levis Kochin provided useful comments as well. I also
thank Elizabeth Case and mydaughter Tamar for excellent editing; they
demonstrated that clear writing enhances clear thinking. Finally, I wish
to thank the Earhart .Foundation for financiaI support.
Xl
Introduction
In the slave societies of-the American South and the West Indies, as well
as in others, slaves consistently - albeit rarely - bought their contracts
from their owners in arder to redeem themselves from slavery. In these
societies the law afforded owners virtually absolut rights over their
slaves, only seldom granting any legal rights to ~he slavs themselves;
consequently slaves were not legally entitled to own the pfoperty neces
sary for self-purchase. There was no legal barrier ar authority to stand
in the way of owners' retaining both slave and freedom money. Never.the
less self-purchase; whereby slaves acquired legal rghts to their own la
bor, dd occur.
As will be elaborated upon in Chapter 7, the' study of propert'y rights '.
and of the costs of transacting can yield an explanation as to why slaves
were able to buy their freedom; such explanations an be tested against
the facts. The property rights model I develop in this book can provide
explanations of an array of such phenomena, which standard economic
theory cannot successfully address. These explanations range from iden
tifying the re.asons behind the choice between wage and piece-rate con
tracts to pinpointing the conditons under which charity is more eff1cient
than profit-seeking behavior.
In the followng chapter, I shall define "property rights" and introduce
sotiteof the'certttalideas'ofthis book. In Chapter 2, the examination of
the gasoline shortage of the 1970S illustrates the usefulness and impor
tance of the property rights framework and familiarizes the reader with
its mechanics. Chapters 3,4, and 5 present the property'rights mode! and
its main organizational implcations. Chapters,6 thi:ough Ia expand the
mode!and apply it to various problems, including tights formation, slav
ery, and resource aIlocation in non-market settings. Chapter I I summa
rizes the discussion presented"in'Chapters'I-I0"n~i,pr0simts,some.gen~."
eral conclusions.
-," '
I
I
The property rights model
THE DEFINITION OF ECONOMIC AND LEGAL RIGHTS
What are property rights? The term "property rights" carries two distinct
meanings in the economic literature. Qne, primarily developed by Al
chian (:1965, :1987) and Cheung (:1969), is essentially the ability to enjoy
a piece of property. The other, much more prevalent and much older, is
essentially what the state assigns to a person. 1 I designate the first "eco
nomic (proprty) rights" and the second "legal (property) rights." Eco
nomic rights are the end (that is, what people ultimately seek), whereas
.. lgliiglitsare'th' irieiiiis to achieve the end. In this book I am concerned
primarily with economic iights. Legal rights pIay a primarily supporting
role a very prominent mie, however, for they are easier to observe than
economic rights.
I define the economic property rights an individual has over a com
modity (or an asset) to be tbe individual's ability, in ,expected terms, to
consume the good (or the services of the asset) directly or to consume it
indirectly through exchange. 2 According to this definition, an individual
has fewer rights over a commodity that is prone to theft or restrictions
on its exchange.
The notion of rights is closely re1ated to that of residual claimancy.
The residual claimant to, say, an apartment house is its economic owner
in that he is able to gain(here.by.exchange)Jromanincreas(!jJ::uh.evalue
ofthe building, whereas he loses from a reduction in that value. Being its
owner, he is motivated to take any action that will, net of its cost, in
crease the value of .the property. The residual claimancy from an asset or
an operation is often shared by several individuais. An important propo
sition, to be elaborated on in Chapter 3, is that in order to maximize the
value of rights, a person's share in the residiilshould increase as his
1 Ellickson (I99I) makes the same distinction and elaborates on the role of"the
enforcement of rghts wthout the assstance of the state.
2This definition follows that by Alchian (I96S), Alchian (I987), Alchan and AIlen
(I977, pp. II4 and I98), Allen (I99I) and Cheung (I970).
.::tr~(~p;';.tf.:M!5~((..:m:;tt.:. .::L~~:
. '.-:. "::;,~.~.~,:r.:8,<.:i.,.,, "}.',,Z ,",';',",_,,,-,, v'i, ,;'$,_,L-,
.\1',;,;,.-i~~_,
. ,--Un'
.'~. ,h
Economic analysis of property rights

contribution to the me:tn output increases, and it should fall as bis contri
bution decreases.
,
The economic rights people have over assets (including themselves and
other people) are not constant; they are a function of their own direct
efforts at protection, of other people's capture attempts, occasionally of
formal and informal non-governmental protection, and of govemmental
protection effected primarily through the police and the courts. 3 Legal
rights are the rights recognized and enforced, in part, by the government.
These rights, as a mIe, enhance ecoI).omic righ~s, ,bur the, frrp.er are nei
the!" necessarynorsufficieilt for .the e~istence of th.e ltier,.Amajqrfunc
'tion legl dghts is to; acco~qd~te:.~d:,partya4Judication.ap.d.eu,.
forcement. In theabsence f these safegards, rights may still ,be valued,
but assets and their exchange must then be self-enforced. Squatters are
less secure in their rights to the land they occupy than are legal owners
not beca use they lack deeds but beca~s'e less police protection is expected
for such holdings. Agreements based on ,goodwill are examples Df ex
change not supported by third-party enforcement.
.,
As defiued here, praperty rights are nor abs~l~te and can be changed
by individuaIs' actions; such a definitiori, theu, is useful in the analysis of
resource allocation. The past faiIure of economists to expIoit the proper!y
rights notion in the anatysis of behavior prbably stems fiom a tendency
to consider rights as absolute.
The concept of property rights is closely related to that of transaction
costs. I define transaction costs as tlie costs associatei!. with the :ransfer;
capture, and protection of rights. 4 If it is ~ssumed th~t for any ass!:!t each.'
of these costs is rising, and that both the full protection and the full trans
fer of rights are prohibitively costly, then it follows that rights are never
complete, beca use peopIe wll never find it worthwhile to gain the entire
potential of "their" assets. In order that the rights to an asset be complete
or pedectly delineated, both its owner and other individuaIs potentialIy
interested in the asset must possess full knowledge of 'alI its vlued attri
butes. With fuU knowledge, the transfer of rights to an asset can be
readily effected. ConverseIy, when rights ate perfect1y delineated, praduct
. nformationmust be costless to obtain and the (relevant) costs of trans
acting must then be zero.
When transaction costs are positive, right!) to assets wll noi: be per
fectIy delineated. The reason is that, relaJ:iv' to their vaIue, some of the
3The distinction sometimes made between p'r~perty rights and human rights is spu
dous. Human,xights are simply part of a person's property rights. Human rights may
be difficult to protect, ox:..to. exchange, ..,but..so"3:re.. r:iglits ,to..many..other: ,assets .See"
Alchian and Allen I977, p. Iq.
'.
,'. .
4See
Barzel I9 82 and Allen I99I. What:Jensef.l aI).d Meckling (I97 6 ) define as
agency cost is what is defined here as transaction cost"
4

attributes of me assets are costly to measure. The'efore the attributes of
such assets re not fully known to prospective owners and are often not
known to the current owner either. The transfer of assets entails costs
'reslting from both parties' attempts to determine what the valued attI
butes of these assets are and fram the attempt by each to capture those
attributes that, because of the prohibitive costs, remain poorly deline
ated. s Exchanges that otherwise would be attractive may be forsaken
because of such exchange costs.
An illustration of me costliness of exchanging rights and meir effect
onre$.?,uIce allocation may be drawn fram tp~.,:lrft:,<?t4?H~g~:,f.(;)?~~~~:::.:,::,,::::,,:,::
'c:;; .;i1"pl~yerslby theNational FootbaULeague(NFL)~By dFaftinga player, a
.
team acquires the exclusive'hegotiation i:lghts for his se:vkes, inclusive
.
of the right to transfer to any other NFLteam. Every year the twenty-'
eight NFL teams selecteligible college pIayers in a predetermined se
quence. It would seem mat the team with the right to, say, the twentieth
selection would choose the pIayer among those not yet drafted whose net
value to a'ny of the teams is highest. Given the diversity ofboth pIayers
. and teams, the probability that the team with the right to the twentieth
selection will also be the one placing its highest value on any of the re
maining players, is the same as any other team's, that is, one in twenty
"eigiii:':Were me costs of exchange'arnong teams low, the prabability of
that player being tra'ded' wotild'therilietWenty-'se'Ven'intwenty-'eight; 'The"
observ.ed trading frequency of newly drafted players, however, is much
rowei than a low transaction cost model predicts. This cost of trans
acting, at least, does seem to be considerable.
. What underlies this costliness of transacting? What are the factors that
prevent peopIe fram realizing the fuII value ofilieir assets? Commodities
have many attributes whose leveIs vary from one specimen to anether.
Measuring these levels is toa costly to be comprehensive or entirely accu
rate. How difficuIt it is to obtain full information in the face of variabiIity
.fundamentally determines how difficult it is to delineate rights. Because
it'is costly to measure commodities fully, the potential of weaIth capture
, is present inevery exchange. The opportunity for wealth capture is equiv
alent to findng property in the pubIic domain. A commodity lies in the
"public domain" when the resources needed to acquire it accrue to no
one. 6 As viewed here, some wealth splls over into the publc domain in
every exchange, and individuaIs spend resources to capture it. This is
characterized as ."capture" because here, in contrast to a market sale,
the original owner does not receive what the recipient expends. Whereas
pepIe always expect to gain fram exchange, they a1so aIways spend te
.. "5si~i'i~~~;;~~id~'~~ti~;;~i;:;otdbbrat(:don here) apply to the protection ~'f assets.

6 As discussed in Chapter 2, the waiting time that people spend in line to acquire a
"free" good accrues to no one, and therefore such a good lies in the public domain.
'T}-,;:::o

sources on capture. Individuais maximize their (expected) netgains, the
gains from exchange as conventionally perceived net of the cost of ef
fecting exchange.
The sale of cherries illustrates the phenomenon of wealth capture. Ob
vious problems of information present themselves when cherries are ex
changed. Customers must spend resources in order to determine whether
a store's cherries are worth buying and which particular cherries to buy.
Store owners who allow customers to pick and choose cannot easily pre
..y,c:nt"t;hc:m.from.eating"cherries after they have already decided whether
'9rriOt'tobiiythem~norcaritheyp:revent customer~' careless handling of
cherries. Indeed, th prOcessof picking and choosing itself llows wealth
capture in the form of excess choosing. 7 The fact that the same cherry
may be inspected by multipie customers indicates that some of the cher
ries' artributes are placed in the public domain. The high cost of informa
tion results in transaction costs - costs that would not arise were the
owner and the consumer of cherries the same person. If informaton
about the cherries were costless, their initial owner would not have to
relinquish any rights, and pilfering, damage, and excess choosing would
be avoided. In reality, such public domain problems are unavoidable;
people can take steps, however, to reduce the associated losses . .one of
the main- tasks I will undertake is to discover some methods to reduce
such losses.
Physical opeJ;"ations within, and on the ftinge of, an organization such as
a firm usually involve many commodities and, correspondingly, many
attributes. Severa! individuais share in ownership of the attributes, each
owning alone or with others some subset of these. Stockholders own
some of these attributes, but definitely not all of them. For example, a
firm (or, more accurately, its stockholders) that has a service contract for
a copier to which it has the title does not have fuII economic rights over
the. c;opier . The firm is not the only party that gains when the copier
perfoims.welI and loses when it does noto The service supplier is the
.residualdimant from'd1eservicingop~ration,gaining if it provides good

servke and losing if the svice is poor; it: is thus "apart ownec of the
copier. Among t>~her partial owners is the manufacturer, which is liable
for certain damages the copier may cause. Others are the employees, who
are able to put the copier to personal use without charge; they are also
p~rt owners of economic rights, though not of legal rights, since in prac
tice they have a claim on some of the copier's output. Here, too, restric
tions my serve to separate rights and p-!event free rides. In Chapter 8 it
wiU be shown that such restrictions do not necessarily artenuate rights
but instead may enhance them.
FACTORS THAT AFFECT
.Tl!E ALLOCAl;'ION OF OWNERSHIP
..
DIVIDED OWNERSHIP OF COMMODITIES
Net gains from exchange can often be increased if the original owners of
commodities transIer only subsets of the commodties' attributes while
retaning the resto Cases where only a subset of rights is transferred are
common; for instance, this is so in all rental agreements, as it is in any
sale subject to guarantee.Exchanges that take this form result in divided
property rights for single commodities: two or more individuais may own
distinct attributes of the same commodity. As will be elaborated in Chap
ters 4 and 8, restrictions on the behavior obhe()wne,r~"ml;l,y be imposed
in order to enhI:;~~th~~ep~~~ti;;~ ~(th~ir individual economic rights.
Incomplete separation makes attributes common property, relinquishing
them to the public domain; if they are in the public domain, resources
are spent
their capture.
Not only is ownership of commodities often divided; ownership of
what appear to be the assets of an 'rgahization may ~{! divided as well. 8
on
7Barzel 1982..
8Alchian (1965) recognizes that ownership of commodities and of organizations
may be divided. Posner (1992.) discussed property rights; he, too, notes that ownership
can be divided.
. ,. .
...... .
One f the most celebrateq propositions in econmics is the Coase Theo

rem: When rights 'are we1Ldefined and the-c~st of transacting is zero,
resource allocation is effident and independent of the partern of owner
ship (Coase 1960). Were rights well defined everywhere, much of eco
nomics, including this book, would be superfluous. Because the cost of
transacting is positive, delineating and enforcing rights is costly - prohib
itively' so if done to perfecton. In this section the proposition will be
extended to consider the effect of positve transaction cost.
The ability to receiv:e the ncome flow generated byan asset constitutes
part of the property rights .over it. The value of an asset is lowered when
non-owners are inclined and able to affect its income flow without bear
ing the full costs of their actions. The maximization of the net value of
an asset, then, involves the ownership or ownership pa:t:tern that can most
effectively constrain uncompensated exploitation. The kind of ownership
partern to emerge depends on the variability in value of such assets.
It is relatively easy to ensure the rights to an asset when the service
.fk>w k-generates..can-.be..readily. asce.ctained. because. it. is "ea.sY:.1;().imp().sj'! ..
a charge commensurate with the level of service exchanged. When the
..service flow is known and cnstant, it is easiest to ensure that rights ove.c
me asset are also certain. If the flow is variable but fully predictable (for
7
Economic ana/ysis of property rights
instance, if the service is the amount of electricity solar panels produce

function will adjust for the effects of careless driving. The net gain in
as the length f the day changes), rights are still easy to ensure, as they
using the renral market, then, is less than it would be were the two parties
are if the flow is not certain but is unalterable (for instance, the amount
to exercise greater care. If smoothness were costlessly measurable, the
of electricity the panels produce as weather conditions change). It is evi
effect that each transactor had on that attribute could be easily deter
dent that, given the mean outcome, variability and uncertainty may re
mined and accurately charged for. In reality, assessing such marginal
duce the value of the asset but need not affect the certainty of ownership.
charges accurately is prohibitively expensive, and (maximizing) owners
When the flow of income from an asset may be affected by the ex
will choose not to exercise their rights fulIy. Some of the income stream,
change parties, ensuring ownership over it becomes problematic. When
then, is left in the public domain. It is partly recaptured by the exchang
the income stream is variable and not fully predictabl, it is costly to
ing parties, who act differently than owner-drivers would.
determine whether the flow is whatjt.shouldhveHoeeri:lniinypartlcuhii::::::::::::::
:::Whereas::rightscannQtbe fuUy defined economically when both ex
case;'t;ohs~~ilentlY,kjs'~Io:.ostiy todeterminewhether'part ofthe in-o
change partiesare a,bJe toaffeccttheoutorne, only on pattem of owner-~.:;:
. com'e stream has been captured by the exchange parties. The exchange
ship does maximize th'net incometrom the asset(and thus tsvalue to .
parties wilI engage in wealth-consumingc'apture activities beca use they
its original owner). As I have already stated, the general principie under
expect to..gain from them. When the income strcam from the exchanged
.lying the maximizing allocation of ownership is that the greater a party's
property is subject to random fluctuations, and whe~ both parties can
indination to affect the mean ncome an asset can generate, the greatex:
gain by affecting that income stream, the delineation of ownership then
is the share of the residual (that is, ownership share) that party should
becomes imperfect.
assume.
. A special case of great importance for understanding the circumstances
The nominal owner of an asset may seem to have the right to the in
under which ownership can be ensured arises when only one of two ex
come the asset can generate. When the highest income the asset can gen- .
.. changeparties can affect.the..income flow. Making the person who can
erate requires exchange, some of the income potential will be used up in
affe't the flo'I,V .b.e.ar..ftI II.. r.<=s.pon.sip~lity. fo.,:. hjs.. Q~he~..a~ti.Q.ns.c;:psgre~. JlJ..aJ .
the process of effecting the exchange. The net income an asset wll gener
ownership becomes secure. Such a person, being the "residual claimant"
. ...... atti 'di4iends on the delineation of rights,that is, on how secure rights are
to an outcome that only he or she can influence, is the full-fledged owner
over it. Il the case described earlier, where only one person can affect the
. , of the asset.
income from an asset, it is only when that person becomes the owner.of
In reality, randomness is pervasive and both exchange parties can af
the asset that rights become perfectly well defined, and it is only then that
fect the service flow generated by exchanged assets; ownership is there
the income is maximized. This is a case where (as will be discussed in
fore seldom if ever fully secure. For instance, the incomc: .!!tre.am gener
Chapter 4~ rights are allocated dearly and income is maximized only
ated by a rented car depends, in part, on how smoothly the car operates.
under a particular allocation of rights. Since transacting is costly, the
Since used and even new cars are not identical to one another, they are
Coase Theorem cannot be invoked; it becomes meaningless to state here
not expected to run equalIy smoothly. A smooth ride is an attribute that
that as rights are well defined, income is maximized regardless of who
both the owner and the renter can affect. A renter will find it expensive
has these rights.
to determine to what extent the smoothness of the ride of the rental car
reshs from its character and to what extent it results from the care given
THE RELATIONSHIP BETWEEN INDIVIDUALS' RIGHTS
to servicing it. It is expensive to determine the true mean of the popula
..ANDECNOMtcORGAN rZATI O N S
tion from which the car is obtained, and it is difficult to know to whom
to attribute the actual performance. Similarly, the owner cannot tell how
Alienating assets and obtaining income from them require exchange, the
much the smoothness of the rented car's ride has deteriorated beca use of
mutual ceding of rights. Contracts that delineate the terms under which
the way it has been driven and how much it has deteriorated beca use of
legal rights are exchanged govern much of the exchange of economc
ts character. As a result, the owner may skimp on servicingIeptal cars
property rights and are central to the study of such rights. Some con-.
doing less than owner-drivers would and renters may be less careful
tracting parties consist oniidividuals acting.()n their own behalf. Others
with..ienf:d..caisthertlneywould be wth their own. Each party expects
consist of pairs of organizations such as firms, gov.e.rnm.~P.J!!.. I.JJ.!>,.. @c;I..
such behavior of the other. Therefore, the demand function for rented
families. In addition, there are contracts between individuais and sueh
cars will adjust for the effects of inadequate servicing, and the supply
organizations. Because individuais' objectives are relatively clear;
8
whereas those of organizations are not, it is useful to define ali. economic

property rightS as rights possessed by individuais. Ultimately individuais
always interact with other individuais, regardless of whether one or both
interacting parties represent organizations in some capaciry. The pay
ments supermarket shoppers make for merchandise can be viewed as ex
changes between individuais and an organization - between customers
and the store. Such relationships, however, can always be reduced to the
individual leveI. Let us consider the relationship between the cashier and
..... "the' custiners, on the one hand. and between the cashier and the store
m~!1ag!,!r.> nthe ther. Acash~~r in a :tore hasth.erighuo c()llecJ;:~oney~,
fram distbmers who buy merchi:idse nth stre. The cashier; fcourse,'
does not usually retain customers' payments; rather, in exchange for an
hourly wage, the cashier cedes to the store manager rights over his or her
time as well as rights Over the cash received from customers. In turn, the
manager's relationships with other individuais such as the store owners
involve other sets of exchanged rights. As I will demonstrate throughout
1:his book, the funciining of any organization can be similarly reduced
to the ceding of vari9us rights from orie individual associated with it to
another.
The assumption of indiv'idulmaximzation and, in particular;the'as
sumj5tion that individuais maximize the value of their economk' rights' .' ."
are useful not only direc'tly in the analysis of individuaIs' behavior but
also. indirectly as .th assumption uRerlying the functioning of organiza
tions - irdeed, f ali socities. individual maximization implies that
whenever indlviluals perceive that ertain actions will enhance the value
of their dghts, they will undertake such actions. This always applies,
whether the individuaIs operate in markets, firms, families, tribes, gov-_
ernment, or any other organization.
theory of property rights are available. These operational features also
apply to the analysis of constraints.
Because of the costliness of delineating and polidng rights, opportuni

ties arise for some people to capture what appears to be others' wealth.
As wll be demonstrated in Chapter 3, these opportunities adse fram
people's ability to overuse and to underprovide unpriced attributes when
exchanging with each other. Exchange partners may impose restrictions

on one another in order to reduce the leveI of undesired behavior. Conse
:.. gJJ~l}~y,PJ:~~r.!Y::r.~.~ts."",partic.ulru:I,Y.th~ ..r.igl:ltt.<>...c.<>'I.1~?me 'tI hat appears
tO"be .one's own properry:'::::ieftni:ldesubjettto::constraint. The
- charaer arid incidence of theconstraiits are predictable. The analysis
of consttaints, therefore, can be incorporated into the study of property
rights.
OPERATIONAL FEATURES OF
THE PROPERTY RIGHTS MO DEL
The exchange "alue of an asset is a function of the gross income it can

generate and of the costs of measuring and policing its exchange. The
ownership of assets' attributes is expected to gravitate into the hands of
those people who are most indined to <;tf(ect the income flow the attri
butes can generate. The gross income stream (the rlarket value of the
flow of services) an asset can generate, the value of the contributions of
different individuaIs, and the costs of policing and measuring the attri
......... ,....butes.. af. the.asset.determne.hQth..how..strictlf..rights.;to. it.will.pe, c:leline
ated and what ts ownership pattem will be. Since these and similar mag
nitudes are measurable, the ingredients necessary for an operational
THE PROPERTY RIGHTS APPROACH VERSUS
THE WALRASIAN MODEL
The presence of positive ttansaction costs is what makes the study of

property rights significant. On the other hand, in the Walrasiah, perfecdy
competitive model, rights are perfectly delineated and transaction costs
'are zero; It'isuseful;then; to briefly conttast.the two models. A funda
mentai difference between the two concerns the role of prices. In the
Walrasian model, costlessly determined prices !i.uffice for alI al1ocation
problems. when transacting is costly, on the other hand, exchange re
quires non-price al1ocation methods with corresponding organizatons.
In the Walrasian model, when equilibrium is disturbed a new equlib
dum is instalitaneously attained because, given zero' transaction costs,
the cost of adjustment is zero. In addition, Walrasian commodities are
made up of strictly identical specimens, people are fully informed regard
ing the exchanged commodities, the terms of trade are always perfectly
dear, and trade is instantaneous. As a result, neither buyer nor seller has
to make any effort to incur any cost other than for the buyer to dispense
the appropriate amount ofcash and for the sel1er to cede the appropriate
units of the good. Prices alone always suflice to a1l6cateiesourcestotheir '
highest-value uses.
In the Walrasian model, where prices are sufficient for efficent alloca
don, institutions are superfluous; firms; dibs, tribes, or families cannot
enhance efficency. Nevertheless, for a long time economists attempted to
address qtfeStions of organization by what amounted to ad hoc tinkering
with -the Walrasian model. Only recently have economists (and other
social scientists) begun to take notice of the inconsistencies intrinsic to
such anipproach. The model used here explicitly explores the effects of
II
IO
'':;''<:::-''!. '-'<''',}''.'
..,., .....", ...:" .
,_,.'-.Jl'i".1....tt,.,_"'j.',6~.,;h\ ...!1 'r!. _.',',.,
'.\. "
,... '
,~- .. ,$-.,""
positive information cst and the resuIting positive transaction cost on

behavior and 'on organization.
.
When equilibrium is disturbed in a positive transaction .cost world,
price adjustment is not expected to be instantaneous. As long as prices
are not fuJIy adjusted to new conditions, the quantities demanded are
not, in general, equaI to those supplied. Nevertheless, it is possible to
determine how equilibrium will be attained. Where transaction costs are
positive, a whoIe array of activities is required to effect exchange; cash
with which to pay the pecuniary price is helpful but definitely not ~uffi~...
....::::::;;ciffit,Bocause of the compl~xity. of exchange, parties have.rnany Oppor~
. ~turiities to alter thejrbeh,ayioI' ;i:n ofletto g,ain fr9.pithedisrepancy,He..
tween the priceactually chrgedandthe one that would have achived
equilibrium.
.
To illustrate, consider some of the activities requiredto effect pur
chases in stores. Buyers must decide, among other things, whether to
shop during the busiest hours (when, at the going price, th~ quantity
demanded exceeds that supplied), or at off-peak times (when the reverse
is true). They must then obtain ali sorts ofinformation: identify the loa
tion of the desired merchandise; determine by themselves or with the help
of the sellers if the items they seek are available; decide if they are of the
appropriate quality; select the specimens they think are best; ascertain the
price, over which they may haggle; and make payment, not necessarily in
cash. In addition, they may have to take care of warranties and, on occa
sion, exchange the merchandise. Completing purchases, then, involves an.:
elaborate set of operations. More important, the costs and va'uations of
most of these operations can be altered. For instance, at any particular
. time a buyer may exercise return privileges more vigorously, and a seller
may be out of an item that is usually in plentifuI supply, or may unexpect
edly heIp carry the merchandise to customers' cars. When the market
clearing price changes but the nominal price does not, buyers and sellers
may stilI adjust in many ways. They can gain from such adjustments, and
weaIth maximization implies that adjustments will be forthcoming:
Sellers can adjust to a price that is lower than the market-cIering levei
alongvariousmargins. A'seUerwhtfis in control of the quality of the
merchandise or of the number of cashiers per customer wiII adjust along
such margins, especially the latter. For example, supermarkets ~end to
reduce the speed of service at rush houes. In general, sellers who'choose
not to adjust prices or who are prevented from adjusting them may stiU
adjust along Qther.margins. Given wealth maximization, 'the margins
along which they will adjust and 1:he corresponding effects on reso.urce
allocation are predietable.
.....................................................:....::.:
The analysis of non-price adjustments or of property rights Ileed not

be restrieted to the market sector in an economy 01' to market ecn9inies;
I2
on the eontrary, the results of such analysis apply everywhere. They are
as applicable to Hong Kong as they are to China during the Red Guard
era 01' to tribes entirely without a market system. Application, of course,
requires knowledge of the underlying constraints, and sueh knowledge
may be harder to come by in some systems than in others. Although
property rights analysis is usually applied to the capitalist market system
only, it is most usefuJ (and the Walrasian model is least usefu1) in systems
in which market prices are least used and aIlowed to adjust. In Chapter
9 J. will discuss briefly how property rights tooIs may be applied to a non
. pric~ economy. .
.
;'VifruallY . llUgoverninentsplaYc.a:major.role.. with.regard.to...pr.operty
prpertiesarid partici
rights; they.s'o maintain the legal rightto
pate directly in economic activities. In addition, governments are heavily
~nvolved with adjudicating and enforcing contracts. A comprehensive
analysis of the' coles of government is beyond the scope of the present
study; these roles will be touched on in Chapter IO in the course of ana
Iyzing the behf\vior of individuaIs and enterprises.
Cqstoms and mores seem to be additional non-price factors that affect
the allocation of resources. However, the effects of these factors on be
havior and 'On the enfoJ:cement of contracts will be ignored; although the
factors to be considred'iire'iiIlwed to hange, customs'andmores, like
tastes, are assum~d to be stable and accordingly haveno effect.on. the
margin.
vrious
'.
THE-DJSTINCTIVENESS OF
THE PROPERTY RIGHTS APPROACH
An enormous amount of literature written in the Iast quarter century

departs from the Walrasian, costless transacting, model. This literature,
in whichthe costs of information play a major role, is diverse, and thus
far nq single modeI has stood out as the most useful one. 9 Different ap
pr~aches with a 'bewildering array of names proliferate: "agency theory,"
or the "prindpal~agent modd";"market signaling"; "rent seeking";
"bounded rationality"; "asymmetric information"; and "contract the
ory." It is difficult to determine the precise differences between and even
within these approaches beca use, as a rule, many assumptions are onIy
implicit. Moreover, the empiricahvork in the area is too meager to help
distinguish among them.
I sha11 not attempt to sort out these modeIs, instead offering a few
.l?gg~.s.tions as to why I find such modeIs to be Iess appealing than
the pr;;penydihtsmodeC'kshoiil .. t:>itiadedear,though, that the
9Eggertsson (1990) elaborates on this point.
13
';,';;

differences among the -models often seem more a matter of emphasis than
a reflection of different fundamental assumprlons.
The starting point of "agency theory" is that principaIs' maximizing
attempts are frustrated by agents whose objectives do not coincide with
their own. 10 The asserted asymmetry between the two parties is likely to
divert attention fr<;lm the reciprodty of - and perhaps even from the gains
deriving from - exchange. The "rent seeking".approach, whether applied
to market or government activity, tends to ignore (almost to a fauIt) gains
from exchange; it concentrates on peopIe's efforts to capture weaIth from .
"~'eacli::ilie.r:::and'iiegleci:s:tile",oppotttiiiieii::t.gaih:tl1rough "avoifif!.g.
.. waste.1;l> Indeed~ 'it neglects r the possibilit}r that, .indfviduallyand collec~
tively, people take advantage of the opportunities available to avoid
waste. In the process, they tend to exhaust these opportunities, nd the
cost of any further attempt must exceed the gain.
The problems inherent in the models based on asset specificity and on
the opportunistic capture of quasi rent are very different. 12 Such models
usually deal with variables that are exceedingly difficult to observe and
to measure. The proxies required to make such models operational are
even further removed from the desred variables than is usually the case
in economics. Thus, it is particularly difficult to determine precisely what
",it. is .~ll:1t..e.IllP,iri:ll, r~$~~, cQnfirm ,or refute. "Market signaling," like rent
seeking, emphasizes exploitation rather than maximization;13 as with the
asset specificity model, it is difficult t9 formulate empirical counterparts
to the varianles the theory suggests.
In contrast, contracts that use the state's assistance to delineate and
reassign ownership are central to the property rights approach. The study
of contracts formed by maximizing individuaIs, and of the performance
siich contracts induce, tends to reveal the correspondence between theo
retical variables and their emprica I counterparts. Knight (1924) was ap
parently the first to specifically point out the economic role of property
rights, and Gordon's (1954) thrust is similar. Coase (1960), Alchian
(1965), and Cheung (1969) bring operational elements to the analysis. I
wish to emphasize an additional element that helps keep realty in touch:
aconstant nqtiry as to "who owns\\rhat"aiidwfiat~predsly,'it isthat
each party receives and concedes in a transaction. These simple points
are nevertheless worth mentionng because they are frequently over

looked. The r~lative ease of rendering the property rights model opera
14
cional will be made clear in the following chapters.
The property rights model developed in Chapters 3, 4, and 5 will be
used to follow through on Demsetz's (1967) and Umbeck's (1977) em
bryonic contributions to rights formation. In Chapter 6 I shaU try to
show that the property rights model is useful in predicting when new
rights will be created and when existing rights will be plac:ed in the public
domain. I shall also argue that such hanges pervade economic activity.15
Several emprical tes!s of propn.y' rights propositions involving such di
verse activities assla~ry,.the .allocation of water rights in. the western
;United States, ndthe homestead movement will beptesented iii theJol.:
lowing chapters.
'
14 Akedof, in his pioneering study (I970), recognizes the effect o the non
uniformity o cQinmodities ,and o the inadequacy of the received'model of handling
themo
lS Another distincrion of niy study - alt..hough this need not be unique to the prop
erty rights approach _ is that I take.no account of problems of.risk aversion; ali my
attempts to explain behav\or proceed under the assutIlption of risk neutralityo As I .
will demonstrate fn Chapter '3, there is much to be gained anQ. little to be lost by
assuming people to be risk-neutral.
0
0
10Ross (I973) and Jensen nd Meckling (I976) are early proponents of agency
theory.
11 Tullock (I967), Krueger (X974), and Posner (I975) initiated,therent-seeking Jit
eratiire.
12WiIliamson (X975) and Klein, Crawford, and Alchian (X978) initiated the notion
of the capture of quasi rento
... 13 This approach had been initiated by Arrow (I973) and Spence (1973)0
14
..x. .?
The public domain
2
The public domain:
Rationing by waiting and price controls
In the following discussion an elaboration of a single example _ the gaso

line price controls of the 1970S - serves to illustrate the usefulness and
power of the property rights framework. It also introduces tools that are
- pseful in analyzing situations not subject to government intervention, in
. which the price of a commodity (or a cOriUnodity-attribute) differs from
.the (supposed) market clearing price.
Chapter I presented aproperty rights proposition central to this book:
Given inat'property rights are never. peJ:fe.c:~ly ..d,eli[le~t~4 .~9I,l1.e..y.al"e.d.
properties will always be in the pU"bHdmain. In the present chapter the
. natu're of maximization. as affected by properties in the public domain,
:-' will he.exammc;:d and the actual resolutions of severa 1 public domain
issues will be analyzed. Because an analysis of rationing by waiting offers
a convenieni introduction to the subject of property rights, I will briefly
concentrate on such an analysis; subsequent discussion wll provide a
more detailed analysis of maximization under price-controls, which will
highIght major features of the substance and the mechanics of property
rights.
R.ATIO.1'HNG BY WAITING
The rationing by waitng model used here, which is stripped of many

real-world features, is most elementary. Using this model makes it easy
to concentrate on the public domain issue while ignoring peripheraI
probIems. I will use the results of this basic anaIysis in my subsequent
ariaIysis of the price controIs pIaced on gasoline in the early 1970s.
When the government provides commodities at a zero pecuniary prce
and makes them availabIe on a fust-come, first-served basis, which it
...... enforces,..commoditiesteaIloct:edsfrictlyaccording to the order in
which individuais join the queue, and ultimately by the amount of time
individuais spend waiting in line. Even though orderly queues are often
16
encountered, they should not be taken for granted, as the folIowing ex

ample illustrates. Suppose it is publicly announced that a package con
taining a million dollars is to be given to the first person in line at a
particular place. It might seem that anyone who hears the announcement
would rush to be the first at the site and wait for the package to arrive.
If, however, the line is not to be policed, the ultimate owner of the money
is likely to be someone with an armored truck and a machine gun. In the
absence of policing, those who hear such an announcement will probably
not bother to join the queue unless they are abIe to compete effectively
:with.oWn:ets::of.. machine:gtm~.,. :.:::::::::::::.::::: .
. :. cCompetition for the freegood may t~k~';i':;ariety of forms; The specific
nature of restrietionsdelinets the margins of competitlor." ri. the previ-'
ous example, the restrictions are first-come, first-served an.d no policing,
whereas in the current example the margins are firepower rather than
time. Other margins are used in other cases. For instance, higher educa
tion is often provided free or for a low monetary fee, and the margin
of competition for admission is a combination of residency and grade
requirements. The provider of the free good may stipuIate first-come,
first-served; the queue will be orderly provided the appropriate restric
tions are applied. Such restrictions seem to be applied often, and orderly
rationng by waiting s a common occurrence. In the discussion that fol
towsfassriiethaftheqee is poIiced so that it remains orderly.
The price-setting mechanics developed in price-theory texts can be
used to determine the properries of the first-come, first-served allocation.
On the supply side, the government furnishes a fixed quantity of a good.
I will make a single change from the textbook mechanics on the demand
side: consumers exchange commodities for time rather than for money.
Given the fixed supply, forces of demand determine the equilbrium price
per unit of the good being distributed in terms of the amount of time
spent by individuais in the queue. Almost anything that can be said about
money in the standard case appIies to time when allocation is determined
by waiting. 1
One evident and importam difference between rationing by dollars

and rationing by time, how.e.ver,js..thatalltm~mLdllrs.across indi
viduais differs from the corresponding alIotmem of time. If w~idngdme
per unit of a good is five minutes, waiting is the onIy method of acquiring
the good, and if. the. good cannot be traded, then a person will stand in
line to obtain additional units of the good until the value of five minutes
of his time reaches the value of one extra unit. For example, when the
value of the individuaI's time is $12. per hour (i$ifor five minutes), he
will continue to rejoin the line to obtain another unit of the good as long...........
1
See Barzel I974, pp. 73-95.
17
The public domain

as his marginal valuatin of the good exceeds $1. If the price of the good
becomes $:.: instead of five minutes, the analysis proceeds along standard
lines. Given that the good cannot be traded, however, the inframarginal
individuaIs who ultimately get the good are those who vahie it most in
time rather than in money.
Changes in the rules governing the distribution of the good can be
accommodated easi1y according to this model of rationing by waiting.
For i~stanc:e. there ~sno reason to assume that the given commodity will
.be doled out.1n fi:xil btches'''There are other possible rules governing its
. distribritio1,'i~ .individuais inaybeallowed.~s;much,of it:. a;s they ,desir~;
acce'ss tothline mky be limitedtc/once per pe:od or toau unlimited
number of times; and on~e the commodity is obtained, trading it may be
permitted. Each rule constitutes a distinct way of allocating rigbts to the
good.
.
This basic analysis of rationing by waiting yields one key result: A
commodity announced to be free is ef(ectively placed in the public do
main an4 is of no vlue until ownershipis establ!shed. Establishing own
ership rquires that an individual fulfillcertain criteria; in the present
example, the criterion is to spend five minutes.in a. queue. Acquisition of
the commodity consumes real resources over ."ott'iibve the 'resources
used in production. In tis example, ownership is established over one' ."
already produced unit of the. commodity. Methods differ from case to
case, but wha~ever the method by whiGh right~ tire acquired, it may gen
erally be statd that resources m~t be spent to gain possession of com
mdities in the public domin, th'i: individl1al maxirniz@tion applies here
no less than to conventional exchange, and that persons who have a com
parative advantage in the method of allocation wll acquire the good.
PRICE CO.NTROL ANALYSIS
How r.e property rights allocated to a commodiiy that is sold at a price

below the market equilibrium levei? In the model of rationinghy waiting,
queuing is the means by which ownership is established. Rationing by
waiting can be viewed as a special case of price controI.
Real-world price controls differ from rationing by waiting in two im
portant ways. First, whereas the analysis of price control requires only
that the controlled price be lower than the equilibrium price, in the ra
tioning by waiting model I assume that the (money) price is zero. 2 Sec
ond, whereas in the rationing by waiting modl I assume that competi
..........tion:..can..occur.only..through..queuing, ..in.. the ..prire.control.analysis.that:..
. assi.unption is maintained only initially.
S/Q
Ql
Q*
Figure 2..1
...in th~'~~cioiiii:ig'i:iy wirig 'iriodel~' indiViduaIs' acquire' rights "tO" the

rationed comniodity by spending the appropriate amount of time in the
queue. Under prce controIs, rights allocation is more complex, and the
determination of how rights to an asset are actually allocated is essentiaI
for the analysis of the controls. 3 In the remainder of this chapter I will
analyze the price controIs that were placed on gasoline in the early I970s.
Before-undertaking this analyshwever, it is worthwhile to look at a
generic price controI model.
In the generic model, it is assumed that competition initially emerges
as queuing. Consider Figure 2..I, where D is the .demand curve, S is the
supply curve, the equilibrium price and quantity are P* and Q*, and the
controi price is Pc. Assuming that the control price is perfectly enforced,
a discrepancy between quantity demanded, Qu, and quantity supplied,
Ql' known as a "shortage," will arise. Sellers will supply only Qldj;
then, is the quantity available to consumers, a quantity for which they
.
will be willing to pay P 1"
In considering the effects of the price control on resource ailcation, it
should be restated that in the Walrasian model, in the absence of govern
ment regulations, rights are welf afined and allocation is efficient. Had
" tax of (P1-Pd been imposed, the market quantity would have also
been Ql' The standard analysi$ of the welfare, or deadweight 10ss, from
. ). I focs. on below-equilibrium prce controls (i.e., price celings) and ignore above
eqililbritlln price controls (i.e., price floors).
3See Cheung I974, pp. 53-71.
I9
I8
~r~!"
-;: :.:;. : ' :{,~';:'5~{t.;~lt~!~!>7:.:~:~~'::3:~::: :', :';';.: !."
-:>'<'-;'~J~HR~i: :j': ':~ .':
" ".' - '-:." - , .;1 'I!;;: :';1;';: (,;' ;. :)t:'};{\}!~l<: '; ~ tl,~;~td~:"I~{:<. :'~::;< ~;,'.
Q/time
Qo
~,' :"' ,'.;' '. >~:. ,,~:,
,.', '-';' "';, \ '.',... ~'I.:.,
',:. \ l ",'.t;~;>~,Xd,..,?,i;\};:~~U?I~.~','d' ""_ .~.'
' ,-h ~.\,'r'.~ (" ,,'.,' i~. \ ' .
"J'
"I, n ' .
"
The public domain
such a tax is that it equals the triangle berween the demand and supply
curves to the right of Qt; the excess of the amount consumers are willing
to pay over the resource cost of producing the quantity (Q* - Ql) which
is not produced beca use of the taxo This is sometimes also asserted to be
,the deadweight loss from controlling the price at Pc; as the rationing by
waiting dscusson suggests, this is not the case.
Because consumers can pay only Pc in money, they will spend re
sources equivalent to the difference (P 1 - Pd by waiting in line. For ex
"".mpJ~,jtQl.=,!9.9.""fc=}".P.l = $1.80, and the opportunity cost for con
sliinet:nS$Y'p,e:tJi6t;tI1efibuYrS'hth~~g~egate wll spend a total of
$1"OOlf money an<leight hours ih timeto'ptirchseone hundred ,units..
The amount represented by the rea P 1 - Pc X Ql is the dollar value of
the time expenditure. Because queuing is the only margin of competition,
this is the new equilibrium under the price controI. As the discussion
proceeds, various simplifying assumptions underlying the generic mode!
will be dropped. It is essential, however, to descrihe the control itself and
its background.
The Economic Stabilization Act of 1970 gave the president of the
United States the authority to impose controls on prices. On August 15,
1971, President Richard Nixon imposed a ninety-day economywide
.freeze,ofall..prices at their May 1971 leveI. This freeze was known as
Phase I of the price control and was succeeded by several more phases,
each involving som.e voluntary and some mandatory controls. 4 The final
one, Phase IV, lasted int 1974 and was essentially a period of gradual
price deregulation.
For many commodities the price controls caused inconveniences:
Fewer sales were maae on credit, a smaller variety of goods was avail
able, and free delivery was less frequento As a rule, however, shortages
did not arise. In the case of gasoline, the discrepancy between the con
trolled price and the market-dearing price that would have prevailed
without the controls was too large to mask some of the efiects of the
price controls. 5 In the wake oi the eruption of the Arab--lsraeli war on
October 6, 1973, the Organization of Petroleum Exporting Countries
(OPEC) restricted exporrsaiidised'i:hpiceofildepeti6Ieum. Prior
to the war, the world price of crude oi! had been around $3 a barreI. On
October 16 OPEC raised the price to nearly $5 a barreI, and on Decem

ber 23 the price was raised to $U.56 per barre1. 6 TWs drastic price in
crease - more than threefold - coupled with price controIIed to shortages
and queuing in the United States by December I973. Some aspects oi
these shortages may conveniently be analyzed using the property rights
modelo
4 Much ofthe information on the Nixon administration price controls comes from
Kalt (1981) and Rockoff (1984)' In the case of retail gasoline, prices were not explc
itly controlJedj instead, the margins, or markups, were controlled at various stages.
Only the price of crude petroleum was controlled. (This information was provided by
Bruce Peterson of the Ameeican Petroleum Instirute and Del Fogelquist of the Western
Oil and Gas Assocation.) The Cost of Living Council and the Internai RevenueSer-..
vice were the primary agencies involved in policing and enforcement of the controls.
5The meaning of "too laege" is clacified in the last paragraph of the section entitled
"The Minimization ofDissipation" in this chapter_
20
The gasoline transaction
purch~se"of 'gasolin:~'may appear tobea .simpIeoperation. In the ...

.cour~ "orthis chaRter,' howevet;it .Sh6W"d becQme dear th'at gasoline'
. transact:ions have numerolis valued attributes. During the period of price
controls, matket parricipants were able to alter the levels of gasoline
transaction attributes not controlled by the governrnent. For that reason
the actual allocation of property rights 'differed irom what the nominal
controIs seemed to specify. An examinarion of this price control episode
reveaIs the strength oi the property ,rights framework by responding to
the,question f how individuais were able to benefit irom the use or ex
change of commodities under the controls. It also suggests areas for eco
nomic inquiry and is helpful in developing refu1:a.pl~iI11plications ..
-Figure 2.2 sh"ows the change in crcumstances in the gasoline market....
induced by the Arab--Israeli war of 1973-ln May 1971 the price oi regu
lar gasoline was about 35 cents per gallon (P ). The crude-oil price hike
that resulted irom tne war cau'Sed -,a: shift up'an t(5 ihJe left in the supply
of gasoline-, from something IH~e SMay lfJ71 to $D.e<. 1973, as depicted in
'Figure 2.2.7 In the absence oi price controls, P*, the market-clearing
price would have been around 55 cents per gaBon after this decline in
supply.8
What effect did price controIs have on the behavior of market partc
pants? Prior to Cheung's work, t.l1e, price control literature asserted that
a shortage oi (QO~Ql) wouId arise, that the. typical consumer would
acquire a fraction oi the amoult of gasolne desired. 9 One also encoun
ters in the literature the contentions thatthe .shortage WouId beborne
proportionately, randomly, or arbitrarily. Such contentions are without
foundation. The Walrasian framework provides no guidance regarding
hehavior under the indicated conditions, aIthough random allocation
.The
. 6The average per-barrei regulated price of crude oi! in the United States was $3. 8 9
. for 1973 and $6.87 for 1974. See Statistical Abstract of the United States I986, p.
~~
7 Cntro!
of the wholesale price of gasoline complicates but does not

the
,essenee,0f-t:heanalysis., .."" .. "..,,,......,,,,, .. ,,,,,.. ,"..".".. ".. "...........'..','.'.
'" 'sAi; coltcols were relinquished, the ,uncontrolled price of gasoline finally leveled off
at aro'\lIld 55 cents per gallon in the summec of 1974.
.9Cheimg (1974) provides a review of ths literarure.
21
The publc doman
Economc analyss of property rghts
PrelimitU1:ry considerations of property rigbts under tbe controls
$IGaIloII
P 1=O.8S
~=.().SS
IC =0.35
Ql
Q*
QO
GaIlooslDay
Figure 2.2
seems to be the most likely resuIt from eXt:ending that framework. 10 The
view adopted here asserts that in reality people have many margins of
action that the Walrasian model abstracts from. Given these margins of
aoi~ wealih maxii:iiZioii wiU -giii:e a deterrnliiate equilibrium.
. Wealth maximization implies that individuaIs will carry on an activity
untiI (for the marginal unit)net gains are zero. Even' when a price is
controlled, the question 'must be asked: Can the buyer or the seller take
additional steps to obtain or provide another unit at a cost below the
added gain? If the answer is yes, an equilibrium has not yet been reached.
The notion of a ''''market-dearing'' equilibrium requires that alI individu
aIs make whatever moves they wish under the exisring property rights
arrangement. The textbook analysis of a binding price ceiling that con
dudes that a shortage will emerge implicitly assumes that buyers pay in
cash and that they receive the right amount of the right quality of the
commodity at the right time and place. It ignores the possibility of adjust
ments, thus impIicit1y denying that individuaIs maximize ifadjustments
are available. The way in which margins of adjustment were expIoited
under the gasoline price controIs will be examined in the following sec
tion.
lOIf the Ql available units are allocated randomly among the demanders of the Qo
units, then the deadweight loss is not the conventional welfare triangle. Rather, it is
proportionate to the. entire area under the demand curve above Pc (up to quantity
Qo), the factor of proportionality being (Qo - Ql)/Q1'
22
The Nixon administrarion could have exercised severa I options instead

of - or in addition to - simpIy imposing price controIs. For instance, it
couId have required producers to continue to produce the quanrity Q*
or, for that matter, the quantity Qo. No quantity restrictions were im
posed, however. On the demand side, as the shortage became severe,
the administration could have estimated Ql (see Figure 2.2) and issued
coupons for that number of gallons. Had.coup.ons ben. issued, rights to
.. the pl,lrchase of gasotine. would- hav peen .llQeated. Property rights
wouId then have been secure andpeople would not haveneedd tospend
resources to acquire these rig~ts. COUPOIS, however, were not issued.
What property rights system prevailed under the price controI?
Earlier, in connection with Figure 2.I, I made the general statement
that when the maximum price for a commodity is set at Pc, the quantity
of it that will be available in the marketpIace is Q l ' Underlying this state
ment are two important assumptions about propert}t rights, one explicie.
and the other implcito The explie~t assumption is that under price con
trols the sellers' right to set prices is restricted; here sellers of gasolin~
were Iegally prohibited from selling it at any price above 35 cents per
galIon. The implidt assumption is that sellers retai the l'ight to provide
whatever quantity they wish. Given the sellers' marginal cost curve and
the controI price of 35 cents per gallon, the m.axi}I1izi~ .quarttity th~J.'
wouId have offered was QH as indicated in Figme 2.2.
.
Consumers wished to purchase Qo at the controI price, but this qU"ln
tity had no operationaI relevance: None of the forces under the controI
depended on this quantity. Q1 was ultimateIy the quantity that was allo
cated among the consumers. P 1 is the maximum price that consumers
would pay to purchase the (entire) quantity Q1> which in Figure 2.2 is 85
cents per gallon of gasotine. Ql' the quarttity offqed at Pc, was the quatl
rity that consumers wanted to purchase at the higher piice P1- In reality
rationing by waiting turned out to be th equilibradng force, giver that
the maximum price sellers had the right to charge was Iower i:han the
. onebuYerswere willing to pay.
.
Rationing of gasollne by waiting

Why did waiting lines for gasoline materialize in the fall of I973? Short
ages per se are not a cause for waiting. lndeed, scarcity does not imply
"shortage"; iRDecembec...T973;at55..C~'Flts . pel'.gaUon.the:re.woHldhave
been no shortage of gasoline. Th quaritity supplied would have declined
from Qo in May I97I to Q* in t)ecerlber I973, Q" being the quantity
l . . .. --.--...
... -- .... 23-.....
The public domain
Economic analysis ofproperty rights

demanded at that pcice. Gasoline, of course, was "scarce." A commodity
is scarce if something has to be sacrificed to obtain it. Since 50 cents per
gallon had to be sacrificed to obtain gasoline, it must have been scarce.
The appearance of a shortage arose because the price was controlled at a
levei below the market-deacing price. It was noted earlier that there are
many ways to resolve a shortage; waiting is just one of these. Waiting
lines did emerge late in I973. It is curious that the regulators never for
mallyadopted queuing s the method of allocating gasoline. It became
deai', .how.e:ver, that although it was subject to various exceptions and

pdedcop.trols, this wastheonly method of disteibution that the.r.~g9.I~.:...........
.~ )i" wsgoill8 tQ]allow...: ., ,;-,7
~.;:,-,,~
Under tle 'price control, the acrual pecuniary price per gallon of gaso
'line ws positive - 35 cents. Rationing by waiting was analyzed on the
assumption that the pecu,niary price is zero, but as long as the controlled
price of gasoline is held below the market-dearing price, the queuing
analysis essentially applies. Something valued by consumers and initially
.ow:.rred by sellers was IlQt charged for; thus gasoline was partially placed
in the public domain, and the queue serv-ed to establish rights over the
unowned component. Gasoline sellers owned the property rights to 35
cents per gallon of gasoline, and.b.Qy~r.s.c.ould acquire rights to thediffer:.,.
erice bet1'Veen P~ and Pc - which in Figure 2.;is 50 c:~n.ts 'perg~llQR-:-:-:PY.
joining the queue. If the wage rate of the marginal waiter had been $6
per hour (or IO cents.per minute), the market-clearing queue length
would hve beeI five'mil'l.utesjJergalln.H
Given that buxers acqu!red ga~oline by a combined expenditure of
money and time,

conventional demand curve is somewhat misspeci
fied. As it is usually formulated, it shows the maximum amount of money
people will pay for varying amounts of gasoline when no waiting is re
quired, but not how much they will pay in terms of a combination of
money and time. Ir jseasy to construct a modified demand curve in which
the price is stated in minutes per gallon, given that 35 cents per gallon
-:ci.
.. _.
'.
tire
11 An importam complicatien arises wth regard' tothe mechanics of the queue. It
makes a difference if gasoline is rationed by the gallon or by the tank. In mostcases
gasoline was rationed by the (capacity of the) tank. A person who drove a car with a
small tank could get less gasoline for each waiting episode than someone whose car
had a large tank. Since waiting time was independent of the size of the gas tank,
savings associated with purchase size became more pI'l"minent.
According to the present llodel, running out of gas was not the resuIt of the price
controI. By waiting in !ine and then paying 35 cents a gallon, one couId always get
gasoline. Independent of shortages, a person can save resources (time) by filling the
tank less otten, and people occasionally do run out of gas by postponing purchase too
... .long..Under.price.comrols,.the.expectations.a.e;.first;tharpeCipl'WiIIrunout- of-gas

more often and, second, that among automobiles running out of gas, relatively more
prevalent will be those with a small driving range (per full tank). For these cars, the
increase in the per-gallon cost of filling the tank is higher.
24
must also be ,Paid. Such a demand curve displays marginal valuation in

terms of time per unit over and above 35 cents per unit. This type of
demand curve depends not on1y on the valuation of gasoline but also
on the opportunity cost of time. Of two individuais who have identical
demands for gasoline but who differ in their opportunity cost of time,
the one with the low opportunity cost will outbid the other once waiting
becomes part of the price of gasoline. In the discussion that follows, de
mand is assumed to account for the two components of price..
To return to the main issue, the price control reassigned the rghts of
sellersan:Jjuyerst:gsOliie;::The:selIers:had"tlie::agnFto:i:ne::Value'of the .
. gasoline up'to theontrol piice,and theremainderwas 'pla~dinth'
public domain. Buyers could aequire the right to the remainder - the
difference between the control price and the buyers' marginal valuation
by joining a queue. By paying the control peice plus the time price, buyers
could obtain the property rights to a gallon of gasoline. Except for the
fact that byers had to pay a pair of prices, the market for gasoline may
be viewed as having functioned normally. Indeed, there are many unregu
lated markets in which both money and time prices are paid by the buyer.
A person who inssts on eating lunch at .noon in the cafeteria is charged
a money price by the cashier while facinga time price as well, namely,
.. :VV:li~iIlg tiIIlt":~. I.n:t4i.!! (;::ls.e.Jl1~ pe,C.t;lIlia:cy .pt:it:;~..i& .fi?<;~9 by the seller. In the
case of gasoline the money price was fixed by governrpent. 12
What are the regulators regulating?

The preceding analysis contains many implicit assumptions that connect
the waiting p'ri!=~'.!Y!th the control peice. In the next few sections some of
these assumptions will be altered in order to increase the correspondence
between the waiting model and the actual situation.
The approximate average price of gasoline in the United States in May
I 97 I was 35 eents per gaBon. In the analysis, a single control price of 35
cents per ganon was used. The gasoline prce control stipulated, however,
that the maximum price a station could charge was the actual price it
charged in May I97I. These p:-ices were subject tcnsiderable"Vd":
tion. Prices were lower at gas statons in oil-producing states, reflecting
lower transportaton costs. Prices were higher for premi um than for regu
lar gasoline. Prices were lower at gas stations that used low prices as
promotional devices than at gas stations that used other means of promo
tion. Prices wereJo)Ver at self-service stations. A self-service station seI!'
ing regular gasoline at 34 cents per gaIlon in the summer of 1973 would
12Deacon and Sonstelie (1985) expIoit a gasoline price-control episode in rhe afrer
math of the 1973'"'"4 era. They successfully test some of the hypotheses suggested here,
as well as some related ones.
25
The public domain
have had to seU regular gasoline at a maximwn price of 34 cents during
the price control, and a fuIl-service station selling regular at 38 cents
during this period would have had its maximum price fixed at 38 cents.
A price controI constitutes the assignrnent of property rghts; assuming
that the regulators could easily ascertain the actual base price and could
easily enforce it, delineation was dear in one important respect: Each
seller knew exactIy what price he or she could legally charge. 13. In other
essential respects, however, delineation was less dear.
What, exactly, is it that. Qne'prhas~s"in gas sl::ation? "Gasoline" is
:n~t a.suffident ans;w~r~ ~e buye~is. ~~i>~ct~l tOQP~.~ regar4I.1g;such
". thirigs aspuichase frequency aiid to get gasofneat the lciwest riet re
source cost for a given quality service. Like alI transacted commodities,
gasoline has a large number of valuable attribttes. For example, when is
it available? Is the gas.station open nine or twenty-four hours per day? Is
the octane rating 88 or 98? Is tbe station self- or full-service? In addition,
the costliness of the transaction ncludes travei and search time as well as
time in the station. It is essential to have"specific informatio~'bout t).le
regulation of the attributes of gasoline before its effects can be agequately
examined. Ambiguity surrounded the regulation of such attributes ullder.
price controls.
Much of the ambiguty in the scope of controls stems from the great
number and variability of attributes of gasoline. The attributes of gaso
line transactions can be classified into those-. of. the gasolitle" itself .aJ;ld
those of the services provided with the gasoline. One of tpe' attributes of
the gasoline itself is its octane rating. Gasoline is comiflQnly gr~ded as"
regular or premium, depending on that rating. 14 Hence "premum gaso
line" describes a range of products of octane 90 and above, rather than
strictly defining a single product. (I assume, more or less in conformity
with actual practice, that under the price control premium gasoline had
to have a minimum octane rating of 90.) There are other variations
among premium gasolines. For example, Exxon's premium gasoline had
performance additives different from Shell's, and the premum gasolne
".sold in the Rocky Mountains was refined differently from that sold at
sea-Ievellocations. The prce contraI essentially ignored most of the vari
ations in gasoline quality. Because it is prohibitively costly to delineate
rights to alI the valuable atuibutes of a commodity, it is not surprising
that the control specfications were not fully detailed. Correspohdingly,
it is expected that regulations also consistently fail to specifically stipulate

the level oi"certain attributes. Indeed, the real-world price controI speci
fied only the grade of gasoline and largely ignored other attributes, in
cluding the service provided.
In my initial analysis, in which gasoline was implicit1y considered as a
homogeneous commodity, it was seen that price contraIs effectively al
lowed the selIer to retain the right to 35 cents per galIon and allowed
buyers to capture the remaining vaIue (which had been placed in the
public domain) by joining a q1Jeu.e,Thi!!:cQnJu.siQntnust:be:re.examined::
:,jn ljghtqf :the opportunity'sellers had to adjust the quality of their gaso
. lne. Although it set pices according to May 1971 standards, the regula
tion did not require sellers to match the precise quality of gasoline they
had sold during this perodo With product quality as a variable, and with
a regulation that did not specify alI of the relevant attributes, property
rights had become extremely murky by late 1971.
When attributes subject to variability are incompletely specified, the
affected parties correspondingly have some leeway, each according to her
particular circumstances. To illustrate, consider two stations, A and B,
which were selling premium gasoline in the spring of 1971. Station A
sold 90-octane premiunl"for 39 cents per gallon, and station B soId 92
octane ptemium:for'4t 'etits;' Asstted; Wheiithe prkecrittl Was ifi- .
posed, the lowest octane level at which a gasoline was still considered
premium was 90. Station A was restricted to a maximum price"of 39
cents per gallon and to a minimum octane level of 90, thus requiring it
to maintain its octane at the old level. Station B was restrcted to a maxi
mum price of 41 cents per galIon but to a minmum octane level of 90.
Station B could thus lower its octane to below the pre-t"htrollevels while
continuing to sell the gasoline as premium and to charge 41 cents per
gallon for it. If gas stations had to pay refiners 1 cent per gallon for each
unit increase in octane levei, Station B was able to save 2 cents per gallon
of premium gasoline.
As.long as consumers were willing to pay more than 4I cents per gal
lon for premium gasoline (Le., P 1 >0.4I), they were willing to paythe
higher money price for B's gasoline, provided that the time price they had
to pay was correspondingly less than forA's gasoline. Since the time price
"was spent to acquire rights from the public domain and was not trans
ferred to anyone, there was no countervailing loss from the reduction in
waiting time when buying B's higher-price gasoline. The government's
specificaton of rights played into the hands of Statio:\" B. By adjusting
... gasoline.quality.without violating the regulation,it could capture some
of the value of the gasoline that seemed to end up in the public domain
as a result of the prce controI.
""
The preceding analysis is based on some major property rights
13 Actually,
it is difficult to ascrtain precisely the price used by the regulators.
th;;K~ii~~~~~~J~S.p~ii~r:::%rii:cliu~:7At~~~!J\h:*~~~~i~!ni~i~i~iif~~i:f~:
and Methods (ASTM). Brpce Pet~rson of the API reports that the standards are volun
tary, although there are some state.regulations, with varying degrees of enforcement.
No single octane rating is specified'to'distinguish regular from premium.
26
27

propositions. The 10gical derivation may be cocrect, but there is no a
priori reason to accept the modeI that generated the preceding conclu
sions as a good description of reality. This must be determined empiri
cally. Whereas no actual empirical investigation was conducted, the fol

10wing three tests are capable of refuting the preceding analysis. First,
gasoline quality in terms of octane leveis should have declined as sellers
attempted to capture the value that was placed in the public domain.
More specifically, the per-gallon amount of the refining additive tetra
ethyI-Iead used to boost the octane levei should have declined. Second,
:tlieq~ntitY:ofaniikiiockaadi#ves(i:e::~::siihsitiitesforoctane soId sepa

. Cai:ely fiom gasoline) should-haveiric:rsed:slbs~<1uentto,theiiirpositioli~' .
of the price controI. Third, w~ting time in the higher-prlce sttions
should have been lower than in the lower-price stations; moreover, by
ers in the higher-price stations should have placed a greater value on their
time than did buyers in the lower-price stations. The results of these tests
would be a good indication of the credibility of the hypotheses generated
by the model.
An analysis similar to that of the octane levei applies to other attributes
of the gasoline itself as welll,ts to the gas station services. No matter what
specifications the regulator stipulated, whether explicit or implcit, the
s.upplit!r:s.c~lllcf ,iI1~r.e,a!)e"th.~i,:: .Pf.9(i.t; ,Ryjust meeting them, provided that
before the controls were imposed thes~ specifications were exceeded:
Here, too, the reduction in quality of the 'product did not harm consum
ers, who simply spent fewer resources on acquiring the less-valued rights
that were placed in the public domain.
Turning to gas station services, the type and leveI of services attached
to the purchase of gasoline vary considerably from station to station. As
it applied to gasoline, the price control regulation specified nothing about
the leveI of services to be provided along with the gasoline. The response
in this case is expected to have been similar to that of the octane leveIs.
A simple scenario oftwo stations that differ only in terms of the levei of
services provided can isolate the effects of this lack of specification. Sta
tion C seUs regular gasoline for 33 cents per gaUon and provides few
extra services. 15 StBon D seUs the sameregulargasotineft,6cent!r-bin
provides 3 cents' worth of servces per gaBon in the form of pumping the
gas, cleaning the windshieId, and looking under the hood. Once the price
control is imposed, Station Gcancharge no more than 33 cents per gal
lon, while Station D can charge no more than 36 cents. Station D, like
Stati..or!, B in the previous example, has an additional margin.of .adjust
15 Convenient locations and smoothly functioning pumps are examples of services
even low-service gas stations stiI1 provided. Under competition in general, if a station
was selling gasoline at a price higher than what it had paid for it (including transporta
tion), some extra service must have been provided.
28
The public domain

ment not available to Station C. Station D can reduce its service levei to
zero, thereby saving 3 cents per gallon in costs, and still seU gasoline for
3 6 cents, which enables it to avoid losing some of its wealth to the public
domain. Consumers will buy all that station D can seU at 36 cents a
gallon without service so long as the cost of waiting at station C exceeds
3 cents per gaBon.
Gas station owners were able to alter still other margins of their prod
uct without violating the letter of the, regulation. One of these margins
was station hours. Station owne,rs could ,ch00se their hours of operation,
, th~reby'}owering costs 'withoQi: v:iolatiilg thC!.regulatio.n.SelliI:1g gasoline
" is Inorecostly in the middle.>f ilieriiglitthan dlriilg-business:hour~ be- ...
ca~se night workers mustbe pa!d a higher wage and security is more
problematic. Perhaps because complex pricing schemes are cost1y to op
erate, twenty-four-hour stations charge the same price at all hours. The
average cost of twenty-four-hour stations is higher than that of stations
open only during daydme hours; the single price charged by the former
must therefore be higher than that chrged by tne latter:The price con
trol required statio!ls not to exceed the old price but did not require them
to keep their old hours. Most stations that had been opn twenty-four
hours a day quickly shortened their hours of service. Such stationswel'e,
ths able to charge pdces higher thn those charged by others while in
curring the same costS. 16
The modeI here offers a predictiQn regarqinggas statim survival. The..
available supply of gasoline decl:ined duriiIg the era'of price contt6ls; and '
the number of stations seIling it also got smaller. Thse stations that.had
the greatest number of margins at which to adjust are expected to have .
been able to tolera te the price control situation longer than those with
fewer margins of adjustment. Because consumers were paying the same
fuU price (i.e., time plus money) for the same product no matter where
they made the purchase, they w~re unconcerned. abour. paying a higher
money price and waiting for less time 'at the first kind of station ~nd
doing the reverse at the seCond. Thus, seIf-service (or no-servic~) stations,
being the ones having fewer margins at which to adjust to price controIs,
are expected to have been among the first to go out of business. Stations
already sellng premium at the lowest p~ssible octane leveI before the
price controls are expected to have been similarly affected. These implica
tons are testable; although the data for the latter implcation may be
more difficult to' coJIect.
Thus far I have considC;!red only those margins of adjustment that were
open to .seller~. <?~nrde~ ..md,a,r.g;~~s... c::?:'~s.t~~.::t~,\1\'l1,~9:t.?r;tII,~l:lY~.r.s..,?r.,~>~~ .P.:':lY::.
'e'isandsellersco ''.1 Just 'in order to minimize dissipation. Resources
16Eventually most stations reduced their hours
to dispense their gasoline alIoctions.
29
to
the minimum number required
The public damain
Ecanamic analysis af praperty rights

spent in the queue were not received by others, and the existence of
queues indiCates that potential gains from sidestepping them existed. One
con1mOn way to circumvent price controls, and thus to lower the Iosses
therefrom, was to tie the sale of gasoline to the sale of another product
not subject to price controIs. Owners were able to use lubrication and
other gas station services to mask the true price of gasoline to the regula
tors, who did allow for such nes. A customer whose waiting cost for a
full tnk of gasoline was $5 was willing to pay up to $5 above the com
petiti.ve price .of l)1,btQo,when:itw.as:..offeredw.ith:a:full:tankof:gas:..
.and I1o'wi'!itiI1g:The seller who. providedsuch <ln off'ir was. able.to cp-:
-- tursomeof the vlue thai:: had formedy -ben dissipated by waiting. At
no previous time in history had automobiles been so welllubricated.
$}GaIIon
THE MINIMIZATION OF DISSIPATION
The anaIysis of adjustment by seIlers regarding customer service is usefuI

for bringing out an important point developed by Cheung. He recognized
that price controIs attenuate property r'ights and place some potentiaI
income in the public domain. Income is made non-exclusive, and. re
sources"iiCspent to capture it. An equalIy important point Cheung made
.. isthatthe m:x:imizt 'hypthesiii iiripliestht' thparties tif ilietrans:"
action will act so as to make the dissipation a constrained minimum. In
the case considered here, by adjusting their leveis of service downward
as the constraint of priee controIs became binding, sellers were able to
capture a value that would otherwise have been left in the public domain.
This action is a coriponent of the minimization of dissipation.
Consider the following example, illustNiied in Figure 2.3. So and Do
are the supply and demand curves for full-service regular gasoline before
the supply decrease and before the implementation of the price control,
and Pc, 35 cents per gallon, is the market price. That price ultimateIy
became the controI price. The supply then shifted leftward to SI> and Pc
beca me the binding controI price. Had sellers continued to offer full ser
vice, they would have supplied a quantity Q1 for which consumers were
willing to pay 85 cents per gallon. The difference between that price and
the controI price wouId have been dissipated in the form of time spent in
the queue. The total value of the queuing dissipaton is shown by the
verticaIly shaded rectangle, representing an amount of 50 cents per gallon
for Q1 gallons. This 50 centsper gallon was lost in the sense that the
customers' time expenditure was received by' no one. As indicated, the
........ "--. ..sellercould.capture some of this dissipated income by reducing gasoline
quality and services.
.....
Because gasoline continued to be sold by the gallon, the coordinates of
Figure 2.3 have the correct units for the changed product, while the supply
3
'0
Clt
Q2
Q.3Q"
Qo
Gallon!f/l)ay
Figure 203
and demand curves for the new quality must be redrawn. S2 is the new
supply of gasoline, the production of which required fewer resources per
gallon beca use of the elimination of services. Consumers' valuati~n of the
no-service gasoline was less than that of full-service gasoline; Dz is the
newdemand.The"eIiminated.services, however, were valued by consum
ers more than they cost to produce; this is why they were provided to be
gin with. The intersection between the no-service curves S2 and D 2 at
quantity Q3 must therefore be to the left of the intersection between Do
and S1 at quantity Q*. The falI in quantity from Q* to Qj is a reflection
of a cost of regulaton that the adjustments could not Ilminate. The dissi
pation peigallon was reduce.<i.(to 20 cents in the example), aridthe num
ber of gallons of gasoline (Q2) was larger.than.inthe..absence..Qfthe..adjust::_
ment (Q1)' The total dissipation after service reduction:, whlth- is shown
by the crosshatched area (combined with the appropria'tt-: ~'welfare trian
gles"), is Iess than the dissipation without the service reduion..

I stated earlier that before October 1:973 the discrepancy between the
market price 'and the control price was not '~too large"; the meaning of
the latter now becomes clear. Before October :1973 the adjustment in
gasoline quality was sufficient to yield an equilibrium price s low as 35
cents per gallon; thus, no waiting llnes emerged. Given the many margins
available for adjustment, maximization here meant that various compo
nents of service such as windshield cleaning were eliminated. Elimination
beganat the lowest net value service and went up as the discrepancy
between 1:he cont~ol an4 the clearipg price increased. Mer October 1973,
tlle price control.Constraint in' th~.gasoline market wasso severe that
whenseUei:s'l,1dtak~ndvantage,6f allthe:av::cla,ble:'adjustments,
the equilibrium price exceded3'j'ceptS ~galloh.ConsequentIy shortages
ensued and queues were required to ration the available quantity.
Contract choice: The tenancy contract
even
CONCLUSION
The foregoing analysis of rationing by waiting and of price controIs has

sho~n that because 'of 'the comple,pty of transactions, market partici
pants'can adfusl: to many margins besides quantity and price. Maximiza
tion implies that s.uch margins will be explbited, an<l.. tlt.e.p~~t:exn of that
exploitation is predictabIe: P~~ple will use the lowest-cost methods vail
able to them under the constraints to redaim th~ value that the regula
tions pIace in the public domain. As a result of such actions, dissipation
from the regulatioIp> is miniinized. Inthe.Cse of the gasoiine price con
trols of the I 970S; adjustments were made to provide the Iowest permit
ted octne leveIs, the shortest pssible"hours of operatin-for service sta
tions, and the very frequent lubrication of automobiles.
The analysis of a situation in which price is controlled by the govern
ment at a levei below (or, for that matter, above) the market price applies
quite generally to situations controlled J?u.rely by market forces as well.
The similarity lies in'the factthat, with the exception of organized markets
such as the'stock exchange, many market-controlled priees are kept con
stant in the face of constantly changing conditions~For instance, theptice
of coffee changes by the minute on the coffee futures exchange, and within
supermarkets demand changes by the hour of the day and by the day of
the week. The supermarket price, however, often stays unchanged for
weeks. Therefore, not only price but other factors as well must affectthe
allp.cation. Waiting is just one such factor. The main difference between
the analysis of the adjnstment to governrnent controlled prices and to un
changing market prices is the greater leeway sellers have over the critera
by.whic~to aUQCatetheircommodities':inthe' non"CorrtrUed'slta:tln:: ........ ; ......; .....
At the heart of the study of property rights Iies the study of contracts.
Contracts, whether formal or informal, reallocate rights among con
tracting parties. I will focus here on private contr.acts that are enforceable
by the courts and the police. The tenancy contract between tenant and
landlord - between the owner of labor and the owner of land - is rela
tively simple and is thus a suitable point from which to begin the study
Ofcritiacts.
..... Ona' family fa:nn;a:' !iliigleopettt ot a' smgl fafuly' :,:.thewne.r'of
labor - undertakes the bulk of farm activities. Family farming is common .
and relatively simply organized. By studying tenancy contracts in the con- .
text of family farming, it is possible to isolate some basic contracting
problems that may be obscured in more complex organizations. 1 As a
background t:the analysis of the tenancy contract I offer a criticaI review
of the traditional approach to the rellitfonship between tenant and land
lord.
THE SHARE CONTRACT AND'
CHEUNG'S CONTRIBUTION
Price theory textbooks routnely introduce the notion of a production.

function and discuss the marginal product of a factor such as labor for
given levels of such other factors as capital and land. Given the produc
tivity of the factors and the market prices of factors and products, it
is easy to determine both the optimum amounts and the values of the
contributions of each factor. The assumptions that factors are uniform
and that alI relevant information is frdy available usually underlie these
1 Allen and Lueck (:r992., :r993) have done extensve work on famly farm contracts
in the United States.
32
33
Conttact choice

dscussions. In such a setting, the problem of organizing production is
trivial.
.
These textbook assumptions are violated in reality. Problems arise be
cause efficient production requires that many owners of specialized, non
uniform productive inputs cooperate under conditions for which random
factors also affect output. In agriculture, weather, pests, and other forces
affect output differently during different periods and at different loca
tions. In addition, no two pieces of land are identical, npr ar~ al:1Y two
. .yvorkers. Dt':termining the properties Qf each' u~it onput rqmnis exten- . '.
'siye; and,os,tly measurement.;Owihg tpdiveisiryin the, foh::es tl1at~ffet:'
otput, lhe spedficontributionsof indvidtials areeXttniely d'iffitltt
determine. Compensating factor-owners according to the average contti
bution is not satisfactory here, because individuais can mask their oWn
low-Ievel contributions by attributing them to other forces. Such at
tempts to capture wealth make cooperariori among mdividuals costly.
Individuais will organize their transactions to maximize ihevalue of. their
output net of conventional input costs and of the wealth capt'Ure costs,
the latter being part of the costs of transacting. s I shaH explain, individ
uais can gain by organizing their transactions in such a way as to lower
these costs.
The inefficiency associated with the share conttact is a particular mani
festation of the wealth-capture problem. Having been recognized long
ago, it has received a great deal of attentionfrom economists. In a share
contract, a landlord lets a "tenant work the land in exchange for a share
of the output. The tenant's payment to the landlotd may be likened to an .
ad valorem tax, and this analogy has been used to suggest that the share
contract 1S inefficient.
Let me first consider briefly the ad valorem taxo The demand facing
sellers of a taxed commodity is lower than the consumers' demand by
the amount of the taxo Because of the shift in demand, the market equilib
rium quantity under the tax is less than it is in theabsence of the taxo
The tax, then, distorts resource allocation: Under the tax, the marginal
",uoit.isva.luedatmorethan it costs; expanding production would produce
a net gain. The tax, however, creates a wedge between the incentives of
consumers and those of producers, thus preventing the realization of that
gain.
Economists have argued that this tax analysis applies directly to the
share conttact. I shall use Figure 3.1, whih. I adapted from Cheung
(::969~P. 43), to p,r,c;:sent this argumento In Figure '3:1 the value of the
tenant's marginal product ,on..a ..p lot.ofa.givensi<::e,.is M'l'l:.and ..his.orher......
market wage is w: Were the tenant self-employed~ 'sh' or 'he would apply
L" units of time to the farm. The landlord, how.ever"receives a share, r,
of the output. The tenant whose share of the ourput is (i - r) retains only
Produt
labor
~ffffff/,/h",
LT
"LO
.T.enan.t lab.or.
Figure 3.I
(I -r)' ~f his own marginal product and will apply LT units of time to the
fartn to mximi,ze his wealth. For units of labor between LT and L" , the
output value of the tenant exceeds the wage rate; hOWl:!ver, the tenant
will prefer to sll these units of labor service in the.market because her
share in the farm output is less than W per unit of labor. Such tenants,
then, will stop short of producing the output when the value of their
marginal product equals their alternative earning. The shaded area in
Figure 3.1: i8 the alleged inefficiency induced by the share contract, which
is likened to the tax. distortion.
As Cheung demonstrates, this analysis is seriously flawed. Given the
tenant's share of I r, L T may appear. as the tenants' optimal leveI of
labor input. However, if the tenant is frt;e to apply as much labor as he
wishes, he can increase his income by becoming a share-tenant on two
frros and spending half of L T units of time in each. Indeed, the tenant
will do best by working a tiny bit on each of several farros. Even on only
one facm, the tenant earns on average more than W per unit of time.
...ConsequentlY')c"x..=.r.exceeds.:the.tenanes...eqqiH!?rium share. Tenants may'
compete simply by lowering the share they are willing to accept, but then
the equilibrium labor input wll falI as well. The logical conclusion of the
process would entail share-tenants earning no rent, while their input
35
Contract choice
would plummet to zero. This model, then, suffers from an internaI incon
sistency.
Before I turn to Cheung's solution, it should be noted that the tax
analogycontains two implicit- and by no means innocent- assumptions.
The first is that landlords find it prohibitively expensve to stipulate and
police the amount of labor input. The second is that landlords encounter
no cost in policing the receipt of their share of the output. Whereas either
assumption may be a good approximation of real circumstances in any
particular case, the two are unlikeIy to hold true simultaneously. Simi-.
of such obseryable variables as wealth. Even then, the explanation of
sharecropping is still insufficient. In .a zero transaction cost world, share
cropping alone is not an attractive method of sharing risk. Share con
tracting does distribute the crop of a single plot between the two parties,
but there are other ways to distribute variability that can remove more
of the risk. For instance, because two persons residing on opposite sides
of the globe are not subject to common random forces, pooling the risks
between them reduces the risk each would face alone. Under the assump
tion of costless transactions, such popling woull inv()lve nq' added costs
and thus would be practiced.,In reality, ,oI' course"trans'acting is cosay,
:render1ngspme risk~pooling diffiultto effet.'W1ien trnsacting is costly, ':

however, allcontract forms are costly. Shadng iny be selected in i:hat
case flOt only for its effect on risk bui also beca use of some properties of
transaction costs. Moreover, when transactifig is costly, other contracts
may be chosen in spite of their riskiness. Later in this chapter I will offer
a transaction cost hypothesis to explain sQadng; in subsequent chapters
I will provide transaction cost hypotheses to explain oth~r {rms of orga
nization, assuming throughout that peOi?le are ~is~-neutral.2
1~J,"~Yfimplidtiii'thetiaditi6iiaFappioadi::to::jhe:~shire:'contract::ir;i'ihe'
assumptions that the cost of monitoring ouwut is" ruwaYI> zero' and thar
th cost of monitoring labor inpuf isalways prohibitive: these assump':
tions are ad hoc. Models based on ad hoc assumptions are likely to be
refuted when put to an empirical testo
.
In his ground-breaking work on the theory of share tenancy, which he
tested empirically against observed practices in China, Cheung (1969)
makes several points that are pertinent to the present discussion. First,
he notes that sharecropping is unlikely to hve been inefficient, given its
long history of survival. Second, consistent with rhe Coase Theorem, in
a world of zero transactions costs the share contract will yield an efficient
, ()u~c()llle>. .a:s, \!Vll. ()~h,e.r. (;()llt.r!lc:t .f()r:IfiS,-. "~C:!llls(! S9M.r~;tQr~q,n costlessly
add and police contract stipulations in such a way that alI inefpciency is
elimina1:ed. Third, in attempting to add emprical content to the Coase
Theorem, Cheung spells our some of ihe stipulations that were itecessary
for efficiency to have been attained in sharecropping. He finds that prac
tic~:ijn China were consistent with the implications he derives involving
fixed plot size, required leveIs of other inputs, and restrictions orr allow
able crops. Since the impiementation of these conditions consumes re
sources, the question remains as to why share tenancy was nevertheless
utilized. Cheung argues that risk aversion may explain the prevalence of
share tenancy because under share tenancy the landlord and the tenant
share the vagaries of variations in output and in output value. Whereas
Cheung's critique of the received analysis, which claims that the share
contract distorts the allocation of resources, is well taken~' hisrisk~
aversion explanation is noto
Risk aversion does not satisfactorily explain the share contract. Atti
tude toward risk is a matter of taste. If, as we usual1y 'assume, taste is
a personal matter that may vary unpredictably among individuaIs, an
explanation based on ris!c..aversion is not refutable. Moreover, how
someone may act with regard to a risky prospect depends on his or her
other actions and how risky they are - something we usual1y do not
know. Under uniformity..of taste toward risk, risk aversion is a function
36
THREE METHODS OF COLLABO'RATION BE'TWEE,N

OWNERS OF LAND AND OWNERS OF LABOR
Given the total amounts of land and labor, there is some pIot -size or,
more generally, some size distribution of plots that maximizes total out
puto Only a fraction of workers, however, own the commensurate"
amount of land. Those who own more land than they alone can most
profitably cultivate can gain by cooperating with those whose holdings
are toa small. In order to realize the gains, the factor owners must con
tract with each other. I will here consider three methods by which two
owners can collaborate and in which ownership patterns are ,preserved:
the wage contract, (2) the rental contract;and (3) the share contract..
A discussion of the consolidation of ownership wilJ follow.
. Itw.aspQinted.out earlier that individual specimens of neither land nor
labor are uniform. Before analyzing the general case in which both fac
tors are variabIe, I will consider the special cases in which either.the land
or the labor is uniform. Suppose that .1and .is entire1y uniform and un
2 When the sole contractual problem between tenant and landlord is the method of
. dividing output, the role of risk aversion seemS ooth .simple and important. As a rui e,
however, many. sources of variabiliry confront the parties. The effect of crop-sharing
on overall riskiness th~ll.~.Qm.,$..J!:~s."l.r..;nld.,l~lilt imp!tant,.F.QI;..in~tam;~,. t.,iS..lJ.QL .....
at ali <;lear how transferring the maintenance of iriigation: ditches from the landlord
to the tenant affects the distribution oI risk. .
37

changeable. If collaboration is by a wage contract, workers can gain by
shirking, exerting themselves less than they would were they self
employed. Output is subject to variability beca use of random factors af
fecting output directly and because workers' contributions' to output
vary. It is, therefore, difficult to differentiate the effect of change in effort
from random factors. Under the given conditions, the wage contract re
quires supervisiono Since supervision is costly, economy in its use will be
exercised. Workers are expected to take advantage of the opportunity to
shirk. They will not work as hard as wouId..st':If:-t?~pl'?ye.4.:~'?!lc~J:~::A:s::::t.:.
, ,attractive as the opportunitytoshirkmayseem to a wage worker, he or
"sh':isbounftti lose as a result ofit. Wage payment wiIl be adjusted to
the expected reduction in effort; workers are paid, on average, for what
work they accomplish. Such workers would prefer to work harder
and be paid more because they operate at a point at which the cost to
them of an extra unit of effort is less than the corresponding value of
the increase in output brought about by that effort; however, the
cost of effecting such an arrangement exceeds the gain it would gener
ate. The allocation of resources under the wage contract is not optimaI
here.
.. TenantswhO operate undersharecontracts retain a portion - but not
alI - of their marginalpEoducr.Although theincerrtive to shirkisnotas .
strong here as it is in the wage contract, it is induced by the same factors,
~l)d the preceding discussion of the wage contract applies.
. "fenants who colIaborate with landowners by renting their land pay a
-fixed amount for its use. ActuaI output will differ fcom expected output
beca use of random fluctuations and to tlle't!xtent that the tenants alter
their own efforts. Because land is uniform, it does not contribute'to -out
put varability; the tenants' expected output varies only as a function of
their own effort. The tenant is the "residual daimant": Barring bank
ruptcy, the landlord receives a fixed amount, while the tenant receives
. whatever is left over after paying the rent (of course, this diffe.rence may
be a negative amount). Apart fcom the random element, the tenants'
rewards are strictly a function of their own efforts. In the case of uni
form land, tenants will make the optimal effort under the fixed rent
contract.
The analysis of the situation where land is heterogeneous and labor
(and labor effort) is uniform is the mirror image of the one just presented.
When labor is uniform, wage contracts make expected output solely a
function of the quality of the land. Landlords have the appropriateim:en
tiye t.. t:ll:ima~9!i:lI1<i.imPJ:'Y:etheir Iand and will not gain from misrepre
"senng its quality. In this casethe landlords are the residual daimants
and are the only ones affected by their own actions, A rentaI (r a: share)
contract would be inefficient here, producing less total income for the
38
Contract choice
contracting parties.3 The more efficient contract, that of the fixed wage,
would prevail under such conditions.
In reality, neither land nor labor is uniform. It follows that ali three
contract forms are subject to efficiency problems. When discrepancies
between costs and valuations are inevitable, an arrangement in which
such discrepancies are observed cannot be thought of as inefficient. Inef
ficiency implies preventable waste - waste that would not ocenr if people
were to maximize. In an imperfect world, even the best solution is still
:.:s.ubjectto:discr:epancies:.between.marginal costs and marginal valuations;
flOt ali suchdiscrepanciescan be eliminated economi~ally.When we dis-'
cover discrepancies, elur task is to detrmine how resource allocation will
change when the costs of or the returns from reducing the discrepancies
change. In order to be able to analyze changes in resource allocation, a
more detaited analysis of the nature of variability is necessary.
THE EFFECTS OF VARIABILITY WITHIN FACTORS
Because it is commonly, though onIy implicitly, assumed that land is uni

focI11 and unchangeable, it is instructive to examine the nature of land
and the effects of variability in it. Each acre of land differs from ali oth
eis -' everdrom the ones adjacent to it - in a variety of ways: incidence
of rocks; steepness; degree of soil erosion; presence of various nutrients;
levei of mosture; and exposure to wind and sim. Land parcels ais o differ
in terms of access to groundwater, quality and quantity of rtigation ca
nais, avalabilty of pumping equipment, types of roads servng them, and
dstance to markets. Moreover, the ease of exploting such features alio
varies.
"".Land use would be efficient if landlords were compensated by users of
land for the exact reduction in land value. Because land s not uniform,
however, the exact evaluaton of these effects requires measurement at
every spot. The fact that exact and comprehensive measurements are pro
hibitively costly leads to measurements that are neither exact nor com
prehensive. Indeed;certainfeaturesmaynotbemeasured at alI. Con
tracts explicitly delineate some specific attributes of the transaction. They
also implicitly delineate others that are governed by common Iaw. They
3The i~ct ~hat the fixed-rent contract has not been c1aimed to be inefficent suggests
thar land (inclusive of the improvement and equipment that accompanied it) has been
viewed as unchangeable, On the other .hand, .economists have been quick to criticze
Henry George's single-tax proposal, poinring out thar since his assumption that land
was unchangeable was too unrealistic, his polcy conclusions were rendered.use1ess........... .
Had economists been consistent in recognizing in ali their applicatons that land is
changeable, it is less likely that the share contract would have been singled out as the
only inefficient tenure contract. Either the fixed-rent contract would ais o have been
considered inefficient or both contracts would have been recognized as efficent.
Contract choice

do not delineate ali of them. Because attributes that are prohibitively
costly to speciy are part of the transaction and will be exploited. the
contracts between the owners of productive services will not award each
his or her precise contribution. 4 Among the unspecified attributes, some
are subject to control by the buyer and some by the selIer. By "control"
I mean one's freedom to manipnlate the particnlar unspecified attribute
without making marginal payments to others.
Contracting is viewed here as a strictly symmetrical operation: The
parties contract with each other !;>ecauebotli own .productive servces
that stand to increase in tot~1 va1ue a~a".resillt of thecOQperationbetween
them:Each. can contt;ibute: to: 'i:he coilabortidn\ las.~~inethat'c6ritrct
terms are determined by competition and. therefQre, that those contracts
that maximize the value'of each transaction net of ali the associated costs
will prevail. Thus, in spite of the marginal inequalities - the violation of
the Pareto conditions - no true inefficiencies can be present.
$/acre
~+Lr.r
I " .. .. 'il;,;.:j)i'iicre'iiiiOWifl.,:'::::,:.::
..
.."NQ."."l'k"
. exttaCted nutrients.
,. A: percacre cOsI of and demand for (~aluation of) nutrieni
$/acre
fi)
THE FIXED-RENT CONTRACT
In this section, I will analyze the' n~ture of the deadweight losss that
arise with a fixed-rent contract as a result of not pricing a~.tributes at t1}e ,
margin, ana in the following sction I will indicate methods available to
contain these deadweight losses. Suppose a tenant rents a plot of land on
a fixed-rent basis. I ass!lme that soil nutriel).ts ar~ pJaced under t:he ten
ant's control, permitting the extractiOl}.:of whatver amount li or she' .
wishes without paying a m~rginal charge, and that .the mainenanc~ of
land improvements is left under the landowner's control and 'not prced
on the margin either. The three parts of Figure }.2, depicting different
aspects of the land rental, will be used to analyze the nature of the equi
librium and the forces that lead to it; parts A and B reflect behavi9r
within the transaction, and part C relates to market behayior."
The horizontal axis in part A is the per-acre mount of nutrients ex
tracted from the soil and the vertical axis is dollars per aCre. LN is ihe
marginal falI in land value du'e to the depletion of the nut:ients. Alth6tigh'
tothe tenant the nutrient is free, its use is constrained by 'lhe cost of
extraction. CN is the marginal cost of extracting the nutrients - arising,
for instance, from the need to use water to transform the nutrient into a
form that plants can utilize~ C~ + LN is the sum of the two COSts. DN is
the demand for the nutrients reflecting the (marginal) increase in the
>.i;<
:>
iA
= index of quantty of
maintenance per acre
B: per-acre COSI of and demand for (valuation of) maintenance
$/acre
7+T{
D
I
OR
L '" land in acres
Qo
C: market for land under fixed rent contracts

Figure 3.2.
.. 4Holmstrom and Milgrom (I99I) have aI 50 studied complex (multita5k) relation

ships. Using,the"prindpal-agenE..~del, [heyJocus '~n'the ,{risk-averset'agent: 'perfOf'-'"
mance of the interrelated tasks. Tney do n'ot consider the potential for moral hazard
by the "principal," as I do. Bl;waraflo and Kotwal (I98S) present a double moral haz
ard modeI.
.
41
Contract choice
value of output as the amount of extracted nutrients increases. A self

employed hindowner will extract No, the quantity at which DN and
CN + LN intersect. Since the land rent contract does not charge for nutrient use, the tenant does not pay, on the mugin, for the depletion of the
nutrient. He or she will extract NR' the quantity at whch DN ntersects
C N , the latter being the cost borne by the tenant. 5 It is as if the tenant
receives an implicit per-acre subsidy at the rate of Uo that is equal to the
heighr of L N , the marginal &11 in land value, at NR' Since the tenant wiU
".dc.p.lete,tbe nutrient at the rate of NR' whereas the self-employed wo.r~er:

'wiUdepleteifit'th,rate ofNo,the per7aq:eloss dueto the,excessive use ~
ofthenutrierit by'a renter isth sladed area U:?:'l'C: '~~,
'",,"
. ,','
Although the tenant contr9ls the use of the nutrient and .is able to
extract an "excessive" amount, such as NR' per acre, that control is ulti
mately at his own expense. A tenant who could somehow cOl1l1llit to use
only No soil nutrients per acre would produce a lower output. However, ,
he would have to pay a lower fixed rent such that his net income could
increase by an amount up to the area U. As was pointed out earlier, this
is a general phenomenon. On average, those able to shirk, to chet, or to:
enjoy "free" perks must, under competition, pay for the privilege by an
amount that exceeds in expected terms the valu~ of the privilege. The
privilege is granted only because, it is too costly to eliminate.
. Part B of Figure 3.2 shows the per-acre levei of maintenance and im

provements, assuined to be under the landowner's control..The hori
zontal axis is (an index of) the per-acre levei of maintenance. CM is that
part of the cost of the landowner's maintenance activity that affects the
contract-period crop, and DM is the demand for maintenance within the
contract period. Assuming that there are no cross effects between mainte
nance and nutrients, a self-employed landowner will operate at the inter
section of DM and CM, providing maintenance at a rate of Mo. When the
land is rented out and maintenance is not stipulated in the contract, the
landlord, who does not gain from the contribution of maintenance to
current output, will supply the amount of M R.6 It is as if the lanllord
pays a 100 percent tax on the improvement's contribution, which here
a:mountsi::otoperacre.Tiiep~~acre ossfrom maintenance at leveI M R
rather than Mo is the shaded area T.
Part C shows the market for land under fixed-rent contracts. We.re
marginal discrepancies absent, D and S would be the relevant demand

and supply ctrves, and Qo would be the equilbrium quantity. As con
tracts for rented land fail to price some attributes, account must be taken
of the resulting discrepandes, which are part of the costs of transacting.
To simplify the exposition, I assume that the discrepancies occur only in
nutrients and maintenance. The quantity U + T is the combined per-acre
loss to renters and landlords; it arises from their failure to stipulate and
police the use of nutrients and the level of maintenance. Assuming that
.. ' the average and the marginallosses due to the discrepancies are equal,
QR' the quantity f()r which theheight ofthe demand curve exceeds that .
of' th supply curv byu+T,:is;;th'equilbrium anlountof rented:fand::
U + T is the loss on the marginal acre. The total loss on the land being
rented is (U + T)QR' In addition, there is a loss due to too little land
7
rental, shown as the shaded triangular area W in panel c. ,8
. .
METH'ODS FOR RESTRAINING LOSSES
."
Glven measurement costs, the tenant wlll use such nutnents to the pomt at which
the estimated rather than the actual net gan from an extra unir is zero. For simplcity
of exposition, such inccures are ignored whenever the consequences are not ger.
mane to the argumento

""
"
6 M can be greater than zero because current m'iiitnanc'i'Whicn"eiiliinOOi:"tlie: "":"'"'' ....

R
post-contract value of improvements', may also benefit current production. f:;.landlord
will luhricate his water pump during the current period if ir is the most economic,aJ '
way to enhance his income during future periods.
'
Labor and land are complex factors, each possessing many attributes,
but contracts betWeen pairs of owners are, usualIy quite simple. How,
then, are th~ individual ttdb'ies 'conrrolled by contract, and wniiffrces
determine which contr,,\ct will maximize the va:Iue' oI' the" rstces?' A " ."
~eq-rent land co~tract can simply stipulate duration and rent; alterna9
tivly~ it cal b-s detailed as the;. contracting parties wish it to be.
Wher~as coq~ractors are free to stipulate whatever they wish, not alI at
tributes are worth stipulating nd.monitoring. Any attribute that is not
stipulated and that can be varied becomes a free attribute. Tenants who
are in control of such an attribute will use extra units so long as they
generate added positive (net) income; landlords will similarly use attri
butes under their control.
Although by~ssumption he loss associated with free attributes is too
costly to avrt directly, it can'be controlled in two distinct ways. First,
contract 'stipulations regarding attributes related to the ones subject to
excessive exploitation or inadequate provision may be altered. Second,
an altogether different contract may be used, one that controls those at
tributes left uncontrolled under the first contract.
h
' . W + (U + T)Q .s the cost of transaction as I define it in
7In thls case, t e quannty
RI
Barze119 8 5 See also AlIen 199 1 .
..
11 d b the tenant
sPanel C incorporates only tw? ~mspeclfied attrl~utesl.oI:t con~ro t e tt 'butes how
and one by the landlord. Generahzmg for any num er Olnepen en a ri
,
'"''''ever;'isstraightfiW:td~'''''''''''''''"'''''''''''''''''''''''""'"'''''''''' "''''''''':'
, :d' Such a
9In either case, a mech~nism to enforce contract performanc~ ISbrelulr: 'rts The
mechanism is usually provl~ed P3:rt1y by the contractors. and part y y e c u .
existence of such a mechamsm wIll here be taken as a glven.
43
;. {~n~'~ ;:!'~~~5~'~~~~':l~t '.:' !!.~ ~~.!-;~~-,~.;; :;;{~<:~~.:: ':';;" ~;: ii~:~,~ ;-~~;J~ ;., {o: ~n~:::, >.:.: ;,':. ~~ (;,; ,
'.;;':';0'
.,.,.)~
';".'.':.,.H.~_ /-;::~~::1.~i>.

Manipulating related ttributes
In the conventional land-rent model it is implicitly assumed that land is
unchangeable. This implies that the supply elasticities of aI[ its unspeci
fied attributes are zero; under these conditions the fixed rent contract is
efficient (given risk-neutrality, which I assume). However, land can be
altered. Cheung (1:969) points out that when the cost of transacting is
zero, efficiency can always be attained because the transactors can then
costlessly fix the levei of any factor they supply. By fixing alI attributes at
the desired levels".the"Qve.J::::"Qcunder:::utilization::of:hy::atttibtite'-will"oe::,::::::'
avoided;"TwiU'here{aborate on thetheme of operating On the leveIs of,
landattdbutes but will drop th cosdess transacting assumption. "'"
Transactors can gain by constraining their actions so that they reduce
utilization where it would otherwise be excessive and increase provision
where it would otherwise be inadequate. I here concentrate on reducing
theuse of soilnutrients, an attribute" that is free to the tenant. Consider
n attribute that 1S complementary to the soil nutrient but was not ini
tially made part of the exchange contract. Suppose that a self-employed
individual would have used the complementary attribute at a per-acre
rate of W o, and that the landowner can, at a low cost, fix the level of.the
comphiri'ieiliary attribute. Fixing the levei of the complementary attri~)t~t(!., ", ..
at "W' wili" relider the' cosi f siiigdi .fre 'tiiibute .ii:1'd 'i:lieequlib
dum point less elastic,lO. thereby lowering the associated distortion. For
example, suppose that w,.ater-use is the attribute complementary to the
extraction of a soU nutrient. The landlord can reduce the tenant's cost
elasticity for the extraction of the soil nutrient and the associated distor
tion by supplying up"'io W o water per acre.
The effect of fixing the amount of wati"i: W o is shown in Figure 3.3,
which elaborates on part A of Figure 3.2. Each point on C N is arrived at
by using the combination of inputs that minimizes the cost of extracting
the corresponding amount of the nutrient. The amount of water used in
the process of cost minimization of extracting the amount No is W o
When W is fixed at up to W o, the cost of extracting the nutrient shifts
from C N - which is derived under the assumption that the tenant can
adjust the amount of water purchased - to CNIWo, where W is fixed at
W o The latter is less elastic at No and to the right of it. To the left of No,
however, the constraint is not binding, and in that region CNIWo coin-'
cides with
When W 1S fixed at W o, the use of nutrients falls from NR
.....
lOThe general proposition regarding rhe demand" ('spply) curve is thar around an
equilibriuI1l position, fixing the levei of either-a complementary or a substitute good,
c: .. ..theowrrdemand (supply) elasticity will be less than it is when the quantity of the
related good is allowed to adjust (see Slberberg:J:,990 ).
44
Contract choice
$/acre
J .!
.-'
" ' , N=pe!,~ac'r~'amaunl af nulri~llts

Figure 3.3
. to N l . The totalloss is reduced from U (Figure 3.2, part A) to ulWo (a

triangle in Figure 3.3) .
Although W o is the amount of water used by a self-emplQyed individ
ual, constraining W to W o does not quite minimize the distortion. It will.
be minimized when the amount of water is restricted..to an ilIl!0unt
(somewhat) less than W o- As the amount used falls below W o, a new
distortion is introduced in the sense that water use becomes too small."
But as the quantity of soil nutrient extracted falls with the :duction in
the amount of water, the associated loss ais o falls initial1y at a rate faster
than the increase in the new one induced by the insufficient use of water
(the "envelope" theorem). Some value of W - W1> sU:ch that W 1 <WQ: is the optimallevel of water needed.
.
The landlord may operate on additional inputs. He may impose alter
native or additional restrictions, such as stipulating that the tenant must
supply minimal amounts of factors that are substitutes for the (free) nu
trient, thereby further lowering the demand elasticity (and the demand)
for the nutrient. A similar argument applies to attributes controlled by
the landlord. Consider a factor complementary to maintenal.)ce. If tIle
tenant agrees to supply an amount of that complementary factor equal
to (or, preferably, somewhat greater than) the amount a self-employed
tciii:1fWi:iuldhavepivided;thehmdlord's supply of maintenance will
increase toward its optimal amount, thus reducing the loss associated
with that factor. For example, the landlord will provide more pumping
equipment if the tenant agrees to lubricate it.
.
Free attribute losses can be also reduced by manipulating the prices of
commoditiesrelated to the free attributes. For instance,if the price of a
commodity that is a substitute for a free attribute is reduced, th lemand
for the free attribute will decline, as'wilIHthHiiss'Ciated:1oss~A'laridlord'
who wishes to reduce' the tenant's use of an unpriced soil nutrient and
45
'.
Contract choice
finds it too costly to police the use of a complementary fertilizer may

subsidize the tenant's purchase of the fertilizer. In order to further reduce
transaction costs, some commodities that are substitutes for unpeiced at
tributes may be supplied at no marginal charge in order to reduce the use
of these attributes. 11
Duration is another contract feature that affects the parties' behavior
with regard to attributes. A wheatland owner may find a one-year rental
contract to be satisfactory. On the other hand, in the case of an orchard,
the care and maintenance of the trees, which can be most e(fiiep.tJypr.9.:::
videdbyw.e tenant, become free attributes to aone-yeartenant, whereas
lnger-teriri rentl' contract enhances the ,. tenant's incentive for care
and maintenance, thus reducing the tenant's exploitation of these attei
butes.
.
Consider a tenancy contract of one year's duration. 12 Work animais or
tractors supplied by the landlord are most likely to be severeIy depreci
ated by year end if the tenant is in charge of their upkeep. The tenant's
demnd elasticity for the services tbat result in wear and tear is high, and
the associated 10ss is large. If, however, the tenant has to supply such
iteIf!.s; no shirking in maintenance is expected. Generally, the tenant here
gains much"morefrmnaffeo:;:tiiig' the value of thecaPital more than the
landlord does. Because he or she,ratherthan thelandowner;owns thi~r
input, the tenant bears the bulk of the reduction in value its use :J.duces.
" H~ }?); she w!ll t!len. uSe this capital careful1y, which is likely to more than
compensate' for the greater difficulty of raising capitaL
.. ' A similar argument applies to land improvements. These may seem to
be an integral part ofthe land and thus a "responsibility" of tlie"landlord.
The less long Iasting their effects, however, the less inclinl!d the landlord
will be to perform these functions while the contract is in force. The
probIem becomes less acute if shorter-term improvements such as fertiliz
ingand weed controI are assigned to or simply assumed by the tenant. It
is expecte~ that the contract wll assign the shorter-term improvements
to the tenant and the Ionge.r~term. ones to the landlord; the longer the
duration o i:he contract, the further the dividing line is expected to move
toward the longer-term improvements. 13 Since the effects of the improve
ments are s~ldom conflned to a particular perod, each may be required

to provide a minimal amount.
The duration of the contract, which is endogenously determined, often
plays a major part in better aligning the parties' incentives. The duration
of the contract is understood to be the length of the period in which
the parties cede control of some attributes to each other. Changes in the
duration of the contract affect the parties' incentives to exploit free attei
butes. This can easily be seen by comparing a tenancy contract that cov
'er::s:t.he:.planting.bu.t:.not.:J:be.:ba:ry~ng:~esQil.wjtb.a.contract that covers
both. Given.the shorter contnlcLduration,thetenantis notexpected to
do any planting, since planting is a "land improvement": -the landlord
will receive the entire output and the tenant will pay a IOO percent taxo
The tenant's incentive for planting is restored by extending the contract
peeiod to cover the harvest season. The longer the span between planting
and harvesting, the longer the expected duration of tenure. 14
The task of caring for the trees need not be assigned to the tenant,
just as equipment maintenance is not necessarily a tenant's task. Such
assignments are a matter of choice: The contractors are expected to as
sign the provision of particular attributes to whichever party is more suit
able. The existence of such a choice points to another aspect of the loss
.. 'rstt:iriiigptoblm;IOria:ysiri that in the land-rent contract it is dear
which inputs each contractor fumishes: The landowner supplies the land
and the tenant the labor. The recognition, however, that any commodity
is a collection of attributes suggests that the real situation is more com
plex. Improvements, for instance, are obviously not an integral part of
the land, and the contractors are free to decide which party will take
t:harge of any of them. 15 More generally, the contrad:ors are free to de
cide which party should furnish particular attributes.
With regard to each attribute, one might ask: Which of the two parties
will be more indined to affect the (net) value of output by manipulating
that attribute? The principIe I applied earlier to labor and land applies to
individual attributes as well. If the party that will be more indined to
affect the outcome by varying the.level.ofanattri~uteisput in.o:;:ontrolof.
11 The success oi a price subsidy depends on the presence of a particular transaction

cost - th""t of reselling the subsidized commodity. That cost must be high enough so
that there is no gain irom reselling the commodity.
12Recall the assumption that the relationship between tenant and landlord is gov
erned only by the contract; the expectation of renewal is ignored.
13 A landlord may undertake an impro>ement before signing the contract even if it
isoptimllhi.nrtidefti:Kef ii themidlCi'ofi:llecontractpl:iod;since he is not expected
to undertake it after the contract is signed. The rent he can charge, however, depends
in part on the improvements he provides. Allen and Lueck (I992) present evidence on
this point.
46
14The incentive to exploit free attributes ntensifies toward the end of the contract
perodo The more acute this problem, the more likely it is that the parties will sign a
new contract before the olper one.expires. In professional sports,such phasing in of
contracts is common both fo.r athletes and for coaches. The motivation is to prevent
underperformance by athletes as the expiration of their own contracts approaches,
_ .~hile avoiding the overuse of athletes by coaches whose contracts are .about to expire.
15If landlo.rds are legally required to. keep certain improvements present on the plo.t
at the time the contract is signed, there is no need to spell this out in the contract.
Similarly, if a landlord plans to maintain the improvements, she or he will not require
such maintenance from the tenant. So.me contracts may appear to be lacking in detail
but may, nevertheless, diifer significantly from similarly worded tenancy contracts in
areas in which the particular improvements are simply absent.
47
Economic analysis of properly rights
Contract choice
that attribute, thereby becoming the residual claimant of its variability,

Josses will bei minimized. If, for example, land is rented out on an annual
basis, the maintenance of long-lasting improvements wiU tend to be
placed in the landlord's charge, since landlords are the chief beneficiaries
of proper maintenance beca use of the higher rents they will be able to
charge in future periods.
imize the net yalue of the resources is expected to be adopted. Conditions
can be specified, however, under which a shift from one contract form to
the other is expected to occur. For example, the more valuable the land
attributes that are difficult to measure and to police, the more likely it is
that the wage contract will be adopted.
Any contract is subject to problems of non-oprimal use; therefore, no

single contract is best under alI circumstances. Changes in the circum
stances that affect contract choice may be gradual, but the change in
contract form cannot be; either it takes place or it does If9~,." Th~,s~"!l,,,.
" ,c0II!:l.'.arison .between .contract forms, m1Jst"he... '~global": 'tt'alht' values ".
qust b~caIculated in ordeito determine'which contract generates the
highest net gain. Economists are not equipped to make such comparisons
directly. Coriditions can be spelled out, however, under which a switch
from one form to another is Iikely. Turning to the three contract forms,
one refutable implcation of the immediately preceding model is ,that
when market wage rises relative to land rent, the contract form will shift
away from the wage contract, which induces a (relatively) careless use of
labr, to the land-rent contract (though perhaps first to the share con
tract), which induces a relatively careless use of land. As a prelude to
cmpring the differnt 'contract forms;we,should first examine the prac
tice of share contracting.,
Changing the contract (orm.

.
In thepreceding secJion itwa,s shown that. the:fixed:ren~ 'cQIltrat:places

" ;~: evarius land att;dbutes' in ~epuQlic ;doma.iu'an<Ciht,the, assoi~teQ
losses can be lowered by manipulating the prices arid quantities of related
goods. The overuse of those contract attributes that are Iot directly con
trolled may be curbed but cannot be eliminated. It is proper to ask
whether collaboration between owner,s' of land and owners of labor
should occur through a contract that is an, alternative to the land-rent
contract.
. .
Changing the assignment of responsibil.ities cap ~e radically different.
from changing the stipulations of an existing contract. It can instead be
effected by altering the unit by which the transaction is conducted. Such.
a change in units affects the whole structure 'of attribute wnership, or
responsibility, and the associated incentive system. The contract between
the owner of land and the owner of labor can use acres as the basic
transaction unit - the rental contract:" or it can use }iJ.an-hou~s ~s the
basic unit - the employment contract. Shiftjng from one contract tQ the
other constitutes a global change in the tax-subsidy structure; it com
pletely alters the supply and demand of the various unpriced attributes.
For instance, since an employed worker's pay is not, on the margin, a
function of output value, he or she does not gain from exploiting the soil
nutrients. This change will, in turn, affect the nature of the output
indeed, the choice of what to produce. A switch from one contiact foem
to another may even alter the desired characteristics of exchange part
ners, in effect changing their identity. The basic character of the problem',
'howevet,'remainsunchanged. Each party will overuse those attributes
that are both controlled by her or hm and subsidized by the other party,
and each will reduce the supply of those attribu~es' which he or she' subsi
dizes. The wage contract, for instance, seldom stipulates aIl workplace
amenities, and the employer is expected to shirk in providing them.
.. The decision of whether to adopt a land-rent or a wage contract de
pends on which of the two results in a smaller tQtal loss from the mar
ginal discrepancies, ths'-:nXmzing"lhe"nefgaifi"f' the"cooperatoC,"
Leaving aside, for the moment, the discussion of sharing, ir is dear that
the two contracts cannot be compared margimilly: The one that will max
48
ADVANTAGES AND DISADVANTAGES OF SHARING
The share "tenancy contract stands halfway between the fixed-rent and
the wage contract. When sharing, both Iandow,!ler and tenant are resid:- ... _
uaI claimants because each is remunerated bya fraction of whatever the .
output is. At the same time, each gains from shrking: The landowner
wiIl not maintain Iand improvements as vigorousIy as he or she wouId
have'under the wage contract, and the tenant will not work as hard S
he or she would have under the fixed-rent contract. The margins subject
i:o distortion undei theshare contract indude all those of the other two
contracts. In addition, the specification and monitoring of output are
Iikely to consume more resources under the share contract than under
the other contract forms. Although more margins are subject to distor
tion when the corii:i-act calls for sharing, the loss from some margins of
distortion is reduced more than proportionately. Adoptng a new per
spective on the previously discussed tax analogy may demonstrate how
the share contract can ,r:eslllt in a 10wer leveI of distortions than that
"assoCla:i:e(rwrththeother'twocontr~:rs ...
The analysis of tax distortions is standard fare in taxation literature. It
is well known that the distorcion associated with a tax (or with a subsidy)
49
Contract choice
rises as the square o the tax (or subsidy) rate. Thus, the welfare-loss
trangle of a 1:0 percent tax on a commodty is (approximately) four times
as large as that of a 5 percent tax on the same commodity. In the wage
contract, to the extent that supervision is not perfect, the reduction in
effort is a free attribute available to the worker, because the worker is
not penalized for the reduced effort. It is as if the worker were to pay a
.100 percent tax on the increase in output induced by greater effort. Smi
lady, under the .6xed-rent contract it is as if the tenant were to receive a
:,:t:'PfJl'9.~nt,!)!l~~4Y:Q,R~~l,lJ..tr:i~m$.~nd the landlord were to pay a 100
percent tax ou lnd imptovements, yielding returnswithin the contract , '
perodo In a share contract the taxes are reduced from' 100 percent to the
worker's share in the case of the extra effort and to the landlord's share
in the case of the maintenance expenses. Since the deadweight loss rises
quadratically with the tax rate, the rate reduction brought about by shar
ing results in a more than proportionate reduction in losses. For instance,
a fifty-fifty sharing arrangement would reduce the distortions from each
of the taxed attributes to one-fourth of their leveis at the 100 percent tax
or subsidy. The proposition, however, does not apply to the subsidized
attributes, since these continue to receive a 100 percent subsidy. For ex
ample, to the tenant the nutrient is fu11y subsidized under both the .6xed
'" rennind the shatecontract. Whereas ali of these items are taxed or subsi
dized in a share con'tract, compared with only about half of them being
taxed or subsidized when one or the other of the two contract forms is
used, the quadratic relacionship is capable of lessenng the total burden
under the share contract to a levei below that of the burden either of the
other forms generates .. The share contract, however, entails an ncrease
in monitoring costs, which may tip the scales against it.
Monitoring serves to reduce the losses associated with margins of dis
tortion. The monitoring of each margin of distortion, however, entails
.its own start-up cost. Since the share contract is subject to more margins
of distortion than are the other two contracts, its costs of monitoring are
higher. The gains the share contract is capable of generating in reduced
,distortions may not.belarge.enough,.and.the,.sharecontract,may fail to
be adopted. As conditions gradualIy shift, favoring, say, the wage con
tract over the fixed-rent contract, the share contract becomes attractive
as an interme.diate step; nevertheless, because of the required extra moni
toring costs, the share contract may be skipped altogether. 16 As the fixed
rent contract becomes less attractive, the fraction of rent contracts is ex
pected to falI and the fraction of wag coni:racts is expected to increase;
but it is not possible to state, a priori, whether the fraction of share con
tracts will increase or falt.
16 As the value of the contributions of the twO parties becQmes more equal, a sharing
arrangement is more likely to emerge. This is beca use a fifty-fifty sharing formula
tends to yield the highest reduction in distortions and is also simpler to administer
than others.
COSTS OF SOLE OWNERSHIP
I have thus far focused on the costs and gains associated with fixed
wages, fixed rents, and share contracts between owners of land and la
bor., Although different c0l1:tracts encounter different incentive problems,
every exchaiige;, aod 'therefo~e every contract, is subject to some such
prbl~m;' 'llie 'sole-o:wnrship arrangement is free of thosecontracting
problems that arise when'land and labor are not ownd by the same
person. It might appar that sole ownership should be the preferred
method of operation. However, the sole-ownership arrangement is sub
ject to ~o sets of transaction costs ass6ciated with owning alI of the
inputs in a production processo The first set of costs arises beca use the
pattern of ownership of productive nOl].-human assets is extremely un

.likely to match' ful1y the owne;rship pattern of.human skills that would
have genrated the highest output. Total output, then, can be increased
if people exchange productive assets in order to arrive at a better match
of resources. To accomplish this, owners oflaborwo'dhave'i:o borrow
in order to buy the lani:i with which to wqrk. For this to occur, traris~'
acting must be reintroduced ..,. here the transaction be~een borrower and
lender replac!ng that tretWeen owneJ;'of labor. and owneJ of land - and it
cannot be determined a priori that the transaction between the former
tWo would yield a hgher net gain tnan the"transation.between the latter
two.
The second, equally important set of costs of sole ownership exists due
to the losses in specialization that occur when one individual owns and
uses alI of the productive inputs. 17 Although sole ownership does remove
the incentive to shirk; the gaios from specialization are also forgone. In
order to maximize the return from their land, landowners will engage in
such activities as maintenance work and preventjon of erosion~ Theown
ers of labor will invest in suchac.tivities as maintaining and improving
their cultivation skills. One person who owns both assets cannot profit
ably specialize as much as two individual owners of two assets. However,
, to the extent that collaboration will take place, the owners of the collabo
rting factors will also acquire skills in protecting their assets against
, tapq.tre by their partners. For instance, the lan<;lowner will acquire infor
mation about methods of reducing the overuse of nutrients. The notion
,t~at la,'rid-.andlaooral'e'suffident'toproduceoutpnt -isa" gross 'oversimpli"
fication: Many production factors determine the output. Farmers are sel
'17See
Eswaran and Kotwal 1985.
5I
Economic analysis of property rigbts

dom if ever the sole owners of alI inputs. A present-day sole-owner
farmer in the United States would, among other things, have to own and
operate a spray plane and conduct plant research and deyeIopment. As
markets grow larger, the potential gains from specialization should in
crease; any one person should benefit from relinquishing the ownership
of various assets (or attributes of some asset) and engaging in contracting
with owners of other inputs in order to acquire the corresponding ser
vices. The gains deriving from sole ownership must be balanced against
the falling output caused by the commensurate low,er,Ic:!y.e.l,~r~p~~.~~I,iz.~.~..
tion.
'
....,..,."..,........ "'.
""""""", ....",
,.
" .. ,".' ..
SOME Ii.\.!fPLICATIONS
The contracting model discussed thus far can generate many implications
with regard to actual tenancy practices. Because infoqnation problems
are at the heart of the high cost of transacting, I will concentrate on
tliose implications that are a direct consequence of these problems. Two
sources of change in information costs virill be considered, one associated
with the introduction of a new crop and the other with the arrival of new
"",workers., , ,
When a cr.op, !ie,w., ~(),a.I1" g~~, l>,~::tl1~~, PXQ,6.~aJ~l~>infQ".maJim,o.n, h.ow"
well it will do in different locations within the area is more costly to
obtain than is similar information on old crops. Landowners who per
.sonally cultivate only part of their holdings and contract with others to
cultivate the rest of their land know, as a rule, more about their land than
do their tenants. Because they are aIso. the main benefidaries of good
decisions about new crops, the discrepancy in landlord-.v.~.~qs-tenant
knowledge for the new crop is likely to be higher than it was for the old
crop. Tenants who are offered fixed-rent contracts may suspect that the
rent Iandlords' demand is excessive and that the landlords are exaggerat
ing the productiveness of their plots in growing the new crop. Such suspi
dons are hard to allay; hence tenants' demand for land and their associ
atedcoUhteroffers are likely to be low. Wage contracts are free of this
particular problem because landlords who pay fixed wages bear the en
tire new-crop risk, about which they are better informed than are tenants.
Landowners who switch to the n,ew crop are more IikeIy to offer wage
contracts to their tenants than they were with the old, established crop.
In addition, it is expected that: with the passage of time the direction of
the trend will be reversed: The cost of determining the suitabilit:y of land
parcels for the new crop will decline, inducing the readoption of fixed
"'fei'ifcoi'itra:ctswherethy Were preferred before.
Another nformation problem arises concerning immigration. Little is
52
Contract cboice
known abol.lt how workers who are new to an area will perform. Land
owners are reluctant to commit themselves to paying new workers the,
prevailing wage. Given the Iack of knowledge regarding workers' abili
ties and attitudes, the demand for the services of such workers and, con
sequently, the wages offered are likely to be low. A new worker who
believes that he or she is more productive than the wage being offered
indicates can "guaraI!tee" his or her output by offering to operate as a
fixed-rent tenant.'These workers, then, bear the onus of the information
, .... p~ob~~~:.~t}s..e;X:p"ected that relatively more new workers than established
'ones wiU 'Operate aS' 'fixed"'rept tenants. Moreover, new tenants will be
givenparcels that are more'difficlllt to exploit,suh as diose containing
few improvements. Some old-time tenants may acquire a reputation for
being gentle on the land and on improvements; these individuaIs will be
the favored fixed-rent tenants on the easy-to-exploit parcels. Newcomers
do not haVe such reputations and are therefore expected to get land par
cels for which the lack of information makes less difference. Here, too,
the trend is expected to be reversed as the new workers become more
established in the area.
CONCLUSION
The owners of labor and of land can increase the value of their assets by
collaborating. beca use total output is rendered larger than it would be
were they to operate alone. However, effecting the collaboration, is tself
costly because it is difficult to prevent wealth capture whencooperation
is attempted. Measuring each factor's contribution to output is necessary
for cooperationto be successful. Such measurements are costly and there
fore will not be precise. This lack of precision, coupled with the variabil
ity in output due to unpredictable factors such as the weather, implies
that individuaIs can gain at each other's expense and that they will spend
resources in order to capture these gains. Together owners of labor and
owners of land (bolstered by competition from other owners) will adopt
the contract form that generates the largest net output value where max
imization is sbjecftcnverifirialproductiricostsas well as to the
costs associated with the capture of wealth.
Neither labor nor land is uniform; specimens of each vary in terms of
the leveis of their respective attributes. The contract between the owners
will therefore attempt to control not onIy the factors as a whole but ais o
various individual attributes. Someof these attributes may be controlled
directly (e.g., a tenant pays for irrigation water supplied by the landIord);
those difficult to controI directly may be controlled indire"C1:Iy'"hy'"fixing'...... '"'"
quantities and altering prices. A basic principie underlying the maximiza
53

. Source of losses from various contractual forms
Policing
labor
Polidng
land '
High losses
FW
Inrerrnediatelosses
Lpw ar no losses
se
FR, 50
Polidng
output
Lack of
specializing
FR
se
se
FW
SO,FR
so
se
FW,SO
Divided ownership
FW, FR
,-,:sofSole .J:)Wp.erof land and}at;or

'se:'> Shar'titract
.)'"
FW: . Fi.ied~wage contract
FR:
Fixed-cent contract
tion process is that individual attributes will be placed under the controI
of the party that can more readily affecr the net value of the outcome by
nianipulating the atiribute.
The accompanying table ctaIogues and compares the losses associated
with the various contractual arrangements analyzed in this chapter. This
. table dearly. reveaIs that
single' solution 'is' 'best under alldtcm
stances. As a situation changes, the form of organization'will,tend"to"
change as well.
no
'.
54
In Chapter 3, I pointed out that in order to cooperate, resource owners

must cede' to each other control over some of the attributes of their re
sources. Since it is toa costly to price every attribute, their use will be
sub-optimal, and some potential gains will be forsaken as a resulto Sole
ownership of aIl of the resources involved would eliminate the dead
weight loss associated with cooperation. However, sole ownership may
result in yet a greater loss due to reduced specialization; if this is the case,
"it"will'not'be"chosen'.Itmight'appeartht'nhe'case of commodities or
assets that consist of single physical entities rather than those that are
more complex, sole ownership would be routinely adopted. But because
most, if not alI, transacted commodities have many attributes, multiple
ownership of commodities may be preferred to sole ownership. In this
chapter, I shall suggest sources of potential gains frol1l.multiple owner
ship, offer hyp-otheses regarding the form the division of ownership will
take, and discuss the organzations that are expected to emerge to handle
relations among the owners.
The physi~al entities I consder to be commodities or assets typically
consist of many attributes, and often different subsets of asset-attributes
are owned by different individuaIs. For instance, the economic and legal
ownership of every asset subject to guarantee (or ~oliabilityLisdivided
between its holder and its guarantor; similarly, the ownership of every
rented asset is divided between the nominal owner and the renter. Ex
plaining the pattem of such ownership is en~(alto the study of organiza
tion. My thesis is, nrst, that different individuaIs have comparative ad
vantages in owning different attributes, and that individuais will assume
ownership of the- attributes accordingly. Second, as transactors' gaIn
from affecng the outcome of an aspect of a transaction increases, they
will assume more responsibility for the associated variability. That is,
they will tend 'to assume a greater share of, and thus increasingly become
residual daimants to, the income of the attributes they can affect more
55
DiiJided ownership
than they did before. The management of entities with divided ownership
requires organization, which is one of the subjects of the present chapter.
The discussion of divided ownership and its management is also a useful
introduction to the fum, which is the subject of Chapter 5.
zation. Of the resource owners associated with the operation, several

subsets are expected to become residual claimants to different compo
nents of the large-scale operarion. Within subsets containing more than
one individual, however, individuaIs wiIl operate as members of cen
tralizing organizations.
I shall use three capital goods - a taxicab, a large machine, and an
office building - to illustrate the namre of this problem.2 Problems asso
dated with the use of the fust two goods will be described briefly, while
. .. (he t:hi,d will be explored more extensively. The foIlowing examples en
ompass most of the problemsthat arise when equipmentis used, as well ..
as:thediffetentme'tlf6dsfor tesolving them.'

' - ..... "" ........ " ....
Althoughtax1cabsare not large, a cab may be operated by tW or three

fufl.,time operators. Consider, for example, the case in which a cab s
operated by two in~ividuals who share its ownership and drive it in
shifts. The two may seem to operate as full-fledged partners. I suggest,
however, that they will exploit the opportunities available to them to
. -:reduce common-pr.Qperty problems by making certain attributes exclu
sive propetties.
Since the two drivers share ownership of rhe cab, its unpriced attri
butes will be consumed asill.ey'.w:ere half free, thereby causiIlg~.l1ecab
to b,ecome;in part, cominon property. 3 This case corresponds to sharing
in agriculture, except that here they share its us,slleeachofi:hetWo .
parrners drives the caro The sharing, however, need not apply to alI the
atfri{)utes of the:'ciib. For.'instance, j:he time slots allotted to each owner
may be dear.1y delineated between the two; each then becomes the exclu
sive user of the"cab foi: the appropriate sub-periods. Whether or not the
use of gasoline will become common property depends on the accuracy
of the fuel gauge. Assuming it is reasonably accurate, each may assume
responsibility for the cost of his or her own fuel. The division of responsi
bility for the tls.e of the tires depends on how easy it is to keep track of
individuais' mileage (very e<lsy) and tire damage, which depends on the
type of road traveled and driving style (impractical). Thus, tire use is
more likely to become common property than are time slots, perhaps
also more likely than is the use of gasoline. UphoIstery wear and tear is
toa costly to monitor and is likely to become common property. Con
sider, finally, why the engine may become common property. Suppose
that engines designed. for premium gasoline can also run on regular gaso
line but are damaged by its use and, for the sake of illustration, that
. . .
. .
.
..
fi
2 Divlded ownershlp IS not restncted to eqUlpment. Untl! qUlte ~ecently, many
..
.
L.
ts..
eared.to.be.~:whQle.?~.ln
..
lecent.years~:howey.er
.
~t.!.)a$/;leen
found
. ....... nancra alise a p p . .
I .
THE OWNERSHIP OF EQUIPMENT
Scale economies to equipment used in production are often asserted to.

be a major determinant of the size of the fum. (I shall elaborate on the .
::::::::::fiirii:::a:iid:itssize..in Chapter 5.) Inthestapdardmodel the. si2;e o/the. ;.'
(competitive).6rm is ,such' tbat it fllyei:ploitsscaleeconomis~,In.the :'
following discussion I will concentrate on the ownership pattern for
equipment, some of which is large-scale.
.
The exploitation of large-scale equipment requires input from many
individuais. Since these individuais may attempt to gain at each other's
expense, their interactions result in transaction costs. Maximization re-
quires taking such costs into account and reducing their impacto It is
convenient to introduce transaction costs here by considering 'l?roblems .
associated with such a large scale. In the process of analyzing the effects
of these costs, I shall show that the existence of scale economies associ
ated with equipment is neither necessary nor sufficient for the exis:ence
.of a large-scale organization. Although large-scale operations may also
be associated with activities such as innovation and the research required
'for marketing and for the pricing of commodities, I here cnsider only
the equipment's scale.
.
The potential for large-scale economies can be realized by a centraliz
ing organization; alternatively, such economies may be realized by turn
ing each resource owner into a residual claimant of hs or her own opera
tons, alIowing each free, though not exclusive, access to the equipment.
In the latter case, the individuais operating the large-scale equipment
need not alI belong to the same organization; each may assume full con
trol of his or her niche of the operation. 1 They may accomplish this oy
raisng capital, purchasng other nputs, and selling output. Presunably,
"if'alarge"scaleorganizarionSchseri,the participating owners of re
sources have opportunities to shirk. If each participant instead remains
independent, each would have unrestricted access to "his" or "her" pari:
of the equipment as well, rendering it common property, since vaious
attributes of the large-scale activity are then left in the publicdomain.
I shall argue that,asa rule, the maximizing solution to the problem of
"
.. .
..
h
the two forms of orgam.
t e use of large-scale eqUlpment IS a mlxture
. ......of
....................................................................
..... ..
1 In a classic study Allen (1966 [.192.9]) describes how in Birmingham ih d':e' ~'866s .
.
.
.
workers operatmg
mdependently
of each other shared space and power sources In
large factories.
. . .' .
56
that they can be unbundled to reveal. a rights ~tructure of some comp. eXlty. d
d
3The mec hanl'cs of how a free attrtbute avallable. to a transactor wlil be use un er
fixed pay or under a share contract are presented In Chapter 3.
57

engines using premiwn gasoline are well suited for cabs. If drivers who
rent their cabs pay for gasoline, they will gain by using regular gasoline;
as a result, the engines will wear out prematurely. Even partners may
choose to use regular gasoline. Each saves the entire diffrence in price
relative to what he or she would have paid for premium gasoline while
bearing only half the cost of the resulting engine damage. To prevent this,
cabs are expected to be fitted with engines that use regular gasoline. Cabs
will be fitted with engines that require premhun gasoline only if a central
....................
.......i:Z.e.4gJ:'g:!)i:Z.~lJi..n...J#;~i!! .;h;\IJ;g~.Lfu.~Hng.them,.~
.
SevetlW6ikeiS'ifjyoeiequiid':f'mak'e'largenichine ftfuction. If...
all of them share ih. its wnerilhip,thentheabsehc ofconstrintSon ,
individual behavior will permit many of the attributes of such a mae,hine
to become common property. As a result, incentives for sueh activities as
earefuI handling and maintenance may be greatly weakened. As will be
expIained later in this chapter, this problem may be alleviated if many of
the individuaIs who work with the machine beeome employees of the
machine's owner. Not all those working with the machine are expected
to be the owner's employees, however. For instance, the machine's manu
faeturer may have sold it with a guarantee, while retaining the function
of servcing it. In that case; contrary to the standard model, some of
.the indivduals..w:or:king.. with..the maehine are not the employees of its
(nominal) owner.
Where an Qffice building aecommodates many workers, common
property problems such as the use of eorridors and access to utilities need
to be addressed. Still, most of the structure ean be assigned to distinet
individuaIs, and individuaIs' rights can be reasonably well delineated.
The.b.uilding's users do not have to be sevei:ely constrained not to damage
it and can conveniently belong to more than one organization. Before
moving on to discuss the nature and the organization of offiee building
ownership in greater detaiI, Iet me summarize the discussion up to this
point.
.
The preceding examples address common-property problems that arise
when the ownership of goods is divided among individuaIs. These exam~
pIes also illustrate' that the severity oi thcaptureprohleill 'varisf~om .
one case to another and, indeed, that the severity of the capture problem
is not uniform for different attributes of a single pieee of equipment.
Because different attributes of apiece of equipment are not equally sus
ceptible to capture, it may be advantageous to handle its various attri
butes.. differently. Attributes suseeptible to serious commmFproperty
problems, such as equipment lubrication, will tend to be owned by orga
4 By the saroe token, rental cars - even fancy ones - are expected to be equipped
with engines requiring regular gasoline.
58
Divided ownership
..
nizations created to control these problems. On the other hand, attributes
that are relatively free of such problems will tend to be individually
owned. The ownership of a capital good is expeeted to be vested in nei
ther a single person nor a single organization. The maximizng ownership
pattern of individual attributes of equipment is the one that will minimize
the capture loss net of the cost of loss prevention.
Because large pieces of equipment often possess major attributes that

are susceptible to common-property problems, they will, according to
the hypothesis presentep hre;:be owned by some centralizing organiza
tion, an organizanon that mar- be alled a "fum~" This firm may appe.ar
tocotifofrn: to" the reeel~ed furo, enjoying a scale eeonomy in the use oi
equipment. However, unlike the received fum's product, the product of
the current fum is determined by the nature of the common-property
problems it eneounters. When different attributes of a single piece of
equipment are owned by different individuais or different organizations,
each will produce its own output. Tht! capital good "Iarge office build
ing" is a striking example of an asset subject to scale economy. Altp.ough
it is ideaIly suited to determine the size of the standard mo.el of the firm,
its attributes are owned by a number of owners.
.
.
Large office buildings are owned and used in a radica,lly different wa:y'
from that implied by the standard model: The owner of a large office
building is the individual or the organization holding the title to it. Nev
ertheless, that owner does not usually retaintherights to.ll attributes oi
the building, his or her ownersnip being circumscribed. The indi~iduals' ,
working in an offiee-building are seldom the-employe~s of tire owner of _
the building, and the owner who holds the tide to it is~ as a rule, distinct
from the firms that actuaIly use it. Indeed, sometimes the titleholder sup
plies little more than the coordination of the contracts that govern the
use of the building between various firms, each of which owns some of
its attributes.
.
The structure of rights. over a large office building is complex: The
titleholder usuaIly rents out office space, thereby relinquishingtothe ten
ants a subset of the rights he or she has previously he~d; the renters' - the
tenants - become owners of these rights. These tenants, whieh are often
distinct firms, produce dstinct outputs that have little to do with either
the building or its scale.:Moreover, other parties are often granted rights
to other attributes of the building. If the building has been' mortgaged,
the mortgage-holding bank has most likely imposed restrictions on the
building owner, thereby making itself the owner of a subset of rigbts. It
..... m~~t~...th~ . .4~6ItimLt~.'.Qw.ner~~... s.in.cejt.be;Qmes.axesidual.los.er..iLthe .....
landlord, that is, the'titlehol'der, is unable to make mortgage payments.
If the landlord has r<:;tained .a janitorial service, the supplier of that service
5.9...... .
',,-:,__ ~,~,,;:.,~.::::';;':';~~:'~_;"_.-'i:/,;',"..
'*
assumes the liabilty for its operation and is, in turn, the owne;r of another
subset of rights.
Given that severa! organizations hoId rights over an asset such as an
office buiIding, the question arises as to the principIe that 'governs the
aIlocation of these rights. As I have suggested, the structure of rights is
expected to be designed so as to allocate ownership of individual attri
butes such that the parties who have a comparative advantage in affect
ing ,the income flow over the attributes that are susceptibIe to the
common-property probIem will obtain rights over them.
.
,.
".
TRANS FERRIN-G OWNERSHIP
I shall discuss one additional right, namely, that associated with fire in
surnce, in greater detail than others because the fire insurer's role as
residual clamant is easy to grasp and because it seems to counter the
perception of the funetion of fire insurance held by many economists.
'li is corrpnonly ass.umed that building 'owners buy fire insurance be
caqse they are risk-averse and would like to shift the financiaI risk of fire
te those PloJ;e willing to assume such a risk, I, however, assume rsk
neutrality' (WhiCh :is not unreasonable in this"casej~ Risk-neutrality im
plies that the sole objective of thebiiildiriif owner' igardrig' firt; 18 i:h
minimization of the expected net loss from it. What can explain the fire
inSurance.transact{on? I sl1all argue that one explanaton for such a trans
action is 'that fire" insurers are more efficient than are the titleholders of
.. .-buildings as whers of the fire-risk attribute of office buil4.i.~$s.
Not being a fire-prevention expert, an office-buiIding owner would like
to secure the services of someone to minimize the expecl:ed loss from fire.
The specialist could be hired for a fixed wage; however, the owner lacks
the proper knowledge to drect the employee to do the right job. Under
';1" fixed wage, the employee would find it easy to shirk and would have
no strong inentive to ensure that the fire-prevention program is efficient.
It is possble to overcometrusdifficulty by making the specalst responsi
ble for his or her own actions. The specialist who charges a fixed fee
for his services, while assumng the residual claimancy to his actions, s
providing fire insurance at a fixed premium. 5 As an insurer, the specialist
wilI lose from fire and gaio from its absence. He is, therefore, motivated
to take such preventative actions as minimizing fire hazards, conducting
fire drills, and enhancng speedy fire fighting when fire does occur, in
order to reduce the level- or, more accurateIy, the net expected cost of
SThe need for capital to guarantee the actions will be discussed in the next chapter.
60
Divided ownership
fire losses. "I:'he insurer is thus one of the owners of the building: He or
she owns the fire occurrence attribute of the building.6
This analysis of fire insurance constitutes the application of a general
principIe. As a rule, specialists know more about their line of business
than do their customers. They are therefore in a position to charge for
service that is of a higher quality than what they actualIy provide. For
this reason people would be reIuctant t9 deal with them were their ser
vices not guaranteed. Insurance may be viewed as one form of the sale of
guaranteed service. It also amounts to a division of the ownership of the
.::;::::':::":lnsmea::assets::hetvVeeilihe::nomiiiifownersand 'the insurers. themse1ves.
,~;,." Fir~ insuranceis setdom"aH.:encompassing. FuIl insur~~ce C6verage im- '
. plies that the insurer is the sole residual claimant of the fire hazard facing
the insured firm.? It is unlikeIy, however, that only the insurer can affect
the incidence of fire and the loss from it; the insured is also expected to
affect fire losses. The latter usually makes decisions about where and how
flammable materiaIs are stored, whether or not smoking is discouraged,
and how well its employees keep fire escapes unobstructed. Both parties,
then, contribute to the mean effect of fire hazard, and both are expected
to bear some of the effect. Consequently, coverage is not fuH and both
the insured and the insurer become, to varying degrees, residual claim
,apt$,t;ha:Us, .9.WAers of the fire hazard.
There are two refutable implications that could test the hypothesized
role of insurance. Dne relates directly to fire insurance, whereas the other
is Iess specialized. The first states that when a party's contributon to the
mean loss of fire decreases, the fire-insurance contract will be modified
to decrease that party's share of the fire losses. For instance, suppose that
motors are used in the insured oprations. A switch by the insured - due,
say, to a change in relative fuel prices - from gasoline motors, which are
a serious fire hazard, to electric motors, which are less of a fire hazard,
constitutes a reduction in the insured's contribution to fire hazard. Be
sides the reduction in total insurance payments, his or her share in, the
income variability due to fire hazard ought to be reduced; in this case his
or her coinsurance rate should be lowered. li
The send implication concerrisadistiridpatteriifaii:dbute'owner~
ship between two types of condominiums. In the first ali the units are
6The expected value pi sLlh o:wnershp is nega tive; hence the insurer makes a nega
tive payment (i.e., he or she receives a premium) in order to acquire the right to the
fire-occurrence attribute.
7For coverage to be full, ir must indude, besides direct losses,such effects as those
due to lost business, to inconvenience, and to suffering.
8Insurers indirectly perform another service: the policing of employees' diligence in
reducing expected fire losses. A relative increase in some premiums signals to owners
that their employees have become too lax. For a related discussion, see Hall I986.
6r

located in one building, whereas in the second tbe individual units consist
f separate structures. Fire and plumbing problems in one unit are more
likely to spill over to other units in the single building tban among the
separate structures. It is expected that in the single building the apart
ment owners will alIocate the ownership of tbe .effects of fire and plumb
ing problems to the management, whereas the individual owners will
tend to own (and, jf they so choose, to transact for) tbese attributes where
units are in separate structures.
The a~gumrt'preseiii:ed in this' chapter suggests thatthe severity of the
.cotrinon-PFopertyproble,w encourirered as a;res:ult.Qf ,equiprnerlt us.e 'yar~.
iesfrom'tiriepice toanother, as well as ac::ross different attributes of the
same piece of equipment, and that the divided ownership of. the equip
ment permits the separate hndling of various common-property prob
lems. Each problem can receive individual treatment, while areas from
which the coinrnon-property problem is absent may entirely escape the
treatment designed to cape wth common-property issues. The analysis
that leads to the notion: of isolating individl cornmon-property prob
lems .through divided ownership is nly the first step toward the analysis
of the solution of such problems. To this end, I shall next examine the.
ways in which people r.educe the costs associated willi 'common-prperty
problems.
"
MITIGATiNG COi.-tMON-PROPERTY PROBLEMS:

SCALING EQUIP~ENT,D9wN AND USlNG
THE FIXED-WAGE 'CONTRACT
I have thus far concentrated on the multi-attribute nature of commodities

and have pointed out that since not alI the attributes of even large-scale
equipment are subject to common-property problems, it may be advanta
geous to make at least some.of these the xdusive property of the individ
ual users. In this section I discuss methods users employ to cope with
attributes that are subject to common-propertyproblems.
The simplest way to contain the common-property problem in the use
of equipment is to scale it down in size and value. If the equipment is
scaled down to fit a single operator, the capture problem disappears, the
problem of raising capital is eased, and the need for organization is obvi
ated. The sacrifice associated in economies of scale, however, is often toa
large,9.
An example of the scaling down of equipment is provided by many of
...... 9'Tt::;:p~:o;:-;;tenti~rform:Ttipi~shift~is;~~s~~;ifi~~d~~i~~g~~th~'~q~i~~~~~'i~ used
by oJ.lly one person. This potential is sometmes also sacrificed within firms, presum
ably to.enhnce employees' accountability and thus ease remaining common-property
problems.
.
62
Divided ownership
tbe tools de~igned for non-professionals. These are otten low-cost, low
quality versions of superior professional-quality tools. Professionals can
afford the higher prices because they use the tools more intensively. If a
rental market were to develop, non-professionals could then use tools
designed for professionals. The common~property problem would reap
pear, however, since individual users would have little incentive to han
dle tbe tools carefully. The problem seems severe enough that amateurs
prefer to be the sole owners of low-quality tools rather than share the
~~~~;~~~~:~~~n~l:~i~~~~~~~~~~~~~~h~~hd~~~:i~voiv~i~~~::"'"""
nization, is the imposition of restrictions on the users of equipment in
order to reduce their excessive and careless use of it. The wage contract
constitutes a major application of this method. By virtue of rewa,rding
the worker for his or her time, that contract abates the excessive and
careless use of equipment. This is an implicit result of the argument put
forth in Chapter 3. I noted that nutrients are a free attribute to fixed-rent
tenants, who are expected to extract them in higher quantities than
would a self-employed owner. Nutrients may be a free attribute to fixed
wage workers as well. Since such workers do not gain from extracting
'i:ne'nutrients, they have no incentive to extract an excessive amount. Sim

ilaily,fiXed~wagewotkerswhosha:reequipmentdonotexpecttogain
from exploiting it~ and so would tend not to overexploit it.

Were the wage contract strictly an exchange of time for money, em-
ployees would simply provide time without ever lifting a finger as long
as work effort did not directIy generate utilty. Although such employees
would not h:rin the equipment around them, neither would they do any
thing usefu!. It is obvious that the"wage contract is more than an ex
change of time for money; employers must be able to induce employees
to perform various tasks at some minimal pace. Given job specifications
and the levei of supervision, workers are expected to satisfy the require
ments at the least cost to thems~lves. They will necessarily put out less
effort than self-employed workers do because, on the margin, employees
are not remunerated for effort. Since workers under a wage contract are
not paid for output, their incentiveto.overuse whatever equipment they
are provided is curtailed. For the same reason, the wage contract may
also be used to control equipment abuse when a piece of equiPlIlent is
operated solely by one person but owned by another.
Were workers' and machines' exertions proportionate, employers
could fully rely on workers' maximiztion (i.e., doing as little as they can
get.away with) to avert the overuse of the equipment. Some substitution
between workers' and machines' exertion, however, seems possible; at
lOThey also enjoy quick access but forgo the opportunity to share the costs of
storage and maintenance.
63

the very least, workers lack the incentive to service their equipment.
Therefore, joh specifications and supervision must take into account such
opportunities and make provisions for equipment maintenance.
Workers who seU their labor services by way of a wage contract are
engaged in a constrained transaction: They agree to obey a certain range
of instructions. These instmctions are designed to induce the perfor
mance of productive services and to discourage workers from inflicting
harm on equipment (or ou fellow workers). In turn, employers also ac
cept various constraints 0lJ. their own behavior, such as agreeing to a
:::::'::::::'::':::maXimillri::eqiiipmeni:s'j,eed;:providirtg coffee breaks, and permitting
grievance procedures. ".".'.: <.;: .
. . . . ..
.'.
." .
CONCLUSION
Although commodities and production equipment constitute single phys

ical entities, as a mIe they have many distinctly different attributes. The
value of multi-attribute assets is not necessarily maximized if these assets
are owned by single individuaIs; it may be nhanced by allocating owner
ship. over individual attributes according to comparative advantage.
Thus, fire insurerS rather than the nominal titleholders of assets are the
efficie.l1t. Q:wn~f$. of the assets' attribute of fire hazard beca use the former
'rath~r than the latter are the specialists in minimizing fire losses. As own
ers, insurers are ~e residual claimants to fue protection, gaining most by
minimizing the net 10ss from fire.
Some equipment, especially that which is large-scale, may benefit from
having multiple users. In that case, however, some of its attributes may
be subject to common-property problems. The equipment may be scaled
down to avoid these problems. Alternatively, the equipment operators
may be constrained. The wage contract removes workers' incentive to
overuse equipment by rewarding workers for their time. Many persons
may work with a single piece of equipment without treating it as com
mon property. Whereas the received analysis of labor provides no expla
nation for using time as the dominant unit by which labor is exchanged,
the role of the Wageofittactriveftirig eqii:;:m:erifabsemay partially
explain its widespread use.
The labor contract is but one of many contracts imposing constraints
on the transactors; such constraints constitute an integral part of any
organization. The employment contract and the associated constraints
may appear to lead to the theory ofthefirm. However, as will be dis
cussed in the next chapter, constraints are also commoh'in what are usu
ally called market transactions, and therefore the absence or presence
constraint does not generate a clear distinction between operations in the
market and in firms.
The old firm and the new organization
For many years the theory of economic. lrganization was nearly synony
mous 'with the theory of the fum. Standard economics textbooks occa
sionally referred to other forros of organization, such as the family and
government, butthe firm was the only.organization conslstently and sys
temattcally discussed. Nevertheless, the exact functon. of the fum re
mained unclear. Knight's (1921) attempt to add substance to the con
struct called tl}e "firm" did not fare v~ry:Wt'!II"a.Ilcl. Coase~s.(:I937)
revolutionary approaGh to organization required several deides, as well
as his own work on social cost, to begin to influence economists. Both' ol""
these efforts focused. OJ.1 the firmo In".more recent years, Coase's transac
tion cost icl:eas have been explite'by econmists itempting to generate
. a theory of the firm, as well aS py oths;:rs whQ guestion the usefulness
of such a theory.l The insights gained through these efforts have been
considerable, yet our understanding of organization is only in its infancy.
In this chapter the transaction cost approach to organization will be ex
tended.
In order. to appreciate the need for the transaction cost approach
ando its contribution, one must first examine. the received model of the
. firm extensively and critically. I attempt' to show that as a description of
real fums the received mdel is unteriable, and that under conditions
that would allow a textbook firm to exist, its function would become
inconsequential. Turning to general problems of organization, I first con
sider the distinction between market and fum transactions and the sig
nificance of Coase's discussion of liability. I then briefly outline the major
contributions to the theory of the firmo Finally, I describe in greater detail
my own contribution, which stresses the guaranteeing role of equity
....;....;.capitat.......
1 Prominent
among these are Alchian and Demsetz (I97 2 ), Williamson (1975), Jen
8
. sen' and Meckling (197 6 ), Klein, Crawford, and Alchian (197 8 ), and Cheung (19 3)'
65
64
...................
.........

A CRITIQUE OF THE RECEIVED MODEL OF THE FIRM
Underlying the received model of the fum is the production function: a

relationship that tracks the highest output any set of inputs can produce
using the available technology. The firm is simply an organization whose
purpose is to select the optimal points on the production function, ac
quire the necessary inputs, and transform,them into output, which.it seUs.
Its objective is profit maximization: in other words, maximizing the dif
,ference"betWeen:,rev:enU'es.and:expenses;',.Th'e:C'()S't.functi~n;:th::~dual~ ,.oI.",
the production function, plays aIllajorrole inshapingthefiIin.~
,,'"
Actual observations do not suppori.:'i:he notion of "duality" betWeen
the production functon and the cost function. Besides labor, equipment,
and materiais, firms also purchase a host of inputs - induding financiai,
accounting, personnel, marketing and legal services - that are used for
"organization" and "monitoring." These latter inputs seem neither to
contribute to "production" nor to perform technological functions.
Viewing the relationship from another angle, technology does not seem
to be the basis of many firms' organizationl makeup. For example, it is
hard to imagine that technological know-how could be relevant to a
fum's decision as to whether it should increase the number of its plants
, "iia~ whnit 'ddds to exparid;'wherh'ii sholJblid new plants or
acquire already existing ones from other firms. The productin function
is not the "dual" of the cost function if it accounts neither fr ali fums'
expenditures nor for the s'cope of ali firms' activities. What, precisely,
does the fum produce? The implicit assumption employed by the stan
d'~~d approach seems to be that each firm performs only onefunction
and that two firms piacng the same output (the same Q) do indeed
perform the same operations. In reality, most firms perform a variety of
operations, not all of which are identical among those firms that seem to
produce the same commodity.
Suppose the commodity being produced s cars. Its parts must be pro
duced and assembled. According to the notion of duality, the production
function should determine the scope ofthe firmo It should inform''llsas
to whether the production function of finished cars involves only their
assembly, in which case the firm should specialize in assembly, or also
the production of the parts, in which case the firm should perform both
functions. Engineering considerations, however, seem irrelevant to the
question of whether one integrated firm could perform these two sets of
operations at a lower cosI: than could two specialized firms. Indeed,
H
2Duality theory relates costs and production functions. Varian (1992.), for instance,
defines the firm in terms ohhe production function (p.x). He discusses technology in
some detail (pp. 2.-2.2.), and states (p. 84) that it is straightforward to derive the cost
function given the technology.
66
whereas one can presume that Japan and the United States exploit the
same production function, Japanese car manufacturers are less vertically
integrated, buying more parts from other fums, than are their American
counterparts. To speak more generally, the determination of whether to
perform any function in-house, that is, within a single fum, or to have it
performed in the market by separate firms does not seem to depend on
technology. It does not seem that the differences among such firms are
attributable to production function considerations.
One could argue that Japanese cars differ Erom -Amc:;;riean, afs, ap,Q
tha:t~he twO sets' belong indifferent industries.: 1:Iow~vr;'n mu~t be"
ble to spedfy priori \vhaiconstitutes difference btweencoinrnodi
ties. Existing theory provides no guidance on this 'issue. Without sueli
guidance, refutation of the theory appears impossible, sine any dlscrep
ancy across fums between the production and cost functons may be at
tributed to differences among the commodities produced by such firms.
In this light, duality is not useful; it must, therefore,.be coneluded that it
3
is contrived and should not be used to describe real firms. . .
The absence of dose relations between tpe prodution and cost func
tions is further illustrated by noting that some firms are housed in more
than one plant, whereas sometimes the same physical plant houses more
than one firm. The observation that production is organized "in-b9 use "
does not imply a strictly technological relationship.
lmplicit to the conventional analysis.of the firm is the assumption of
the existence of costless information. As will become clar In th' course

of my discussion, this assumption, particularly as it has been applied to
the prices and the attrbutes of inputs and outputs, is the fundamental
reason for the discrepancy between the conventional model and actual
observations. li information is freely available, then factors' contrbu
tons can be easily assessed, and monitoring their performance becomes
superfluous. lt is not surprising that in these discussions monitoring is
virtually never mentioned.
'
Consider the rewards to factors used by firms and the notion of the
residuaLclaiman1::.According to the conventional model, the owner of the
fum makes fixed payments to hired factors and receives the residual from
the operations of the firmo That analysis s problematic: The existence of
residual in the received, competitive model is inconsistent with the rest
of that model, nor is the actual remuneration of other factors fixed in the
way the model requires - per unit of Utpl,1t.
The questi(},n of how fi.xed the remuneration of factors employed by
the fum is will be di,s.-g~~~J,.:lt.~J:"i,~,,~~,i~, ~hapter, For noW let us tum to
the concept of the residual. In the WalraslaninoOl;fdoiS'fpodm::tioIl', .
8
3 My critique of the production function is in the saroe Ji'ein as are Goldberg's (19 5)
. . " '
and Williamson's (19 8 5).
Economic analysis of property rghts

and commodities are homogeneous, prices 'are known constants, and the
production function is known to alI. In that model there are no random
factors and no residuais. Knight (3:923:), apparently the fust to consider
the entrepreneur as a residual claimant, did not indicate how a residual
enters the model of the fumo Random forces are one possible souree. For
instanee, ehanges in the weather produce shifts in the costs of and per
haps in the demand f,?r agricultural products. Such shifts yield new equi
libriurils and correspondingly result in changes in the prices and quanti
.' tles of factors and products. Nevertheless, under the specmed conditions,
factor-paynent must eXllust the ~roduct; there canbe no "rt';~i.u..~V:JIl,
;:'~y(:s~~ Knight waspririia.tilY' concetned -noi:-wlth:facrors',suchasthe.
wel:J.ther but, rather, with hu.rnailforces that contribute to uncertainty.
The received eompetitive model, however, does not appear to be capable
of accommodating sueh unexpected diserepancies between revenues and
costs. I suggest that the eostliness of information ean generate residuaIs.
Specifieation of the precise units by which inputs are employed reeeives
.lmost no attention'ia standard texts, yet in reality such units are diverse.
In the case of labor, inputs are sometimes measured by time and some
times by performance. Labor service is a major production function input
that workerl suppiy. The .1bot.input requi,t:ed by the productionfunction
is in efficiency units: A worker who eontributes t::Nice, a.s..ma}J..YJ{ffii~?'l.y"
units as does another worker also provides twice the amount of labor
inp.ut. In a similar vein, the demand for labor is taken as a function of
lab contrtbutton to ouipui. Ii:the standard formulation of the cost
functiQn, how.ev~r, labor, wh~_explicitly considered, is not usually en
tered by units that correspond directly to its contribution. Instead, the
labor input entering the cost function 1S measured by the hour, and so
labor is accounted for in units of time. Rours can readily substitute for
efficiency units only if the relationship between the two is proportionaL
Because worker:s,.themselves are maximizers, it seems high1y plausible
tht inducing workers to apply their time efficiently requires supervisiono
Moreover, suj:>ervision costs are unlikely to be proportional to labor costs
under all Circumstanes; as conditions chrige, 'therefore, the per-hour
efficiency of labor is also likely to change. Proportionality is unlikely to
be present. The question of whether to employ labor by the hour or to
pay 'on a piece-rate basis becomes relevant here, but it cannot be dealt
with by the conventional mode!.
The use of the hour as the unit of labor input in cost functions is not
justified by standard production function considerations, and it is im
proper to switch from one function to the other unless information and
siipervsion'problems ietakil'irifcoiiiii:: Threason'hiii-sare'hever
theless used in cost functions seems to be simply that the dominant mode
of employing labor happens to be by time, and the assumption of propor
68
tionality between actual hours and efficiency units is made

in order that
4
hours may be emplyed in production function analysis.
The need for supervision arises because the factors contriburing to a
fum's production are not alI owned by the same person . The ownership
pattern of factors used by the fum is seldom explored in the received.
analysis, even though the typical implicit ownership assumptions are not
innocuOus. Firms are assumed to own the capital equipment they use and
to hire labor. Hiring labor means r~nting labor services from the owners
of the capital good "labor." In a non-slave economy, workers allw fums
.' tous:sriie.'.o{:the: :se!Yie:$::th~:f.~p'ital.good,j;;ang.v..r.~e. If fums choose
to ,own their capital equipment r:thertharirrifit~n:rnust:conclude
that rental is more problematic thanownership; itis pla:usibfe,theii; that
the use of labor is similarly problematic. In reality, firms do not own all
the capital they use; capital is .also partly rented. To the extent that firms
are financed by borrowing, they are renting rather than owning the capi
tal. Moreover, firms often rent space and equipment. It is not self-evident
that those firm owners who borrow or rent capital and who employ labor
have interests that coincide with the interests of the oWners of the rented
on
assets as to how these assets should be employed. As a rule, the questi
of whether these interests coincide is simply ignored in the textbook anal
..... ysi:;.<J fthe firmo If assets owners' interests do not coincide, then methods
of recondlig 'i:herii iiiustbe' considered. 5 " ......... ". '.'
Returning briefly to the notion of duality, it is possible to write a pro
duction function for any process, however complex, even for one incor
porating monitoring technology. It seemS unlikely that attempting such
an indusion here would be useful. The usefulness of a production func
tion in economic analysis lies in its economy in describing economic rela
tionships. If rriiiitring were included, this economy wuld be lost. The
inclusive function would depend, for instance, on whether labor is em
ployed by the hour or by the pieee, which, in turn, depends in part on
the market prices both for labor and for the output. Even if it could be
successfully formulated, the complexity of such a function would render
it unrnanageable, and there is little chance that it could be tested.
It is desirable to reexamine;albeit briefly, the relation between the in
dustry's supply function and the firm's cost. Applyingcompaiativestatics
to the received model yields the prediction that firm size will ncrease
when the market price of its product increases.. The service "use of office
4It is sometimes asserted that it is impossible to separately measure individuais'
contributions to ufput. Under the given information assumptions, workers should...
nevertheless;be rewarded for their aggregate effort and not for their time.
sIn some texts, particularly those regarded as advanced, the issue is further obfus
cated since productive factors are introduced generically, with no recognition of insti
tutional differences among factors.
69

space" provides a converuent example for analysis. In practice, the pre
cise nature of the product each of the finns in the industry supplies is not
constant over time and is not identical with what similar finns in the
industry are providing. For instance, some landlords supply parking in
conjunction with offlce space, some charge for parking separate!y, and
stll others whose buildings have parking space avoid getting directly in
volved with supplying tbis service to tenants by transferring the right to
the parking attributes to independent .operators. Q:>rrespondingly, there
is .no reason to expect that aII" exi,sti'ng.firins con:ril:!uting to the market
supply will get Iargei- as the Iila~J.{et pritc.of the use of office space. ses~.
The structure of fums is not independent of market conditions. Organiza-o
tions are structured to solve an array of comnion-property problems, the
severity of which depend in part on the price of inputs and outputs. An
owner of an office building who had initially offered parking along with
office space may choose tosell the parking component when office rents
go up. In that case, the size of the furo, aS"1l1easured by sales.or number
of employees, may fal1, Whereas we siill 'expect the aggregate supply of
office space to increase with iti price, the size of an individual fum in the
industry, as conventionally measured, may fan. It would be prematlUie ato
this poiot to analyze the vertical or horizonta! }ntegration of such firrns.
More appropriate is the question of how people contract in order to.max
imize the value of their resources, which will be discussed later in this
chapter.
.'. .
. . ...
Is there any advantage to retaining the costless-transaction world? Cer
tainly, exploring such a world may be appropriate for som.e anafyses. Ir'
is not appropriate or useful, however, for analyzing the fum. Costless
transacting dispenses with the problems of supervision and divergence of
interests among collaborating asset owners. In the textbook characteriza
ton of the firm, the inputs purchased in the market are assumed to per
form the tasks expected of them autoinatically and fully. 1'his would hold
true if the relation between' inputs and outputs were costlessly observ
able, because then input owners could be remunerated strict1y on ,the
. basis of their contributions. Such costIess observabilty isone of the fea
tures necessary and sufficient for costless transacting to existo Were trans
acting costless, for instance,' the employer 'could costlessly observe
whether or not workers' time was being efficiently used and could com
pensate workers accordingly. But any other method of remu~eration
could just as easily be 'implemented. It would not matter whether the
nominal unit of pay were taken to be the wage, the contribution to out
put;orany-otherunit;ea~hcold..be''C011vertedtoany'Ofthc'otherswith w -",
out slippage. A fum could function smoothly, but the market, where di
rection is strictly regl:'lted'hr F'rices, would perform just as smoothly. If
....7 0
The ald {irm and the new arganizatian

there s to be a model in which fums perform a non-trivial role, it must
incorporate the costs of transacting.
In addition to its Iogical problems, the received mode! encounters dif
ficulties when it is used to analyze actual observations intruded upon by
transaction costs. The cost of pedectly measuring quantities and prices
of inputs and outputs is probibitive. Were such measurements free - and
therefore error-free - buyers of productive services could pay for these
services in accord with, and in terms of units of, the resulting output.
When measurement of output is costly, payillg..input owners by some
observable measure of.~ertioll.riia:y bepreferhle:'to'paYhli::tilem:'by:::a: ...:...................
masure of their contribution. Thus, pay is often a function of input char
acteristics (e.g., dollars per square foot for rental space and dollar per
hour for labor). Consistent with Coase's notion of the fum, with current
usage, and with the view adopted here, pay by input characterizes a
wthin-f1rm operation and pay by output characte:t;izes an across-flrms
operation. The choice between the two types of transaction depends on
measu~ment costs and indirecdy on market prices. 6 The fum, then, may
expand or shrink in the absence of any change in technology. lf this is
correct, the one-to-one relationship between technology and firm cost is
refuted~
Part of theappealorthe. conv.entional.cost .function lies. in jtS'\.l.se p{,

terms that are commonly encountered.in practice. As the preceding dis
cussion reveals, however, the correspondence between the terms used in
the theory and those in actual use is ephemeraL The entrepreneur is not
the sole residual claimant beca use the reward received by those who
transact with him or her are not fi.xed, and because the scope of a firm's
operatons depends on factorsJlurelated to the productiQt\ (unction. It is
difficult to escape the conclusion that the conventional cost function
model is ill-suited for explaining organizational (and probably any other)
real-world problems.
It is by now a well-accepted idea that evaluating the contribution of a .
factor .to output is costly, and that consequendy such evaluations are
not expected to be performed with complete accuracy. In the presence of
inaccuracies, factor owners who rent out their assets for a given rate of
pay will gain by reducing their effort. nder dispersed factor ownership,
.lhe transition from the production function to the cost function inevita
bly involves shirking. Maximizing individuais must take into account the
effects of such shirking and are expected to devise methods to Iower the
associated losses. Therefore, when the evaluation of .performance is
"6ori:ii:e'tht"cl:m:tacttypes discussed in Chapter 3 in conjunction with farm ten
ancy, only that of sharing is output-based. Both the fixed-rent and the fixed-wage
contracts are input-based.
71
costly, the choice of contracts among cooperating factors becomes sig

nificant.
are costly. In o~der to incorporate such costs into the analysis, a precise
definition o( the term "market transactions" must fust be provided.
Two distinct definitions of market transactions are seemingly consis
tent with the common understanding of the termo One is that they are
properly and fully priced transactions and therefore free of distortions.
In other words, in such transactions individual buyers and sellers bear
the full costs of their actions. Whereas Coase does not explicitly suggest
that transacting in the market involves deadweight losses, the asserted
costliness of ~uch trans:i!-ctions mus(iIpply.that some marginal equalities
of the zero tJ;ansacti'on cost ndel are violated. 8 This view of market
.',trnsations impiel by Ca'aseseems to ~et()be a correct view of reality.'
Indeed, it is my contentiori that the Pareto onditions for efficiency are
violated in every transaction. 9 - It does not seem useful, then, to define'
market transactions as those free of distortionif that set ~s empty.
The other, ,riot well-recognized, de:finition of market transactions is
that, once conclu~ed, such transactions leave no obligations remaining
between transactors, that is, ~ese. transactions are go:verned by caveat
emptor. This d~finition'of market transactions.is attractive not'only be
cause price alone affect~ buyers' decisions but also because of its precise
correspondence with a useful and clear-cut practice: Howev!'!~:l~q:welilt:,
emptor transactions are costly ad gov~rn onlr a small fra'ction of the
total volume of trade. Although it is attractive, this definition of market
transactions is often in conflict with the current indiscriminate use of the
not uniquely' defined termo W.hen using die term; it iS ~ecessaty to make
clear one's intend~d meaning and to stay.alert t$?' t~e pitf~lls of u!:rent
usage.
It is useful to relate the notion that market transactions leave no obli
gations between the transactors to Cheung's (1983) analysis of the firmo
Cheung points out that the organizations falling under the label "firm"
are diverse; that the wage contract does not suffiiently characterize
them; and, most iInportant, 'that there is no satisfactory operational
definition of the termo He suggests that economists should' abandon the
firm as a vehicle of their analysis and instead 'should focus' on contracts.
Nevertheless, Cheung retains the distinction between firm transactions
and market transactions. He says that the " 'firm' is ... a way to organize
actvities under contr,actual arrangements that differ from those of ordi
nary product mar,ket'S" (p. 3). Given costly transacting, organizing activi-
THE FIRM VERSUS THE MARKET:

A FALSE DICHOTOMY
Both Coase (1937) and the traditional approach group interactions

among people' in two categories: those carried out in the market and ,
those carried out within the fum. In neither case is the classification ex
::,"',',:::,'::":liast1,ve:'::bf'iii::hotK:'cas~'s::eaai'of:ihe:t\iiio"'ty.pes"o!,.,in!eraction. ~eelJ1s
hnportant alone andstarids in sharpcntrast tothe othei: In bothcases

activities within firms, unlike activities in the market, require organiza
ton.
According to the traditional approach, entrepreneurs buy inputs in the
market and transform them into output, which is then sold in the market.
As I have pointed out, under the assumption that fums ar entrepreneurs
possess perfect knowledge of market conditions, of all attributes of in
puts and outputs, and of the production function, organization within a
firm is innocuous. Any operation that workers or other resource owners
accomplish within the firm is, in essence, an exchange that could just as
f!~~ilr ..b,f!pert()r~<::~)n"tl).~,m,axket without recourse to such organiza
tions. Precisely the same results .lhat are obtained by firms can be
achieved without recourse to such organizations. For instance, any
worker may operate independently,- remunerated strictly by the value of
output (from which damage to equipment owned by others is netted);
since the true value of any component of a worker's contribution can be
costlessly assessed, the worker can also be directly remunerated for it.
..... T here is no compelling reason for workers to become employees, re
warded indirectly by the hour. It is true that there is no harm in em
ploying workers by the hour, because under given conditions their hourly
contribution can also be evaluated costlessly. But, as was previously
stated, this equivalence between workers' employee status and indepen
dent status is precisely what renders such organization innocuous.
Coase adcipts a radically differeUt'pproachn"attempdngtOexpliri
why some activities are in the market, guided by prices, while others are
in the firm, guided by orders. He argues that transacting in the market
incurs the cost of discovering the appropriate prices and that operating
within fums, where the employer has the right to arder the employees
_ ~pd to restrict their actions, is a means of reducing thecosts} Coase does
not follow through fully on his own assertion that market transactions"
7 Coase does not explain how employers acquire the knowledge underlying ther
orders. Since such knowledge must be based on prices, employers' action does not
obviate the need to discover prices.
72
BNote that becaQse of the costs of conductng market exchanges, the net amount
the seJler receives' is less than the total amount spent by the buyer: Money prices
convey only partial information about the tenns of exchange.
'
"..... """,,,~The,,:vi<:w,,that,,aside.f:rom,QCcasioRal,~extemalities,::"maJ:Ckets, ar.e.fr.ee ,of. distortion ,'" ,
is pervasive, I believe 'that' this view is at the root of ali sorts of confusion, especally
when particular :distortions are considered to be exceptions and therefore to call for
exceptional measures, where,as in reality they are instances of the general case.
73
Economic analysis or property rights
The old firm andthe new organization
ties in the product mai'ket - at least non-caveat emptor transactions

of the desired Qutput: flushing a radiator, treating a sprain, or fucing a
leak. Charging on this basis seemingly places the transaction in the mar
may require iIDposing restrictions on the transactors, restrictions that are
of the same character as those in the firmo
ket. The other type of contract, however, seems to be an employment
Since caveat emptor transactions are costly, they are used only under
contract since it chrges by the providers' time. These contracts place
narrowly circumscribed conditions. Would-be buyers of commodities
such transactions in firms, alheit very short-lived ones. Pigeonholing
whose sale is subject to caveat emptor wiU not part with their money
transactions as being in the market or within the firm proves not to be
before either inspecting the commodities sufficently to convince them
, very illuminating. It would be nstructive to follow Cheung by attempting
selves that they are not throwing their money away or, alternatively, satto explain the types of contract that may be expected to be employed.in
isfying themselves Qhheseller.s':reputatioo;':which:req:uires::that:the:seU,:":::,, :'. :':,' ::::::::.:iffer~nt.sJ::uations. Given the objective of this chapter, however" it
,. ersh;:}ve"Pl'ev,iollsly'investdinthat reptation.J~ Transactions in which . .
.would be even more instructive to determine what characterizes C6littcts .
'. tht belong in firms arid t analyze the structure of firms. With tbis in
.sellers' incur obligations (i.e., guarantees) are an altrnatve to caveat .
emptor transactions; such transactions are, however, accompanied by re
mind, the following section begins with a brief description of the existing .
strictions on the transactors. The within-firms transactions, then, are not
models of the firmo
unique in imposing restraints on the participants.
In most transactions, particularly those which are highly valued, the
CURRENT THEORIES OF THE FIRM
seUers' obligations continue after the sale is completed. This is the case
when sellers guarantee sales and when they become liable for product
Alongside Coase's (1937) groundbreaking contribution, e~onomiclitera
ture offers two additional prominent transaction cost-based explna
malfunction. Indeed, these two types of obligation often apply simultane
tions for the existence of firms. One is by Alchian and Demsetz and the
ously, highlighting the fact that different attributes of a transacnoJ;l are
other is by WilHamson and by Klein, Crawford, and Alchian.' More re
subje'Ctiodistinct problems and are differently constrained. These obli
'" 'gatnsare pattsof crttits tnat specHi Wnat 'hf -rhe parl:es' a'gri' .
cent transaction cost-based insights are offered by several authors.
Among these are Grossman and Hart, Hart and Moore, Holmstrom and
to cede to the other. Contracts may also restrain the partes in order
Mlgrom, and Milgrom and Roberts.
.
to enhance their ability to meet those contract obligations that are not
As was stated, Coase views the firm as a means to economize on the
discharged at transaction time. The use of constraints means that price is

use of prices. The effident collaboration of resource owners 'through the
not the sole means of allocating resources. Implementing and polidng

market requires the determination of the prices of factors and of interme
constraints requirt!"an organization, and different kinds of constraints

diate products at whch the parties would exchange. Such determination
require correspondingly different orgaiiztions. Cheung's suggestion

is costly. Coase argues that organizing production within firms, where
that we study the contracts governing what are considered to be firm

the employer instructs the employee as to what to do, saves on these
operations should be extended to indude alI constrained operations,

costs.
whether they are in the firm or in the IParket.

Alchian and Demsetz (1972) hypothesize that the firm is a means to
The employment contract, wherein a worker agrees to be ordered by

realize the economies of team production, and that members of the team
his or her employer, is the one Coase singled out as characterizing the
arebound.tothe,firm.bya..nexus,of contracts. In team production, the
firm. This contract, however, is just one of an array of methods of con

output of one member cannot be separated from that of others, but the
straining transactions. In the discussion of the tenancy contract in Chap

effort of team members is observable. When the inputs are used within a
ter 3, I showed why the share contract and the fixed-rent contract are no
firm, the firm consists of a residual claimant-supervisor, who:is ar its
more in the market than is the wage contract, even though only thelatter.,
center, and others whose output is difficult to observe and who produce
is said to identify a firm's operations. Many other contracts do not seem

the firm's output under a fixed-wage contract, supervised by the: residual
to faU neatly within any single category. Consider the services one secures
claimant. Becuse'the supervis.or is the residual daimant, he 01' she is
from an automobile mechanic, a doctor; 'r a plumber. Contracts for

properly induced to maximize the value,of.theourput'ofthe:teftlll":" .
.. suchservices take at least two basic forros. In one the charges are by units
Both Williamson (1975) and Klein, Crawford, and Alchian (l~n8) sug
lOStrictly speaking, even a transaction solelyguaranteed by reputation is not apure,

gest that efficient production may require inputs that spedaliie. in. their
instantaneous market transaction, since the other sde of such a transaction is the
specific endeavors. If such inputs are owned by different individuaIs, then
future retaliation against the owner of the reputation.

74
'.

the parties m~y be able to capture each other's specrnc values in the
. course of attempting to cooperate. They argue that fums tend to integrate
. vertically because within the organization the rationale for capture, and
consequently the deadweight loss associated with it, disappears. In my
own study (1982) I also offer a reason for vertical integration. largue
that part of the value of those intermediate outputs that are costly to
measure may be captured when their ownership is transferred. Therefore,
fums that produce such intermediate outputs are likely to integrate up
stream to avoid.,the ca,pture losses. This explanation of vertical integra
,tion i~. ql:!.it~_siInilar, to, but!10t:identical, with, thatJ>L\Vil1iam~<;)ll. and
thiitofKlel:i,<:rawfrdrid'i~:1chial.. ,.~,,,~;,:':r,,,,,,~~,
, , ' "~,i,
. Grossman and Hart (19.86) and Hart and Moore(l:990) focus on asset
ownership. They state that'such ownership gives the owner' control over
ali its attributes and makes him or her the residual claimant to the income
they generate. By combining the ownership over arrays of assets, the firm
saves .on contracting costs with individual asset owners. By merging, the
fum also gains controI 'i:hat it does not hav.~ under exchange between the
met.:ged fum!t. Milgrom and his associates approch the complexity of
the incentive structure from a different angle. Both Milgrom and Roberts
(1990) and Holmstrom and Milgrpm (1994) elllphallize.themul#dimen
sionaliiy of tasks perform d by workers and the corresponding need fQl;'
coordinated incentives and constraints when the workers operate as em

ployees.
.' .
'
.
In and of theJl'se1ves, t!tese expla'nat-ns satisfctrily'aecount neither
for the scope of the firm norJor ts.b~undad.es. There.appears to be no
empirical work that demonstrates tht the size and structure of actual
firms conform to these models. Casual empiricism does not provide sup
port for the notion that these are the main reasons behind the existence
of actual firms. For instance, it is difficult to see how the size and struc
ture of both Safeway and the comer.grocery store can be attributed to
forming .prices, tcam production, specific capital, the measurement of in
termediate output, or the tasks' multidimensionalty. Regarding the
boundaries of the firm, Cheung' points out thembiguity sric::iidirig
what constitutes a firmo He argues persuasive1y that if we cannot dassify
activities as firm or non-firm, then there can be no operational theory
of the fum: The preceding' explanations do not seem to be successful in
determlning the boundaries of the firm.
I propose that the guarantee function of equity capital is subject to a

scle economy that might contribute .to an explanatioQ of both the size
and the scope of the firm, In addition, this function provides a means of " ' ,
"'defiiIiig'tne'oiindiiiiesofini:!'fiiiri:'Gaianteeproblems'iise'only'wnere"",'
var.iability is present. The allocation of variability must be discussed be
"o
76
fore the guaralltee function can be tied to the fumo It is appropriate,
however, to first relate the guarantee function to the Coase Theorem.
THE COASE THEOREM, VARIABILITY, AND LIABILITY
In his momentous article on social cost, Coase (1960) demonstrates what
has come to be known as the Coase Theorem: When property rghts are
well defined and transacting is costless, resources will'be used where they
are most valued, rega.r:d1(!s.s o(:vv.l1:~~ll"<?!..!h.:~.~~.n..s.~.<;:~<>,r.~.. ~~~~~~Ji.a.l:>W!Y, ....."..,,
,Jr.. .bJ~,,r .~er::effectsnn:theothedl1n:'the.courseofhtsdm:dhstt!nion~'

>'h~bdiigsthe questionof liability tO.theJore but fails t address tne ques
tion of the conditions under which liability problems arise. Variability is
a necessary condition for liability. Variability can always be eliminated;
it ean be explicitly priced, ot it can be taken care of by sorting commodi
ties into homogeneous groups.,Liability arises only because variability is
too costly to elminate. A producer of bottled soda would not stay with
his product if he knew that all the bottles would explode. A slight but
uniform defect can be thought of as a lability. Still, the problem such a
liability ereates can be fully resolved ahead of time by adjusting the price
of..theproduct. In both cases, the product is not subject to variability,
.. apd1i~b.nyproble.m~ .are .abse.nt..asare.problems .of.delineating property
rights. Despite the fact that variability is present, rights are also deline- .
ated when each specimen is priced according to its attributes.
If property rights are to be well defined, the person who benefits an
other must be fully rewarded by the beneficiary; conversely, the person
who harms a,oother must ful1y compensate the harmed person. By this
criterion, a contributor to variabil~ty. .mpst assume the full effect of his or
her actions H rights are to be fully delineated. This condition is satisfied
in two of the cases of cooperation between owners of land and of labor
discussed in Chapter 3. In one, land but not labor is unHorm, and the
contract between owners is of fixed rent; in the other, labor and labor
effort but not land are uniform, and the contract is of fixed wage. In
these two cases transaction costs are not assumed to be zero, but property
rights are well defined because the method of pay S'tisfies the conditiori'
that the factor owner who can affect the outcome bears the full effect of
his or her actions.
In each of the two cases, rights are well defined only beca use the 'ppro
priate contracts are used, that is, the contracts are for particular assign
ments of liability. Were the fixed-wage .c:ontract chosen when land is uni
form but labor is not, property. .rights would also not be well defined,
.. 'However, costless transacting is a sufficient condition for clearly defining prop
erty rights, renderng redundant the requirement that property rights be well defined.
77
The old {irm and the new organization
nor would resource use be efficient. The allocation of variability here

determines whether or not rights are well defined. Therefore it is mean
ingless to state that if rights are well defined, resouece allocation is effi
cient regardless of who is liable (or who bears the effect of variability).
In general, both parties to a contract can contribute to the variability
in outcome. Since the individual effects cannot be costlessly isolated, as
a rule property rights are not well de6ned. A fundamental proposition
here is that as the effect a party exerts on the yalu,e, of, th~ ():utcome n
creases, rights wiII bebetter defined if tha,t party ssumes a lai"ger share:,.
of the v<lriaoility of outcome, .T.his s tb hypothesiz~dgui41I}gprinciple:
behind the formation of contracts that govem 'the operations of an orga~
nization, as well as behind the determinarion of when a party will ~sume
a larger share of the variability, thereby becoming more of a residual
daimant.
contract" does,not, however, connote just a single arrangement: Workers

may be hired on a daily basis and paid a spot wage; they may be hired
for life for a fixed sum; or they may be hired on some intermediate basis.
In addition, the employment contract may contain such features as esca
lation cIauses, schemes for severance payment, and requirements for ad
vance notice of layoffs. The "fixed wages" employees receive are not
truly fixed either per unit of output (as duality clearly implies but seldom
explicitly acknowledges) or per unit of time. Each of these contracts ex
poses workers to a variability that differs from that:9fanyo:fi:he'other' .................. .
.contrcts.Employers are likewise sbjeiio~a"vriabilty that depends on
the particular contract'chosen, because they are exposed to the comple

mentary or remaining variability.12
The considerations that apply to labor also apply to both the prices
.. a nd the quantities of all other purchases and sales. Regarding variabiljty
in quantities, firms may purchase their trucks and then bear the effects of
varying durabilities across trucks, or they may rent the trucks, shifting
the variability il) longevity to rental agents. In the same fashion, employ
ers who pay uniform hourly wages to non-uniform workers bear the vari
ability in performance. among the worJq~r.,,:whereas workers who are
pid &yi:h 'piece be~u more of thie .Yli~i~l>ility .in tll.eir. .()'Yf!. p~r.fQrl."O.ane ..
As a final example, buyers'bear the effects of variability in product qual
itY for purch,,!-ses subject to caveat emptor; if the purchase is not govemed
by ayeat l Jl,etor, sellers bear at least some of that variability.
If the function of ownership is indeed to assume responsibility for vari
ability i rder to increase joint income, then holders,of corporate equity,
who are usually passive participants in the operations of their firms, are
not expected to become the residual daimants to -the systematic compo
nent of the variability in the operation of a firm. Rather, an array of other
transactors who can affect the outcome is expected to assume the effects
of components of the variability. Fire insurers are expected to become the
p.rimarY residual claimants of the effects of fire; the wholesale supplier of
acommodity is expected to sign a long-term, fixed-price contract that
guarantees the constancy of the price of his or her commodity, and thus
to become the residual daimant to fluctuations in that price; the buyer of
bad debts is expected to become the bearer of the variability in the repay
ment rate. Each of these resource owners who transacts with the firm is
expected to become at least a partial owner of the line of activities he or
she controls. Similarly, a salesperson rewarded by commission is more of
an owner of his or her operation than is one who is paid a fixed wage...
..".Anin-house.lawyeris'less"auowneroftheoutcome of legal action than
THE ALLOCATION OF VARIABILITY AN.D

THE RESIDUAL CL2\IMAN:r
Each of the operations in which a firm is involved contributes to the

income variability to which it is exposed, and each of its contracts with
the various parties it deals with allocates the overall variability b~iween
them. A party is expected to assume more of the variability, that is, be~
come more of a residual daimant as itS" effect on the mean 'outcme in
creases. Efficiency is the Sole motivator for this hypothesis (since the
model here assumes risk-neutrality): Parties re expected to assume more
of the variability when their gain from affecting the outcome increases,
thereby guaranteeing a larger share of their own actions, which could
otherwise become damaging. When the parties guarantee their actions,
their incentive to take advantage of exchange partners is curtailed.
Consider the variability surrounding the operations of a firm engaged
in the productin of a commodity. Ultimately someone must bear the
effect of every component of the variability in the outcome of the firm's,
activities, just ..as someone must bear the outcome of any action. Any
activity resulting in variable outcome must have one or more residual
daimants. Many input owners can affect the mean outcome of a firm's
operations; depending on their contracts, ,each will bear some of the ef
fect of the variability associated with his or'her inputs. For instance, as
.... vvlls previously discussed, a firm that buys fixe insurance, as many firms
do, is allocating.some of the variability it faces tothe insurer.
It is conventionally asserted..thatthe..contraetuaroblig-atonemploye:rs,
assume to pay wages to employees insutates'th~~e ~wners of labor ser
vices from the effects of variability: Emplqyees Ieceive fixed wages and
employers bear the entire variability in firm operations. The term "wage
12 Each contract will induce a different performance. The total variability also de
pends on the contract chosen.

s an outside lawyer pa1d by the hour, who' is, in turn, less an owner of
the variability in outcome than is the lawyer operating on contingency.
The modeI here is testable. For instance, it yields a prediction as to the
type of legal services one is to employ. In the case of someone who seeks
such services, the more he or she can affect the outcome by his or her
own behavior, the greater is that person's expected share in outcome vari
ability. Thus, an explicit dispute about money between a firm and a party
it deals with that depends primarily on the legal aspect of the argument
is expected to be handled by an outside lawyer on a contingency fee hasis,
notby~tlie:fiiiiY'swn:in::fiouse:eounseC:':::"::::::::::::':::',::::',:::::::,::::::::::::::.:::::,." .., "..
It does not seem possibIe to statepreCiselY'whichactivities':shbId be'- ..
designated activities of the firm and which sho{dd noto Only when trans
actions are subject to caveat emptor is the separation between transactors
complete. However, caveat emptor transactions are the exception rather
than the ruIe. For alI other transactions, contracts impose restrictions on
the parties' be'havior, to some extent creating within-firm transactions.
The strength of such ties is not uniform, and thus some contracts are
more within firm than others. Here, too, it is useful to try to determine
conditions under which the ties will be strengthened or weakened.
The preceding illustrations of the allocationof variability faced by the
fir II1 .s.h.a.re. ,:l ,c()l11IIl()1l..tJ:1r.elicl., .:E.llc:h ,paJ;"tY Jha.t .aells, ,se~vices to the firm
and can affect the outcome of the collaboration is expected to assume
sme or alI of the associated variability. Moreover, to d~e extent that a
party transacting with a nrm affects the product it produces, it is also
expected to guarantee the outcome to the buyers of the product. Not
..... discussed thus far is the question of why the firm itself would assume
any variability. The next section contributes to the theory f the firm by
addressing this' quest{on.
THE FIRM' S ABILITY TO. ASSUME VARIABILITY,
AND THE ROLE OF ITS EQUITY CAPITAL
In the discussion of the assignment of variability among transactors, I

have argued that to maximize the gins from transacring; theiifiewards
individuaIs receive must be commensurate with their net contributions.
The net contribution is not always positive. It may even be nega tive, as
when one party causes property damage or harms another. The leveis of
the transactors' own wealth, however, constrain their ability to compen
sate others when tht::.r.net contributons are negative. For nstance, the
payment promised to a factor employed for a fixed rate of pay may di
verge at any time interval from the spot market price of that factor. When
the reward falls short of the party's contribution, that party must agree
and be able to finance the difference; when the reward exceeds the contri
80

bution, the oth~r party to the exchange must agree and be able to finance
the difference. The parties, then, must have the means to make good on
the guarantees, and people differ in that ability. Because these problems
seem most crucial in the exchange of labor services, I will take my illus
tration from the practice of contracting for labor service.
It might be advantageous for woi:kers to become the residual claimants
where their actions could impose a high cost on others. For example,the
product they produce might be subject to liability pr~blems. Workers
who operate valuable equipment, such as jets,mardamage it,.or ~f they
use power tools their actions couIdharm fello'-Y V\(orkers. Th-e value.of
theassociated variability,however, inight6e hirgerthan worker,s are able
.. to guarantee. As a ruIe, suppliers of labor services are severely restricted
iil their insuring ability, particuIarly since agreeing to use the value of
their future labor services as a lien is not legalIy enforceable: Doing so
would constitute slavery. Therefore, although the potential gain from
guaranteeing their actions may be large; many workers do not have the
wherewithal to effectively do so.
This dearth of guarantee capital ~n be rectified by trade between the

workers who are short of such capital and individuaIs who possess it.
However, the owners of capital will assume the guaranteeing role only if
they can also constrin the workers such that their incenti'lte to induce .
liahility problems, to damage equipment, or to injure fellow workers is
curtaled; otherwise the guarantee function cannot pecorne.profitable. '.
Here again the wage contract ancl the accompanying supervision mar be ',' '
u~d.
,
The capital involved in such a guarantee is "equity" capital. I defin
the firm by its guarantee capital and by the scope of its guarantees. The
scope of the firm comprises the set of contracts whose variability is con
tractually guaranteed by common equity. capital. The {irm, then, is a
nexus ofoutcome guarantees. This definition is ope~ational. It is possible..
to determine when the amount of a firm'sequity capital that guarantees
its operations has increased or shrunk. Moieover, the modeI can predict
when the firm will expand its operations. Finally, the modeI enables us
tOditefrtlntHhe oprimal debt-equity ratio for a firm - a question that is
not within the scope of Cheung's suggested mode of investigation.
Consider a publishing firmo If, for a fixed' amount, it co~missions a
writer to create a manual, then the variabllity in the sale of the manual is
entirely within the firmo If, however, it signs a straight royalty contract
with the writer, onIy part of the transaction is in the firm, since the writer
will bear'a'share of the variability in the sale,of the book. Assuming that
the Iatter ventuie'scceds~aiia:'tht"i:he"pubHs1iei'proeeas'i:opubisli ..
more of the writer's work, the model here. predicts that the new contracts
will gravitate toward a higher (lump sni) 'hrance and a smaller royalty
8I
rate, placing it more within the fum. This is 50 because as the writer gains
the value that ontracts can generate from a given set of resources. For
experience, his or her work becomes more predictable; correspondingly,
instance, certain types of workers may be most productive if provided
the publisher is better induced to advertise the book if the marginal pay
with substantial on-the-job training. Workers may be reluctant to invest
in themselves unless their future remuneration is guaranteed. The suff

menU to the writer are smaller. Of course, the publisher has many other
contracts. Fixed-wage contracts with employees are mostly in the fum,
ciency of a fiem's equity capital is a necessary condition for it to be able
while contracts with salespeople' who work on commission are less so.
to provide such a guarantee for its employees; ceteris paribus, the higher
The latter are likely to be even less within the publishing firm when used
the outstanding equity'capital, the doser to optimal the leveI of training
for sales abroad, where the publisher's expertise relative to that of the
will be. Equity capital is productive, and its amount wll be expanded to
salespeople is less than for domestics.1~~k.P.1J.Us.b~!'$~~fJ,tr.1;$:::with::::: :::::the:point:at:whiclrthecost.ofexpanding it by one more unit brings an
bookstors may stipulateth~outrighnle of the.books or, alternatively, .

qual1y valuedimprovemilt in contract termS Df the. employed-factors,
including borrowed capital. The guaranteerng function, therefore, deter

a buyback of the unsold copies. Inthe latter case, the trarisaction partilly . .
remains within the publishing firm, since it bears part of the risk of poor
mines (at leastin part) the optimallevel ofequity capital. A firm may be
sales and, of course, the publisher must have the means for the buyviewed as the set,or nexus, of contracts guaranteed by the equity capital.
back. 13 Expansion is a final ingredient in the discussion of the publishing

The use of equity capital to guarantee the firm's activities is subject to
firm: lf it chooses to expand into an experimentalline of publishing, the

both economies and diseconomies of scale. These scale factors play a
debt equity ratio is expected to be reduced, since the chance of large

major role in determining fiem size. The fundamental force leading to
losses increases. Before concluding this discussion of the fum, I shall offer
scale economies arises from the fact that by its very nature the occurrence
of what is being guaranteed is random, and therefore the capital that
a "few more observations on the allocation of variability between firms

and some of the parties with whom they transact.
<4st~nds by" to guarantee one prospect can be used to guarantee others.
... Althoin owners oflabfcailot guarantee large potentiallosses, they

I
Put somewhat differently, providing a given guarantee leveI to a larger
can readily financeoneparticuiar'obligation"-'to's'Upplylabotsrvkes"!

voliriieorprospects requires a less than proportionate increase in the
when the market wage exceeds the contract wage - simply by showing
!
amount of gurantee capital as long as the prospects are not perfectly
.. ' up for work. 14 Owners of labor are, therefore, more likely to enter into .
correlated. ThIs factor, then, favors large-scale firms .
Guarantee capital is not capital itself but the command over it. When
. 'contracts in which they ale required to guarantee the difference between
their market wage and their actual wage than they are to enter into cona guarantee has to be effected, the capital is transferred from its current
tracts that require them to guarantee otner possible effects of their behavowner to the beneficiary of the guarantee. The ease of .transfer is im
ior. The inability to personally gUtlrantee large losses appliell" to many
portant here, but it is costly as well. The guarantor who holds enough
professionals, including lawyers. For this reason, one expects only large
cash to be able to effect the guarantee loses the return that an asset would
law firms to undertake large contingency cases requiring a substantial
have provided. On the other hand, the guarantor who holds physical
amount of legal services without the guarantee of reward. in particular
.i
capital will incur a cost of transfer when effecting a guarantee; as the size
.
of the payment increases, the cost rses per dollar of transfer, since the
cases. Smalllaw firms may be unable to finance the up-front legal services
required in such cases.
assets that must be used for the transfer are progressively less liquido
Owners of productive factors other than labor may also lack sufficient
Although the .guaJ.'antoris.assumedto berisk-neutral; because of this
wealth to guarantee their actions fuUy. Any asset may be used to provide
lquidity problem he or she is expectedto act as if risk-averse.
a guarantee. Equity capital, however, is a factor specializing in guaranDiseconomies of scale to guarantee capital arise whendifferent indi
viduals take part in assembling it. lf individuaIs simply guarantee each
tees. ..
The buyers or employers of any productive factor are able to affect the
other's prospects without constraint, then the incentive for caution is
outcome of the transaction and are therefore expected to guarantee their
weakened. They may pool their <:::tpit:al, as is done in stock corporations,
actions. The better their ability to provi de such guarantees, 'i:h higher
thereby resolving the free-ride problem. But whenindividual shares are
....13 .... .:...... .,..
..... ...
. .
.
.'
..
small, ownership and control tend to diverge. In either case,thisproblem
. The pu~hsher demonstrates such an ablhtr by slmply. selbng on credlt, whJch
limits the size of the equity capital and consequently firm size.
mlght expiam why wholesalers often seU to retallers on credlt.
. ...
. . .
.'
' . .
. 14In other words, workers simply continue to work at the contract wage, and the
One final pomt worth r,nentlOmng IS that the role of eqmty capl~al may
dlHerence between that wage and the market wage accrues to their employers.
be usefully contrasted wlth that of the share contract. The eqmty firm
82
'I
'"

uses its equity capital to guarantee its contracts with resource owners
who seU their 'ser,!,ices to it. Under the share contract, the cooperating
, resource owners simply divide whatever the outcome of their efforts
turns out to be, thereby eliminating the problem of guaranteing factor
remuneration. Sharing is likely to emerge when the provision of guaran
tees among factors is difficult (as is the case when the cooperating factors
consist mostly of workers) and when their output is easy to divide. The
contingency contract is a form of sharing common in legal work. Under
this contract no glJarantee c~pitaL, is nee.del, bJlt the spplier of legal
services must be able to self-finance 'theJegal wrk since remunerationis
'no.tgu~rlteed.~;<..
">".';
".,:,
, , : .
The formation of rights
;.<::-',
CONCLUSION
The received model -:- in whichfirms are essentially a production function

phenomenon and the miJ?num point of the average cost function deter
mines the competitive firm size - is '!fls!ltisfactory. If transa,ction costs
were zero, such firms .could a~ise but would be of trillial importance.In
any case, there seems "to be virtually no correspondence between such
firms and those actually observed. Following Coase (1937), r suggesnMt.
firms, r at least organizatons, result from pOl;litive transaction costs.
Contractors must agree on a formula to allocate the outcomes of their
interactions. When such outcomes are variable, the contractors allocate
the variability among themselves. I J:iav argued 'that stich no"cation" s at
the heart of organizatio!). The central principIe !1nderly.ing an o~ganiza-. _
tion is that the greater the inclination of a transactor to affect the mean
outcome, the greater the claim on the residual the transactor will assume.
Most activities are subject to many sources of outcome variability, and
different resource owners or sets of owners may assume different parts
of the variability. Guarantees are required in the presenc~,of variability.
Not all resource owners possess the requisite amount of capital to guar
antee their actions. This is especially true for those whose main asset is .
human capital. The ownrs of equity capital cooperte with capitl-poor
'owners. The former guarantee the contracts of the various other resource
owners within a single organization and may be viewed as the owners of
the corresponding firms. Equity capital is subject to both economies and
to diseconomies of scal~. These help determine the size of the fum.
84
. 1
Having analyzed the causes that allow properties to be kept in the public
domain, I can now address the issue of the f>rmation of economic
(though not necessarily legal) rights. It might be tempting to trace the
pattern of currently existing property rights holdings to its point of origin
to determine how and why it carne about, yet such an effort would be
futile. The ability to consume commodities, including those necessary to
'" sustain life, implies, the,possession of rights over them. Once this is under
stood, .it, b!eome,s, ,dear, that one,annot, ,e~pe~to ~;u.s,oyex,anY, !,:yj4ene,
of a pre-property rights state, since it is not possible to endow a pre
property rights state of affairs with meaning. In order to gain a toehold
on the evolution of property rights, one must start with the simultaneous
emergence of life and property rights and then consider a world where
some rights are already in place..One must resort to something less dra
matic than, but similar to, the physidst's big-bang .tho~y. Once some
rights are already in existence, it is possible to explore their evolution
with respect to changes in economic conditions and legal constraints.
Inferences about the creation of property rights may be drawn by
studying instances of anarchy or violent upheaval that necessitated radi:.
cal acts of rights redeflnition. A spectacular example is that of the Cal
foinia gold rush, described and analyzed in detail by Umbeck (1977).
Rich deposits of gold were discovered in California in 1848 when i:he
region was under U.S. military occupation, just days before the signing
of the peace treaty between the United States and Mexico. The gold
bearing land was not privately owned, and according to the treaty it be
came the property of the U.S. government. The United States flnalized
the transfer of power from Mexico to itself by abolishing the Mexican
law pertaining to mining rights on government land, but it failed to enact
'aliw"regulii:ing'i:heprivate acquisition of rights to mineral land until
I 866. Although the U .S. governrnent was the nominal Qwner of the gold
bearing land, it lacked sufflcient power to enforce its ownership or main
85

tain order. This impotence was exacerbated-in that the thousands of for
tune seekers virho descended on the Sierra foothills to pro'spect for gold
during this time of anarchy included many members of the American
military who had been stationed in California and had subsequently de
serted their posts. The gold prospectors were on their own, since there
was a complete absence of legal constraints governing the gold-bearing
land in California. Umbeck describes how these prospectors peacefully
established rights over the deposits in one mine after another and a~apted.
::::::::::::J:Q:u~w.ir.WJI~:!:;mces as they arose..
... . .
.
Mthough theprocess of formingright.s to ,the gold-:bearfuglandws", .
not subjeCt to much violence, it consumed substantial amounts'of r:" .
sources. Rights delineation. was difficult to effect under gold-rush condi
tions beca use of the high cost of gathering relevant information. The situ
ation the gold prospectors encountered was entirely neveI. Litde
information was available to determine the precise criteria by which dis
putes would be settled and ownership ascertained. Rights that 'are ini
tially in the public domain become well defined when it is )?ossible .to
determine who the ultimate owner will be.The conditions of the Califr-'
nia gold rush made this determination exceptionally cosdy.l In addition,
in the absence of state courts and of the knowfl procedures under which
they operated, predicting who would win any particular dispute was dif
ficult.
In spite of its appearance, Umbeck's case is not one of'a primordial
creation of rights, since itis concerned with the private creation of legal .'
institutions where those previously provided by the state have c'eased to
be available. In another study (1981) Umbeck is careful to note that as
chaotic as the gold rush was, some rights were defiued alI along in prac
tice, particulady those related to human assets and to personaI belong
ings, which induded guns,
A SCENARIO FOR THE EMERGENCE OF RIGHTS
..... As.:w:as.pre:v.iously.stated,it.s.notpossible to trace the initial emergence

of rights. Instead, I shall sketch a model that illustrates how, out of a
Hobbesian beginning, economic and legal righrs may have been formed, 2
While not necessarly "realistic," the model is capable of generating re
futable implications,
.
1 In an oral cOlll.liuriition, Umbeck stated that casual evidence supports the notion
that miners' eamings equaled their altei"tive wages. If so, the potential ;net valueof
the gold was dissipated.
...............................................:..H....::.:.:..,...::.. H":'
2 Attempts are often made to' describe the "ideal" state. This is not my intento
Rather, I am concemed only with how the state actually evolved.
. .

Imagine an onset of human history in which diverse individuaIs are
distributed over space. They use their strength and wits to survive, to
maintain ownership over themselves when not enslaved, and to acquire
possessions. There can be no legal institutions under these conditions;
therefore, legal rights do not existo People, however, have economic
rights over their possessions. individuaIs interact only to do better for
themselves. With the passage of time, they accumulate information about
those .surrounding them. As they become acquainted with one another's
patterns.' of behavior, they begin to engage in exchange. As long as legal
institutionsareabsent;, there existsno third party to compliridiviiils .
to perform; ali exchange agreements have to be self-enforced.
. Exchange, of course, permits specialization. In this scenario, it may be
assumed that individuais specialize in the activities in which they have a
comparative advantage and engage in trade whenever it is profitable.
Of special significance is the exchange between individuais specializing
- in production an~ those specialiiing in protection. The exchange
agreemen.ts must spell out.the amou.Qt of commodities the producers will
furnish and the nature of the protection the protection specialist will pro
vide. The latter requires.. .the..".-............
delineation of th~ properties that.
will receive
protectioI.
. in this Hobbesian world, the legal delinatiri fiighis' emerges

through self-enfQrced agreement~ between individuais with a compara
tiv'advantage-,in-protectiDn and tllOse with a comparative advantage in
production. The parties enter into such contracts because, given maximi
zatian, ead gf them- expects fo gain from the exchange. By delineating
rights and adjudicating disputes, the protection specialist erects the lega .....
machinery of the terrtory he protects and becomes the "ruler" of that
territory. The legal rights he delineates enhance the economic rights that
already existo Protection, the delineation of rights, and the erection of
legal institutions - which are some of the basic functions of the state
are thus seen to be consequerrces of the quest for private gain. Given that
delineation and protectio~ are costlytoproduce, however, they are not
expected to be carried out to perfectian.
Hobbes argued that people will nstall a ruler (whom he called a king)
to' protect themselves against their own predatory inclnatons. Rulers,
too, are expected to have predatory inclinations and might covet the
wealth of their subJects. Moreover, rulers are expected to be powerful.
Subjects, then, will seek protection from the ruler. Ir seems plausible that
there would be economies of scale in the protection of sets of neighbors
t againstea'Chother~mdaga-instoutsidersbythesa:iiieprotector.The ruler
,
specializes in the management of force. Given the scale economies to his
operations, protection specialists are Iikely to become more powerful
l.
1
j
l!
1
87
The farmatian af rights
than each of their clieIits. Before installing a king, subjects are expected
of law; consequc;:ntly, we see that history has been dominated by dictator
ShipS.5
to have formd a collective-action mechanism to control him. 3 This is a
non-Hobbesian feature of the model.
Constraints placed on the mler must accompany his installation. 4 One
THE COMMONPROPERTY/PRIVATE PROPERTY
likely constraint Is on the size of the force under his personal command;
DICHOTOMY
others inc1ude the form of his remuneration and the length of his tenure.
By employing him only partIy on the basis of a residual from the outcome
U~til recently, most economists had not explicitly adopted a property
of protection but mostly for a wage, and by also supervising him, subjects
rights framework from which to analyze economic problems. The earliest
will have made the ruler largely an employed, supervised manager of the
and most notable exception is Frank Knight's discussion of social cOst ".
protection effo,rtl'atherthanIt&:'piinia:ij"iesiaU"ardaim-aiit~::Maiiy:me(f~ :::.::::.:..: :::::::::(i9Z4l:In-his analysis of the useof roadshedemonstrateddds~ve!y:the ...
vrai:Y.::states in Italy iqtported a mIei (podesra), typkally for a one
. rle of ownership in lhe allocationof resources. In a similar v.ein, several .
deeades laterH. Seott Gordon (I954) analyzed the e.ommon-property.
yearperiod, and paid him a :6xedreward. Sueh eontrols and remunera
tion methods redueed the possibility that he could organize a takeover
problem of flshing in international (public domain) waters. After Knight,
and amass enough wealth to finance it.
and even subsequent to Gotdon's eontribution, economists did not con
It also seems plausible that there would be seale economies in the delin
cern themselves much with property rights. Economists; infrequent use
--eation of borders between neighbors and the adjudication of disputes
of property-rights considerations may reflect a belief that such consdera
tions are unlikely to produee useful results. Indeed, because most prop- .
that might occur. Moreover, the delineation of assets that the mler pro
vides for protection is eomplementary with that required for trade, so he
erty does not appear to be common property and, more~portIlt.. be
is the low-eost provider of the service. To lower the cost of erp.ploying
cause the transformation of what is c1early eonsidered common property
him,.hj~)~ubjects are expected to encourage him to provide legal services
ioto what is dearly considered private property is rarely observed,
aIlclt().fac:ilit.a.teC:()Iltl':ict: .tradf::,tl1:itis,. t.ra~.~t~.~~w.ak..~l' "!lS~ ..o. thixd7...
property-rights notions as expounded do not seem to be especi~lly use fuI.
party adjudication. The more standardized the traded commodities, the
The perception that property-rights considerations are not useful in
easier it is to trade by contraet and the greater is the advantage of the
the analysis of resource allocation seems to stem from.an all-or-nothing
scale economies to delineation. The more homogeneous an area, the
view of rights. Both Knight and Gordon assumed that property rights are
either present and perfectly well deflned or totally absent. They neglected
larger the expected state size; likewise, the more standardized the goods
produeed in the f.ea, the greater the expeeted seope of the state.
the possibility of an intermediate state in which rights are oo1y imper
fectly deflned.
In this scenario, where subjeets maintain tight control over the protec
tor, there must be a balance of powe;'between them, along with safe
The usual eharacterization of commodities as homogeneous entities,
often with only one attribute, makes it easy to condude that commodities
guards to preserve the balance. The safeguards cannot be perfect, how
ever, and externaI shocks due to such events as foreign conquest or
are either owned or not owned, and that there are no intermediate states
plagues wll upset the balance. The likely outcome of ao upset balance
of ownership. Such a view seemS to have been bolstered by the. assump
is a dictatorship. Once a dictator takes over, the state's economy is
ton that economic rights are equal to legal rights and that the litter are
likely to decline and stagnate. Nevertheless the initial dictator and his
either present or absent. Moreover, the position usually taken has been
suceessors are likely to ding to power. It is difficult to return to the mIe
thafpiopertyiightsarTirgely,:,:,:perhaps entirely - created and enforced
by government. Correspondingly, it has traditionally been asserted that
3The organization preceding the emergence of specialized protectors as described
it
is the government's fauIt that rights are left in the public domain, sub
here seems similar to organizations that have been observed in a number of anthroplJ.-..
ject to "open access.,,6 Knight and Gordon implied that if the govern- .
logical studies.
ment had turned roads or flsheries ioto private property,.1;he assocated
4 Given the English model that served Hobbes, the notion of a king (and a hereditary
regime) seems natural. The city-state of Venice, however, is a better example. Starting
its independent existence around A.D. 600 andencling in I797 with the Napoleonic
S().Qquest, Venice seems to be the foremost"c~xample of a constitutional, or rule of law,
regime dat did not turn into a dictatorship. The ruler position in Venice (th.e
dogeship) was not hereditary, and the powel?of the doge was rigidly constrained.
5 Another
relationship between pwperty rights and the state is posited by North
(I98I), who emphasizes the effect of clear delinearion.on.grow-th......:.....;...:.......: ..................

6 Similarly, it is often asserted that it is the government's duty to fllr piotect its
citizens against theft.
88
89
The formation ofrights
common-property dissipation would have disappeared. This view is easy

to accept if one believes that commodities are one-dimensional, either
owned or not owned. However, one then wonders why alI resources are
not always owned. The existence of theft has been recognized as an ex
. ception to the view that rights are pedectly well defined. The notion that
rights are not well defined in general has not, however, been pursued.
Rights that ar:e explicidy delineated by the state constitute only a small
fraction of alI legal rights. The rest are delineated contractually by their
owners in the process of exchanging them. In a contract transaction, the
buyer becomes the legal owner of the commodity sold by the selIer where
the contract (often supplemented by common law) delineates the attri
butes of the commodity. For example, a buyer of a ticket to a play is
legally entitled to certain services delineated on the ticket. Contracts,
ECONOMIC RIGHTS AND LEGAL RIGHTS
I however, seldom if ever delineate alI the attributes of the transaction.
The v~ry~ucce~s of Umbeck's study derives",ill"p,a..:r.t.~J!()mtheuniqeness::,:"' ':I,::,,::'The::huyer:bec:omes:the':~'Conomic'O\v.ner::notonIY?,.f the attributes the .
,of the Cahforma gold rush. Umbeckis ble to explain therole of violence'
!
sellerlegallyagrees to dehver but also of othets where, beca!lseoHactors
or, more ~c'curately, the threat of violence when the state's authority is
! such as reputa~ion, the 'seller chooses to deliver even ~ough he or she is
absent. H1S results, however, do not apply easily to more orderly circumnot legally obhged to do so. In the case of a play, the tlcket holder has no
~ta~ces. A~ a rule, in an ~lready functioning society, the creation of rights
legal recourse in case.of.a shoddy perform~nce, but he or she knows t?at
IS an ongomg processo RlghtS are created in the presence of state authorthe theater company IS hkely to do a good Job nevertheless. When buymg
ity, which has a con;tparative advantage over private individuais in the
the ticket, the buyer here becomes the claimant to and therefore the ecouse of violence and tends to discourage its private use? When a state
nomic owner of the expected leveI of performance.
authority is in place,. the role of allocation devices other than violence is
To clarify further the distinction between economic and legal rights,
greatly enhanced. As I shall argue, as economic conditions change proplet us consider inventions. Many inventions, especially minor ones, are
erty ri~ht~a,r~ .. t:()1t~tantl:y created and abandoned; therefore the~e is a
developed in~o sal~ble p~oducts without any legal protection. The own
need.for an analysis that fits contin\li I1g,5.111()<:>rh. (;h.aQg~~ jIl qp.,<;litim:'J.s. , ' " ,ers, ()~ o~?~r. .1n.~~I1tl()ns,sl~p'I:r..~o not seek l~gal .protection beca use the
It IS .useful here to reiterate the definition of economic property rights
.
cost of acqumng and exerclsmg s~ch protectlOn IS greater than the gain
o~fered in the introductory chapter. These defined an individual's "ecothey would generate. Still, inventors often gain from their unpatented
:-nomic rightS" Ovet an asset as the individua/'s ability to direct/y consume
inventions, at least for a short span of time. Although these inventors
. the s'!rv.ices ?f.. the asset, or to consume t indirectly through exchange.
lack legal rights over. t~eir ~nventions, they. have econo~ic rights over
. Th~ delmeatI(~n. of property rights is itself subject to indhdduals' optimithem. To pursue the dls:mctl0n f~rther, con~lder the afterhfe of patented
zatlOn. Descnbmg what the property is and protecting it consume re,
drugs. A drug patented m the UmtedStates IS legally protected for seven
s~)Urces, and perfect delineation is prohibitively costfy. Hence property
.,.... 'teen years. ~e:e, as in the general case, the legal ri?hts tend ~o enhanc.e
nghts are never pedectly delineated. Moreover, transacted commodities
the economlC nghts. When the patent for a drug expIres, genenc competI
. "have many attributes, and the rights to differem attributes of a given
tors often emerge. Nevertheless, the originally patented drugs typically
(physical) commodity or to different attributes of a transaction are not
sell at a large,premium relative to the generic ones. In spite of the loss of
all eqally well defined.
legal protection, inventors retain substantial economic rights over their
Legal rights, as mightbeexpected, are a major factor in terms of their
drugs. The reason seems to be that many buyers find it too costly to
effect on economic rights. "Legal rights" are defined as what the governascertain theproperties of the generie,medicinesandal'ewillingtepaya
n:ent delineates and enforces as a person's property. By granting legal
pre~ium for the established brand-name ones. The rights the inventors
r~ghts, the government participates in defining and protecting economic
retam result from consumers' ignorance.
nghts. Legal delineation"is likewise both costly andincomplete. More
over, individuaIs have a comparative advantage over the government in
THE DELINEATION OF NEW RIGHTS
various delineation activities and actually undertake many of them. Cor
.'The seed for the analysis of rights creation in an ~ngoing society was
r~spondingly,. individuais' behavior must be considered in the study of
.. .. nghtsJotmatrrc ................"" ........", ... ,.. ,..
planted in Demsetz's (I967) study of the Montagnais Indians of Labra
dor. Demsetz's poim is so simple that it now appears to be self-evident:
7EIlickson (1991) shows how rights may be established within existing srates, but
New rights are created in respon~e to new economic forces that increase
without recourse to third-parry enforcement.

the value of the rights. According to this view, rights in the sense of the
9
9I
Econo.mic analysis of property rights

ability to gain from property are largely. a matter of economic value
rather than legal definition. Demsetz hypothesized that as the value of a
common-property resource increases, people are more likely to establish
rights over it. SpecincalIy, he noted that prior to the Europeans' arrival
in Labrador, when the value of beaver pelts was low, beaver habitats
were held as common property. When the European market became ac
cessible, the value of beaver pelts increased and beaver habitats were con
verted to private property. Demsetz did not, however, explore the nature
of the break between the old and new concept of rigl;:tts; despite th!'!.nov
elty pf \1is observations, he failed to.. follow 'throgh:systematically.8
Though some economists(arid other ,~odal scient~sts)have applied.D tm1:"
setz's ideas,1:hey have not extended hismethodolgy. I have chosei to
expand and elaborate on this embryonic analysis of the formation of
rights in order to show how individuaIs tend to delineate iights routinely
more carefully as the value of these rights increases and less so as their
value declines.
'
People acquire, maintain, and relinquisn rights as a matter of choice.
For example, after the discovery of golq in California, indi~iduals found
ir worthwhile to delineate very accurately their iights to certa in gold
bearing propertiel>. After extracting only the relatively easily mined gold,
they often chose to abandon their claims. IndividuaIs take such. actions
directly in the private sector and indirectly, through the state, in the pub
lc sector. People choose to exercise rights when they believe the gains
from such actions will exceed the costs. Conversely, p"eople fal t exer- .
cise rights when the gains from owning properties re deemed insuffi-.,
cient, thus placing or leaving such propertes in the public domain. 'What
is found in the public domain, therefore, is what people have chosen not
to c1aim. However, as Demsetz pointed out, when conditions change, a
piece of property considered not worth owning may be newly perceived
as worthwhle; conversely, what was at nrst owned may later be placed
.
in the public domain. 9
IMPERFECTLY DELINEATED RIGHTS
Assuming that ownershp is not attenuated, the legal owners of commod

ides are free to exercise their rights over their cOIlI11odities in any (legal)
way they choose. What causes an mperfect delineation of rights, then, is
the choice of owners not to exercise alI of their rights. Since rights that
. 8In the sarne.article Demsetz seerns to vacillate between the positivistic view that
rights are created in res.p..()~.e..J9... e..!1lIli,.n~:Ut.iQ.nluand . .the.nnrmative.:~de.w..hae
govemrnent should enhance private rights.
. '" '. .'.
'Libe'cap 1989 analyzes common-pool problems i.n a similar vein. He focuses on
the political bargaining for the legal delineatio'n 9f rights..
are not exercis~d are placed in the public domain, it follows that people
deliberately place some of their properties in the public domain. Two
examples will illustrate this point. Both restaurant owners who supply
their patrons with "free" salt and owners of movie theaters who charge
the same price for better or worse seats, thereby providing the differential
free of charge, place some valued properties in the public domain. Pa
trons capture the rights to free salt by consuming it to the point at which
its marginal value to them is zero. Restaurateurs price menu items high
enough to cover the cost of the salt; however, they still re~~9\lil)1l.. ~1l.e. ........................ .
marginal unit of salt to the public dOl"l?:~i~h ..fiip.:C:~Jhe..~romaJ;ginalcharge
~-,-:t()paJ;J:<:ms isless thantne cost ofthe margiIlall.lnit. Theaterowners forgo
the differential in value between better and worse seats, while moviegoers
capture the right to the better seats by getting to the theater early enough
topreempt the occupation of such seats by others, with the value of wait
ing time of the marginal person in the queue equal to the differenc~ in
the value of seats.
Owners are ilot prohibited by law from exercising their rights and im
posing marginal charges for each of their commodities' atttibutes.
l- Rather, for some attributes they deem the returns to be less than the
. , cost.~,.::rI;l~J::9I'its of mposing marginal..charges consist of measuring or
metering and polidng. 10 We..t:e slllU::h.a.fge~iIl1P9~e.d, hj'; r.e.p,uudQJ;. the..
rest:auranteurs would take the form of higher prices, net of the cost of
the salt, received for meals they provide, whereas the theater owners
1
... ";
wiuld b rewarded by higher revenues from ticket sales. Buyers, of
!
_ course1 "Y0u1d have to pay extra for the higher-quality seats and sepa
rately for the salt they consume. Still, they would.gain more from the
lower meal price, and their aggregate net valuation of the movie would.. '.'
still be higher. The owners, however, deem some of their rights toa ex
pensive to exercise and choose to place them in the public domain. l l
Since owners cannot capture such values without incurring even greater
costs,.their actions are not dissipating. Even while seemingly placing such
values in the public domain, owners are still able to extract some value .
.Ih the case of niove theaters, for instance, owners who seU alI seats at
the same price have the choice of either preassigning the seats or leaving
them unassigned. It is assumed that they choose the method that yields
the higher net income, thus reducing loss to capture.
Cases in which oWners place attributes in the public domain, as in the
preceding iIlustrations, are ubiquitous. Salt is just one of the many "free"
10 An additional cost is that of convincing patrons that prices will not be rased after
they enter the premises.
f.. .. .. .P-Asimilardiscu1'Jsionis presentdinAlchianafld AlIen (1977); see especialIy Chap
ter 5. Because of the cost of exercising rights, the polcy prescription tbat alI rights
should be made private and enforced by the state is inconsistent with individual max~'
imization and is, at best, innocuous.
92
93

attributes available to restaurant patrons. Another s the opportunity to
eat at rush hour while not paying the owner a differential above the non
rush hour price, and while capturing the valued rush-hour time, as a rule,
by waiting or by rushing ahead of others. Patrons also do not' pay, on the
margin, for the amount of time they occupy space in the restaurant and
for the levei of commotion they create. Many similar opportunities are
available to supermarket shoppers, who can also capture the value of
better than average produce or fieat by increasing their efforts at se1ec
JiQ~hf!'Uy):w:b.~R],"e.nters..ofequipment are charged by the day, the inten
..H.......
H ..rH.H.....H..HlfH.. l.. d'.. .......H.... I...... f "b
12'
'.
Slty o use, ltse mu t~ ImenSlOna, IS a reeattn ute,

When the price people arewillin~fto py for servic::e iIiereases, returns .
from its better delineation also increase. If, for)nstance, the value of alI
theater tickets were to be doubled, the difference in valuation between a
bad and a good seat would also double therefore, the return from pricing
the difference would increase. Umbeck (1981) pointed out, however, that
in the new situation the gain from theft would also be higher, as people
would gain more from stealing the difference. They might, for instance,
buy tickets for the lower-price seats and then attempt to occupy the
higher-price ones. A prior reasoning does not yie1d the outcome that
policing expenditures should increase more slowly than the gains irom
... more.detailed.pr.icng. Although owners controI which methods to em
ploy for protecting their rights,13 thieves are also free to use whatever
methods they see fito Thus, the claim that rights will be better delineated
when the returns from mare accurate pricing increase is not always true,
Although one cannot make unambiguous statements regarding the
levei of delineation in reaction to the increased value of the asset, one can
make such a statement regarding the effect on delineation of a change in
the cost of metering or policing. The incentive to steal is a function of the
value of the target commodity but not of the costs of meterng or polic
ing. When metering or policing costs decline, there is no reason to expect
the gain from theft to increase. Therefore, if the costs of metering or
policing a service were to decline, rights to it would clearly be expected
to become better de1ineated and theprice oftb.l':ser:v:i~WJl.ld conform
more closely to its v~iue,14
. .. .... .... .... ..' ..
H
.....
12Chapter 3 includes a discussion of attributes that are placed in the public domain
for each of the tenancy contracts,
.
13 Cheung (1977) provides an example of an unusual policing method. He argues
that theater owners in Hong Kong underprice the more expensive movie-theater seats
in order to get them fully occupied and that such occupancy constitutes a relatively
cheap method of policing, In other words, ownei:iipay holders of expensve seats a fee
(in the form of a lower ticket price) to perform the policing funcion.
14 A confirmation of this mplication - observing a more detailed price"structUre'
when the costs of policing fali - would also confirm that t:ansactions consst of many
attributes whose leveis vary from one specimen to another; otherwise, one price per
dass of transactions would suffice.
The formaiion of rights

Regarding Umbeck's point, shouldn't owners have an edge over thieves
or other interlopers? On the contrary, examination of the detlnition of
economic property dghts suggests symmetry. Anyone who expects to be
able to benefit from an asset, be it its legal owner or a thief, is its (at least
partial) owner. When the market value of the asset increases, each of its
current owners - and possibly some new ones as well- may capture some
of that increase. Consistent with Umbeck's observation, there is no a pri
ori ~easQn to. e:li'pe.t that one particular owner will prevail.
O;msider herd f attl. Soppose that in the absence of then, it would
gen~rate ,nanhual incorleof $ioo.The legal o'Wnerspends. directlyand
indirectly, $20 a year on protection. At a cost of $5 a year, however,
thieves are able to steat .$10 a year's worth of cattle. Given his $:ro a year
loss to theft and $20 protection cost, the owner's net annual income is
$70. The total net income the herd of cattle generates is $75, of which
the legaI owner owns $70, thieves own $5, and $25 are lost. Suppose
tht the market price of beef goes up anq the potential net nnual income
.Erom tb.e herd is nw $200, Th~re is no a priori.reason to expect that the
net incnie of the legal owner will more,than double. At the higher value,
organized criminais may attempt to steal the entr.e herd, so it is conceiv
able that the owner's net inconie could even'drop'to'zero:
The increase in the i~come potential of t~e asset has another effect ri
.
the behavior of its owners. It increases .the aggregate gajns from coopera
!'
tion among tttem, whieh'in turn'is e~pCcted to lead t:Q-ttte better delinea
tion of the sset. In the cattle example, if the thieves were members of
nmadic tribes or transients, the' g.in to the owner oi buying them oH
and subsidizing their settlement in a faraway spot becomes higher. This
solution amounts to the implicit reallocation of the rights to the eco
nomic value of the cattle. In the process, the delineation of rights over
the cattle indirectly becomes clearer. Their gain from theft is thus Iow
ered. The sucess of such methods;'however, req!lires a supply of thieves
that is not very e1astic. Similar is the case where a store>s merchandise is
stolen mostly by employees. When the valueof t.he merchandise goes up,
the gains from cooperation between store owners and employees in
crease, and employees are expected to be made fuller residual claimants
to the income they he1p generate. This may be achieved by changing their
. status to part owners by using commissions as rewards.
i..
. Owners who place properties in the public domain may be able to
,
lower their losses by restricting access to thes~ properties. A restaurant
.
owner~ for example, will allow access to "free" saIt to patrons but not to
,......;. ..:nQn-p~irons.Thestate;'too;mayl'estrictaccessto~free 1"guods;"Recrei
ational fishing and hunting onen require licenses and may be subject to
catch limits. In some cases the state does not find it worthwhile to impose
rstriCtions. Here there is open access to the public domain, as in fishing
95
'
Economic analysis
of property rights
The formation
beyond states' territorial waters or in consumption of air. IndividuaIs, of

course, maxirlize subject to whatever constraints they face.
of rights
contract time tl}e fence, which is not sufficiently valued to be explicitly

mentioned in the contract, is simply ignored. Suppose, moreover, that
while the contract is in effect, a highly valuable use for the fence lumber
is discovered. Because rights to the fence are not well defined, a conflict
DISPUTES AND THE FORMATION OF RIGHTS
regarding its ownership mayerupt.
Owners of commodities may choose to retain them or to exchange them.
Regarding the capture of the rights placed in the public domain, I have
Exchange is subject to contracts to which the parties obviously agree.
asserted that whatev~r the cri teria are for capture, individuaIs will meet
It may be puzzling, then, that disputes over ownership erupt at all. A
them as long as the gains from doing so exceed the costs. These criteria
preliminary discussion of the effects of chan!Ses}l1.~()l1~.i~ion~.~~~re.~e.~ .......... ,.. ,... ~~~...~~.~.,t.?...fi:~ .. ~J:1(;: ..J?.l:l1"ticulars of the situation. In the case of the fixed- .
lineation is- incomplete will reveal'Y1l.atcauses.. dispute~an(Hiowthey'a:re"""""rentterrant'whos~'con:tractdes not constrain the extrationlevel of soil
., .. , se'ttled..
.'
.. , .............:... '.
."
".
.
nutrients, the criteriori' iS theappropriate method and intensity f cultiva-
Commodity owners decide whether or not to place attributes in the

tion; in that of the single-price movie theater, the criterion is time. 15
publc domain. Theater owners, for instance, may price ali seats equally
These cri teria, however, do not necessarily remain intact when the gains
one week and then adopt a more detailed pricing scheme the next; they
from capture increase. In prticular, parties who initially only implicitly
are free to alter which rights they retain and which ~hey relinquish, berelinquish rights to an attribute to the public domain may cIaim that
cause they continue to own the asset. The sale of theater tckets constithey retain partial or complete rights to the attribute and may attempt to
tutes a remal contract of space in theaters, and owners can form new
compete with their transacting partners in recapturing the relinquished
contracts as' older ones expire. When owning an attribute becomes prefrights.
erable to placing it in the public domain, the commodity owner wilI make
In the case of a contract that does not clearly delineate some rights
whose value has increased, a conflict may emerge. In these cQntracts the
the appropriate contrac.tchanges at contract-renewal time. However, an
attributethat .is l1.tll,!! ..p.llblic. .cJ.()Il1:a~I.1. -:w.hile .th~ .o.lt9p..VaU~.1?tUUn........ own.ers. oLassets relinquish to their exchange partners subsets of their
force can be claimed only by spending resources.
rights to the.assets. The initial owners, whose actions (or, more likely,
In the case of theater tickets, the status of those seats that increase in
inaction) hav~ implied that they have relinquished rights to an attribute,
value while the old price is stll in force is dear. The advertising of particmay now contend that these rights are their own, but their transacting
uIar pricing schemes for specific durations is part of the contract between
partners may make the same daim. These considerations apply most
an owner and patrons. There is no ..dispute here between owners and
clearly when the parties operate explicitly under contract; they may also
patrons; any breach of contract aside, dUring the adver!i.~~p_period the
apply to informal contracts and to relationships such as those between
rights to tickets at the old pric are relinquished by the owners. These
neighbors. Consider neighbors who possess a hedge that separates their
rights are not relinquished, however, to particular individuais. Since the
properties. InitiaIIy they may have elected, at least in practice, to leave
value of the seats is higher than it was before, competition among patrons
the hedge in the public domain. Changes may induce them to attempt to
for these seats will intensify. When the value of the seats is higher, the
capture some attribute of the hedge, however. For example, arare bird
gain from avoiding such resource-consuming competition is also higher;
may have built a nest there. Here, too, a dispute may emerge as the value
of property previously placed in the pubIic domain increases.
sirtt these rights are already in the public domain, it is not necessarily
.
The transctorsjrisideredheie:reoperatinguIlclera contract, possipossible to avoid competing for them.
In the polar cases of a fuIIy owned commodity (or of ful1y owned attribly an implicit one. One issue for them to consider is how the court might
butes) and a commodity placed entirely in the public domain, the comallocate the disputed rights and what costs they would incur in the at
modities continue to be owned and unowned, respectively, when their
temptto influence these decisions. The parties wilI compare their predic
values change. Disputes, incIuding those that have a good chance of endtions of court decisions and of the associated legal costs with those of
ing in litigation, may occur in intermediate cases. In these Ol1ies the con
15 As stated earlier, theater owners have the choice betWen leaving seats within a
tracts between pairs of parties simply fai! to speIl out stipulations to attrisingle-price category unmarked or .marking them individually. In thidofmef'case;'pa~
"hut1dht"seer:n-edto'beoflittle value at contract time but whose value
trons must wait for the theater doors to open; in the latter case, they must wait for
increases before the contract expires. Consider a landowner who rents
ticket sales to commence. The latter form favors individuais who are willing to com
mt early. Owners, then, ndirectly decide whch type of patrons to cater to.
out a piece of land with a deserted wooden fence on it. Suppose that at
96
97

such other methods of settling their disputes as arbitration or entirely
private settlement, and they will select in each case the method they per
ceive as generating the highest net gain. It is true that the would-be plain
tiff in a court case has the power to force the comt's resolution of a
dispute. When a dispute 1s actually settled out of court, it is because a
would-be plaintiff has perceived this alternative to be of a lower cost
than settling in court. Tbe other disputant, however, may provide com
. 1?cnsation, perhaps in the form of concessions, so that the method of
osettlemenf involving the lowest total cost ,(including the costs of negotiat, " hig the ,compensation) will besdected~ Apartfrom the costsooftrans~
~ctlng,the'Coase theoreIh holds i:~ue ~v~n h~re. As i~ obvi6us:~hatever
the parties' dedsion, it affects rights qelineation in their own case. As I
will now suggest, the parties' decision may indirectly affect delineation
in general.
'00
Private contracts affect the delineation of rights in at least one more

way. As conditions change, contract stipuIations that had been attractive
in the past may cease to be useful. Since the common law tends to absorb
features that recur in private contracts, it is likely to have incorporated
features deemed attractive in the pasto The courts are expected to rule
accordingly in litigation where the parties have faiIed to stipuIate on vari
ous features of their transactions and have therefore implicitly accepted
the common-Iaw stipuIations. When writing new contracts, however,
contractors may explicitly stipuIate whatever they wish; as Iong as the
stipulations are not in conflict with basic prIncipIes of the law, the courts
. wiH respect the new sti puIarions~ As new stipula"tions are written into
_ contracts, the common Iaw becomes exposed to them, tends to take them
into account, and gradually replaces the oId, less desirable stipuIations
with the preferred, newer ones. The common law is continually revised
in the direction deemed desirable by private parties.
The courts participat~ in rights ~elineation in two ways. The first is indi
.rect: When the parties choosetosettletheirdisptes without resoitirigto
the co'Urts. their actions are influenced by their perceptions'ofhow the
courts would have acted in their dispute. The second is direct: The dis. putes aFe actually.settled by. the courts. The balance of this section con
siders the second componentof rights delineation.
In countries ogerating- at least in p'art under common law, such as the
United States and England, common-Iaw court rulings serve as prece
dents for new rulings. The courts serve to resolve disputes. When private
disputes end in common-Iaw courts, the resolution of the particular dis
putes contributes to the production of a public good, namely, the delinea
tion. of rights in situations similar to the one litigated. Since court rulings
become precedents for similar' cases, litigants are resolving others' dis
putes. 16
Private contrators play several indirect but crucial roles that comple
ment those of the comt. One role relates to the gains that result from
anticipating and avoiding disputes. Because disputes and litigation are
costly, contractors gain if their contracts, ,anticipate potential trouble
spots and provide for them. When such contracts do nevertheless reach
the courts, court rulings are likely to delineate rights dearly beca use they
are dealing with careful1y crafted contracts. This effect is enhanced by
. ,[Qon;:esoof..sele.ctivity"which,partial1ydetermineo,w:hichdisputes.will.be.liti
COMMON PROPERTY
I have stated that econonllsts tend to classify ownership status into all
or-nothing categories, the latter. being termed "common property"
property that has no restrictions placed on its use. The term originated
in the English vllagers' practice of using certain areas for, among other
things, collectively grazing theiranimaIs and cutting .firew.o.od.Thecucc.
rent meaning of "common property" certainly does not fit the English
villagers' actuaI practice, as shown by Dahl~an (1980).18 Dahlman's de
scription makes it clear that the village cOmmOIl. was open only to the
villagers, not to outsiders, and that the villagers' own rights were stinted.
They couId neither add livestock to the herd nor cut whatever amount of
wood they wanted.. On the contrary, they were allowed to place in th .
17Landes (1971) was apparently the first to argue that disputes result from errors
o excess optimism. Priest (1980) applied the notion to the decision to litigate, and
Priest and Klein (1984) conclusively demonstrated the effect empirically.
18See also Ostrom 1990.
16Private rights are constrained by both common and statutory law. I shall not
discuss the forces that affect statutory law, since that would require an analysis of
legislative behavior beyond the scope oi the present work.
98
99
,; ,.-":.:}~; ~.' . ~: ;';:?'::<:~,~:x:;;;~~,~,~~R--:1}:':,l,,~:::: ,~:::!~,:.:
gated. Focusing on the case where parties litigate for direct (mostly fi
nanciaI) gains ooly, disputants go to court only if they are optimistic
about the outcome. 17 Indeed, between them they must err in the direction
of excessive optimism. A court ruling that is expected to be too ambigu
ous to truIy settIe a dispute deters the parties fcom litigating. Only if
disputants expect a ruling that will dearly delineate rights, thus incurring
few added future delineation costs~ will they litiga te. Among alI- potential
litigants in a given class of disputes, self-seIection will bring out the actual
..... i.........
~t!g':lJ.:l~~ :v.v:~(;)O ~~Pt!~t.::~:rt:II!~K:t.~;;t~'\I\lmS~/j\a,r.Iy: 4:e.~in.:~t,~:J:~gl1ts. . ~llt . h.(l.<i.
. previously been in dispute.
'FliE ROLE OF THE COURTS

IN THE DEl:INEATION OF RIGHTS
~~TI:;:~ ; >~f}~.;;"}~XttH,~I)f.;~H!.~ b::~~'~r::-
';"';" ::,{;. "'~,'; ~,({:;~5,;;.~h:.S}hr~,~r,rmtn;:-:!';~\i!j.~LHJ.~'4~: .-~~:A;~<~ -..;;:;'; ~':';' ,:.~':A~:t".
-~_.~.~~._"""'-'!

herd only a set number of animaIs, and aUvillagers were restricted as to
the amount of wood they could cut. Whereas that Iand \yas held in com
mon, its use was directly controlledby the villagers, partly through vot
ing. It was certainly managed as private property.19
.
Regarding current practices, properties under governrnent controI are
sometimes tagged as "common" or as being in the "public domain. It is
improper, however, to view such properties as being unowned. Proper
ties that in economic (rather than legal) terms are owned by no one are
deprived of any value. The view is sometimes expressed thilt such propet'- ...
':::"::::::":::::ties::wouldbepositively valued were they diverted tO.priVl}te ownership.-.
. .A d()seJy relatd view holeiS' tht tnetransfer,govemment pr'pny:'to'
private ownership will necessarily increase its value. A priori reasoning,
however, is incapabIe of demonstrating that private ownership is neces
sarily more efficient than govemment ownership or that designating
properties such as those in mid-ocean as private will yiel" net benefits.
If, as I have argued, metering and policing are expensive, the~'private
ownership, as compared with a zero transaction costs state, is never free
of dissipation. As long as access to, and use of, public property is subject,
as it usually is, to restrictions - such as prohibition of huntlig of young
game animaIs - one cannot conclude that rights would be better deIine
ated under private ownership than they are uhder publc ownersliip. The
distinction between common property and property under government
control will now be examined in the case of the private use of public
roads.
.
Roads are economic goods typically held in the public sector: The con
ditions that implicitly underlie Knight's (I924) analysis of private owner
. ship of roads are those such that private entrepreneurs can determine and
collect the optimal prices and police the use of the roads costlessly. In
practice, pricing and policing costs must be considerable. In comparing
private and publc ownership of roads, it must ais o be recognized that
public roads are not, in fact, managed as common property. Besides re
strictions on features such as the' safety and size of vehicles, road users
are required to pay various fees and taxes, the gasoline tax being the
.. mostsignificant;<
The gasoline tax is a device for rationing road use; the higher the tax,
the lower the demand for roads and the lower the levei of congestiono It
is, however, a rather blunt device for optimizing the conges.tiol leveI.
Because the tax payment is proportional to the number of gallons pur
chased, it fails m.distinguish, for instance, between peak-hour use and
off-peak use; it makes the wrongdistinction, in terms of con~estion 'c;osts
U'Dahlman supplies considerably more detail on the management of the c(Jmmon.

users impose, between more and less fuel-efficient cars. However, be
cause of the costliness of pricing and policing, market prices are subject
to similar shortcomings. Correspondingly, in their pricing schemes res
taurateurs do not distinguish between peak and off-peak hours and be
tween fast and slow eaters. It is incorrect to conclude on a priori grounds,
then, that the value of roads will increase if they are made private.
If priva te ownership is not always superior to government ownership,
it is appropriate to consider the hypothesis that effidency motivates plac

.' .. ing assets in the public sector. A partial justification for this hypothesis is
offered i the discssion of non-market allocation (see Chapte,J: ..9,). Effh.
dencY. onsid~~ations",ill dictate how theyare used. It isexpected, for
in$tanc, that as the costs of,' or the gains from, monitoring public-sector
attributes increase, the use of public~sector attributes will be increasingly
re~tricted. Testing this proposition will also test the more fundamental
hypothesis that the maximizing forces in govemment are the same as
those in the private sector. The foIlowing analysis of the deIineation of
. -property rights involving the North Sea among the surrounding countries
(like the examination of homesteading in Chapter 8) will illustrate the
preceding discus~in and provide some evidence regarding the operation
of governrpent. 20
"
THE CONVERSION OF THE NORTH SEA
INT-Q ow::~:n~D PROPERTY
In I9S8 the..CQnventwn on the Continental SheU was signed in Geneva
(see Dam 1965). The provisions 'of the convention divided among the
countries bordering the North Sea 21 some of the commonly heId att~i
butes of that sea, particularly those related to minerais. Two factors had
been working to enhance the value of the North Sea in the years preced
ing the agreement. First, underwater drilling, which was becoming more
widespread, was declining in cost; second, various signs were emerging
that the region contained natural gas and crude oil reserves. 22 The coun
tries surrounding the North Sea could conceivably have unilaterally ex
tended their territorial rights toward the middle of the sea. Oil compa
nies, however, were not going to invest resources in searching for oil
uniess they expected their potential legal ownership and, concurrently,
20Lueck (I989) demonstrates that state action regarding wildlife is consistent with
the hypothesis that its objective is to maximize the value of wildlife.
21 Belgium, Denrnark, France, the Netherlands, Norway, the United Kingdom, and
......West.Ger.many.......
'c.. : .... , .. , ......
22For example, gas was discovered in the NetherIands and beneath the North Sea
near the United Kingdom.
IOO
rOI
capital asset is rented, the unit cost of measuring the reduction in the
value of the asset declines as the rental period gets longer, since these
costs are independent of the length of the rental perodo Determining how
productive the asset is requires only one measurement of the output; de
termining how intensely it has been utilized requires just a pair of mea
surements - one at the beginning and one at the end of the rental period.
The availability of such scale economies, with regard to both a transac
tion's size and its duration, reduces the costs of rghts delineation and,
therefore, the loss othe.rwise :.ssoc;iated with' placing attributes in the
public domain. ....
.
An ntirelydiffe~ent'methodof lowering the.costs associated with plac
ing attributes in the public domain is to induce the parties to act as if the
attributes were owned even if they are not:When supermarket shoppers
are allowed to choose items such as apples, they are in a position to
capture the value'of the. befier apples, which are sold at the same price as
the inferior apples. Sellers .ke deliber,ilte action to make the displayed
apples appear uniform. Indeed, were all buyers to choose iahdomly from
the available selec60n (and given competition amng sellers), the cost to
consumers of apples of a given average quality net of.tle expense of
chuosing would be less than it is when consumers actually do .p"ickand
choose. 23
.
Here are three predictions that relate to delineation: (I) As rental
length increases, the falI in the rental fee ..will exceed th.e:decline.in .uni.t ..
cost, owing to the sheer cost of finding additional customers; (2) th more'
costly the measurement of the asset, the greater the price decline wiEh
duration'of rental and the fewer the rentals relative to sales; and (3) the
more diverse the product offered for sale, the more prepackaging will
occur, and, in addition, the sellers who are more reputable will more
commonly engage in prepackaging.
their econon.llc ownership of that oil to be secure. The preceding discus

sion suggests that the increase in value of the oil resources of the North
Sea generated forces to better delineate rights over it.
By reaching an agreement, the countries involved gained ownership of
segments of the sea. They could then either exploit their sea rights di
rectly or grant them to private parties and let those private concerns ex
ploit them. Subsequent events proved that the formal agreement and the
accurilte delineation of borders was ultimately of great value. When the
:No:rth::Sea::countries:::canvened::to.::establish:::rights.over the sea, no one
knew where oil would beJound,soitwas easy toarriveat a formulai:hat
would give each country the territory nearest to it without generatirig
much dispute regarding the precise setting of borders. The formula actu
ally selected was that any point on the sea (and on the sea bottom) be
longed to the country to which the point was closest.
As it turned out, many of the major oil and gas discoveries lay close to
the border between the Norwegian and the United Kingdom sectors.
Since the border was precisely marked, ownership of these finds was not
in dispute. There is little doubt that without the agreement oil companies
would not have searchd in that area. The value of the clear delineation
is further illustrated by the following observation. There is a deep trench
in the Norwegiiin sectorfihe North Sea. Laying a pipeline across the
trench is prohibitively costly. Some f the Norwegian oil deposits are on
the United Kingdom side of the trench, which seems to make the United
Kingdom a more naturl owner of that area than Norway. Consistent
with the Coase Theorem, however, once rights were delineated, there was
little difficulty in developing the area. Indeed, some of the Norwegian oil
.. 'is !>hipped by pipeline to the United Kingdom.
MEANS OF ENHANCING RIGHTS
One reason attributes are placed in the public domain is that it is toa
costly to measure and police all the attributes of a transaction. Transac
tors may attempt to capture attributes that are notadequately measured;'
or they may engage in excess measurement in order to reduce capture
costs. In order to maximize the gain from exchange, transactors are ex
pected to seek ways to curb sueh costs. One such method is to exploit
scale economies in measuring; another is to discourage duplicating mea
surements.
- . "Some measuring and policing costs increase less than proportionately ..
to the number of units in a transaction. For instance, as a rule, less unit
measurement is necessary when all transacted units are obtained ftom a
single manufacturing batch or from a given field than when they are ob
tained from several batches or from different fields. Similarly, when a
CONCLUSION
By their own actions, individuaIs are able to control and to affett the
delineation of their rights over "their" property. IndividuaIs wilI exercise
such control as part of their maximizing processo Whenever individuais
find the existing level of:delineation to be unsatisfactory, they will alter it
until they are satisfied. In the same sense that individuaIs are always in
equilibrium with r:egard to their asset holdings, they are ais o in equilib
rium with regard to their. rights over their assets. At any given moment,
..... theiLghts..r.e..~.Q.PJ:ei~.~lY.9.~.6.~.d..rha.r..~hey....gQr .:w.:i~h.r. l:t:,tI,lg~.J4~m,.
Economic conditions;' however, are constantly changing, and with
23
See Barzel I982.
...................I.'?3. ...
I02
il
t(.;:):,':-:' ::':~~..'.:'::~-. ~~~ ~:<\i~:iJ:::;'~!;~:~:~:::..;'~;
< >": . :: ;:." "';;.
"::.~ :~~!{:~'.~t:~ \' :.;.
. :: ~.,;
',-~' .,':.' .:; .~.~: ::::.;i~~.\~;;:a~::~,::: :~BHt~~n~~:: l~:':'~:' -(;;f:':~~Jh~ nt':~~>;~}}H,;:;;::~>:.
.' :~~ ,:.; :~"'!: '1-:':':
. !~'.'; ;,;.1:::: i;~ ~;:v;,:I:U~,tn;n<.~1:~!1H8t}:~:!~lK;~;~~.:~'~:
::;; <~.:: :.' ;.:
.". ::;:~-;::::,~::!>~:X"<: .:'.. :-;~'"', :... ,' ,~..:,-
'... , ' ,,'::
,H::~::i~~;;:'.~'; .:,
';'~ Z',.: :.;::+>~
"1

them the equilibrium property-rights delineation is changing as wel1. As
rights to conimodities possessed by individuais become more valuable,
the individuais wilI delineate these rights more thoroughly. As the value
of rights to commodities that lie in the public domain increses, people
will tend to spend more resources to capture them and to turn them into
private property. Such transfer from the public domain to private owner
Slavery
ship is sometimes effected by individuaIs and sometimes by the state.
. When the vaIue of rights to those commodities that are in the process
of being exchanged increases, disputes between the exchnge parties may
erupt. The resolutionof .disputes results in the de1ineati(}n;oftli:oori:';:::::~:::::::::: ....................................................................................
tested rights. The courts pai:.icipate in tfedeHReai:io~;. of disputed rights;

and the cOrrlmon-law courts interact with individuais in such determina
tions. Individuais choose whether or not to go to a common-Iaw court,
and they wi11litigate new cases until rights become, in their perception,
By the beginning of the twentieth century, slavery had largely disap
well deflned.
peared. Although the institution is extinct, comparing slaves and free la
When the costs of metering and of policing assets or assets' attributes
borers and exploring the forces that permitted slavery to flourish under
~xceed the valuatons, such assets or attributes will be relinquished into
certain circumstances can shed light on present-day institutions. It is par
the public domain and become common property. Such common prop
ticularly illuminating with respect to the understanding of the evolution
. erty is property people choose not to own. Both the English common and
of the status of women and general policing and ownership practices.
gover:n!D:.l1~ J,r()perty in general are vaI u~.q,.d. .their use is constrained;
as a' rule, they are noi:' really cOl11I11().Il.l;'r()p~t:o/: ........................ .
THESTAVE CONTRACT
Labor services, routinely exchanged in the market,l are subject to con

tract. The typical contract for the services of a fre worker transfers a
rather narrow, usually short-term set of attributes from the labor owner
to its buyer. Slavery, too, may be viewed as a labor contract - one, how
ever, that gave slave owners extensfve rights over their slaves. In the case
of forced slavery, the contract extended over the slave's lifetime. The vol
untary slave contract typically specified a shorter duration and gave the
owner fewer rights over the slave than did the forced slave contract.
Forced slvery was initiated by theft - free people were captured and,
as the term suggests, were forced into slavery.2 VoIuntary slavery was the
resuIt of an. explicit contract - a contract to which both parties agreed,
presumably in the belief that slghrigthconttatfwou:ld'bebriefidiilto
each. In some cases voluntary slavery resulted when people who had
posted themselves as loan collateral defaulted on their contract and lend
ers assumed ownership overthem. Indentured servitude, a form of volun
tary slavery, was a direct method of repaying loans. Such loans often
served to finance the passage of indentured servants frem Europe to
America. To repay the loan, indentured servants worked in America'for
lThe terro "market" is used here and throughout the remainder of this book ac
cording to its conventional meaning.
2S ee Bean and Thomas I974.
I04
I05
-!i
Slavery

a number of years virtUally as slaves. The duration of the servitude was
determined in an auction in which the winning bidder was the person
who bid the shortest number of years of service and bought the loan
conttact from its previous holder - often the captain of the passage ship.
The servant worked out his or her debt to the final lender, usually an
American farmer.
The term "conttact" normally implies' a voluntary reIationship, in
deed,.3. relati<?nship.from wh~~h both parties expect to gain. The term fits
v:olunauy slavery :wellbut le~s satisfactory in the case of forced slavery.
. It Is der.that
persons',having be fored intoslavery, did notex
pect to gain 'from their change in status. Therefore, if theterm is appro
. priate at all to forced sfavery, it applies only to the period of time after
which the persons had already become slaves. Even then, such usage
stretches the definition of "contract" to its "limito NevertheIess, it will be
seen that the notion of a slave contract is useful in analyzing the institu
tion' of slavery.
.
free
is
S/da'
en
~k
oI
"\:
,m
Hours of leisure/day
Mil
SLAVES' SUPPLY. OF LABOR

Figure 7.1
Slaves were exttemely P90r people. A recognitio'oheic'povi:ty is es

sential if various slave prctices are to be properly understood. The utili~ ..
zation of slave labor per unit of time may be explored through the con
ventional labof-Iesure '-dibice analysis: Figure. 7: I illutrates such an
analysis, except that in <!lddition it ,recognizes the effect of fatigue. Ir is
assumed that beyond a certin poi'i:itin the working dy,.fatigue becomes
dominant and that the slope of the budget line will become positive; more
hours will lead to lower output. A free worker with budget MM will
choose to be at point A, which lies on U, the highest indifference curve
that the worker can attain. A person subject to a budget constraint MM
obtains his or her entire income from""the use of his or her own time. A
budget' constraint M' M' implies that the individual has a dai1y income
equal to the vertical distance between MM'andM:M' from sourcesother
than the use of his or her own time.Similarly, M"M" indicates a negative
non-Iabor income, that is, a debt-payment obligation. A positive income
elasticity for leisure - a highly plausible relationship for a person whose
earning power is low and who is a net debtor - implies an increasing
su pply of labor as the budget consttaint shifts inward. Given such income
elstic~ty, as the fraction of a person's potentialJabor income that must
be used' for debt repayment increases, the person accordingly increases
t:ne:qua~ityoflaborheorshesupplies;3. ..................
3The supply of labor also depends on the intensity of work, which is usually viewed
as.given- !;lu'!; is actually a choice variable. I abstract from it here; for a more elaborate
analysis, see Barzel 1977.
Workers incur maintenance expenditures - the minimal expenditures

on items (e.g., food, medicaI care, shelter) that will sustain them at a
given exertion leveI. Such expenditures most likely increase with the
amount of labor supplied. Maximum net income from work is the largest
present value '(ir labor income a worker is able to produce net of the
present vaI.ue of his or her maintenance expenditures. Tllat particular
labor effort associated with producing the maximum net present value
also permits the largest possible debt payments. Such a leveI of exertion
defines the largest debt that a person can pay while earning just enough
to sustain himself or herself. A free person with a commensurate debt
will presumably choose to operate at such a levei of exertion. Policing
costs aside, a forced slave can be viewed as a persou who operates pre- ...
cisely at that point. The owner o" a forced slave will' require the effort
that, net of maintenance expenses, will yie1d the highest present-value
income stream. Thus, a forced slave may be compared to a fr~e per!>on
who has contracted to operate in exactly the same way the slave is re
quired to. A forced slave is deprived of his or her entire net present value.
Such slaves essential1y operated under-a'ontract, the terms of which have
....just;l:>~en described. It is a contratt in the sense that it characterizes the
amount necessary for manUInission, which is the act of buying out that
contract.
In the era when the law permitted slavery, the question of manumission
I06
I07
Slavery
was pertinent. As indicated in the opening paragraph of the introduction

to this book~ the analysis of self-purchase poses a major puzzle: How
could a slave, another person's piece of property, purchase bis or her Own
contract? In order to resolve this puzzle, transaction costs, particularly
policing costs, must be considered.
convince their future employers that they were more productive than they
actually were; since it was dif:ficult to ascertain their true productiveness,
their wages would consequently be bigher. Slaves, on the other hand,
would have done best to convince their owners that they were no good,
since little would then have been expected of them. s Slave owners, then,
had to spend resources to figure out how productive the slaves were and
how hard they could be driven, as w.ell as to actually supervise their ef
forts or oUtput..
. AssUriting lhat the supervision of effort is subject to diminishing mar
'ginal p(Qducclvty, intheir supenrision effQrt owners would have stopped
:short'!extracting the m,aximum output of whichslaves were capable.6
The difference between slaves' maximum output and their actual output
became the slaves' property in practice. As a rule, it carne not in the same
form as the product they produced for their owners, but, rather, in the
forrIi of reduced effort. Still, sIaves were able to convert some of this
, poteritial into m~terial goods.
Owners had -to choose between s"upervising their slaves' output, which
is cmparable to what employers have to d when the'free workers they
employ work by the piece, and sup.ervising their effort, which is compara
ble to what e!Ilployers must do when thyemployfree woiker's by the
hour. The latter re-quired continuous supervision, for whenever slaves
were not watched they had little incentive to perform. The former re
quired ovrseeing 9utput as. to .i!~i"quantiry, gUl!1iey, and, indirectly, its
effect on -bther productive inputS such as equipmenr. The supervision of
slaves' output also required'-utput quotas; left to their own devices,
slaves would have produced as little as they coufd. Moreover, quotas
could not have been set by simply observing past performance, since
slaves' incentive to produce little during the demonstration period would
have been strong indeed. Quotas were subject to error; toa high a quota
would have resulted in the destruction of slaves. Owners selected the
quota that would 'maximize their own -wealth and were expected to en
force the production of that quota but ,no more. This means that owners
left their slaves with the difference between the quota and the maximum
they could produce, a difference slaves could exploit. Given the two
methods of extracting output, owners could offer supervised sIaves the
option to work under the quota method, setting the quota to exceed the
supervised output. With this as an option, some slaves would have cho
THE COSTS OF SUPERVISION
Since slaves were the full-fledged legal property of their owners, ali the
::',:::::::::lnC'i5me:they:::cu:ld::generate.waslegally their.owners'. as welLqwners .
. who took anything from .their slaves vvere. simply taktig somethIig that,
was legally theirs to start with. Owners had the legal rght to take; they
also had the might. How, then, could slaves accumulate wealth, some
times to an extent that enabled them to purchase their own contracts?
Had both slaves' capabilities and the net output they produced been cost
lessly measurable, wealth-maxmizing slave owners would indeed have
extracted every last ounce of productveness from their slaves. Under
these conditions, which would have permitted owners to assume owner
shp in the most complete sense of the word, it is ironic that sIave status
itself would have been inconsequential. Had policing been costless, the
owner could have obtained the same income whle allowing tlie slave to
'operate asa.'free person; for instance, he could have done this by recev
ing an explicit'payment of the same value as that generated by the slave's
services withot requiring slave status. The free person, however, would
still have had to work exactly as a slave would in order to be able to
make his payment, or as a debtor would to pay off his or her debt. 4 This
is consistent with the Coase Theorem, which states that when rights are
clearly delineated - as they are here - resource allocation is independent
of the ownership pattem. In reality, the evaluaton of inputs and outputs
is costly, and policing is requred to induce effort. As we shall see, accu
mulation by slaves is thus possible.
When the net output of a free worker can be determined at low cost,
it is advantageous for him to operate independently, since this eliminates
the problem of incentive"Suchwasnotthecase'with'slaves; the labor
services slaves could provide be10nged to ther owners. Even when output
was easily measured, slaves would have gained from producing less and
thus hadto be induced to produce more. The distinction between sIaves'
and free workers' incentives became even more acute for tasks that free
workers performed by the hour. Fxee.workers would do best if they could
, 5,slaves had to be careful not to do toa good a job of deceiving their owners, snce
owners who
their slaves were nor worth their cost would have found rhem
4Had ali rights been well defined, which requires costless transacting,...not.. only,.......
would slave status have been rendered devoid of significance but forced slavery would
never have arisen in the first place, since one person would never have captured the
rights of others.
"::expeiliib'le;'"
.. .... .
6This statement applies to the average slave. SIaves who were mistakenly asked to
:il>. ITIore than they were capable of doing were in serious trouble, compared with
other'slaves, to the extent that they sometimes died of exhaustion,
Io8
r09
>I
Slavery

sen to operat~ under the quota, since for them the quota ohviously was
not excessive. Though they allowed slaves to retain output in excess of
the quota, owners were aware that the levei of the quota could have been
ncreased to begin with. Since owners' confiscaton of slaves' accumula
. tion would have been equivalent to hreaking the contract, confiscation
would have defeated the purpose for whch the quota was established to
begin with. In pursuit of their self-interest, owners permitted slaves to
own and to accumulate.
........ SlvS:seld6m::becamed(;ilLNveitheless:~::tlie:fret;:Hmethey:wereable
. to g:linenabled them, among other things, to grow vegetables,' fish and
hunt, and also to steal. Their owners and neighbors were ready buyers
for what slaves had to seU. Slaves who were ultimately able to buy their
contracts were, as a rule, household slaves, sometimes well-trained ones.
Because their tasks were diverse, they were difficult to supervise, a diffi
culty that inreased with the skill of the performer. Household slaves,
especially skilled ones, had better opportunities to accumulate than did
sla ves who worked in the fields.
The need for supervsion and the desire to economize on its cost made
ownership of slaves less than fully delineated. Slaves were able to capture
some of these undelineated rights - in this c.ali~..rigt.,s.Jo themselves. Suc
cess~ paitlyde'
sidnsin'feigninii' hi~bility, on the Qne hand, and to
activities such as fishing, on the other, as well as the luck of having errors
made in their favor, eventually enabled some sIaves t buy their own
contracts.
.
".
productivity was lowered. Ohviously, they preferred more palatahle

food, which they would have substituted, in part, for less appealing if
more nutritious food. One important tool owners possessed for minimiz
ing maintenance costs was the ability to control consumption directly;
they attained this controI by simply supplying consumption items in kind
rather than providing slaves with a budget, which would have permitted
choice. Thus, slaves' diet in the American South contained much com
and sweet potatoes, then considered nutritious but less desirahle than the
more expensive wheat and white' potat9~s eveD.por. Ieee farniers chose
to consume. Ow-ners' control of shives' couswriptiori; however,was not .
absolute. Slaves would, on occasion, avoid taJ.<ing medicine, since -in their
eyes remaining iI1 was sometimes preferable to returning to' work. SIaves
also traded food rations for alcohol on occasion. That prviding con
sumption in kind was intended, at least in part, to restrict choice is dem
onstrated by the fact that owners flOt only refrained from supplying alco
hol but, most tellingly, also took various costly steps to deny slaves access
to it. Drinking reduced sIaves' productivity, and owrfers made a con
certed effort to make drinking expensive to slaves.
.
ENFORCING PROPERTY RIGHTS

IN THE PRESENCE OF SLAVERY
to
The legal ban on both forced and voluntary slaver,y i8 now.universal; in . :'
earlier eras, various restrictions were placed on baih practices. Voluntag- .
slavery was a contractual arrangemerrt into which neither party wold'
enter if she or he did not expect to gain from it, while under forced slav
ery it was owners who gained. Eventually all states abolished slavery,
most of them voluntarily. Needless to say, slaves had littIe direct influence
on the abolition decision. Issues of theft may have been partIy responsible
for the abolition decision. Because of theft, the institution of slavery, both
voluntaryand forced, mayentail costs that hnpinge on free persons, who
might consider them excessive.
. TWQJm:msof theft are associated with the institution of slavery. Al
ready mentioned is the fact that theft was the initial step in forced slavery.
The other form of theft occurred when slaves escaped; the escape tumed
what were to their owners valuabl~ assets into assets deprived of some
or all of their value. The loss that owners incurred due to abolition must
be weighed against the policing costs associated with theft.
Coercion of free people into becoming slaves has large1y been a conse
quence f raids.nd .. Qf.full~.fledge.d ...wars."among...nations..For..centuries.
West Africa was subject to raids alid ",s i:hemain source of slaves for
the American South, the Caribbeari,a.nq.- S9qth America. Since losers of
......SLAVES' CONSUMPTION
The sharp disparity between interests of the owners and those of slaves
was not confined to slaves' effort and its supervision; it also affected con
sumption practices. Owners could not drive their slaves hard unIess they
made sure that they were properly maintained - that the nutrition, medi
cai care, and other services si aves received werecommensurate with the
effort required of them. Correspondingly, as Fogel andEngetmanCI9?2:) '.
have shown, in the United States slaves received what by nineteenth
century standards appears to have been good medicaI care and a nutri
tious diet. 7 Owners, however, had little incentive to let their slaves con
sume food other than that induded in the least-cost diet, or to let them
consume other servjc~s beyond the lowest-cost ones required for mainte
nance. Slaves,Qn the other hand, were not necessarily harmed when their
7Fogel and Engerman, however, do not interpret their findings on nutrition the
way I do, nor do theY'systematically apply transaction cost notions to the analysis of
slavery.
IIO
.;.
1i;,; :
'.,<; ';i',-:-:;-;,.t;,(,. ~:.;,'\ ':: -
'-i-. ;..:;';;<' ;~':l;_:> ,:::':;;' :o;. ~(~':f~;~~:~; '.':~ ..~;.:"
~~1 ,";~{i.~::m~,!trra(:'H~:,!::\S!:;~'~:!:'; i~;"~. ~~? . ~'~~.~.~ ~:;;: ~~,:.;-? :,~~,~~: -~: -;);.}.;? <::.~)
Slavery
wars and raids were not the winners' counnymen, one would not expect
states to outlaw that source of slavery. However, raids in which free per
sons are captured and turned into slaves against their will can also occur
within a country. Free persons will take steps privately to protect them
selves against enslavement; they will also use the state to enhance their
rights. The most obvious use of the state machinery is to make enslave
ment of one's countrymen a crime. A more extreme step is to prohibit
slaveryaltogether. 8 Enslaved persons would have found it exceedingly
.. difficult.todemonstrate that they were forced into slavery in a society
'tht p~ITJ:tted slavery. The returns !rom suchtheft, then, are higher
where shlvery is perm:i~tedthanthey re where slavety is prohibited'X:on::'
versely; the threat to free persons of being turned into slaves is relatively
low in countries that prohibit slavery,9 The fear free people have of being
turned into slaves may partially explain the ban on slavery?O
A different kind of cost arises in the case of escaping slaves. It seems
to be generally true that certain costs of protecting property from theft
. are assumed by the.state. In Chapter 8; I suggest as a general proposition
that activitk:s will be banned if the police protection they require is very
costly. If, as compared with the private value of slavery to slave owners,
the cost to the polce of rec()v(!r.illge!)~aped slaves was excessiy.~,J,hlVery
. would be bnned~'
....
The higher the dange.r to a country's residen'ts 'ofbdiiii'forcedhito
slavery, and the higher the cost of.recovering escaped slaves, the more
liklyt is th~t slavei:y will.be prohibjted. For a long time in the American
Soutb, being a slave and being black were almost synonymous. l l Free
people: nearlYa.Jl of thm white,-did not have to fear becoming enslaved,
and escaped slaves could readily be identified and captured. As the popu
lation of free blacks grew, however, the distinction between a free persoii
and a slave became more difficult to establish. 12 Consequently the costs
of slavery increased, and the net gain from the institution declined. 13 If
slavery were legalized in our own society, the distinction between free
people and slaves would be more difficult to establish and the problems
of the forced enslavement of free people and of capturing escaped slaves
would become even more acute than they were in the Old South. It is not
surprising, then, that in many societies slavery was severely restricted and
that it is now universally prohibited.
.
The demise of indentured servi rude may have been due to forces simi
lar .W .~hQ~(!.t.httt..te.4..tg.m"'..c!..e.J..11.i~e..(J..f.~h'!.ye..l:Y.~ ..At.Jt.~.~f.l.::e.p~jf.l,-.!p.e popula
8Consistent with this view is the observation that raids to acquire si aves dedined
after the colonization of Africa. Enhancement of property rights within colonies in
creased the colonies' value to<the colonizers.
9 Isolated cases of slavery are nevertheless reported in the press.
.
10 According to' Jewish law, Jews were allowed to keep other Jews as slaves. The
law required, however, that ali Jewish slaves be freed by the onset of every Sabbatical
year."lf during a Sabbatical year one spotted a Jewish slave, it was c1ear it was iIIegal.
This constraint on slavery reduced the gain frommegally enslaving other Jewish indi
viduais, and it is consistent with the hypothesis presented in this study.
11 The identification of sIaves with a particular people is the apparent source of the
term "slave." To the Slavs the term "Slav" apparently meant "person." Most of the
Slavs present in andent Rome were slves, and the Romans used the term meant for
.Q.t:Ie.to.identifyHthe..other... HH ....................................H................H.... :....::..:.. :.. H.. HH H........ .
12Southern states imposed various restrictions on manumission, perhaps in order

to prevent the formation of enclaves of free blacks. These efforts were not entirely
successful.
112
tin:ift.A:ffierka WaS'sparS';"aiid'sefvahts;'tlest'ffiif'GfmanY;'Were
easify distinguished from "free people. Ovei time, the population grew'
and diversified,making escape easier and perhaps also making the en
slavement of free people easier, thus lowering the net benefit of the insti
tution to the originally free population.
By the end of the nineteenth century, slavery had largely been abol
ished, perhaps because the costs of policing the institution exceeded its
rewards. 14 The current prohibition of slavery implies that each individual
is the owner of the capital asset embedded in himself or herself. The
abolition of slavery was accompanied by the transfer of such capital
assets from the previous slave owners to the'slaves themselves. The prohi
....... bition .orslavery .also. entails .additional.r.estrictions.on contracting: Essen
tialIy, when workers contract for the supply of their own labor services,
only short-term contracts aJ;"e legally enforceable. These additional re
strictions may also reflect the attempt to make the theft of people more
costly.
CONCLUSION
Even though forced slaves seem to have been stripped of any rights, in
practice ownership over them was not complete. Owners had to spend
resources on supervising the work effort and consumption pattern of
their slaves and on preventing their escape. Such efforts were subject to
increasing costs, and attempts to economize on these costs included
granting slaves various rights. In some cases slaves' 'output'ratherthan .
their effort was supervised. The attempt to lower the cost of supervision
then included granting slaves the right to part of the output or of their
own time. These slaves, though legally ther masters' property, were able
to accumulate wealth and occasionalIy to buy their own contracts.
13 Some slaves woo .were, in fact, freed by their owners chose to reta in the legat
status of slave, presumably to retain the owners' protection and to reduce the chance
of being forced back into sIavery.
14The insttution resurfaced under a new guise in the nineteenth century, with the
importaton of hinese workers, a newly distinct set of people, for railroad construc
tion. Like its predecessor, it did not last longo
II3
Wealth-maximizing constraints on property rights

ist market eeonomies. For instanee, store owners are sometimes prohib
ited from opening at night or on Sundays, automobile operators are not
allowed to transport paying passengers unless they acquire a "medal
Hon," owners of water rights in the western United States are not free to
seU the water to others or to divert it to new uses, and the size and loca
tion of structures that may be erected on urban 10ts are restricted by
zoning regulations.
In these examples the govemment imposes the restrictions; but restrie
. tions. are not the exclusive d0l11:<l~nofgove'iiment;Smehomewners::::::
aresubje.topriviecovenants - restrictions superimposed on those of
governmental zoning. In' addition, stockholders who own. corporations
are severely restricted as to what they may do with their properties, and
fixed-rent tenants are often restricted with regard to the crops they may
grow. These sets of private, voluntary restrictions were imposed by the
original developers of such projects. Both the landowners who intro
duced the restrictions and the developers of the housing projects and of
the corporations presumably believed that the restrictions would increase
the net value of the projects to their buyers, thereby increasing the total
netamount they themJiJelv.e,s could extract. If this is so, the apparent in
'consistency .1J.e.t\iVe,e.l1."Y.elllt.ll ..Illa.J{iII1i~<lti()Il. ~IlcJ,. ,(}\\TI1eP'llip restricti()ns
must be just that - apparent and not real.
Wealth-maximizing constraints
on prQper.ty, rights
",
" , '
-o
The economists wh-;>first adopted the property rights framewo~k in their

analyses often considered any, government restrictions on those rights
("attenuation of rights") to be undes!ra~le. A person's ability; to realize
the potential value of.his or her property depends on the extent of his or
her property rights, which consists of the ability to use (and exclude), to
alienate, and to derive income from the property. The abili'ty, or P0"\I\Tt:l" ...
to exclude prevents the ptoperty from becoll)!ng cOnllnon 'property, and
the ability to alienate and to derive income permits the realization of
gains from exchange. 1 Since restrictions generally reduce freedom of ac
tion, restrictions n a person's prQ'peity righis reducl! tl1. 'value of.the
property to its owner. Thbs it wuld appear that such restrictions are
harmful. 2 In this sectiOIl I will demonstrate that' restrictions that seem t
attenuate rights in fact often help to delineate rights more clearly.
Restrictions, that is, non-price allocations, can have no useful role in
the Walrasian model. According to that model prices alone can direct alI
resources into their highest-value uses and will yield efficient allocation,
since price adjustments are costless 'and rights are wellde6ned. Restdc
tions, then, are at best superfluous and may well result in lower 'output.
The unfavorable attitude by economists toward restrictions on,oratten
"uation of, rights may reflect the implicit applcation of the Walrasian,
costless-transacting model.
In spite of the Walrasian mdel, restrictions' on individuaIs' freedom to
do what they wish with "their" property are widespread even in capital
1 The terms of alenat:ion - whether price discrimination is legally permitted,
whether sale is subject to price control, and how easily buyers can collude when deal
ing wi tbme--seller--,-aff-ecttJ:,e:'Ifalae,-eji,dle ,r-ighu,--...... ,,--..............,............
............,. --.........
2 An owner is not hrmed bi' restriction if it existed before the property was
acquired. The owner i~ harmed only jf the restrction was imposed subsequent to
purchase. The latter situafon i's'likely to involve the use oi resources to abet the
harmful effects.
''I T
4..........
DIVIDED OWNERSHIP AND
RESTRICTIONS ON OWNERS
As I stated previously, commodities may be viewed as collectons of many

attributes whose leveIs tend t vary from one specimn'-o{ a commodity
to another. IndividuaIs sometimes choose to divide the ownership of
commodity attributes among themselves because the most efficient owner
of one attribute is not necessarily the most efficient owner of other attri-'
butes of the same commodity. When different attributes of a commodity
, are owned by different individuais, a special effort is required to exclude
each of them trom using attributes belonging to any of the others. Orie
way to effect exdusion is to impose restrktions on the ways owners may
exercise their rights such that the commodities will not become common
.property too easily. As viewed here, one function of restrictions on com
modity ownership-is to secure rights better by making exclusion easier to
enforce. Such restrictions tend to prevent non-owners t.rq,m consuming
those attributes that belong to others. Restrictions ou non-owners cannot
..oeusefully'Viewed as attenuation.
The way a refrigerator is owned is a straightforward,convenient exam
pIe of the role of divided ownership and of the corresponding restrictions.
The sale of refrigerators to final consumers does not constitute an out
-..,

right transfer, since the manufacturers retainresponsibility for - that is,
they remain the owners of - the attributes that are subject to warranty
and those for which they are liable. 3 It is clear that manufacturers are
more efficient owners than consumers of the potentially IethaI Ieakage of
the coolant, beca use it is their actions during the manufacturing process
that largely determine whether or not the coolant wilI escape. Manufac
turers are also the more efficient owners of th~ longevity of the motors.
The longevity of motors depends, to a great extent, on manufacturing
procedures about which manufacturers are iruormed at a low cost and
,:consumers:are:liiilntormed:;:itefrigetator:hyers;,then;become the' own-'
crs oIoniy a subset ofrefrigertors' attributes.Given that therrotor is
subject to a warranty, they do not bear the entire cost of the motors'
short lives; given liabilty, they do not acquire ownership of attrbutes
such as "leakage of coolant." The advantage of divided ownership be
comes evident when alienation is desirable. This is the case nor only when
a new refrigerator is transferred from manufacturer to wholesaler to re
tailer and ultimately to consumer; it is also desirable when a used refrig
erator is offered for sale.
Consumers who decide to seU their refrigerators will transfer to buyers
only those attributes that they own. If they were the owners of the
~tt.I:H).lH~"Je,9.kag~ pf Qolant/~, they. would encounter difficulties in ar
ranging a sale contract, since they would be able to provide satisfactory
information about that attribute only at a high cost to themselves.
The manufacturers, on the other hand, who are the efficient owners of
"leakage of coolant," remain liable for it; they continue to own this attri
bute. The refrigerator owners, therefore, are not involved with the trans
Jt;r. C!f the "leakage of coolant" attribute;and the sale of those remaining
refrigerator attributes that are easier to alienate is correspondingly facili
tated.
When ownership of a commodity a physical entity - is divided, the
owners of some attributes may have easy access to those that are owned
by the others and may be able to treat them as free. For instance, buyers
of a guaranteed product are likely to be less careful with it than they
would be were ii: not waranteed; to them; theseiVicecOveredbyi:heHwaJ:~
ranty becomes a free attribute. In Chapter 3, in the course of a discussion
of the tenancy contract, I showed how restrictions can reduce the cost
associated wth free attributes. Restrictions se em to play a simlar role
here. For instance, refrigerator manufacturers retain ownership of liabil

ity and warranty attributes only f consumers subrnit to restrictions on
such things as abuse and commercial use. Such restrictions help to isolate
the attributes 'owned by the manufacturers from encroachment by con
sumers, thereby lowering dissipation, or capture, costs. The ownership
of refrigerators by consumers is attenuated by such restrictions: Those
who use their refrigerators commercially stand to lose the warranty pro
tecrion. When a warranty is voided because the warranteed product is
being used commercially, a consumer is being penalized for failing to
heed a restriction. Not attenua.ted, fqrexampl.e, is the consumer's owner
ship of the right.to decide wha-t'Joods to sto:(e in the .refrigerator; he
ol'she ownsthis p:uticular:' ttribute. Indeed, beCilse the restriction on
commercial use reduces capture costs, it inreases the net value of the
original transadion; in other w'ords, ir" increases the value of the re
maining attributes.
Restricrions m\Jst be enforced in order to be effective; 'sueh enforee
ment is costly, hut not uniformly so. For instance, commercial use of
refrigerators can be' easily proved; therefore, manufacturers make the
warranty conditiQnal on private se: Plain carelt;ssness is toa expensive
to police, and eareless behavior is not restrained. ConveFsely, the dura
tion of the warranty is shortened when careless behavior is expensive to
plice - paint on refrigerators is gua:rante~? for a much shorter' tiie than
are motors.
.
The value of a transaction is another factor that helps to determine.
whether constraints will accompa1y it: Ii 10w-value.-ciansactions, the _
eost of managing restrictions cn easily exeeed the associated gains. Nor
mal1y, as the value' of transactions increass, more iestrictions wi1l De.
imposed and more resources will be devoted to their enforcement. Not
surprisingly, both the manufacture and the use of commercial airplanes
are associated with more restrictions than are the use and manufacture
of automobiles. As a rule, airlines contraet for the manufaeture of spe
cifical1y designated airplanes an monitor the performa.nce of the xp.anu
facturer by stationing their own engineers on site. Airplane manufactur
ers, in turn, keep track of the airplanes they have producedafter'they are
delivered to airlines.
.: RESTRICTIONS AND
THE' COST OF POLICE PROTECTION
Restrictions on ownen;hip may perform another function, related to the

protection of assets agamst theft. Because assets are always in danger 'of
3The manufacturers retain ownership of income streams that have only zero or
nega tive values. They are the residual c1aimants, i.e., responsible for those nega tive
income streams. Payments to the manufacturers for these potential services come
ahead of time, when the original transactions are concluded.
'being"stolen;ownership':ls.neverentirely"secure~'Thefrcan,take'many"
II6
II7
forms, and prot,ection is also multidimensional. The state, through the

police and the courts, provides protection against theft. The state, how
ever, is not necessarily the most effi.dent protector against alI aspects of
theft. For instance, cal' owners can protect their cars cheaply by parking
. in well-Hghted areas and by not leaving their engine running; They can
also buy private protection in the market by installing theft-deterrence
devices in their cars and by parking in lots with attendants.
Two features of the state's provision of poIicing services bear on the
question of how to Iower the resource cost of protection~ First, the state
commits itself to assist in th.e.p.rotection.oUegalIy..delineated.properties....
against.theft;.. Secorid;polieseivies arsuppliedatno.margiliaIcharge
to individuaIs. Individuais may attempt toreduce their own tforts at
protection and rely primarily on the police even where the retums on
their activities do not ;ustify the associated costs of policing, 01' where
they can provide protection less expensively than the police. The state
. Can reduce such excessive use of police services by imposing certain re
strictions on individuais. AccordingIy, the state stipulates that automo
biles must be equipped with locks and that individuaIs must lock their
cars when they leave them. 4
In light of the preceding discussion, the next three sections aredevoted
to the'interpretation of three sets of restrictive practices. The first section
.... deals.with restrictions. imposed bythe government on -water ownership
in the western United.States;5 next I explore government restrictions on
homesteaders; and finally I discuss private restrictions imposed on two
firms, one of which tums.Chrysler hardtops into convertibles, and the
other of which engages in the long-term storage of a truck manufacturer's
replacement parts;"
RESTRICTIONS ON WESTERN WATER OWNERS
Two distinct legal doctrines govem the use of water in the United States.
In the eastem states, the prevailing system, originating in English com
mon law, is riparian, permitting landowners reasonable use of water
from rivers that run by their properties. Water is relativ.ely abundant in
that region (as it is in England), and so naturally the rights to it are rather
loosely defined. 6 In contrast, many of the western states are arid, and the
prevailing system is appropriative, granting individuais more complete
4Lueck (X989) suggests that the state restriction on trade in game-animal products
may be explained by the attempt to reduce poachiJ:lg.
5This section relies heavily on Miller (I985).
6CRose(x99 0 ) analyzes changes within the riparian system brought about primarily
by the emergence of large-scale use of water for power for which previous rights were
not well defined.
. ....

rights to water. 7 The flow of water in westem streams .varies significantly
both seasonally and annually. Priorities concerning the water in these
streams are relatively dearly defined. 8 The more thorough delineation of
rights to water in the West is consistent with its aridity and with the
correspondingly higher value of water there. The rights individuais have
over "their" water are severely restricted. The balance of this section
concems the function of the restrictions on water ownership in the
~est.
.
.'.
Themostsevere restrictions on owners of rights to water in the westerl
states regulate how the water is to be used. Most westem sta~es prhibk
owners from using water for any purpose other than the one it originally
served; the owners of rights to water are not allowed to seU them unIess'
the sale ncludes the land where the water is used, and they are not even
free to use the water on their own properties in any way they wish.
Like other commodities, the commodity "water" is comp~sd of many
attributes. Recognition of that fact is crucial if one is to understand why
the delineation of rights to water is costly, why owners of water are not .
allowed to do whatever they wish with it in order to mximize' its net
value, and why owners' rights are attenuated. A special characteristic of
water is that its abundance (and quality) aownstream vary according to
how individuais upstream use it. Whether or not the individual receives
water depends, in part, on how much of the water that individuais with
greater seniority have the right to divert seeps back in"to the stream - i..iJ.
other words, on what portion of the water withdrawn by those wi:h
higher priorities is returned to the stream. One person's re~1ized right to
water depends on the mode others take in exercising their 'own rights.
Were measurement costless, individuais would presumably be granted
rights to remove a certain net amount of water from a stream. A person
with a right to a net amount of 100 acre-feet of water could, for instance,
withdraw 250 acre-feet to irrigate a crop if it was clear that T 50 acre-feet
would seep back into the stream. Measurement, however, s not costless;
it is much cheaper to measure the amount of water removed from a
streamthaniti:s tmeasurt'heamount retumed to it by seepage. Similar
considerations apply to other attributes- of-water. It is easier to measure
attributes such as salinity and temperature for the water withdrawn from
7 Anderson and HiU (I975), among the first to systematize the a'nalysis' of property
rights, offer an embryonic explanation of the transition from rparian to appropriative
rights to wat.er.in.the western states.
.
8The priorities reflect the initialorder of the individual c1aims on water in the
streams. The amount of water originally granted-.to ..indi'liduals..was...in .accord.with...
the amount actuaJly used in their operations, usually farming. As is pointed but in the
next section, individuaIs' participation in rights protection seems easier when one
owns just what one uses.
Il:8
. :.-:::;;.i:~:_; ':': ,\ :',:, : :'\ ,.. ,.t, ..1 I,
... ',',',
.<. ,':-' ;,1.6t~!.~{(~..\,,~t5"",r.~..~~.-:.,-'. ;..': ,<.:,.,.
,'. :.','
'/.~.> .''':'.~.'",'
i,,' i,',
" . ',' ... ;;, '>.',1,'.

a rivec than it is to measure the same anributes for the water in the return
flow. 9
The original owners of watec rights to a given stream presumably used
the water foc a specific purpose, for example, to irrigate an orchard. The
precise contents of the grant of a low-priority right is a function of what
" h er-pnonty
" ' h o ldecs are alI owe d to d o. PeopIe Wlt
. h n'ghts to Wlt
. h
the h 19
draw watec from a stream who are restricted to a specific use have rights
that are delineated moce narrowly, and are consequently less valuable,
.than tli~y 'Wlli~d be"in the absence of the restrction. As a result of the
iest:riction; ht;wever, the rights ()f thosepeoplewith Iower .prioritiesand
the rights !;ti the two groups c;ombined are more:deady delineated. In
.other words, whereas the rights of people with high priorities are attenu
ated, the rights of those with lower priorities are thereby enhanced. The
more dearly rights in general are delineated, the more they are valued,
since If;SS is expected to be spent on their capture.
Two obsrved practjces lend support to the hypothesis that the state
imposed restriction on water use is intended to ncrease the value of wa
tet.rights. ::rbe first regards the inter-organization transfer ofwater. Mu
tual ditch companies and irrigation districts are permitted to use water
in a way that seems to evade the regulation of wtertransfer:Thesetga
nizations of neighbors are pennitted to combine the water rights of their .
members. The restriction on transfers within each organization is
.avoided, enabling mem~rs. toimproy~_the aliQcation. of water among
themselves. Provided that ihe pattem f water se is nt ltered, the re
tum- quantity (and quality). of the--~ater is likewise- n~t altered. The
greater flexibility available to these organizations, then, does not signifi
cantly impinge on the rights of others. Aliowing water transfer within
ditchcompanies and withn irrigation districts is therefore consistent
with the explanation that the purpose of the more general constraint on
transfers is to enhance tJ:te clear delineation of rights.
The se~ond practice concems the distinct Jorm' of the constraint on
wate,.. use that is applied in Ne~ Mexico. 11.1. tnat state waterrightli are
defined in terms of consumption, that is, the ne!" amountof water re
tained by the owner. Among the western states, New Mexico is one of
the most arid, and it consequently values its water highly; so it is not
surprisipg that it finds the costs of the added measurement that cqnsump

tion rights entail to be worthwhile. The relative ease with which owners
of water rights in' New Mexico may obtain the regulator's permission to
seU water is consistent with the interpretation that constraints play a role
in better delineating rights. 10
HOMESTEADING
Before I776, allland in the colonial United States had been initially de
clared to be the property of the British crown. Bridsh authorities em
ployed vacious methods ..o.disp.ensing.Jand,...including."granting ..lt ...to...........
. s tatesftradini :companle~;aiid i.ndlvi.dua1SwhoCiossedthAtlaiidc~'The" . . ..
tiansfet of power in I776 was accompanied bya change in the land
release policy. The federal govemment of the United States adopted a
more unified policy of managing its land holdings, one element of which
was to discourage "preempton" - the priority right of the illegal settlers,
or squatters, to buy the land OR which they had settled.
Until the Louisiana Purchase in I803, most U.S. government-owned
land was in the thirteen original states, and market sales of govemment
land were routine. During the nineteenth century the U.S. government
took possession of vast amounts of land in the West. Abundant as the
la:nd"was, the govemment made a concerted effort to prevent its becom
ingcommonproperty. Asthepopulation grew;thegovernmentcontinued
to sellland in some areas. In others the govemment periodically released
for settlement blocks of land that were not handled in a conventional .
commercial fashion. The govemment could have auctioned off these par
cels of land or set a price, or a price structure, at which to sell whatever
acreage buyernvanted to purchase. Instead, passing various acts, the best
known of_ which is the Homestead Act of I862, the government either
set a price substantially below the market price or charged no price at
ali. At the same time it imposed severe constraints that those claiming
land had to satisfy before they could gain the right to seU the land.
A possible explanation for these govemment restrictons is that they
were imposed in order to induce settlers' self-protection against raids
where such protecton was cheaper than direct pr.otection by the state._
Had all unsettled land been placed in the public domain or made avail
able for sale, so that anyone could have settled in any unoccupied area,
the cost to the state of protecting settlers would have been high. Such
could have exceecleef the
costs, particularly in isolated or hostile areas,
.
net value of the newly settled areas to thelr owners. On the other hand,
'9B'
.
h
I
f
h' .
f
ecause It IS eaSIer to measure t e vo ume o water r an It IS to measure some o
its.other attributes, it is relatively easy to see that a reduction by one owner of the
10]ohnson, Gisser, and Werner (I9 8I ) di~uss water rights in New Mexico in a
amo.unt of.~at~r returned to th~ stteam will not reduce the amount of water avaiJable
t?):li,~r::p..rl?rBltY,{)~~<;:S ..~~~.l1:.f"t~e.!_~r.eI~.()r:;.atfe~far:t~er.d.?:~I::lS!r,el'I~I:n.-. td!J.~r.>:t.llae)()~~f~$imilar vein. In a study of water rights in four western states, Gsser (I 995) elabo~ates
prlonty.u~rs .. ut a reuuCUon In tue qua Ity o returne water Wl . ten. t? re ur:;e ~ne
on the analysis I offer here. He demonstrates a strong trend toward a c1earer dehnea
~~~er quahty. av.a~lable to .al! those downstream, regardless ~f thelr pnontJ:. Thls.lmton of rights to water in New Mexico 'and a fitful trend in Arizona, California, and
phe!> tba~,. prlOntles remammg constant, the value of the nght to water in a glven
'Ti
s
stniam is likely to decline the farther downstream it is.
exa "
I20
I2I
Economic ana/ysis of property rights

the total cost of protecting an area and the share of the government in
such cost dediried as the density of settlers in the area increased. Simi
larly, these costs dedined if the new area abutted previously settled areas.
Whereas the governrnent routinely seemed to take upon itseff the pro
tection of its people and their property, the govemment as the owner of
the land could relatively easily withhold protection from areas it chose
to keep unoccupied. By categorizing whoever was occupying empty land
as an interloper or a squatter, the governrnent reduced its obligation to
protect such people. Because the govemment w~s tli oWrie.rf the land,:
. the decision as to whetheran areawasvaluableenoughto justifyits'

.. settlement rested with itself. It could take into accoUntthe cost of protec
tion in determining whether to open a new area for settlement.
The restrictions associated with homesteading, particularly those im
posed by the 1862 act, are consistent with.the hypothesis that their func
tion was to induce self-protection. Dense settlment enhances self
protection and economizes on policing and military costs in general. l l
Correspondingly, homesteading was practiced in areas where'lndian na
tions were most likely to contest settlemeni by whites; not in areas where
such a threat was absent. A central feature of homesteading was for the
govemment to select a unit,of land large enough (I60 acres) to support
a family and to restrct each settler to one such unit. 12 A less obvious but
equally important feature was the attempt to ensure that the land would
actually be densely occupied. This may help explain not only tbeJow
pecuniary price (zero in I862) of the land13 but also the'federal govem
ment's policy of waiting to open new land for homesteading until1:he
land became so valuable that a large fraction of the units open for home
steading had a good chance of being daimed. 14 This underpricing policy
is probably what generated the excitement associated with the opening
of new lands. 15 Homesteaders had to satisfy two additional constraints
l11n the first edition of this book I considered only self-protection. The more general
hypothesis is due to AIlen (I99I),

.
121n later acts applying to the more arid land farther west, the unit was increased
W'up'Uf6"4<:)"'actes;'
13Settlers, however, could preempt the land for $I.2.5 per acre after six months'
residence.
4
1 This argument paralIels Cheung's (I977) regarding:the phenomenon of owners
"underpricing" the better movie theater seats in order to get those seats fully (or
densely) occupied, thereby securing effective polidng by the patrons themselves.
IS Ali homesteads in each newly opened area were 'released by the government under
diii same set of constraints even though they were not llequally valued. Allocation
~:r~~~;v:~uc~~:~~:tb~~: h:riI~Ftei;rl~<:6id:;f~~~r~littis~fii~~'~:k'th~;:d'!th;........
gun; henc the term "Sooner, Conversely, the.l,?west-value homesteads remained
undaimed. The Graduation Act of r854 provided for progressive price reductions for
such unclaimed land.
.
Wea/th-maximizing constraints on property rights

before they were granted ful1 ownershp of the land: (I) They actually
had to occupy the homestead for at Ieast five years, and (2) they had to
improve the land before acquiring the right to seU it. These two restric
tons are consistent with the objective of dense settlement: They make
leaving costly. The limited duration of these requirements presumably
reflects the desire of the government to enhance efficient cultivation - to
permit the exchange of land in order to reach holdings of an efficient
17
farm size 16 and an efficient levei of labor input.
Texas is unique among the states in that it joined thelllli<>'ll~S.:lil,J;"~Pll.b..~:.
lic, not a territory, and the public land therewas"heldby the state rather
than the federal govemment. Although it faced serious protection probIems, after it joined the union the cost of protection was borne by the
union as a whole. Prior to joining the union, the state subsidized and
restricted new border settlers. That the purpose of the policy was to en
hance protection is demonstrated by the fact that after joining the union,
.lhe state made virtuaUy no land grants to settlers, instead selling land in
order to release it.. 18
.... P.RIVATE CONSTRAINTS, MAXIM.IZATION, AND
VERTICAL.. .I.NTE..GK~:n.9N ..
Making convertibles for Chrysler

In the two p~e~ious sections I have attempted to explain the role of gov
ernmentl onstraints on private ownership. The constraints discussed in
this section 19 are strctly those between private parties. In I984 - and
most likely in other years as well - Chrysler cORtracted with another
firm (seemingly Cars and Concepts, Inc.) to turn its automobiles into
convertibles. That firm did not act as a subcontractor; it purchased the
auto bodies from Chrysler and then performed the conversion. By pur
chasing the auto bodies, the conversion firm presumably became the
owner pf these automobiles; nevertheless, its ownership was severely con
strained, First, the contract between the two firms stipulated that only
Chrysler could purchase the converted ca.rs. Second, therepurchase terms
16The size of the individual holding and the consolidation of holdings presented
particularly acute problems where the land was used for grazing, which requires large
blocks of land.
17 Allen (I99 I ) elaborates on the hypothesis that the purpose of the Homestead Act
was to reduce the cost of protection by government. He ingeniously exploits differ
.... ences.in..ronditions..across..the:Unite.d.S.tates.. (;ipd ..()ther countries) and the changes'
fcom the 1860s to the I930S to test the hypothess, which his results conclusively
support.
18The Texas account is taken from Allen 199 1 .
19The information in this section was provided through casual conversation.
I 2
....................

were stipulated ahead Df time; Chrysler agreed to buy only cars whose
conversions met its specifications and at a predetermined price. Finally,
the conversion fum guaranteed the conversion; when selling the convert
ibles to the public, Chrysler guaranteed them but turned guarantee work
concerning the conversion over to the conversion fum. It may appear,
then, that the rights of the conversion fum were attenuated. Upon further
examination, however, it becomes dear that the restrctions effectively
delineated ownership.
.
Of course, Chrysler could have operated very differently. It could have
"'crea:rd:'a'separ'ate':departmen:t:or:divislon::to"ma~e::convern6fes;In1:hat
case,however,the head ofthat department would riecessarily Ilave been

a Chrysler employee. Since employees' pay is not strctly a function of
performance, the incentive of the manager to economize would not have
been as strong as that of the residual daimant to the conversion opera
tion. Induce:p1ents such as bonuses and stock options would have helped
but could not have resolved the problem fully.
The constraints Chrysler imposed on the conversion fum provided the
latter with a strong incentive to perform the conversion effidently with
out impinging on Chrysler's own profits.lt would have retained any cost
reduction it was able to effect and would have lost the value of cars that
. .(;()uJdIl9.t . p.1!=!~~ ~h.e..J;~p'-:lJ;(;h.a.~~. sp~#ia,tiQnS., Thf:;d~cal success of such
vertical disintegration must have hinged crucially ou the transactors'
. ability to cheaply measure the conversion-related attri!;mtes of both the
cars Chrysler first sold and thase it later repurchased. 20 The relative ease
of measurement of these attributes permitted the conversion firm to be
come, in effect, the owner not of whole cars but of the conversion attri
butes of these cars. Far from attenuating the rights of the conversion firm,
the restrctions p~;~itted that firm to become the owner of a subset of
attributes and then to assume the consequences of its own operations,
thereby largely removing its incentive to shirk.
Storing truck parts

For severa I years Sajac, a firm specializing in 10ng-termstrage;stOred
spare parts made by Paccar for the Kenworth and Peterbilt trucks Paccar
manufactured. The contract between Paccar and Sajac stipulated, among
other things:
20 Accordng to Wi1li~son's (1:975) and Klein, Crawford, and Alchian's (l97 8 ) .
view, it may seem that the conversion firm was earning quasi-rents capturable by
Chrysler.lf measurement of the pertinent magnitudes is indeed easy, then disputes can
be decided in the courts relatively cheapJy. The fundamental probJem is nor thar of
the capture Df quasictents but, rather, that of the measurability Df rhe attributes Df the
transaction, which in this case did not pose a serious problem.
12 4

Sajac would pay Paccar the serap price for the parts.
For a four-year period after receiving a batch of parts, Sajac would
make the parts available only to Paccar.
3. Beyond the four-year-period, Sajac would have the option of scrap
ping the parts or continuing to store them and to make them available
to Paccar.
4. For parts Paccar ordered back from Sajac, Paccar would pay 90 per
cent of the accounting cost.
5. Paccar could refuse to accept parts it. determined not to be in good

condition.
' .
".
....
...
1.
2.
This arrangement was advantageous to Paccar for income-tax purposes,
since the InternaI Revenue Servce (IRS) had stipulated a few years earlier
that firms such as Paccar could not write off the cost of parts they did
not actually sell. By selling the parts to Saiac at just a fraction of the
accounting cost of these parts, Paccar was able to write these parts off.
The IRS, however, challenged Paccar's 'write-off of the parts transferred
to Sajac, claiming that the transactiQn did not constitut a sale because a
sale requires a change of ownership, nd coditions such as items 2-=-5 in
the list ensured that Paccar would remain the true owner of the parts.
Granted that the tax' incentive induced the arrangement at least in part,
wht accounts for its odd features?

-.
The view that commodities are made up of many attributes and that it .
is sometimes advantageous not to have a single persbn orasingle Hrm.
own all of them suggests that there is no definitive answer to the IR..S
challenge.21 Paccar did remain the owler of some attributes of'the pa'itS;
but Sajac became the owner of others, namely, the storage attributes.
Sajac's gains came primarily from lowering the costs of storage and from
selling the parts back to Paccar, Sajac apparently made efforts in both
respects: It maintained facilities for low-cost, long-term storage, and it
devoted resources - essentially making sales pitches t0 bring to the'
attention of Paccar's replacement-parts division the availability of vari
ous parts it was storing even though it appeared that Paccar continued
tObwntheparts.
It seems that Paccar's main objectives in restricting Sajac to selling the
parts back were to maintain its own reputation and to prevent Sajac from
taking a ride on the Paccar brand nam by selling truck drivers low
quality parts bearing Paccar's name: Paccar could have protected its
brand name by storing the parts direct1y.or by renting space from Sajac
in whichto store the parts, thereby retaining control of their quality. In
neither case wouMtheincentivesfor'e:ffici~t"$torage hllvebeenasstrong....
21The court, however, ruled for the IRS.
12 5
as they were in the actuaI case, where the reward, on the margin, be
longed entirely to the fum in charge of the storage.
than one person is a major source of side effects b~ause the own~r of a
subset of a commodity's attributes can relatively easlly consume wIthOUt
charge some of the attributes belonging to oth~rs. It i~ toa expell;sive,
Vertical relationships
however, to price and police alI the effects assocIated wIth transactl0ns.
.
One method of reducing the costs of the attempted capture by some peoThe relationships betwe~n Paccar and .the storage finn and betw~en
pIe of others' rights is to impose constraints on the for~er: One farmer
Chrysle: and .the converSlon firm shed hght on the problem of ve~cal
might be able to capture another farmer's water by sWltchmg t~ a ~rop
integratlOn. Llke other firms, both Paccar and ChrysIer are onIy parnally .
that returns Iess of the withdrawn water to the stream. The Iatter s nghts
vertically integrated. In our daily lives""e.ar.~,~c:>~~<:l:~c:>::o~se.J::Yi~g:,~,igl:1I;)!:::::::""::Cn:b'e"pr6teCted:;:hwever."bynot allowing the former to use water to
.integrated !J:ms22 that wefail,tQ'Qskwhy every individu.al do~snot. operirrigate ne~ crops. Such resttictions serve to ma~e rights clearer, thereby
ate as an mdependent one-person fum. The precedmg dIscusslOn of
enhancing them even though they may superficIally appear to attenuate
therelationship betweeI). me two pairs of firms points to some of the
them.
.
As the costs associated with a restriction rise, new methods of pro
' gains resulting from operating independently. When people assume ownership, they also assume responsiblity for their own behavior. On the
tecting rights are expected to emerge. When the value of a new crop or
other hand, when people work as others' employees, or when other proof the water itself increases, I expect that protection of the rights of
duction factors are rented out, .shirking is to be expecte~. When two indelower-priority owners of water rights will be enhanced th~ough the use
pendent fums perform operatl0ns that would otherWlse have been perof a more expensive and more accurate method of measurmg the water
formed within a single firm, however, a market exchange is required. As
used by high-priority owners. The more accurate metering of the actual
I previously argued, any exchange consumes resources, and effecting fuU
use of the water itself should permit the relaxation of indirect constraints
'alienation is often expensive:'Correspondingly, most exchanges involve
on the use of water such as the choice of which crops to grow, and
only subsets ofthe'attributes'ofthegoodsinvolved;In'the'caseofthe'
"'i:he'gr~~t:i~g"to own;rs ofwater rights greater freedom in trading those
truck~parts storage, Sajac, which bo~ght.the p~rts, seems to have operrights.
.
. .
ated mdependently of Paccar, but ItS operatlOns were severely conThe restrictions on water use and the restrlctlOns on homesteaders
strained. Nevertheless, these operations were much less constrained than
have been government-imposed. The framework in which they were dis
cussed is the same as the framework for the private restrictions Chrysler
they would have been had they been carried out within Paccar. Sajac was
more of a residual claimant to the stO'ge operations th~n a ~acca: diviand Paccar imposed on their exchange partners. In e~c~ ?f t?es~ ~ases I
sion would have been. PresumabLy the costs of .the partlal'ali~natl0n of
assume that every decision is ultimately made by maXImlzmg mdlv~duals.
the parts first from Paccar and then back to It - were relatlvely low;
In none of the cases are marginal costs expected to be equated wIth the
theref~re Sajac, the s~orage firm, assu~ed responsibility for the s~orage
corresponding marginal valuations. In each case, however, f?rces. indl:
ope(atlon. The costhness of tra~sfernn~ other spare-f.>arts attnbutes
ing toward such equalization are assumed to be present. ThlS pomt wlII
seems to be the reason Paccar restncted SaJac to return of the parts rather
be elaborated on briefly in the next chapter.
than,aII.ow:ing it to dispense them without constraint.23
CONCLUSION
The. costliness of transacting makes undesirable side effects a virtually
inevitable consequence of exchange. Ownership of coinmodities by more
22 A supermarket, for instance, is a firm that seUs a huge array ofpfoducts and
whose emplQyee:s perform ,a wide variety of vertical (as weU as horizontal) operations.
"""23The"amblgti'igdng wnO the true owner of the parts was, and consequently
the nature of Paccar's scope of operations, ilIustrates Cheung's (19,8,3) observation
conceming tne difficulty of determining what a firm actually is, and points to the
advantage of concentrating on contracts.
I26
': ,,;'
.:\ :::;.:c~;:[~;'l~;{ T~ ;;.' ,;.!~i:;~~./,t';:;'~~~]3:'-J~~~~~':
~,;,.~:,
;':'. ,~:,:i~~B;;~~',X.;'y,,",
""~,,,
..\,'
",~;.;
<'. :"/
;'~T
:.
::.1:' !( ,.!.>. '.'!; _'o '_.
>'
:'
','~,'
Property rights and non-market allocation
The property rights approach to the study of economicshas been pro

moted by market-oriented economists, who have used it occasionally to
demonstrate the superiority of the ~arket. Contrary to the perception
that property rights tools may best -be used to analyze the market econ
omy, where allocation is performed largely if not entire1y by prics, these
tools seem to be uniquely well suited to analyzing r~surce allocation
in. non-market settings. Under the :Walrasian approach, where r:iglits,are.,
perfectly de6ned, nothing is lost by quickly dispensing with the topic of
property rights, for there is little to say about them within that model,
where prices determine everything. Indeei!, the Walra"Sian model ma'y
provide satisfactory answers ~ many p~oblems i~ c:pitalist etnomls,
where prices play a vital role in economic life.
Those economists who have contributed most to the study of propertj
rights tend to be strong advocates of unregulated markets. They contend
that people and the economy thrive when left to their own devices, and
that government intervention tends to reduce wealth. In the market, the
argument goes, prices move resources to their highest-value uses; when
prices are not given the opportunity to perform their function, misalloca
ton results. Governmen't intervention is deemed acceptable in such areas
as national defense~ police, the courts, and the rnoney suppJy; such inter
vention is said to be desirable only inasmuch as it faCilitates the function
ing of markets. Despite the prevalence ofthis reasoning, government reg
ulatons cannot be dismissed on a priori grounds. In this chapter, I
explore some ofthe conditions under which government regulations may
complement o' enhance the actions of individuais.
For non-market economies, where market prices are eschewed or sup
..,....... P:r.~~.!>.t::4,-.t.h~.:Wa.:1.J:.~~i<'l.J.l.,~Q4~J ,i~.. iJ.lc::a.:p.~I:>I~,c:>f..e.~p~<l,iJ.l~I:l~ ,.h?,v.v:,.!~~,2tl,J:<::~S.., ."
are allocated. Herethe' property rights approach attains the utmost im
portance in the:analysis,of decisions about aUocation. The irony is great,
for champions o{',the' .free market have developed tooIs that are most
128
powerful when th~y are used to analyze non-market - including social

ist - economies. 1 Although I believe that the property-rights approach
applies to alI human behavior and to all human institutions, any serious
attempt to demonstrate this would prove toa lengthy and toa speculative.
In support of the assertion I shalI, however, fust offer a brief discussion
of the applicability of the property rights approach to two specific areas
of non-market alIocation: (I) allocation by voting in market settings,
where it is shown that individuaIs sometimes choose to bypass the market
and instead adopt non-market alIocation devices, and (2.} alIocation by
voluntary charitable behavi()r.,::whereit:dea'ilY:demonsli'atessn::iidvaii:':,::":::,'..:::' ..........."",'.'",
tage over markt"beh~vi~: I will then briefly examine the function of
private property rights and the method of inducing people to perform in
a non-market economy.
ALLOCATION BY VOTING
On occasion individuais allocate resources by voting - a mechanism that
explicitly bypasses the use of prices in favor of non-price allocation. Indi
viduais who use markets retain controI of how they use their wealth,
m.'aintaining fuU d'lscreton over what to purchase. Within voting organi
zations; onthe otherhand;individualsare'subjeet to'constraints' imposed',
on them by their felIow voters. Yet by their own behavior individuais
demonstrate that they value some of these constraints even though they
reduce their freedom of action.
Voting is used not only in the public sector but also in many profit
seeking settings, including shafeholder corporations (mostly to elect of
ficers) and in condominiums (to make an array of'01'el'ating decisions).
The origins of organizations such as corporations and condominiums, in
which voting is used, lie in the operations of entrepreneurs. In the case of
cortdominiums, for example, developers typically erect the housing units
and related structures and complete other prepanltory work before sell
, ingthe units to individual buyers. Developers do not have to seU housing
units as condominiums. Another option, one that maintains the owner
occupier tax advantage, is to seU the units to individual buyers, who op
erate independendy of one another. It is presumed that developers will
generally choose the method of sale that promises the largest difference
between the aggregate selling price of their units and the costs they incur.
In fact, condominium developers are offering buyers packages that
consst of both the physical structures and the rules that-will govern some
....aspectsc,ofthe.prospl'!ctive owners' future behavior - rules that include
1 As Wing Suen ponted out to me, the irony is compounded
the championing of
the use of prces in socialist systems by such eminent economists as Lerner and Lange.
12 9

decisions by voting. These developers offer the units as condominiums
rather than independent ones because they expect their buyers to pay a
higher price for condominiums than they would if they were to purchase
the units independently. Clearly, condominium buyers valle packages
that constrain them to allocate resources by voting (e.g., on whether or
not to build a swimming pool) more than they value deals that, by con
tras~, allow decisions regarding the supply of such services to be made in
'the market.
Although no attempt will be made here to explain the rationale Jor:.
:'sl1ch beha.'Vlt,2 individuais reveal through their behaviorthat theysome:"
times peefer non-price allocation to allocation by price.3 The us.e .of
prices, then, is not always the most efficient method of.allocation. The
next section points to OJle advantage of non-market allocation.
DONATED BLOOD VERSUS PURCHASED BLOOD
An attempt to demonstrate that unhindered markets are b~st is found in

Kessel's (I974) analysis of blood donation. The puzzle he ttempie to
solve was why, when blood was purchased, the incidence of hepatits
among transfusion recipients was higher than when it was donated. Kes
sel examined both the mechanisms used to provide blood and the appar
ent advantage donated blood had over purchased blood in avoiding the
transmission of hepatitis. He reached the condusion that'a more vigorous:
pursuit of profits would nave secured higher-quality blood in the market:
Despite his masterful command of theory and evidence, Kessel's explana
tion of why the market for blood did not function adequately and why,
in fact, blood purchased in the market was no bargain is not compelling.
Missing from Kessel's model is the recognition that effecting market
transactions is itself costly. Alternative mechanisms, where resources are
allocated by inducements other than price, can sometimes generate a
higher net gain than market transactions. The analysis of non-market
organization for the supply of blood will provide a different explanation
...as.towhy..thequalityofdonatedblood is higher than that of purchased
blood.
Blood, like other commodities, is a collection of atttibutes whose leveis
vary from one specimen to another. In particular, some bloo~ spcimens
2See Barzel and Sass I990, for a detailed discussion of this topi~.
3Developers whoconstruct housing units in unincorporated areas, to Jje .sold as
condominiums and to be managed bY',:r homeowners' association, are basically .devel
oping whole poltical units. In this case incorporati'011'l"S'an''pti't'fr"tesi4ents:"P'i'i':"'"
sumably developers take this into account. Here a political unit may emerge as a
consequence of the quest for profit. Buyers implcitly choose the associatei:! poltical
restrictions.
.
.

are infected with hepatitis while others are free of it. The lower the prob
ability that a batch of blood is infected with hepatitis, the more valuable
it is. In general, buyers would not pay a high price for the low-quality
specimen if they could cosdessly (or at least cheaply) assess it, as the
Walrasian model assumes. but this is not the case in the market for blood.
By simply posting the two prices at which they would be willing to buy
the two types of blood. buyers are unlikely to secure the desired qualties.
Because sellers prefer selling their products at higher prices, blood buyers
.need'to be able to determine which grade of blood they are getting if they
.wish t avoid paying the higher price for the inferior commodity;lnthe
time period Kesselinvestigated, however, the test for hepatitis infection
in blood was virtually worthless, so it was difficult for buyers to distin
gitish between the two kinds of blood. As a consequence, the market
for. blood perfonned poorly, and the proportion of infected specimens
purchased was high.
Despite the inaccuracy of tests, knowledge of whether any blood speci
men was or was 'not tainted with hepatitis was not a1ways difficuIt to
come by; because people often know - or at least suspect - when they are
carriers of the .disease. The problem is that impersonal markets are ill
suited for extracting the' information and dei:rmining the qillity f such
a 'commodity as blood. Because information aboutblood 'jsnotcostless'"
to ali concerned, the "alue of market exchanges of blood is greatly low
~re.d. The jnfQ{llation i~ free tp. subset of individuaIs - the sellers - but
the market is 'unlikely
extract he information costlessly because seIl
ers,who can.gain by conceaIing}t, are abIe to do so.
Using a monetary reward happens to be disadvantageous for securins
hepatitis-free blood. The incidence of hepatitis among drug addict.. is
high because they tend to infect one another by sharing needles; these are
the same individuais for whom cash for blood is a particularly attractive
.trade.4 The attempt to purchase blood in the open market, then, is likely
to atttact a relatively. large proportion of carriers. 5 Separation is never
theless possible, but: J:lotwithin the unrestricted market setting. Obtaining
blood from donors insteid of purchasing it for cash alters the selection
criteria of suppliers because it tends to cheaply screen out people who
k:n.ow they are carriers. Would-be donors must be persuaded to donate.
They respond because they wish to he1p. (lther human beings. People who
to
4The fact that transacting is costly and needs facilitating suggests that cash is not
as neutral as it is often thought to be. If, for example, blood sellers were paid in non
transferable tuition vouchers, it seeInshigl1lyl*~Iy..~h.-at...tll.e!r.a.c.ti()l1,?finfected addicts
"""amongsellers"woilloeS'malIertlin 'ir was with cash as the means of payrr/ent.
5Kessel identified various methods of screening out individuais who are Iikely to be
carriers. These methods, however, are costly. Sellers of blood were generally not
screened, whch implies that the cost of screening must have been toa high.
I3 I
C&:,~.':;";~'"_'=~"~.:'L>.";z{So2l~:t:,::~~:;~~L~.,;c;,,;B~i:,~~i:;.:2;;:~:~~;:;~~~~~~~~~~:ifi~;~l:i8E!~~
Praperty rights and nan-market allacatian
know or suspect that they are carriers are aware that their donation will
prises, and some government enterprises may actually seem to operate in
do harm. They are simply expected not to donate.
that way. Even when profit maximization is the stated objective of the
af course, cash markets are not used by accident; they ~co?omize on
govemment enterprise, however, the identity of the residual claimants
some of the costs of effecting exchange. The use of donors m heu of cash
remains unclear.
markets incurs costs that are absent from cash markets. For insta~e,
The difficulty of identifying such individuaIs and the common claim
volunteers are not generalIy the lowest-cost suppliers of the commodlty
that government is generally wasteful and inefficient both suggest to
or service they donate. The costs and gains of using markets as compare~
many that private-property rights are absent from government opera
with other allocation methods, such as charity, differ across commodltions. Indeed, former communist countries claim to have abolished.
ties. Given the difficulty of testing for the presence ofhepati~is and the .
. private-property rights, at least with regard to the mean~ '?f l":rduction:~ .
HN virusthatcauses.:AIDS:,;:blaod:is:.:one:commodityJor:whl'Ch::the::ad.:.:::::::::::::..lshal1rgue that private propertjr rights must exist m'a functioni+ig econ
vntagesf non-market over market allocation are evident. .
omy? I shall also argue 'that the notion that govemment is .inefficient
The cash-market sellers' ability tpgain by knowingly passmg off lowcannot be correct. After some general comments on.individual maximiza
quality specimens as high in quality .is common to many co~odities
tion, I shall try to determine what can be inferred from the sheer existence
besides blood. Nevertheless, charity is not expected to be used m all such
of a government-run enterprise, using a city bus syst~m as the example. I
shall then consider the meaning of an arbitrary detail of its operations _
cases. The precise nature of the quality problem is likely to differ from
one commodity to another; and each should generate its own charitable
the activities of bus drivers - and move on to more general aspects of the _
response. Mechanisms other than charity may be more effective in securoperation of such a system. I shall not inquire directly into what the
ing the desired quality for some of these commodities.
proper areas for government activity are; rather, I am using a propertyAlthough a donor's motive for c~oosing o~e .charity over ~nother
rights approach to offer a glimpse into the way a gov~rnment.fun~tions.
seems to be a matter of taste, one slmple pre.dlctlOn c~n be d~r~ved: .lt
The assumption of individual maxirp.ization is heavdy e~lOlted m ~cQseems reasonable to assume that, like other actlOns, charlta~.leJpv1Il&.wtll
nomics literature that features an analysis of profit-seeking enterpnses
expariWherithe'gairistgeiieiatesbeco~e'rigei:6 On. the issue .of
and private consumption. I wish to examine the implications of individ
blood, it is predicted that in the span of time after the dlscovery that
ual maximization for government operations. Because government.s are
'.
blood could be infected with the HIV virus but before effective tests for
run by people, government activity ultimately results from the interac
it were devised, the amount of blood donated relative to that sold for
tions of maximizing individuais. Contrary to the implied assertion by
cash increased substantialIy.

some economists and other scholars, one consequence of maximizing be
.....
havior is that government actions are never deliberately wasteful or ca
pricious. If "waste" means that some individuais lose from an action
ALLOCATION BY GOVERNMENT
from which no one else gains, such an occurrence is inconsistent with
maximizing. Whoever takes any kind of action must expect to gain from
The government of every country plays an important role in economic
it; indeed, the perceived net gain must always be the largest one available.
activity, and some governmertts' roles are enormous. Ali governments
An action that appears to others to have been wastefu! must, neverthe
engage in non-business activities, such as conducting foreign affairs and
less,
have been e:xpt:(;tedt()gellerate a gain to the person who underto~k
operating the courts, and virtually alI also engag~, to varying degrees, in
Ti:. Moreovr~sch a person must not have believed he or she could gam
more businesslike activities, often conducting operations that in some
more by acting differently. The resources under consideratiQn are ulti
other countries are run privately, usually for profit. It is occasionally ar
mately alIocated to a particular use by whoever is in cOlltrol of them.
gued that governments should seek profits when they manage their enterThe logic behind the allocation is straightforward. $uch resources have
6This account does not offer an explanation as to why people give to charity, and
alternative uses, each with its own valuation. The ability of potential
<
it does not attempt to predict when people wOllld give more or less. It does, however,
shed light on the social value of induc:ing sllch a mode of behavior. To make the
notion of giving operational, it is.enough to assume that charity is a commodity that

provides utility to some individuais, and then to apply to it the law of demand: The
lower its unit cost, or the higher the.llllit benefit, the more units will be "consllmed."
I3 2
7In the aftermath of the breakdown of Eastern European commllnism, many of the
functionaries of these regimes who contnl:ledto-(')pe:t'ate.jn-theiTold-in~i'tu.tions-actu"
ally landed on their feet. This implies that these individuais hd sllbstni:ial economic
rghts.
users to bid for the resources, however, is subject to constraints. For ex

likely to be puni~hed. So long as the city bus system is not totally paraample, the constraints may be that bids may be made only by citizens or
Iyzed, property rights over it must exist. lO Various observed bus-system
by party members. Similarly, bids may be restricted in form(e.g., lecture
activities, which I outline in the following paragraphs, imply the specific
existence of certain private property rights.
fees or promises of future high-paying jobs, open cash bids being prohib
The fact that riders use city buses is an indication of the existence of a
ited). Whatever the rationale for the constraints, 8 a maximizing control
ler of resources will allocate such resources to the person who makes the
whole system of private rights. The activities of employed bus drivers
highest bid as determined by the controller. The winning bid generally is
illustratethe mechanics of this system. Employed drivers will be fued
not the same as the one with the highest value in the absence of the.
unless they perform some minimallevel of services. The drivers engage
constraints. In this regard, resources may.appear.tobewasted;.:pan:ic'~::::::::::::::ht.::~~..bl'lg~:.:w.~~b:.Jh.~i.I'::s.IJ~ryis.f.?,:il.'4e.xfhanges constitute a reas-.
>larly in the eyes of those whoare not' ware:ftheconstraints. Given the .
signment of property rights. Here the drivers acquire the right to a wage,
constraints, however, other potential users of the resources that appear"
while relinquishing some rights over themselves by performing driving

services.
to have been wasted must not have bid high enough for them. Applying
Drivers may be asked to do more than the speci6ed minimum; if they

this reasoning somewhat more generally, it can be concluded that the
lower the perceived net gain from an action to the individuaIs who have
are to perform beyond the minimum, they must be given an incentive.

Such an incentive need not be higher pecuniary pay; it may take the form
the right over it, the lower the chance that such an action wilI be taken,
no matter how bene6cial the action is supposed to be or how large its
of a better chance for promotion, a more convenient work schedule, or

an easier route. Whatever its form, no extra effort will be forthcoming
expected gain1i absent the constraints. 9 One must bear in mind that the
higher the potential benefits of an actiori, the more the beneficiaries are
without it. Drivers have at least some rights over themselves; they con
abl~.to~id.!()~ it; "ability," however, II1~YlJ(!difficult to translate into an
trol, that is, they own the rights to, the levei of effort, and theyexchange
ctual bid.
. .
rights over particular effort leveIs for other rights .

......Employeddi-ivetsanti'(operate in isolation. Someone in the bus sys
. In order to be able to nlyze>iridvi'dufbhavIorin' a..goviiimnt
organizatiqn, it is necessary to address the relationship' between private
tem must gain from forming the contracts that induce the drivers to per
ownership :}nd government activity. Because maximizing people will act
form busing services. Smilarly, the mere fact that drivers are given routes
only when they expect to gain from their actions, one must be able to
to drive and schedules to maintain implies that someone has been in

. determine who gains and who loses from government actions in order to
duced to perform the functions of creating the routes and schedules. Ir
connect the actions with ownership. Selecting, fo~"~xample, a dty bus
can likewise be inferred that ce.ttain individuais are induced to maintain
system, one must ask who owns it, that is, wh6 has at least some of th' ... .
buses; otherwise the buses would not be in operating condition. As long
power to consume its services, to obtain income from it, or to alienate its
as bus lines operate, then, a whole array of people must be rewarded
assets. The "city" is not a satisfactory answer, because it does not identify
for performing individual functions. The bus system may be managed
the individuais who gain when the buses are running on time and lose
bureaucratically and may function sluggishly. Still, some property rights
'when they are noto Answering "the dty" denies that such individuais
must be granted to the individuais assodated with it; otherwise no service
exis~ and therefore implies the claim that there is no residual daimant to
whatsoever would be forthcoming.
the operation of the bus system. Yet if no one gains from improving the
Governments often seem to.set.outp.uttar.getsatleveIs..s.uchthatat the
operations of the bus system and no one loses by letting the system deteri
margin valuations differ from costs. In communist countries such goals
orate, t must lie in the public domain. Allowing the bus system to deter
were often stated explicitly. The constant shortages characteristic of such
orate requires less effort than maintaining it, so it would cease to function
regimes arose from prices that had been set lower than was' required to
if it lay in the public domain. Similarly, it cannot be true that the bus
lOThe claim, made by leaders of communst states, that private property there had
system lies in the public domain when access to its assets is constrained.
been abolished and that alI pcoperty belonged to the..l:l~a.t~seems to have been an
attempt to divert attention fcom who the true owners of the property were. ~y.dently
For instance, in a functioning system, attempts to commandeer buses .are
8The constraims are, of course, also imposed by individuais who are maximizing:'
9The same considerarions apply to the process of selecting government bureaucrats
and government projects - and ultimately to the selection of the government tself.
these owners also owned the rights to the terminology. Ir is ironic, ar least in rerms of....
the rhetoric, that communist states had a hard time keeping resources such as air and
rivers from getting imo the public domain (the ultimate in collective ownership!). As
a result, the value of these resources was reduced to a levei much lower than the one
they occupy in capitalist states.
I35
beca use the individual would reap the difference between the gains and
the cost. Moreover, such a service would be expanded until, on the mar
gin, the cost equaled the gain.
In a poltical system, given the way such systems usually operate, the
effect of demand forces is less direct, but it is definitely not absent. The
system of rewards in government seldom compensates individuais for the
full private gains they generate. An operator in the poltical realm unable
to claim :1:00 percent of the residual would stop short of the levei of
service a private operator would reach. Although the same forces that
bridge gaps created by price contro~~..t.e.~<:f.~<>,pJ:e:yentgapsbetweenmi~::.
ginal valuations and marginal costs from growing ever larger, a full
fIedged residual claimant is less handicapped than one who has only a
partial clairn; the full-fIedged daimant will consequently produce what
appears to be a more effident outcome.
Why would an operator not be allowed to daim :IOO percent of the
residual? MQre generally, why are individuais not always allowed free
rein to become residual daimants? Prohibitions mustperform real func
tins. The superior who has constrained the operator presumably has the
power to impose constraints. 12 Returning to the example of suburban
. busser:vice:Thesuperi:or may have had.no.interest in increasing the oper
ator's wealth but could still have been able to increase his or her own
. wealth by selling the righ~t~~~~~~'th~'~~b~~b:Iii~d~rto'aiiswer"i:he '
questiono regarding the absence of a full-fIedged residual daimant, one
nrust also-ask what prevented the superior from selling the right to the
opeJ:ator or, even better, to the highest bidder. I conjecture that the an
swer lies in the fact that the seller's wealth depends.not only on the pecu
niary price but ais o on features of the buyer a?d of the exchanged pr~J2: .. _
erty.
In general, maximizers may choose not to seU an asset or a franchise
to the highest bidder. There are many reasons for this reluctance, ali of
which may be viewed as resulting from the presence of side effects. For
instance, manufacturers sometimes pay their salespeople a commission
. ora salary insi:(iiid of selling them the merchandise outright. By not grant
ing salespeople full residual-daimant status, manufacturers ensure that
11S
non-market economies a
11 d"
.
d"
. .
their incentive for capturing wealth from other salespeople is tempered.
ome,
,.
re ca e
comman
economles; a command
"
b
eC~)flomy IS ~ne In
WhlCh the planner imp?ses an.utput targetalong with the control
~n a?other case,persons w hO ~annot fu ~ Iy gu~rantee t helr
acUons m~y e
pnce. Imposmg as the term may sound, It will be obeyed only if it sets low targets.
llldllled to take unduly large nsks. The lllcentlve to undertake excesslvely
risky projects where some of the risk is borne by third parties is reduced
Because mputs are neve: uniform, and beqluse of random fluctuations, the output
ta~get asth<:mcommander
set must,
as a filie,
short
ut. Other
'f h d " k
II
d to become f u Ii -fied ge d resl'd ua I .
wIse
the case ofwIII
slave
the comma
d fali 'lI
. of Imaximum
b out.Ph'
I te. eclslon ma ers are not a owe
bl'
rvJ>
n er WI mcur osses y pUnIS mg those
I .
-
lll1:a .e to ?bey the.,,Qmm.t!d..b.e.c.a.us~jUs.:.excessiv.e...If..o.urputjs,costly:, to..measure",. , .......... , .....~.,a.I~~:?~:. ,..... ...... :,..,:.. ........ . . ; .
~hl.rklllg wlth r.egard tO,the target ou~put may :ls,? ccr. The quantity supplied, then,
One reason resldual-dalmant nghts are not granted may apply pnmar
clear the market and from allocations that simply seemed arbitrary. Gov
emment enterprises were affected by such policies in various ways. How
would the operations of the bus~system repair shop have been run under
such conditions? Presumably, from time to time the repair shop would
have been hampered by a shortage of parts. As Iong as buses were run
ning, we must condude that individuais in the repair shop were being
rewarded for getting the buses to work and that the rewards were larger
when the repair-shop services were better. Repair-shop personnel would
have gained by having parts on hand, and conseql;lently.should nave been
willing to spend resources in order.to secuie them'-They might, for in
stance, have aitempted to trade wn rhe repai,r !moI? ~f another city's bus
system or with a truck repair shop. Alternativly, they might have 'offered
a special reward to the parts producers for furnishng extra. parts. Such
producers, as is generally true under price controls, would not have been
likely to produce at full potential ~or the controlled-price segment of the
market. They might readily have been induced to expand output, how
ever, if the reward had exceeded the controI price. l l
The details of sum operations cannot be.determined by an armchair
economist, but maximizing behavior implies i:hat the discrepancies be
tween marginal valuation and marginal costs must generate forces to
ward their elminatidn. Indeed, as was shown in the discuss!on of price
controIs in Chapter 2, once the added adjustment of transaction costs is
accounted for, the discrepancies must be elminated entirely. This attain
ment of equilibrium simply follows from maximiztion. Th equilibri'um:# ..
itself, however, is likely to differ from equilibria reached under diffe~nt
sets of constraints. Given the government-imposed restrictions, the ad;
justment costs may be so high that the final output may lag drastically
behind the corresponding market outcome.
The advantages of allowing residual claimants to operate in an econ
omy are dear. In the context of new opportunities for the bus system.L ..
suppose it is discovered that the residents of a new'suburb'are willng to
pay more for bus service than this service would cost. If an individual
were in charge of deciding whether or not to start service, and if- he or
shewerealsothe sole daimant to the residual, service would be provided
~s expected
to lllcrease lf the reward IS larger;
wiII n o t be f or th commg
. I'f'It
1&
&imply commanded.
. _an.mcrease
..
.
..
. here
.'
. authonty;
.
12 For slmphclty,
the supenor
IS assumed to possess the ultlmate
people intermediate in the hierarchy are ignored.
I3 6
:I37

i1y to despotic regimes.The explanation has to do with the concentration
of wealth thai: free enterprise entails. Bad luck may cause the holders of
residual daims to experience financiaI ruin, whereas good fortune can
render them rich. When many opportunities to assume residual claims
are made available, at least a few individuaIs are likely to become rich.
Rich people, particulady those whose wealth is not easy for others to
keep track of, are in a position to finance coups. Such people pose a
threat to the despot. This, I suspect, is one reasOn why dictators are often
.::::av.ene.:tcr.:free:ente.rpris.e..communist regimes' harsh treatment of "profi-'
teers" may be a case in point. The suppression of opportunities thatmay'
enrich some individuais is costly to detators, who could instead allow
their exploitation while collecting a commensurate franchise fee. It is not .
that dictators are not assumed to be maximizers. Their longevity and the
security of their status are what is viewed as valuable to them. They are
willing to sacrifice pecuniary gains they could otherwise obtain by auc- .
tioning off various residual rights, because these would have posed a risk
to their security. Instead, they prder to operate bureaucratically - a less
lucratve option but one that promses greater longevity.
CONCLUSION
The distntion between the private and public sectors is not to be

equated wth that between the presence and absence of private property
rights. Such rights are necessarly present in both systems. The distinction
lies instead in organization, particulady in the incentives and rewards
under which producers tend to operate. In the private sector, producers
have the opportunity to assume the entire direct effect of their actions. In
the government sector, people assume a smaller portion of the direct ef
fect of their actions. Both systems reflect the outcome of the actions of
maximizers, and both must operate dficiently.
IO
Additional property rights applications
In previous chapters I demonstrated the expediency of using the property

rights approach to explain various aspects of such specific phenomena as
gasoline price controls and slavery, and to develop a general approach
to, among amer things, non-market allocation, the maximizing role of
restrictions on private property rights, and, in the context of farm ten
ancy, the choice among various' forms of controI. The property rights
rramework can be applied to ther p.rbletnsas well. tshall now briefly
consider several of these problems, beginning with an analysis ()f d:l(:in.di~.
viduals' abilty to protect themselves against losses to monopoly, pro
ceedi~ to. a di&cq~sion of the relationship between property rights and
theft and of prperty riglits in relatiOIo innovation and to price infor
mation, and conduding witl]. an e~ploration of property rights as they
relate to wildlife.
PROTECTION AGAINST LOSSES TO MONOPOLY
Monopolies are said to result.in resource misallocation, l which takes two
forms. The first and better-known type of misallocation arses because
'monopolies produce "toa little," charging 'prices that exceed marginal
costs, whereas the second type arises in the process of cieatirig monopo
lies. Would-be monopolists' spend resources in order to attain monopoly
positions, and such expenditures are dissipating. The magnitude of these
capture costs is c'omparable to that of the expected profits of lhe monop
oly. Naturally, monopolists have economic rights towhatever gains they
obtain. Since monopolists' gains exceed their contributions, their gains
. occur, in part, at the expense of others;thus,. they also seem to have the
right to harm other people. Yet such rights are not exercised indiscrimi
:nateIy,andItlsworthwhileto'de"teiiriinetneCi:irist:nces"ii"ndec"whith" .
'. such power is likely to be used.

1A
.....................I:3.l).................. - ..................
............................................... .
more detaiJed discussion of this problem may be found in Barzel 1994.
I39
Economic analysis Df property rights
In general, a propeny that can be captured must lie, at least in part, in

the public domain. Moreover, what lies in the public domain must have
been relinquished by the previous owner. In the case at hand, if someone
is able to capture a monopoly position or, more accurately, if someone is
able to capture, at a resource cost, the rights to the monopoly gain, then
the monopoly position itself must have been lying in the public domain.
In order for these rights to have entered the public domain, individuais
must have relinquished them in the first place. I am here attempting to .
delineate conditions under which individuais would allow this to happen
where monopoly is attained bypr:edatoIYpridi:ig~Iripreaiitory:picig~::
an initiaUy competitlvfdndustry is taken over by a predator, who monop
olizes it by temporarily pricing the target commodity at less than it costs,
thereby forcing the competitive producers to leave the industry or to seU
their facilities cheaply to the predator. This method illustrates the general
principie behnd opportunities for capturing mon0l'0ly gains. 2
In the competitive industry that exists before predatory action, con
sumers are able to purchase the target commodity from many sellers at a
competitive price; this ability is threatened by the predator. Recalling
that property rights a,re defined as an individual's ability to gain from the
consumption and exh~n.ge of goods, it seems proper to ask what gives
consumers .tll~a.~.iIi1:y~() ()1J~ai.n a..go()cl.a.t..tlle.. C;()IllP.et.tiye pricl!: .AIl~i-:- ..
monopoly laws aside, consumers surely do not have a legal right to the
competitive price. Consumers, however, can acquire the legal rights
and consequently the economic rights by the simple expedient of slgn
ing long-term, competitively priced contracts for the commodity while
the industry is competitive. TheY...might choose to commit to such long
term contracts if they fear monopolization bya predator. Long-term con
tracts will also benefit the competitive sellers: It is difficiJi to ruin sellers
who have signed such contracts, since they do not have to sell alI their
output at the predatory price; for the same reason, they command a high
acquisition price from the predator.
As long as the extra costs of arranging long-term contracts over and
above those of spot exchanges are less than the perceived loss to monop
oly, threatened sellers and consumers will gain by establishing rights tO
supplyand to be supplied, respectively, at the competitve price. 3 When
the cost of protection, whether through long-term contracts or any other
21 here take no account of the controversy surrounding the logic of, and the evi
dence for, predatory practices.
3The successful predator will gain more (in present terms) fronl t\.le ultimate mo
nopoly pricing than she or he willlose from below-cost pricing during the predatory
...period.o&.from..buyingouu;ompetitors at terms attractive to i:hem. Conversely, con
sumers and preyed-upon fums will be the combined net losers from such predation,
and it seems highly probable that consumers' losses from the 'monopoly price will
exceed their gains from the initiallower price.
I4

means, is less than the expected loss to being monopolized, the would-be
monopolist cannot gain from the monopolization and therefore will not
attempt it. In this sense the would-be predator does not possess the right
to such a gain. Since the totallosses the would-be monopolist inflicts on
others exceeds his or her gain, as long as contracting costs are not high,
the rights competitive buyers and sellers possess tend to be secure. As
suming that in the absence of defensive action the predatory practice is a
real threat to competitive buyers and sellers, it is expected that the less
;~tb:.iU~JJ~!r.ge long-term contracts in an industry, the less likely it
isthfiliiridustr)TWiUbesbject to monopolization.
In summry, the ability of consumers and competitive sellers to ex
change at competitive prices depends on their own efforts to enforce the
appropriate arrangements. By and large, they are expected to be success-'
fui because monopoly tends to be "dissipating" in the sense that the mo
nopolist's gains are less than the costs he or she imposes on ochers. Still, ..
when enforcement costs are high, consumers' and sellers' rights to be
served at the competitive price are likely to be relinquished, and a preda
tor may then capture the right to a monopoly position.
THE RELATIONSHIP BETWEEN
PROPERTY RIGHTS AND THEFT
The existence of theft makes the distinction between economic and legaJ
rights clear; it also highlights the notion that economic rights are never
absolute. Thieves lack legal rights over the goods they steal. Nevertheless,
they are able to consume the goods, to exdude others from using them,
to derive incme from them, and to alienate them. Each of these capabili
ties is an attrbute of ownership. The lack of legal rights may reduce the
value of these capabilties, but it does not nullify them.
The fact that thieves have rights over stolen property implies that the
current owners of property that might possibly be stolen do not have full
rights over "their" property. Owners cannot be certain of the future use
of such properties. The economic rights they do have depend, in part, on
the protectloneHorti:radebYthestate. iliseighcii also depend on the
measures owners take to protect themselves from theft: The more they
are willing to spend, the more secure their rights are expected to become.
The dfort they make, however, is not expected to deter all theft. For
instance, fences around orchards are not made to be totally insurmount
able; the cost of making them insurmountable exceeds the gain. Here,
too, individuaIs choose to leave some rights inthe public domain. When
the probability that thieves will steal o.ne's apples is"po"S"iivii'tneri'rie:
has only partial ownership over the apples.
Private-protection methods are as varied as are commodities them
14 1 '
selves. Owners 'of apple 'orchards may empl'oY guards to reduce theft;
ence between. total c'onsumers' valuati'on and t'otal inn'ovat'ors' C'osts.
they may place trees farther fr'om their property b'oundaries than they
Each of the potential patterns of activity leading t'o a particular innova~
would in the absence of theft; and they may grow less valuable apples
tion will generate its own net present value. The highest value will be
that are less appealing to thieves. Indeed, landowners might grow entirely
generated by the innovative activity that satis6.es two conditions: that it
be free of duplication and that it be undertaken at the time that yields
different crops if theft became a serious problem. 4 Those owners who
wish to employ others t'o tend their orchards may fear employee theft.
the highest consumer valuation net of innovating costs. One market force
Changing a contract's form is expected to change the incentive to steal.
mat tends to bring about the realizati'on of these conditions is competi~
If, fot example, an 'owner forms a fixed-rent contract with a worker to
ti'on am'ong innovators in recruiting customers f'or their particular inno~
tend the orcharcl>.tlle:PfJ~mr~Il~'Wrk!;;,;~thdtsimplydjsappear:s;'::::,::.::::::::,,::yti9.!lspefore expending resources on the actual inn'ovatio~..5 lri order,
"'The'notiornhat theft is a manifestation of the imperfect delineation of
to attract potential customers away from competing inrtovators, not nly
rights can be illustrated by reconsidering movie theaters. Of the attributes

will innovators tend to cede to customers ali the gains fromthe irinova~
theater owners relinquish to the public domain, two relate to the differ.
tion, but the winning nnovator will also be the one who performs the.
ence in value among seats. The first relinquished attribute is the differ
innovation activity dosest to the time that maximizes its net present
ence in value between the better and the inferior seats within a price class
value; only then can the bid for customers dominate that of competitors.
of seats. People can capture the difference in value by arriving early and
When the cost of recruiting cust'omers, whether directly or indirctly, is
occupying the better seats. To the extent that policing is imperfect, a sec
low, customers have, in practice, the rights to the gain from the irnova
ond attribute that is partially relinquished is the difference in value across
tion. As in the case of predatory practices, the easier the advance con- .
. '
price classes. Where seats are sold in several price classes, as is common
tracting, the better the delineation of rights.
Innovators who possess unique talents do not need to cede to custom
o~tside the United States, buyers of lower-priced tickets can cpture the
diffrnce to the extent that they are not prevented from occupying
ers al1 the potential gains the innovations.generate. Such inno.vators may
... higher-pricedseats; The'state'tklsparththeefteiiiiitffort onlyiri
encounter another problem, however. Every customer may try to obtain
the latter case, sinc6 jumping seats constitutes a legal infraction, whereas
a bargain by offering the lowest royalty payment. Since the innovators'
marginal cost of serving extra consumers is zero, the}'" may be willing to
selecting a certain seat within a price dass obviously does noto The eco
nomic logic of the two types of capture, however, is the same.
relinquish ground rather than lose customers. IndividuaIs' incentive fur
bargaining under these conditions implies that the difference between the
maximum consumers' vaIuation and the marginal cost lies, in thepublic
domain.
The potentiaI loss from such bargaining can, however, be low~
- PROPERTY RIGHTS" TO 'INNOVATIONS
ered if the parties can be compelled not to bargain. Therefore, if the op
In contrast to other areas in economics, the importance of property rights
tion of selling the use of the innovation to different consumers at different
considerations to innovations has been widely recognized by economists;
prices can be eliminated, either privateIy or through legal prohibition of
nevertheless, some major rights issues remain unresolved. One such issue
price discrimination, then the innovator's rights over the innovation will
concerns the economic rights innovators can expect to have over their
be partially restored. Maximizing sellers will equate their constrained
innovations. When a uniquely talented individual develops an innova
""marginaLre.venue.totheir(zero).marginal cost and will price accordingIy.
tion, it seems plausible that such an individual will b' ~ble to obtain the
The superimposed uniformity of prices -r()yalty rates in this context
rights to it, since he or she will control the 6.eld. On the other hand, if
delineates rights, and their determination is free of direct capture costs.
many individuais are able to develop the same innovation at a similar
This arrangement is not entirely free of cost, however, because when a
cost, it may appear that none has a right to it and that, in their competi
single price is prescribed, the welfare triangle that couid have been
tion for the gain from the innovation, its economic value will be dissi
avoided under price discrimination is relinquished to i:I1e,pubIic domain.
pated; nevertheless, in this case rights ma}' also be weU defined,
A useful measure of the net present value of an innovation is the differ
5yu (19 81 ) discusses at length methods'used"fr"'suttf'fecfuti~g;"Detse;"(968r"
4Reportedly some avocado orchards in.the San Diego area have been abandoned
beca use of increased theft.
I4 2
was the first to consider a:dvance contracting for selling commodities whose produc
tion is subject to deelining costs.
. ,
'.

PROPERTY RIGHTS TO PRIC INFORMATION
Coase (I937) notwithstanding, economsts seldom realize that determin

ing which prices will clear the market is a resource-consuming activity.
Were such prices easy to determne, no serious errors in setting prices
would occur, yet significant pricing errors are widespread. Some con
certs, for example, are poody attended, whereas tickets for others can be
. obtained on1y from scalpers: Likewise, some artists seU out at their gal
lery openings, but others see an entire show go by without selling a thing.
..:La:s.t;..some.ne:w::srock::offerings..are..instantly snatched up, while others
turn out to be duds. Such examples suggest that setting prices correctly
is costly. Providing price estimates is.costly as weU and must generate a
reward; since the estimates are subject to error, speculators may capture
the value oi the difference. Resources will be spent both on the capture
of price information and on the prevention of such capture.
It mght be expected that prices would be set by individuaIs who spe
cialize in doing so, yet on examination the direct sale of price estimates
proves problematic. If the producers were to purchase price estimates for
their commodities, they would also require a guarantee of the quality of
the servie they were buying. Some of these commodity producers, how
ever, could take advantage of the fact of the guarantee by lowering prd
'dq'litjr"ncfShlfting part of the guarantee burden to the pricing spe
cialist. This difficuItY may be avoided if the producer of the commodity
also sets lhe price, which.might explain why the two activities are often
performed within the same organization.
Some commodities and services are not as susceptible to the problem
of guarantee abuse as- others, and for these the producer of the price
estimate may guarantee his or her service, thereby becoming the residual
claimant to variability in the value of the service. An important exampIe
of the provision of price services by independent specialists may be seen
. in syndicates of investment bankers, which advise business flrms about
the price at which to seU new stock. At the heart of the transaction be
tween a syndicate and an issuer of stock is the implicit price guarantee.
Essentially, a syndicatebuystheentirestockflotationattheagreed-upon
price and offers it to the public at a price not exceeding a predetermined
ceiling. A syndicate that errs by overestimating the market-c1earing price
will be the oneto bear the effect of its error. The syndicate, then, main
tains the property right to its price estimate.
Pricing errors may have an addition:;\leffect on behavior. The ultima te
buyers of a new stock may either rely on the recommendation of the
syndicate or devise their own (or use their advisers') estimates of the"mar~''''
ket price of a stock and act accordingly. The lattr action constitutes a
duplication of effort, since the information is a public good that has al
I44

ready been produced by the syndicate. The lower the costs associated
with acquiring a stock, the greater the demand for it; the demand is ais o
higher if buyers believe that, on average, their suppliers offer them bar
gains. They cannot, however, expect every new stock to be a bargain.
Since some stocks are expected to be duds and others are expected to be
bargains, buying on1y the Iatter and avoiding the former would be highly
prdfitable. Information on new offerings. that aUows a speculator to
avoid some of the du.ds while concenttating on the bargains should en
able him.ot heI:' ~o' arn a posit~ve returno Success by such a speculator,
. however, wilnower the retum to uninformed buyers; they will be forced
to buyrel~tively more'duds, snce a relatively large fraction of the bar
gains will aItea4y have heen purchased by the speculator. The demand
for stocks by buyers who do not acquire information on individual issues
bet who are aware of the average returnon new stocks will be Iower
. when they must compete with speculators, resulting in lower stock prices.
In orderto protect their rghts from being captured by specuIators and
to prevent an advene shift in the demaI'ld facing them, syndicates must
d~ter speculators from acquiring information. Tliis may explain the re
strictons on the number of shares of a l1ew issue individuaIs or organiza
tions .are allowed tc? purchase. If would-be speculators can use their nfor
mation to buy only smaU blocks of shares, they will seldom flnd the .
information worth collecting. Thus, the restction seems to protect the
rights of syndic.ate~ to th;eir. costly 'pri<:~ !nforma~io~.
. ,
PROPERTY RIGHT'S TO WILDLIJ?E.
WildIife, by its very nature, s fleeting. How can one establish rights to
it?6 It would be simple to let one person own the entire habitat of a
wildlife stock such as a herd of deer. Some large hunting ranches in the
United States do contan the entire habtats of several wild animaIs, and
some African game preserves are home to whole herds of many different
species. Seemingly, the reason these ranches re so large is to encompass
endre habitats. Likewise, many primtive sodeties seem to have owned
the entire habitat of wildlife stock tliey hunted or flshed.
A wild'ife stock that oc;cupies and traverses a land area is only one of
several attributes of the land; the others include the capacity for growing
wheat; yielding mineraIs and groundwater; and offering sceruc' vantage
.. points, as well as providing habitat for wildIfe. Each implies a distinct
parcel sjze and shape for optimal use. Wildlife stocks themselves are di
"erse, varying greatly both in vaIue (pests being negativelr..Y.<:i1lle<:iLat:ld ..
Ticlij;-sl'ze..onhe..ti-ltoryi:heyoccupy;Weretransactlo'costs zero - nd
'~The'dis(;,ussion
in this section borrows heavily from Lueck I989'
145

taking account oi the iriteractions among the attributes - each would be
optimally exploited and, consistent with the Coase Theorem, the owner
ship pattern would be irrelevant. Since transaction costs are positive, the
ownership pattern matters. The ownership by single persons of entire
habitats of migrating birds that travei across vast areas and of wildlife
stocks that cover a wide swath predudes the ownership pattern that ac
commodates many small farms and other efficient-size land holdings. We
would expect the ownership pattern that emerges to conform primarily
to the most valued use oi the land. This implies tht in nany ihstanes
the non-wildlife use of the land would' 'take precedc:!nce. 'Governnent~
however, can take action so as to enhance the value of such wildlife
and i1. would appear that it does.
The current legal doctrine governing wildlife in the United States as
signs the regulation of wildliie and its partial ownership to the states and
sometimes to the federal government. Private individuaIs are also legal
owners of some wildlife attributes, as well as being 'the eonomic owners
of many of them. The allocation of rights to wildlife in Calla<a is similar
to that in the United States. In England, however, :these rights are mostly .
assigned to private individuaIs. The difference is consistent with the hy
pothesis that these governments attempt to maximize the value oi wild
life.
'.
Although the population density in England is considerably higher
than ir is in the United States and Canada, the average farm size is rnuch
larger in England than it is- in the United States and Can'da. 7 The habitat
required to support the most valuable wildlife stocks in these three cClun
tries displays the opposite pattern. Large mammals are rare in England,
and game birds and fish tend to be local there; whereas in the United
States and even more so in Canada the habitat of large mamrnals is vast,
and many game birds and fishes migrate long distances. Whereas English
law is the basis of Iaw in the United States and Canada -law in the latter
two, especially Canadian law, sti1l has much in common with that in the
former - the wildlife components oi the American and Canadian laws,
diverge.drastically.from the English ones. In England, where holdings are
large and wildlife habitats are modest, wildlife is not controlled bythe
state to a great degree, individuais holding the bulk of the rights to it. In
the United States and Canada the law is l11uch more restrictive regarding
the rights to wildlife individuaIs have, and the states, provinces, and fed
eral authorities retain a much greater' ~egulatory power than do their
English counterparts. For example, the restrictions on the length of the
7For example, accordiiigroLueck'(i9'i'9f"i;1"i899>':the:';;;~~~"f~~~~i;'~'i~'th~""""
United Kingdom was 390 acres; it was 134 acres.in th,e United States" (p. 312.), and
"private landholdings are even smaller and mor4i'-scatteced in Canada than in the
'
"
United States" (pp. 316-17).

season, the bag limit, and the trade in wildlife products are much more
severe in the United States and Canada than they are in England. It is
worth noting that within the United States and Canada the rights re
tained by the states and provinces are relatively more local than those
assumed by the federal authorities. Moreover, consistent with the notion
that the purpose of the regulations is to maximize the value of wildlife,
private ownership to wildlife in game iarms and fish ponds is routinely
granted. Whereas in the absence oi government regulations, much of the
roaming wildlife in the United States and Canadawouldbe:commo'if.::' .."""",,,,,,
, property, the law reduces the overexplittirifSuch wildlfe, thereby
increasing its value to individuaIs. The need for such laws and regulations
is much reduced in England, where individuaIs' land holdings conform
quite closely to wildlife habitat.
The property rights modeI: Recapitulation
II
PeopIe have discretio.n cver hcw to. delineate rights, and as part of their
maximizaticn effcrt they do. sc to whatever degree they desre. Rights are
never perfectly delineated, hcwever, beca use the fact that ccmmcdities
ru;:e not unfcrm and are e:x:pensive to. measure makes perfect delineadon
prchibtively cO!rl:Iy. Nevertheless, since 'pecpIe do. what they deem
"best," rights may be sad'tc be always weU delineated.
Thcse prcperti~s 'that'pecpIe chcse.nct.tc delineate are in the publie
dcmain. $uch properties indude much of the wcrld's oceaDS;they.also...
indude the cccI air in air-ccnditicned shcpping maUs, which is not
ch~rged fcr.cn the margin. Properties in. the pubIic dcmain can be aug
..nented or diminished.:'Asthe values cf t:ommcdities and cf commcdity
attributes cl)ange, .anl as the costs cf delineaticn and cf protectcn
change, people's deciscns regarding whaf to. leave in, what to. relinquish,
and what to. redaim frcm the public domain change ccrrespcndingly.
The public dcmain is ubiquitcus; innumerable ccmmodity attributes
are placed in it. Any service nct fuUy charged fcr on the margin is at least
partly relinquished tC.the public domain. Owners cculd charge fcr such
services; but the extra returns cften do. nct justify the extra ccsts. Fcr
instance, ccncert-hall cwners relnquish to. the public dcmain the differ
ential in valuaiicn amcng' equaUy priced seats. When the value of
ccncert-hall seats falls - which might happen when the hall is used by a
lccal chcir rather than a famcus cpera star - seats are expected to be
priced in lesser detai!. The differences in value amcng seats that were
formerly differential1y priced are relinquishedto the public domain. Pa
trons are expected to. daim that differential through the expenditure cf
rescurces. Resource cwners attempt to.. maximize the net values of their
rescurces: They attempt to. crganize their actions sc that, eteris paribus,
'they'riiiriii:i:ii:~'tlielo'sse!drompladnfttrlbuteshl'thep'bcdomaii .......
In order to gain from exchange, pecple must spend resources on trans

ferring rights to. commcdities. Contracts are expected to be structured sc
14 8
as to. minin:ze the ccsts cf exchange cf given transacticns. In fcrmng

their ccntracts, transactcrs have a chcice of units by which to. regulate
their transacticns; fcr instance, labor can be exchanged by the hcuc cr
by the piece. The use cf each cf these units relinquishes different subsets
cf attributes to. the public dcmain: the per-hcur effort when labcr is scld
by the hcuc; the care cr quality cf the cutput when labcr is scld by the
piece.
In additicn to. having a chcice cf units, contractcrs may also impcse
restricticns cn the way they ccnduct their exchanges in crder to.. reduce
the:amoottneY'spend:to:capturefroiii::each:othei:::Refdgerator' produc'"
ers, fcr example, cIten make the sale of their merchandise subject to. a
warranty, while at the same time prchibiting buyers from using their re
frigeratcrs ccmmercially. The warranty service is a free attribute to. buy
ers in that they.are nct as heavily penalized for carelessness cr fcr extra~
heavy use as they wculd be if they had to. pay fcr repairs themselves. The
restricticn serves to. reduce such usage. The mcre valued transactions are,
the mcre attributes are expected to. be priced, cr the more ccmprehensive
the restricticns are expected to. be.
Since exchanged ccmmodities are nct tinifcrm and are nct fully mea
sured, the value cf the exchange is variable. A ccntract allccates the vari
.a,b,ility. i.n.. ~ . tr llI1:sa,qi()I1. ~()n.:g ~~t! traI1S;i.QQrf?. As .transactors alter the
units by which they effect exchange, and as they alter fhe asscciated re
. stricticns, they alsc divide the variability in cutccme a!llcng themselves
differently. That allccaticn cf variabilty which maximizes the ccntrac
tcr's wealth is the cne wherein the ability cf a ccntractcr to. affect the
value cf the mean cutccme cf the transacticn is pcsitively related to. the
share cf variab~l~o/. ~e cr she will assume.1t is expected, fcr instance, that
refrigeratcr manufacturers will assume variability in inccme by guaran
teeing mcre ccmprehensively thcse aspects cf their refrigeratcrs' perfcr
mance that are mcst affected by their prcducticn prccedures yet least
likely to. be affected by ccnsumers'acticns. As was ncted earlier, whereas
a refrigeratcr's paint is guaranteed fcr cnly a shcrt pericd, the warranty
cn its mctcr remains in fcrce fcr several years.
Ali sales, with the excepticn of scme gcvemed byc<aveatemptot;t'::
quire crganizaticn. In sales subject to. caveat emptcr, buyers need assur
ances that they will nct walk away with wcrthless merchandise. If buyers
check directly, the degree cf crganizaticn may be trivial. Nct trivial, hcw
ever, is the rescurce ccst cf such exchanges. Ncn-caveat emptcr sales,
that is, sales in wpi,h the transactcrs have impcsed ccnstraints cn each
cther, require real crganizaticn in crder to. pclice and to. mcnitcr the
ccnstraints. The nature cf, and the ccsts asscciated with, such crganiza
ticns vary withthe pricing methcd used. A change in conditicns, such as
a change in the valuaticn cf the transacted ccmmcdity, is expected to.
149

change the method by whieh the commodity is sold, as well as the organi
zational structure governing its exchange.
Who will post the priee at which to commit to trade, the seller or the
buyer? Ths theory yields the mplieation that whiehever party can better
prediet prices will be the one to post them. Whoever posts a priee subjects
hmself or herself to exploitation; his or her exchange partners may en
gage in excessive priee prediction, discover predietion errors, and take
advantage of the willingness to deal at a fi:xed priee. It is further implied
that the person who ean more readily predict the priee will be the one
who will have to assume the consequenees of posting it. Pricepr:ediction:
is subject to economies of scale; the more units'lfedtiliat price, the
lower the unit eost of the predietion. Thus, as the size of the buyer vis-
vis the seller inereases, the more likely are buyers, whose unit cost of
prediction has declined, to post prices at which they agree to purchase
what sllers wish to seU.
Because ali but the lowest-value transaetions are subject to constraint
and r.equire organization, only a small fraction of transactions are "in
the market,". as this term is usually understood. The frequently asked
question as to which transactions will take plaee in the market and which
will remainVl1itllin,~Il~Jirm is not likely to reeeiY~..JJs.eful answer, how
ever. Firro transacnons arenot uniform, and some of them are more in
the firm than others. A more fruitful qiiestion co~ceriis 'th~de:e~~i~~tion:
of the forms of organizations that will govern different kinds of tiansac
tins and E- the forces thatwill bring about ehange to these organiza
. tions.
'. .' The ~omple~iry of eommodities and activities causes ownership pat
terns to be complex as well. The most effident owner of a particular
eommodity attribute is not necessarily the most efficient owner of the
commodity's other attributes. It may be advantageous to split the owner
.ship of a eommodity among several individuais. Beeause the eommodity
is itself not physically split, eaeh owner, if not properly constrained, may
find it easy to consume some of the others' unpriced attributes. In the
prevouslymentioned restric:tlon on the commereial use of refrigerators,
'the owner of the attribute "warranty service" restriets the behavorof
the owners of some of the refrigerator's other attributes.
Organizations, including business firms, typically engage in a multi
tude of activities, including' policing and monitoring constraints. In the
case of the business firm, many factors affect the variabilty in its income.
The priee of"each commodity it buys or sells can fluctuate, each specimen
,<:)Lt,Ile."<:91J.l!ll:<:)g!~i~s..i~,, e.;:o,changes.QJay ,diff.eI,f,t:om .()thers, and its ineome
may depend on whether or not sueh phenomena as fires, earthquakes,
and foreign eonfiseations occur. Eaeh of these instanees of income vari
15 0
ability may be borne by a different party. A firm may purchase a raw

material on the spot market, or it may operate under a long-term, fi:xed
price contract for it. Assuming no breach of eontract, in the former case
the buying fum bears the effect of fluctuations in the' price of the com
modity; in the latter case its contract insures it against fluctuations. These
considerations apply to aU the firm's sources of variability. In each of
these instances it is expected that the party that is better able to affect the
mean outcome will tend to assume the associated variability. For exam
pie, it is expected that a raw-material supplier that has some power to
,:::::set:.ifs:pfice::is::m:fe::likety"w::sign:::fixed+p.rice,,10ng7term eontracts than
one notpossessing sueh power.
Similar considerations apply to labor servces. Workers, who can affect
outcome-value more easily than the purchaser of the labor seI'vices ean,
are likely to operate as independent contractors selling output rather than
labor. At the other extreme, purchasers of labor services will assume vari
ability in outcome by paying a fixed wage on a long-term basis. It is to
be expected that as the market wage of workers rises, they will gravitate
toward self-employment. 1
When two parties agree to a formula for dividing future income vari
ability, one may emerge as the winner and the other as the loser as the
outcome,Qf.the,t.r~nsactiQn,unfolds. Because the loser could gain by re
neging on the contract, each party demands assuranees from the other
that the contract will not be bre.aehed. A necessary though insufficient
condition for such assurances is that a party be able to meet his or her
obligations. Fixed-wage suppliers of labor can readily guarantee perfor
mance even when the market wage exceeds the eontract wage, because
they own their own labor services:The employer of sueh workers must
be able to ensure wage payment when the market wage falls below the
contract wage. Equity capital specializes in providing such assurances.
More generalIy, it seems that equity capital is assembled (and augmented)
in order to guarantee all the contracts signed by the firmo These contracts
may, in tum, be viewed as eonstituting the firm; in this sense shareholders
are the owners of the firm, and the firm is a "nexus of contracts."
CONCLUSION
I have attempted to demo~strate how the property rights transaction eost

model can generate a better understanding of the allocation of resourees
and of the interaction of this allocation with econom:ic'organization. The
literature that assumes that the costs of transacting are zero and that"JI
1
Barzel 1987.
151
'< '. :;,';:<:;'~::~~:~:','>:: .: .. \~>~\::;,~~d::~.'! ~.,_:;~;~'<"'.:R\f!;~::i,5!,~:jt~~;'H~;Uf .'~!':';~'>!;-':,.' '.;,~. <. ",.~: ,';' .'!';',1:";':":"~':.' '!.:
-><':"." :., <~,,\.,
; .. '
,r;,;, ',:..;, : ~,'!:,k<i!.: i<i
::~ ~::~:;-'~
c:;, ",
","-,,-,,~~,~~~~:::.::..~.:>:.
"\;~-~''.,<:':";:j~"~,,::":;';';';";;;"-;~::~!Mg\"x:,;,;:-::{_~~~.~\",I::n[l~~";; -;;-::':'1 :.:' :~:J'~'
;",;.
<'i "li; :,'~' '-:

property rights. are perfectly well delineated IS incapable of dealing with
a vast array of actual observed practices. Particularly glaring is the inabil
ity of such an approach to explain why exchange parties would ever im
pose restrictions on each other. The property rights approach is capable
of addressing such issues.
Many approaches to the analysis of economic behavior do not explic
itly assume that transaction costs are zero, but they also do not inquire
as to what the precise rights of the parties are. It is my mpression that.
economists who neglect property rights considerations are prone to rlak
ing. ~ll:lp~i(;i!:ll~~lIll:lpti.(;)l'ls. ~llt.llr.~:(),f.t;~ :l1:()~,:~~II::~~l'l?"~h.:i~:Pl:~ll~l1:g,,. .
'results that are hard to accept. It is quite common to find cases where
within single models some transaction costs are implicitly assumed to be
zero (e.g., it is routinely assumed thatmonopolists know precisely what
their demand is) while others are assumed to be prohibitively expensive
(e.g., it is routinely assumed that price discrimination by monopolists is
prohibitively costly);2 The, approach that insists on asking who owns ev
ery particular attribute of a commodity and what "owners" can actually
do with "their" commodities seems to come doser to the root of transac
tion costs and is, therefore, less likely to fall prey to untenable assump
tions.
,
Final1y, consider the applieation of the property rights approaeh to the
.. ,,ditribtinof 'gilis' fcom ' t"d'. 'Miiliy' gds 'are' valied 'less by their
current owners than they are by other individuaIs. Who owns iliese po
tential gains from trade? the eompetitive, zero transaction cost model,
the distribution of the gains is costlessly determined. The costless infor
m~.~ion (or the uniformity of commodities) necessary for sue h competi
tion is, however, seI dom encountered in reality. Opportunities for people
- to gain at the expense 'E"fhers seem ubiquitous. Whereas individuais
are always ready to expend resourees to increase their share of the pie,
they will also seek out methods and organizations by whieh to better
delineate rights to it, thereby dividing the pie without shrinking it toa
much.
A considerable portion of this book has been devoted to analyzing
sueh behavior. It is fitting to eoncludeby visiting one unlikelyplace
where the time and effort of haggling over the distribution of the gains
from trade are eHectively avoided: an Egyptian bazaar. To quote Werner
(I987), in Caro's "principallivestock market, camels take center stage .
. . . [T]he camel market's own King Solomon ris] Muhammad Abd al
Aziz.... Sales are conducted one-on-one - one buyer, one seller and one

camel at a time... With an acutely discrminating sense of camel flesh
[Muhammad] sets a fair price... His authority is usually sufficient to
settle any difference." Transaction costs are near zero in this particular
market. Nothing, however, comes free. Here the cost in quesrion is "a
small margin for [Muhammad's] commission" (p. 1:32')'
In
'.
2 Another example occurs in asymmetry-of-information modeIs, where one set of

individuais is, as a rule, implcitly assumed to be costlessly informed, while for others
the information cost is assumed to be prohibitive.
1:5 2
I53
.J.'-f:,,1 "', c;,rH...."'''''
Barzel, Yoram. 1994. "The Capture ofWealm by Monopolists and the Protection
of PropertyRights." IntemationalReview of Law and Economia 14: 29 2
References
49
Barzel, Yoram, and Tim Sass. 1990. "The Allocation of Resources' by Voting."
Quarterly Journal of Econo1?'tics lOS: 745-7 1
Bean, Richard N., and Robert P. Thomas. 1974. "The profits of the Slave Trade:
The Fishers of Men." journal of Economic History 34, no. 4: 885-9 1 4.
Cheung, Steven N. S. 1969. A Theory of Share Tenancy. Chicago: University of
Chicago Press.
Cheung, Steven N. S. 1974. "A Theory of Price ControI." journal of Law and
Economics 17, no. I: 53-72.
Cheung, Steven N. S. 1977. "Why Are Better Seats 'Underpriced'?" Economic
. . ' . . ..
Inquiry 15, no. 3: 5 I 3-2.2. .
Cheung, Steven N. S. 1983. "T\1e. Contractual Nit1.iJ;e of me Firm:~ journal of
Law and Economics 2., no. I: 1-22..
'.,
.
Coase, Ronald H. 1937. "The Nature of th Firm." Economica 4, no. 3: 3 86
Akerlof, George. 1970, "The Market for 'Lemons': Quality Uncertainty and the
Market Mechanism." Quarterly journal of Economics 84: 488-50.
4 0 S
Alchian, Armen A. 1965. "Some Economics of Property Rights." Il Politico 30,
Coase, Ronald H. 1960. "The Problem of Social Cost." journal of Law and
no. 4: 816-29. Reprinted in Armen A. Alchian, Economic Forces at Work.

Economics 3, no. I: 1-44
.
Indianapolis, Ind.: Liberty Press, 1977.

Dahlman, CarI J. 1980. The Open Field System and Beyond: A Property Rights
Alchian, Arnien A. 1987. "Property Rights." In The New Palgrave: A Dictonary

Analysis of an Economic Insti~ton. Cambridge: Cambridge University
of Economics. New York: Norton.

.
Press.
.
Alchian, Armen A., and William R. AlIen. 1977. Exchange and Production, 2nd
Dam, Kenneth W. 1965. "Oil and.Gas Licensing and me North Sea." Journal f
ed. Belmont, Calf.: Wadsworth.

Law and Economics 8, no. 2.: 51:"76.
'.
.
Alchian, Armen A., and Harold Demsetz. 1972. "Production, Information Costs,
Deacon, Robert T., andJon Sonstelie. 1985. "Rationing by Waiting and the Value
and Economic Organization." American Econmic Review 62, no. 5: 777-95.

of Time: Results from a Natural Experiment." Joumal of Political Economy
PQuglas.W. :l9.9:la... :'Homesteading.. and..P.roperty ..Rights; or 'How the

93, no. 4: 62.7-47
.
West Was RealIy Won.' " journal of Law and Economics 2,4, no. I: 1-23.
Demsetz, Harold. 1967. "Toward a Theory of Property Rights." Amercan Eco
AlIen, Douglas W. 1991b. "Wkat Are Transaction Costs?" Research in Law and
nomic Review 57, no. 2.: 341-S9
.
.
Economics 14: 1-18.
Demsetz, Harold. 1968. "Why'Regulate Utilities?:" jornal df taw and'EcotrO-m
AlIen, Douglas W., and Dean Lueck. 1992. "Contract Choice in Modem Agricul
ics II, no. I: 55-66.
ture: Cropshare Versus Cash Rem." journal of Law and Economics 35, no. 2:
Eggertsson, Thrinn. 1990. Economic Behavior and Institutins, Ca'mbridge:
397-42.6.
.
Cambridge University Press.
Allen, Douglas W..,.and Dean Lueck, 1993. "Transaction Costs and me Design of
ElIickson, Robert C. 1991. arder Without Law. Cambridge, Mass.: Harvard
Cropshare Contracts." Rand journal of Economics 24, no. I: 78-100.
University Press.
Allen, George C. 1966 [1929]. The Industrial Development of Birmingham and
Eswaran, Mukesh, and Ashok KotwaI. 1985. "A Theory of Contractual Struc
the Black Country, 1860-1927. New York: Kelley.
ture in Agriculture." American Economc Review 75, no. 3:352.-6 7. . ....
Anderson, Terry 1., and P. J. Hill, 1975. "The Evolution of Property Rights: A
Fogel, Robert William, and Stanley L. Engerman. 1972. Time on the Cross: The
Study of the American West." journal of Law and Economics 18, no. I: 163
Economics of American Negro Slavery. Boston: Little, Brown.
Gisser, Micha. 1995. "The Evolvement ofWater Rights: Testing the Barzel Rule."
79
Arrow, Kenneth J. 1973. "Higher Education as Filter." ]oumaIQff'blic..E.I;Q7
.Mimeographed.
nomics 2, no. 3: 193-216.
Goldberg, Victor P. 1985. "Production Functions, Transaction Costs, and the
Barzel, Yoram. 1974- "A Theory of Rationing by Waiting," journal of Law and
New Institutionality." In Issues in Contemporary Econorttics, ed. E. Feiwel,
395-402.. New York: Macmillan. .
Barzel, Yoram. 1977. "An Economic Analysis of Slavery." journal of Law and
Gordon, H. Scott. 1954, "The Eonomics of a Common Property Resource: The.
Fishery." j oumal of political Economy 62, no. 2: 124-4 2.
Barze1, Yoram. 1982. "Measurement Cost and the Organization of Markets."
Grossman, Sanford, and Oliver Hart. 1986. "The Cost and Benefits of Owner
journal of Law a'ftitEconomics 25, no. I: 27-48.
'ship: A Theory of Vertical and Lateral Integration." journal of Political Econ
Barze1, Yoram;198 5. "Transaction Costs: Are They Just Costs?" journal ofImti
omy 94;' n0;"4: 69 1 '"7I'9"......... 1.....:..... :.... :.............."....... .......................................................... ",," ..
tutional and Theoretical Economics 141, no. 1:'4-16.
Hall, Christopher D. 1986. "MrketEnfoi:cd Information Asymmetry: A Study
Barzel, Yoram. 1987. "The Entrepreneur's Reward for Self-Policing." Economic
of Claiming Races." Economic !1JQuiry'2.,4, no. 2:27 1-91.
Inquiry 25, no. I: 13-16.
154
155
"'''''1 \..-,. "", . . Lo< ..... ...,
Keterences
Hart, Oliver, and John M60re. 1990. "Property 'Rights and the Nature of the
Firm." Journal ofPolitical Economy 48, no. 6: III9-58.
Holmstrom, Bengt R., and Paul Milgrom. 199I. "Multitask Principal-Agent
Analysis: Incentive Contracts, Asset Ownership and Job Design." }ournal of
Iaw, Economics and Organization 7: 24-52.
Holmstrom, Bengt R., and Paul Milgrom. 199+ "The Finn as ao Incentive Sys
tem." American Economic Review 84, no. 4: 972-9I.
Jensen, Michael C., and William H. Meckling. 1976. "Theory of the Firm: Mana
gerial Behavior, Agency Costs and Ownership Structure." Joumal ofFinanciai
Economics 3, no. 3: 305-60.
'
Johnson, Ronald N., Micha Gisser, and Michal Werner. 1981. "The Definition
of Surface Water Right and Transferability." Joumal of Iaw and Economics
24, no. 2: 273-88.
.
::::::,::::::::KIl..!;!:iJ~~p4,P,,;I;98I. The Economics and Po/itics of Oil Pria Regulfltion: Fedral,
Policy in the Post~Embargo Era. Cambridge, Mass.: MIT Press; .
Kessel, Reuben A. 1974. "Transfused Bl09d, Serum Hepatitis, and the Coa'se
Theorem." Joumal of Iaw and Economics 17; no. 2: 265-9' .
Klein, Benjamin, Robert G. Crawford, and Armen A. Alchian. 1978. "Appropri
able Rents, Vertical Integration, and the Competitive Contracting Process."
Journal of Iaw and Economics n, no. 2: 297-326.
.
Knight, Frank H., 192I. Risk, Uncertainty and Profit. Boston: Houghton Mifflin.
Knight, Frank H. 1924. "Some Fallacies in the Interpretation of SociaJ Cost."
Quarterly Journal of Economics 38, no. 7: 582-606.
.
Krueger, Anne O. 1974. "The Poltica I Economy of the Rent-Seeklng Society....
American EconomicReview 66, no. 3: 291-303.
,
Landes, William M. 197I. "An Economic Analy~is of the Courts." JOlfmal of
Iaw and Economics 14, no. I: 61-108.
Libecap, Gary D. 1989. Contracting for Property Rights. Cambridge: Cambridge
University Press.
Lueck, Dean. 1989. "The Economic Nature of Wildlife Law." fournal of Legal
Studies 18, no. 2: 291-324.
Milgrom, Paul, and John Roberts. 1990. "The Economics of Modem 'Manufac
turing: Technology Strategy and Organization." American Economic Review
80, no. 3: 511-28.
Miller, Kathleen A. 1985. "The Right to Use vs. the Right to SeU: Water Rights
in the Western United States." Ph.D. diss., University ofWashington.
North, Douglass C. 198I. Structure and Change in Economic History. New
York: Norton.
'
Ostrom, Elinor. 1990. Governing th Common: The Evolution of lnstittions
for Collective Action. Cambridge: Cambridge University Press.
PosIlel,r'f.Richl'~~~lA.~ "J:9..7,5,~ ,~Ihe,Soc.i~.1 qs.t of Monopoly and Regulation." Jour
na o Po itica Economy 83, no. 4: 807-27.
Posner, Richard A. 1992. Economic Analysis of Iaw. 4th ed., .Boston: Litde,
Brown.
Priest, George L. 1980. "Selective Characteristics of Litigation." Journfll afLegal
Studies 9, no. I: 399-42I.
Priest, George L., and Benjamin Klein. 1984. "The Selection of Disputes for Liti
'
gation." JouriilfLegal Studies q, no. I: I-56.
Rockoff, Hugh. 1984. Drastic Measures: A 8#t.QrY...gLW.QgfUl/nd.P..r..k,?'CQntmls, ..
in the United States. Cambridge: Cambridge University Press.
'
15 6
Rose, carol M. 199 0 "Energy and Efficiency in the Realignment of Common

Law Water 'Rights." Joumal of Legal Studies 19, no. 2: 261-9 6.
Ross, Stephen A. 1973. "The Economic Theory of Agency: The Princpal's Prob
lem." American Economic Review 63, no. 2: 134-39
Silberberg, Eugene. 1990. The Structure of Economics. 2nd ed., New York:
McGraw HiU.
Spence, A. Michael. 1973. "Job Market Signaling." Quarterly Journal of Eco
nomics 87, no. 4: 355-79
Stigler, George J. 1966. The Theory of Price. 3rd ed., New York: Macmillan.
Tullock, Gordon. 1967. "The Welfare Cost of Tariffs, Monopolies and Theft."
Western Economic Journal 5, no. 3: 224-3 2.
Umbeck, John. 'I977. "The California Gold Rush: A Study of Emerging Property
Rights." Explorations in Economic History 14, no. 2: 197-206 .
. Umbeck, John. 19 8 r. "Might Makes Right: A Theory of the Formation and Ini
tial Distribution of Property Rights." Economic lnquiry 19, no. I: 3 8-59. ,...
.Varian, Hal R. 1991. MicroeconDmic Analysis. 3rd ed., New Yik:'NortoO::"
Werner, Louis. 19 87. "North to Cairo along the Schorching Way of the Forty."
Smithsonian (March): 120-33Williamson, Oliver E. 1975.Markets and Hierarchies: Analysis and Antitrust
lmplications: New York: Free Press.
WIliamson, Oliver E. 1985. The Economic lnstitutions of Capitalism. New
York: Free Press.
Yu, Ben T. 19 1h "Potencial Competition and Contracting in Innovation." Jour
nal of Iaw and Economics 24, no. 2: 215-39
'.
157
lndex
'.
.'
ad valorem tax, 34
agency theory, 14
Akerlof,. George, I 5n '
Alchian, Armen A., 3, 4n, 6n, 14,
65n, 75-6, 93n, 12.4n
Allen,.Douglas W., 3n, 4n;33n, 43n,
46n, 12.2.n, 12.3n
Allen, George c., 56n
Allen, William R., 3, 4n, 93n
allocation by government; 132,.,.8'
Anderson, Terry L., 119n
Arrow, Kenneth J., 14n

asset specifidty, 14 .
...... ..
"'.....
....
Barzel, Yoram, 4n, 6n, 17n, 43n, 76,

.
103n, 106n~ i3011, 139n, 151n
Bean, Richard N., 105n
blood, donated, 130-2.
......................
Dahlman, Carl J., 99, 100n
Dam, Kenneth W., 101
Deacon, Robert T., 2.5n
Demsetz, Harold, 15, 65n, 75, 91-2.,
. 143n
dispute resolution, 96-9
dissipation, minimization of, 2.9-32.
divided ownership, 6-7, 55-64, 1 I 5

17
donated bloQd,..I3,q-2.
. ~conomiCdghts d'flned,3 ~ 90 ""',.,.",.
Eggertsson, Thrinn, 13 n
Ellckson, Robert c., 3, 90n
emergence of rights, 86-9
emergence of state, 86-9
Engerman, Stanley L., I LO..
equity capital, 80-ij.
Eswaran, Mukesh, 46n, 51n
California gold rush, 85-6
.. ,<:aveat emptor transactions, 72.-3, 80

firm, theories of, 75-7
fixed-rent contract, 40-9
charity, 130-2Fogel, Robert W., lIO
Cheung; Steven N. S., 3, 14, 19n, 2.1,

30, 33-6,44, 65 n, 73-6,.81,. 94n, Fogelquist, Del, 2.on
12.2.n
Coase, Ronald H., 7,14,65,72.-7,

George, Henry, 39n
84, 144
Gisser, Micha, 12.1n
Coase Theorem, 7, 9, 36, 77, 98, 102., gold rush, 85-6
108, 146
Goldberg, Victor P.,67n
command economy, 136n

Gordon, H. Scott, 14, 89
common law, 98-9

Grossman, Sanford, 75-6
common property", 56,62.-4,89-90,

guarantee capital, 80-4
...... 99-IolO ".se8,also,public..domam".o.'e;.,:.....

,
Crawford, Robert G., 14n, 65n, 75-6, Hal!, Christopher D., 6m
I2.4 n
Hart, Oliver, 75-6
159
lndex
lndex
Hill, P.J., lI9n
. Hobbes, Thomas, 86-8
HoImstrom, Bengt R., 40n, 75-6
homesteading, x2x-3
indentured servitude, x05-6, u3

innovations, 142-3
insurance, 60-2
Jensen, Michael C., 4n, qo, 65n
Jewish Iaw, lI2n .
Johnson,R,onaldN., 12In ..
predatory pricing, X39-4X

price control, x8-32, x36
price information, x44-5
Priest, George L., 99n
. property rights
to prcs, X44-5
to wildlife, x45-7
see tllso economic rights defined; le
gal rights delined
public domain, 5, x6-3 2, 97
Walrasian model, II-x3, x9, 2X-2,

66-72, x14, 128
Warner, Michael, X2xn
w.ater ownership, restrictions on,
IIS-2X
Werner, Louis, x51
wildlife, 145-7
Williamson, Olver E., 140, 65 n , 67 n ,
75-6
Yu,Ben T., 143 n
quasi rents, 14
Kalt, J9seph P., 20n

rationing bywaiting, X6-18
Kessel; Reuben A., X30-2 .
'rent seeking, q
Klein, Benjamin, x4n;~5n, 75-6, 99n; rents, quasi, 14
X240
Roberts,John, 75-6
Knight, Frank H., q, 65, 68, 89, xoo
Rockoff, Hugh, 20n
Kotwal, Ashok, 40n, 5 In
Rose, CaroI M., lISn
.Krueger, Anne O., qn
Ross, Stephen A., 140
Sass, Tnn, X30n
labor.input, 66-7~
slaves",I06-8
share contract, 33"':7, 49-51,58-9
signaling, market, X4
Landes, William M., 99n
legal rights defined, 4, 90
Silbetbrg, Egeri'; '44 n
slavery, x, 15-13
liability,77
slaves' consumption, IIo-II
Libecap, Gary D., 92n
Lueck, Dean, 33.n, 46n, lOIn; II8n,
'. slaves' supervisipb cost, 108-10
q5n,~46n
.'
. :" Sonstefie,jon:2:5 lt
spe.cializing.. gains from, 5 1-2.
. market signaling, 14
Spence, A. Michael, I4n
market transactions defined, 73
Suen, Wing, I29n
Meckling, William H., 4n, qn, 65n
supervision costs of slaves, I08-xo
Mlgrom, Paul, 40n, 75-6
supply, fiem and industry, 69-70
Miller, Kathleen A., lI8n
monopoly, 139-41
tax; ad valorem, 34
M~ore,John,75-6 .
tax distortion, 34, 49
thett, 90, 111-13,117-18, 141-2.
.National Football League, 5
Thonias, Robert P., IOSO
non-market allocation, 128-38
transaction costs, 5
Nonh, Douglass c., 89n
defined,4
Nonh Sea, rights to, 101-2
Tullock, Gordon, qn
. Os:rom, Elinor, 99n
ownership
allocation of, 7-9, 48
divided, 6-7, 55-64, 115-17
"Pterson, Bruce, 2.on, 2.6n
police protection, I 17-18
. Posner; Rchard A., 6n, I4rt
,.
Umbeck, John, 14, 85, 90, 94-5

Varian, Hal R., 66n
venical integration, 12.3-6
voting;I'Oo; 129;;':36.' .
wage contract, 48, 63-4
variability in, 78-80
I6I
160

Barzel - Economic Analysis of Property Rights

Uploaded by

Copyright:

Available Formats

Barzel - Economic Analysis of Property Rights

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Barzel - Economic Analysis of Property Rights

Uploaded by

Copyright:

Available Formats

POLITICAL ECONOMY OF INSTITUTIONS AND DECISIONS

James E. Alt. Harvard University

Douglass C. Nonh, Washington University of St. Louis

Other books in the series

Mithael Laver and Kenneth Shepsle, Making and-Breaking Govemments

Brian Levy and Pablo T. Spiller, Regulations, lnstitutions, and Commil:n1tm-1.

Leif Lewin, Ideology and Strategy: A Centuiy of Swedish Politics (English

Gary Libecap, Contracting for Property Rights

Elinor Ostrom, Goveming the Commons: The Evolution of Institutions for

J. Mark Ramseyer, Odd Markets in Japanese History

J. Mark Ramseyer and Frances Rosenbluth, The Politics of Oligarchy:

PUBLISHED BY THE PRESS SYNDICATE OF THE UNlVERSITY OF CAMBRIDGE

The Pitt Bulding, Trwnpington Street, Caril bridge. United Kingdom

CAMBRIDGE UNlVERSITY PRESS

The Edinburgh Building, Cambridge CB2 2RU, UK http://www.cup:cam.ac.uk

40 West 20th Street, New York, NY 10011-4211, USA http://www.cup.org

10 Stamford Road; Oakleigh, Melboume 3166, Australia

Yoram Barzel 1997

This book is in copyright. Subject to statutory exception and

ISBN 0-521-59275-5 hardback

Series editors' preface

Series editors" preface

The CambridgeSeries on the Political Economy of Institutions and Deci

...<.... ':~-"':: ,.. ::>"

.: &(;~';I,,~;':: \.~ :<".;:' '::-:

':~.\~. ><;~i.i'~! ~~I:,\;.~~;g~:i~;{;:.:~t~!:;':'/;::~;:;/I.:; 'I( :~~::

Preface to the second edition

Preface to the first edition

The intellec.tual content of "property rights," a term that has enchanted

THE DEFINITION OF ECONOMIC AND LEGAL RIGHTS

. '.-:. "::;,~.~.~,:r.:8,<.:i.,.,, "}.',,Z ,",';',",_,,,-,, v'i, ,;'$,_,L-,

Economic analysis of property rights

The property rights model

.. "5si~i'i~~~;;~~id~'~~ti~;;~i;:;otdbbrat(:don here) apply to the protection ~'f assets.

The property rights model

Economic analysis of property rights

gains from exchange as conventionally perceived net of the cost of ef

The sale of cherries illustrates the phenomenon of wealth capture. Ob

vious problems of information present themselves when cherries are ex

changed. Customers must spend resources in order to determine whether

Physical opeJ;"ations within, and on the ftinge of, an organization such as

a firm usually involve many commodities and, correspondingly, many

attributes. Severa! individuais share in ownership of the attributes, each

owning alone or with others some subset of these. Stockholders own

.residualdimant from'd1eservicingop~ration,gaining if it provides good

.Tl!E ALLOCAl;'ION OF OWNERSHIP

DIVIDED OWNERSHIP OF COMMODITIES

commodities transIer only subsets of the commodties' attributes while

common; for instance, this is so in all rental agreements, as it is in any

sale subject to guarantee.Exchanges that take this form result in divided

distinct attributes of the same commodity. As will be elaborated in Chap

ters 4 and 8, restrictions on the behavior obhe()wne,r~"ml;l,y be imposed

in order to enhI:;~~th~~ep~~~ti;;~ ~(th~ir individual economic rights.

Incomplete separation makes attributes common property, relinquishing

Not only is ownership of commodities often divided; ownership of

what appear to be the assets of an 'rgahization may ~{! divided as well. 8

One f the most celebrateq propositions in econmics is the Coase Theo

Economic ana/ysis of property rights

The property rights model

instance, if the service is the amount of electricity solar panels produce

':~.\~. ><;~i.i'~! I:,\;.;g~:i~;{;:.:~t~!:;':'/;::~;:;/I.:; 'I( :~~::

.. "5si~i'i~;;id~'~~ti~;;~i;:;otdbbrat(:don here) apply to the protection ~'f assets.

in order to enhI:;thep~~~ti;;~ ~(th~ir individual economic rights.