Content uploaded by Karen Clay
Author content
All content in this area was uploaded by Karen Clay
Content may be subject to copyright.
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?Annu. Rev. Sociol. 2000. 26:525–46
Copyright c
2000 by Annual Reviews. All rights reserved
THE CHOICE-WITHIN-CONSTRAINTS NEW
INSTITUTIONALISM AND IMPLICATIONS FOR
SOCIOLOGY
Paul Ingram
Columbia Business School, Columbia University, New York, New York 10027-6902;
e-mail: pi17@columbia.edu
Karen Clay
Heinz School of Public Policy, Carnegie Mellon University, Pittsburgh,
Pennsylvania 15213; e-mail: kclay@andrew.cmu.edu
Key Words economy, commitment, exchange, state, organization
■Abstract Thevariantofnewinstitutionalismthatisourfocusisapan-disciplinary
theory that asserts that actors pursue their interests by making choices within institu-
tional constraints. We organize our review of the theory around its behavioral assump-
tions,theoperationofinstitutionalforms,andprocessesofinstitutionalchange.Ateach
stage, we give particular attention to the potential contributions of sociology to the the-
ory.Thebehavioralassumptionsofthetheory amount to bounded rationalityand imply
transaction costs, which, in the absence of institutions, may frustrate collective ends.
The principle weakness of these behavioral assumptions is a failure to treat preferences
as endogenous. We categorize the institutions that arise in response to transaction costs
astowhethertheyarepublicorprivate intheirsourceandcentralizedordecentralizedin
theirmaking.In detailing the resulting categories of institutional forms, we identify key
interdependencies across the public/private and centralized/decentralized dimensions.
The new institutionalism is in particular need of better theory about private decen-
tralized institutions, and theorists could turn to embeddedness theory and cognitive
new-institutional theory as a source of help on this topic. The dominant view of institu-
tionalchangeisthatitisevolutionary,drivenbyorganizationalcompetition,andframed
by individual beliefs and shared understandings. Sociology can refine the change the-
ory by adding better explanations of the behavior of organizations, and of the processes
by which institutional alternatives come to be viewed as acceptable or unacceptable.
INTRODUCTION
Ronald Coase (1998, p. 73) likens mainstream economics to studying “the circu-
lation of the blood without a body.” The new institutionalism that is the focus of
this essay is about the body. Its fundamental assertion is that actors pursue their
0360-0572/00/0815-0525$14.00 525
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?
526 INGRAM ¥CLAY
interests by making choices within constraints. This assertion can be expanded to
identify three characteristic elements of the theory. First, it holds that actors are
boundedly rational in the sense that they pursue a broad set of self interests, but
with limited knowledge and cognitive capacity. Second, institutions are defined as
the rules, combined with their enforcement mechanisms, that constrain the choices
of actors. These rules include the laws of states, the policies of organizations, and
the norms of social groups. Third, institutions ideally constrain actors such that
their best choices are consistent with the collective good, enabling, for example,
mutually profitable exchange between actors. However, there is no assumption
that ideal institutions will exist, and globally inefficient institutions sometimes
persist because they favor particular actors who have the power to defend them.
Thetheoryis pan-disciplinary,andmajorcontributionshavecomefromeconomics
(Coase 1937, Williamson 1975, North 1990, Greif 1994), political science (North
& Weingast 1989, Ostrom 1990), law (Ellickson 1991), sociology (Nee & Ingram
1998), and anthropology (Ensminger 1992).
This “choice-within-constraints” framework is only one of many theories to
be called new institutional (Fligstein 1997). We do not attempt in this essay to
give full treatment to all new institutional theories (for broader new institutional
reviews, see Scott 1995, Stinchcombe 1997, Nee 1998). Instead, we try to give a
more complete treatment of the accomplishments and shortcomings of the vari-
ant of new institutionalism that is our focus. However, other theories come into
play as sources of solutions to the theoretical problems with which the choice-
within-constraints framework is currently struggling. Particularly likely sources
for intellectual exchange with the choice-within-constraints framework are the
more cognitive version of new institutionalism, from sociology’s organizational
theory, and the related theory of embeddedness from economic sociology. Indeed,
it is one of our main assertions that sociologists are well poised to make contribu-
tions to the new institutional theory that we elucidate.
This essay has three main parts. In each, we treat a core component of the the-
ory, describing and critiquing its current state. We begin by detailing its behavioral
assumptions. These assumptions amount to bounded rationality, and they imply
the foundation problem of the new institutionalism, that exchange is associated
withuncertaintyandrisk.Uncertaintyandriskinexchangecreatetransactioncosts,
whichfrustrate efforts to attain collectiveends (from simple gains from trade to the
fruitofcomplexcollectiveaction).Akeyweaknessof these behavioralassumptions
istheir inattention to the origin of preferences. This is the missing linkinalltheories
based on behavioral assumptions of rationality. On this topic the new institution-
alism can use the findings of embeddedness theory (Granovetter 1985, Uzzi 1996)
and cognitive institutionalism (Meyer & Rowan 1977, Powell & DiMaggio 1991).
We then denote the major classes of institutions and explain the mechanisms
of their operation. Institutions can facilitate the attainment of collective ends that
transaction costs might otherwise prevent. They do this by aligning individual in-
centives to be consistent with collective ends. We propose a classification system
for institutions based on two dimensions—public or private and centralized or
decentralized. Public or private refers to who makes the institutions, with public
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?NEW INSTITUTIONALISM 527
indicating the state and private indicating organizations or individuals. The cen-
tralized/decentralized distinction captures the degree to which institutions are cod-
ified and responsibility for enforcement is centralized with identifiable functionar-
ies. These two dimensions define three classes of institutions, public-centralized,
private-centralized,andprivate-decentralized(therearenoinstitutionsthat are pub-
lic and decentralized in the literature on which we focus). We describe in detail
these three classes, and we consider the interdependence between them.
Finally,we consider the question of institutional creationandchange.Currently,
an evolutionary explanation of institutional change is most prominent. Processes
of the selection of social systems (groups, organizations, states) and political com-
petition over institutions push the institutional framework gently in the direction
of transaction cost efficiency. “Gentle,” however, is key to describing these evo-
lutionary pressures, and consequently, new institutionalists are not surprised to
observe inefficient institutions that persist for very long periods of time. Missing
from current ideas on institutional change are realistic views of the role of orga-
nizations in the process and how beliefs and shared understandings affect change
of other types of institutions.
BEHAVIORAL ASSUMPTIONS
To denote the behavioral assumptions of the new institutionalism, it is useful to
break down a simple summary of the fundamental behavioral assertion of the
theory: “actors pursue their interests by making choices within constraints.” The
significant classes of actors are individuals, organizations, and states. Individuals
constitutethe other classes ofactors, and are the focusof our discussion ofinterests
and choice. Organizations are defined broadly as groups of all types, whether they
are social groups, coalitions, or corporations, structured to pursue some collective
purpose. States are a special type of organization “...invested with the authority
to make binding decisions for people and organizations juridically located in a
particular territory and to implement these decisions using, if necessary, force”
(Rueschemeyer & Evans 1985, pp. 46–47).
These three classes of actors each produce their own form of institutional con-
straint: individuals produce private-decentralized institutions, organizations pro-
duce private-centralized institutions, and states produce public-centralized insti-
tutions. In this sense, it can be said that actors lead a double life in the new
institutionalism, pursuing their own interests within constraints, while producing
constraintsfor other actors. The interplay betweentheactorscanbestbe understood
as a three-layered hierarchy, with states superordinate to organizations, which are
superordinate to individuals (Nee & Ingram 1998). States constrain organizations
and individuals that are their subjects, and organizations constrain the individuals
that are their participants. There is also upward influence in the hierarchy, as actors
try to affect the institutions that constrain them.
This hierarchy of actors is useful to illustrate some key differences between
the choice-within-constraints framework and the cognitive institutionalism and
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?
528 INGRAM ¥CLAY
embeddedness arguments that are prominent in sociology. Those latter two lines
of new institutional argument recognize the same classes of actors that we do,
although in both cases they see all of these actors as embedded in something out-
side (culture and networks of social relations, variously). This embedding adds
other influences, beyond the immediate action frame, on the actors and their re-
lationships. The choice-within-constraints new institutionalism that we focus on
strips away those outside influences, and relies on the interplay among individuals,
organizations, and states to explain institutions and behavior. One of our goals is
to illustrate for sociologists the advantages of this narrowing, although we also
recognize many limitations.
Moving from actors to interests and choice, the new institutionalism views indi-
viduals as rational in the basic sense of making choices that further their interests.
The theory, however, interprets both interests and the rationality of choice much
more broadly than neoclassic economics does. For example, the new institutional-
ism distinguishes itself from neoclassical assumptions of rationality by attending
to “cognitive costs” of decision making. The pursuit of benefits is limited by indi-
viduals’ capacity to retain and process information; in other words, individuals are
boundedly rational (Coase 1937, Simon 1957). Furthermore, information is often
costly (Barzel 1989). These two factors create transactions costs—the costs of
writing and enforcing contracts—because individuals cannot foresee at the time
ofwriting all of the contingenciesthat might be relevant, nor can they observe all of
the actions of their partners. Also, transactions costs give rise to the possibility of
opportunism (Williamson 1975).
The possibility of opportunism and its role in the new institutionalism deserve
particular attention because opportunism has been the basis of criticism from
sociologists and because it contrasts with the emphasis on trust in the network
embeddedness approach to economic sociology (Granovetter 1985, Uzzi 1996).
Often, the opportunism-trust debate takes place on rhetorical ground, with new
institutionalists criticized for unwarranted cynicism (Ghoshal & Moran 1996) and
respondingwithwarnings that “there are real dangersinadopting the more benevo-
lent construction” of human nature (Williamson 1994, p. 81). Such debate quickly
reaches an impasse, and the gains from intellectual exchange would be greater if
the starting point was a recognition that the new institutionalism does not rely on
the assumption that individuals will engage in malfeasance, but rather on the as-
sumption that they pursue their self interest. From the assumption of self interest,
the interesting question becomes “Why shouldn’t individuals engage in malfea-
sance?” As we explain, the new institutional answer is that sometimes institutions
constrain individuals so as to remove the gains from malfeasance.
The problem of credible commitment emerges from the possibility of oppor-
tunism,andis at the core of the newinstitutionalism. It isillustratedby the dilemma
faced by a kidnap victim whose kidnapper has a change of heart and decides to set
her free (Schelling 1960). The victim gladly promises not to reveal the kidnapper
tothe authorities in exchange for her freedom.However, the kidnapper realizes that
once the victim is free she will have no incentive to keep her promise and reluc-
tantly decides the victim must be killed. More generally, the problem of credible
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?NEW INSTITUTIONALISM 529
commitment is faced by any party to an exchange who wants to promise in the
present to do something in the future that may not be in his or her interest to do
when the future actually arrives. The problem is endemic because in almost every
exchange there is at least a moment when one of the parties has control over all or
most of the goods and must decide whether to follow through on the agreed upon
bargain or make a grab for more.
Although the costs of rational decision making have been thoroughly analyzed
asthe basis of transaction costs and the needforinstitutions,interestsorpreferences
have been largely ignored. The new institutionalism applies a “thick” view of
preferences, which often includes, in addition to material goods, ideology and
social goods such as status. In practice, however, nonmaterial preferences have
been given short shrift in the theory. This is illustrated by North’s (1990, p. 22)
characterization of the trade-off between ideological and material preferences:
“...where the price to individuals of being able to express their own values and
interests is low, they will loom large in the choices made; but where the price one
paysforexpressingone’sownideology,or norms, or preferences is extremelyhigh,
they will account much less for human behavior.” This is too weak a concession to
the role of ideology, particularly in the face of evidence such as that from Poole &
Rosenthal’s (1996) analyses of the whole history of US Congress roll-call voting,
which indicated that ideology explained more of the variance in voting behavior
than did economic interests.
Another limitation of the behavioral assumptions is that the new institutional-
ism, like most other theories based on rationality, suffers from the absence of an
explanationfor the origin of preferences. This is one of the areas in which sociology
is poised to make a critical contribution to the theory. Economists have generally
takenacomplete set of well-orderedpreferences as a starting assumption,although
preferences are almost certainly endogenous. Thus the explanation of preferences
has been left to others (DiMaggio 1990). Veblen (1899) explicitly recognized
the relationship between preference and social structure, and more recently soci-
ologists have argued that preferences are socially constructed (DiMaggio 1990,
Friedland & Alford 1991). Certainly, the status and social identity implications of
objects explain much of their appeal to consumers. Sociologists are also able to
help with explanations of how actors form strategies to achieve their preferences.
Strategy formulation is inhibited by bounded rationality, and it is clear that actors
do not weigh the full set of means to achieve their preferred ends (Cyert & March
1963). Recent organizational sociology has explained, for example, how corporate
diversification or mergers become accepted and popular strategies (Fligstein 1990,
Dobbin & Dowd 1997).
OPERATION OF INSTITUTIONS
If the main problem derived from the new institutionalism’s behavioral assump-
tions is transaction costs, institutions can be seen as a partial solution. Institutional
rules and their enforcement mechanisms can structure the interactions of actors
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?
530 INGRAM ¥CLAY
such that the costs of acquiring and processing information are less debilitating,
and the opportunities for malfeasance are reduced. This proposition raises a num-
ber of questions. Who makes rules? How are the rules made? Who enforces the
rules? And how are the rules enforced? In this section, we briefly discuss each
of these questions and then examine the literature on three broad classes of insti-
tutions.
Institutions are typically categorized as formal or informal (North 1990, Nee &
Ingram1998).Weuseamorefine-grainedcategorizationbasedontwodimensions,
who makes the rules (the state or some other entity) and how are they made and
enforced (in centralized or decentralized fashion). We use this two-dimensional
categorizationscheme because we thinkitisbetterthan the formal/informal catego-
rization for (a) differentiating the most relevant distinctions between institutional
forms, (b) emphasizing the ways that different institutional forms are interdepen-
dent, and (c) relating the different institutional forms to the actors with which they
are most closely associated.
States make rules in a centralized manner, but so do other entities. Many clubs
and other organizations, like the American Medical Association, create rules in
much the same way as a state does, through the actions of one or more appointed or
elected members. Rules, however, may also arise in a decentralized way, through
interaction and taking the form of norms. Norms may specify that neighboring
rancherslook out for oneanother’scattle or that coworkers cover for one anotherin
anillness. In what follows,we refer to institutionsasbeingoneof three types: public
(effectively, public-centralized), private-centralized, or private-decentralized.
If institutions are constraints, it is not sufficient for rules to exist; they must
also be enforced. Enforcement can be by the party who was harmed (second-party
enforcement) or some other party (third-party enforcement). Public and private-
centralizedinstitutions typically havethird-party enforcement. Forinstance,theUS
government punishes individuals who are convicted of federal crimes. Similarly,
the California State Bar Association can vote to suspend a member’s license to
practice law. Except in rare cases, the breaking of the rule did not harm the federal
governmentor the California Bar.Incontrast,private-decentralizedinstitutions can
haveeither third-party orsecond-partyenforcement. For instance, among medieval
merchants, one merchant might refuse to interact in the future with a merchant
who had cheated him, a form of second-party enforcement. If other merchants
also refused to interact with the cheater because of this episode, then punishment
would also be of the third-party type.
A related issue is how rules are enforced. Many institutions offer some form
of dispute resolution to ensure that a rule has actually been broken before the
sanctions are applied. The government maintains a court system for this purpose,
trade groups maintain arbitration panels, and, even in decentralized institutions,
the parties may at times turn to an impartial third party. Sanctions, if they are
deemed necessary, can take a variety of forms, both economic and noneconomic.
Economic sanctions can include fines, restitution, or penalties that impair future
earnings, such as restrictions on economic activity. Noneconomic sanctions can
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?NEW INSTITUTIONALISM 531
includeloss of certain rights,status,orinteraction with certain groups.Forexample,
social groups often punish indiscretion with ostracism.
Against this background on the nature and enforcement of institutions, we now
examine in detail the three institutional forms we have identified. We emphasize
the ways that each institutional form operates to affect the choices of actors. We
also identify interdependencies among the forms.
Public Institutions
There are four major ways that the public institutions provided by the state can be
understood to affect its choices and those of organizations and individuals. First,
the state may smooth exchange between its subjects by providing institutions
that allow them to make credible commitments. This can be achieved if the state
provides a legal system to protect property rights, decrease transaction costs, and
enforce contracts. Spicer (1997) illustrates the role of such a system by examining
a case in which the state did not play this role, the mass privatization of financial
markets in Russia from 1992 through 1996. During this period, regulation of
organizationsthat had formedtoattractconsumersavingswasweaktononexistent.
It was unclear who had regulatory authority in the market, and the contenders to
supply regulation lacked the necessary resources. A number of well-publicized
and unpunished scams by organizations in the industry reduced the willingness
of Russians to transact with financial organizations. The institutional weakness
was recognized by the executive of a US bank operating in Russia, who expressed
a wish that the state would regulate his bank more heavily. Such a hope sounds
unusual coming from an American banker, but it indicates how the state enables
the credible commitments of subject organizations by constraining them.
There is quantitative evidence of the role of public institutions for enabling
credible commitments. Some studies exploit changes in laws governing specific
industries to show that increased legal constraint on organizations causes them
to flourish. Studies of populations as diverse as US health maintenance organi-
zations and telephone companies, Toronto day-care centers, Niagara Falls hotels,
and Singapore banks have demonstrated that their failure is reduced by increas-
ing government involvement in monitoring, certifying, authorizing, and endorsing
their activities (Wholey et al 1992, Barnett & Carroll 1993, Baum & Oliver 1992,
Ingram & Inman 1996, Carroll & Teo 1998). The effects of broader changes in
publicinstitutions are seen in Ingram & Simons’ (2000) analysisof the effect of the
formation of the Israeli State on the failure rates of workers’ cooperatives in many
industries.The transition fromthe weak British Mandate for Palestine tothe strong
Israeli State caused a radical improvement in the institutional support for credible
commitment and a corresponding 60% decrease in organizational failure rates.
Stone et al (1996) use, to exciting effect, a comparative analysis of the institu-
tions of Chile and Brazil as these institutions affect their countries’ garment indus-
tries. Brazilian laws governing the economy are so complex and their enforcement
so expensive that “...[public] institutions could not be relied on for dealing with
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?
532 INGRAM ¥CLAY
day-to-day problems in regulation and business transactions” (100). By contrast,
Chile is “... regarded as having a relatively well-defined property-rights system
and a liberal economy” (102). The comparison between the two supports the claim
that effective public institutions facilitate economic exchange, but also indicates
that private institutions may often be a decent substitute. In some areas, Chile ap-
peared to have lower transaction costs than Brazil. For example, customer orders
weremuch more likely tobe renogotiated before deliveryin Brazil,and, as a result,
Brazilian garment manufacturers were less likely to produce customized goods. In
other areas, however, private institutions have arisen in Brazil to offset the ineffi-
ciency of public institutions. Starting a business in Brazil requires the negotiation
of complex government rules, but specialized professionals handle the details in
∼7 weeks, for an average cost to the founding firm of $640, comparable with the
time and expense of starting a business via Chile’s simpler process. And Brazilian
garment manufacturers rely heavily on credit, despite the cumbersome and ex-
pensive contract enforcement procedures they face. Apparently, they substitute a
reliance on reputation, facilitated by institutions of credit agencies and customer
credit references. The emergent picture of an interdependence between public and
private institutions presents a rich set of questions for the new institutionalism.
This issue reasserts itself repeatedly as we discuss the various institutional forms.
The second key feature of public institutions is whether the state can credibly
commit to not subsidize subject organizations when they struggle. The recent
transitionsfromstatesocialismhavedemonstrated that, absent such a commitment,
entrepreneurs will direct their energies towards “holding up” the state treasury
rather than to producing economic value. As Stark & Bruszt (1998, p. 119) put it,
when the state hears organizations’ “...siren cry, ‘Give me a hand, give me your
hand,’ it must be bound to respond not simply that it should not, or that it will
not, but that it cannot.” McFaul (1995) describes the problem of state subsidies to
enterprises in Russia, which, by the spring of 1993, were estimated to be 22% of
the gross domestic product. According to McFaul, state weakness was the cause
of Russia’s failure to make credible its commitments not to subsidize. Stark &
Bruszt (1998) refine that position by emphasizing that state strength is relative
to the strength of other actors. Their analysis of transformations in east-central
Europe shows that efforts to credibly commit against subsidies were sometimes
implemented not by strengthening the state, but by attempting to weaken subsidy-
seeking groups by limiting the size and interconnectedness of organizations.
Thethird keyfeatureofpublicinstitutions is an outgrowthofthefirst two. Astate
that is strong enough to guarantee the property rights of its subjects and to resist
their calls for subsidies is also strong enough to appropriate their wealth (Weingast
1993). Unless the state can credibly commit itself against such appropriations, its
subjects’incentivesforproductiveeconomic activitywill be greatly curtailed. This
is aptly illustrated in North & Weingast’s (1989) account of the Stuarts’ impact
on the economy of seventeenth century England. After coming to the Crown in
1612, the Stuarts exploited their subjects in numerous ways: they sold monopolies
(at the expense of industry incumbents and potential entrants), they sold special
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?NEW INSTITUTIONALISM 533
dispensations from laws, and they even committed outright theft, as in 1640 when
they seized £130,000 that private merchants had placed in the Tower of London
for safekeeping. Evidence that citizens recognized the parasitic nature of their
sovereign and incorporated it into their economic calculus comes from the Stuarts’
inability to raise debt through normal means (they resorted to “forced loans,” the
payment terms of which they subsequently violated). These abuses [among other
causes(Carruthers1990)]led eventuallytotheGlorious Revolutionof 1688,which
resulted in numerous institutional changes to reduce the Crown’s capacity to act
independently of Parliament and the courts. This loss of Crown autonomy had,
however, positive implications in that it enabled the Crown to make a credible
commitment not to appropriate subjects’ wealth. The value of this commitment
can be seen in the dramatic increase in the Stuarts’ capacity to borrow funds—
governmentdebt grew17-fold inthe 9 years after therevolution.North &Weingast
argue that the improvements in Crown finance were mirrored by improvements in
private financial markets.
Strikingas this account is, to attribute the state’scapacity to makecredible com-
mitments to a constitutional division of powers is to finesse the central issue. If a
state is powerful enough to enforce property rights, why need it be constrained by
the “parchment barrier” provided by a constitution? Why doesn’t the state simply
breakthe law? Weingast(1993) examinesthisquestion with a game-theoreticanal-
ysis.He considers a state that decides whether to transgress against each oftwocon-
stituent groups, which in turn decide whether to acquiesce to or challenge the state.
The repeated game has multiple equilibria, but the normal diversity of interests
among constituents makes it more likely that the game will result in an equilibrium
in which the state transgresses the rights of some constituents and retains the sup-
port of others. The “credible-commitment equilibrium,” in which the constituents
challenge the state for any transgressions, and the state therefore abstains from
transgressing, can be expected only if there is substantial agreement among con-
stituents about the appropriate role and limits of the state. “[T]he foundations for
institutional restriction fundamentally rest on the attitudes of citizens.” This recog-
nition of the normative underpinnings of public order may sound to sociologists
like a second Glorious Revolution. Certainly, it is sociology that is best positioned
to explain how individuals come to share attitudes towards the state, and indeed
this question has been the subject of recent sociological effort (Meyer et al 1997).
Fourth and finally, some institutions provided by the state are not to be under-
stood as part of a grand effort to facilitate the credible commitments of actors, but
rather in terms of the distributional battles over zero-sum interests (Knight 1992).
These may be the battles between suppliers and consumers, as shown in Dobbin &
Dowd’s (1997) analysis of the effects of competition policy on railroad foundings
in early Massachussetts. Or they may be the battles between rival organizational
forms without apparent efficiency differences, as for thrift-savings organizations
that fought as much in the legislative arena as in the market (Haveman & Rao
1997) or national coffee roasters in the United States, that derived a competitive
advantage over regional roasters through an international treaty (Bates 1997).
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?
534 INGRAM ¥CLAY
Although our categorization of institutions as public/private and centralized/
decentralized suggests two types of public institutions, centralized and decen-
tralized, this discussion of public institutions has been exclusively about public-
centralized institutions. In the choice-within-constraints framework, the
public-decentralized cell is empty—there is no literature on state-level institutions
that emerge without central authority. New institutional sociologists, however,
have addressed that cell, describing the equivalent of national and international
norms of state structure and behavior that emerge through cultural and associa-
tional processes (e.g. Fligstein & Mara-Drita 1996, Meyer et al 1997, Dobbin &
Sutton 1998). The failure of the choice-within-constraints theory to account for
the influence of national culture and world society is one of the costs of its re-
liance only on influences within the immediate action frame created by interacting
individuals, organizations, and states.
Private-Centralized Institutions
Private-centralized institutions may be divided into two types based on the nature
of their effects on actors: those that govern property rights, and those that enable
transactions.ThefirsttypeisillustratedintheAmericanWest of the nineteenth cen-
tury, where individuals formed claims clubs, cattlemen’s associations, and mining
districts to protect their property rights in agricultural, ranch, and mineral-bearing
land. In each case, they reaped or tried to reap the benefits from excluding others.
Members of claims clubs in Iowa prevented speculators from buying up the land
they had been squatting on (Bogue 1958). Ranchers in the American West who
ran their cattle on federal land kept out later arrivals by excluding them from the
roundup (Dennen 1976). Also, during the California Gold Rush, groups of min-
ers established mining districts on federal land that allowed individuals to mine
without interference from others (Umbeck 1981, Clay & Wright 1999). All of
these groups had some type of written rules, although, as Clay & Wright show for
mining districts, written rules were often augmented by norms or even supplanted
by them. Enforcement was at least nominally centralized, with members of the
claims club, cattlemen’s association, or mining district using third-party enforce-
mentto protect theirproperty rights. Thisenforcement took differentforms in each
case: for claims clubs, it was physical force against speculators; for the cattlemen’s
association, it was exclusion of outsiders from the annual roundup, an economic
penalty; and for mining districts, it was removal of interlopers by physical force.
By enforcing rules against speculators buying their land, later entrants running
cattle on federal land, or new miners seizing property held by others, members
provided outsiders with incentives to respect their property rights.
The formation of private-centralized institutions to govern property rights is not
justa historical phenomenon. Ostrom (1990) documents avarietyof contemporary
institutions that have arisen to manage property rights over resources that are
common property (that is, resources that are held in common by many actors).
Someorallofthe individualswhoholdthe property in common make the rules, and
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?NEW INSTITUTIONALISM 535
enforcement is through third-party imposition of economic and social sanctions.
One difference between these institutions and the ones in the previous paragraph
is the types of incentives that the rules and enforcement create. In this case, rules
and third-party enforcement of these rules create incentives for members not to
overutilize the resource, rather than for outsiders to respect members’ property
rights. For example, in Torbel, Switzerland, alpine meadows are held in common
by the 600 villagers. Grazing rights are allocated based on the number of cows that
a citizen can support in the winter. During the summer, a group of herdsmen tend
the village cattle in the communal meadow. Milk from the cows is used to produce
cheese that is allocated to families based on the number of cattle that were grazed.
Elected officials hire the herdsmen and impose fines on households who misuse
the commons by sending too many cattle (Ostrom 1990, pp. 60–65).
Private-centralizedinstitutions do not always arise toaddresscommon-property
problems.Johnson& Libecap (1982) address the questionofwhy these institutions
donotarise, in a study of the shrimp industryintheGulf of Mexico. In this industry,
over-fishing is an ever-present problem. During the 1930s, 1940s, and 1950s, an
industry association had managed shrimp stocks on the Gulf Coast by setting
price floors that made it unprofitable to harvest small shrimp. Union members
were given incentives to adhere to the price floors through fines, and the union
pressured packers not to buy below union price or from nonunion shrimpers.
In 1956, however, the Fifth Circuit Court of Appeals found the union to be in
violation of the Sherman Act, making further efforts to limit entry or effort illegal.
This illustrates the interdependence between institutional forms, because it was a
public institution that defeated a private-centralized institution that was designed
to solve the problem of over-fishing. In other cases, private-centralized institutions
cannot solve common-property problems because of the difficulty of monitoring
compliance with the institutions or because of holdout by actors with a stake in
the common property (Wiggins & Libecap 1985).
Examples of the second role of private-centralized institutions, to govern trade,
date at least from the rise of merchant guilds in the late-medieval period. Although
some authors have interpreted these guilds as a vehicle for merchants to capture
rents by limiting entry, Greif et al (1994) provide evidence indicating that their real
purpose was “...to allow rulers of trade centers to commit to the security of alien
merchants.” The guilds created incentives for rulers to respect member merchants’
rights by credibly committing to end trade if these rights were violated. Guilds had
and enforced rules that required members to punish by ostracism any members
found to be trading with rulers who had violated members’ rights. Third-party
enforcement by the guild of these rules through the imposition of fines or other
sanctions created incentives for members not to trade with such rulers, thereby
making a guild’s commitment to end trade credible.
In the twentieth century, a private-centralized institution has arisen to facili-
tate trade in the diamond industry. Bernstein (1992) examines the rules that gov-
ern transactions and the reasons why diamond dealers have chosen to enforce
these rules through a private-centralized institution and not through standard legal
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?
536 INGRAM ¥CLAY
channels. As in the foregoing cases, members are governed by formal written rules
that represent the codification of, and are supported by, industry norms. Bernstein
finds that the use of arbitration panels and mandatory prearbitration conciliation
is a response to members’ need for speed, secrecy, and specialized knowledge
of industry laws and norms. The industry enforces arbitration decisions with the
threat of suspension of membership in the diamond bourse. In the case of non-
compliance, a bourse faxes the individual’s picture to all other diamond bourses
worldwide. Informed of noncompliance, members then refuse to trade with the
individual in question because of the risk that they will be cheated. Through this
reputation mechanism, the institution creates incentives for members to adhere to
industry rules and norms in their transactions with other members.
The most ubiquitous private-centralized institutions that govern trade are orga-
nizations that internalize transactions. In his seminal 1937 paper, Coase took up
the pivotal question of why organizations exist. His central insight was that the
governance of exchange within organizations as opposed to markets depended on
the cost of transacting in each type of institution. In more recent work, Williamson
(1985) and others have systematically investigated the effect of information, op-
portunism, and asset specificity on the governance of exchange, concluding that,
in some transaction environments, exchange is more efficient within an organi-
zation than the market. Furthermore, the prevailing public institutions influence
the attractiveness of various governance arrangements (Nee 1992). Ficker (1999)
illustratesthe relationship between the environmentof public institutionsand other
factors, and the market/organization trade-off in the evolution of the Mexican Cen-
tral Railroad. The Mexican Central Railroad was founded in 1880 into “a country
characterized by economic backwardness and an incipient and precarious insti-
tutional framework.” The company initially pursued a strategy of building main
lines and depending on market transactions with railroads and other types of trans-
portation organizations to supply them with freight. These market transactions
did not materialize, however, owing to the weak Mexican infrastructure, and the
difficulties of organizational and technological coordination. In response to this
failing of the market, the Mexican Central Railroad switched to a strategy of inter-
nalization, extending its trunk lines and building branch lines to supply itself with
freight.
Private-Decentralized Institutions
Institutions of the third form—those consisting of private-decentralized rules such
asnorms—werewidely ignored until recently,no doubt because they are much less
visiblethan centralized institutions.Anexceptionto the neglectofthiscategory was
the pioneering work by Macaulay (1963) on contract enforcement by executives.
Macaulay showed that, when problems arose in a relationship, business norms
dictated that the two parties attempt to resolve the problem informally rather than
go to court. This was true even if going to court would have provided a higher
payoff for the plaintiff. Plaintiffs had an incentive to agree to informal resolution,
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?NEW INSTITUTIONALISM 537
because developing a reputation for litigiousness would adversely affect future
business dealings within the community.
Private-decentralized institutions have also been seen to support long-distance
trade in the absence of a legal system. In long-distance trade, merchants can often
profit from using other merchants as agents to sell goods, collect debts, and so
forth. This agency relationship, however, raises the possibility that the agent will
actopportunistically,keepingsomeor all of the monies owed. The Maghribi traders
in the eleventh century western Mediterranean (Greif 1989, 1994) and American
merchants on the nineteenth-century California coast (Clay 1997a, 1997b) over-
came this problem by forming coalitions, which allowed exchange to flourish. In
both cases, merchants in the coalition conditioned future use of other merchants
as agents on those merchants’ having acted in accordance with group norms in the
past. For instance, when a Maghribi merchant was accused of cheating in 1041–
1042, he found that “people became agitated and hostile and whoever owed [me
money] conspired to keep it from me” (Greif 1994, p. 925). Merchants checked
on agents’ past behavior and verified trade-related information in letters to one
another, creating an information network. By tying future economic gains to past
behavior as an agent, merchants were able to ensure that the future gains to mem-
bership in the coalition were greater than the gains to cheating and being punished.
Punishment was decentralized in that each merchant, having learned that an agent
had violated group norms, then had an incentive not to interact with that merchant
in the future because he expected the merchant to cheat him. The incentives cre-
ated by this reputation mechanism enabled merchants to enter into and enforce
contracts in the absence of a legal system.
The ability of private-decentralized institutions to support “order without law”
goes beyond trade. For instance, Reid (1980) shows that, on the Overland Trail
during the nineteenth century, individuals acted as if laws existed, carrying pre-
vailing legal norms with them, even though there was no legal system to enforce
them. The threat of social and economic ostracism on the trail created incentives
for individuals to adhere to the norms that led to order. In addition, Ellickson
(1991) documents Shasta County cattlemen’s use of norms that dealt quickly and
effectivelywith the ongoing problem ofcattle that strayed and damaged neighbors’
property. By tying the prompt return of stray cattle and other forms of cooperation
to adherence to norms, the institution created incentives for cattlemen in the area
to act in accordance with the norms. There were laws regarding stray cattle, but
few cattlemen knew them, and fewer still relied on them.
Nee & Ingram (1998) argue that decentralized-private institutions, which exert
the most immediate control on individuals, determine the effectiveness of insti-
tutions of the other two forms. Whether pubic- or private-centralized institutions
have their intended effects will depend on whether private-decentralized insti-
tutions encourage individuals to accede to or oppose them. Heckathorn (1990)
presents formal models to show that the coercive efforts of states may fail if af-
fected populations develop “opposition norms,” which supply social rewards for
opposing state coercion. Homans’ (1950) reanalysis of the Bank Wiring Room
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?
538 INGRAM ¥CLAY
shows a similar result for the private-centralized institutions maintained by the
Western Electric Company. There, norms in the work group influenced employees
to engage in behavior that was the opposite of what the organization’s pay system
tried to encourage.
Embeddedness theory, from economic sociology, is also informative about the
operation of private-decentralized institutions. Indeed, at the surface level of ob-
servedcauses and effects,thetwo approaches are sometimesdifficultto distinguish.
ForexampleMaurer &Haber’s(1999) new institutional analysis andUzzi’s(1999)
embeddednessanalysis both findthat organizations’ access tobank loans is greater
if they have social relationships with their bankers. There are differences, however,
in the behavioral assumptions that these two theories use to arrive at predictions.
Newinstitutionalists believethat the significance of interpersonal relationshipsisin
their capacity to create social sanctions, which give norms their teeth. Participants
adhere to norms when to do so is justified by a comparison of current and future
payoffs for adhering to or violating the norm (Ulman-Margalit 1977). In contrast,
embeddedness theorists argue that social relationships have a logic of their own,
which is sometimes divorced from the rational calculus of benefits (Granovetter
1985, Uzzi 1996). New institutionalists could gain by incorporating something of
the embeddedness view of relationships, because even rational-choice theorists
concede that interpersonal relationships rest on more than calculated self-interest
and that norms are sometimes internalized (Coleman 1990).
It is also true that embeddedness theory could gain from the new institutional-
ism, because the latter theory has gone farther towards understanding the role of
interpersonal relationships in the broader institutional framework. Greif’s (1994)
seminalstudycomparing the institutions that theMaghribiand the Genoese used to
support long-distance trade in the Mediterranean during the late Medieval Period
illustrates the limitations of private-decentralized institutions (embeddedness). We
previously described the Maghribi coalition in which trade was embedded in a net-
work of coethnics with close informational ties. In contrast, in the twelfth century,
the Genoese used vertical agency relations under which merchants were almost
never agents, agents could be of any ethnic background, and information was not
shared. Merchants began to rely on a public institution—the legal system—to en-
force contracts. Thus trade became more anonymous, allowing the Genoese to
respond effectively to the geographic expansion of trade. In contrast, the Maghribi
merchants’ reliance on the information network and the need for merchants’ to be
of a specific ethnic background and to act as agents may have limited their ability
to expand geographically. The lesson here is that there is a limitation to managing
exchange through relationships, because relationships can only be stretched so far
while maintaining the characteristics that make them effective. For some forms of
exchange, particularly those over large spans of space and time, centralized insti-
tutions may be more effective than private-decentralized institutions or embedded
relationships.
Afinalissue,bothfortherelationshipbetweendecentralized-privateinstitutions
andother institutional forms andfor the relationship betweenthe new institutional-
ismandother sociological theories, concerns the beliefsandshared understandings
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?NEW INSTITUTIONALISM 539
held by individuals. It is becoming increasingly apparent that these mental con-
structs affect the operation of both public- and private-centralized institutions. For
instance, North (1993), comparing the performance of public institutions from
Western economies to those making the transition from state socialism, attributes
variance in performance to differences in the mental constructs of individuals in
differentsocieties. In addition, Williamson(1975)arguesthat the appropriate mode
of governance in an organization will be conditioned on the prevailing “contrac-
tualatmosphere,”by which he means the receptivenessof individuals to alternative
governance modes. Despite the importance of beliefs and shared understandings,
the new institutionalism that is our focus has little to say about how they are ac-
quired. Organizational theory’s cognitive new institutionalism, however, has made
this one of its central questions (Meyer & Rowan 1977, Powell & DiMaggio 1991,
Scott 1995). It contends that individuals come to beliefs and shared understand-
ings through processes of social construction. It offers another institutional form,
cognitive institutions, which act as interpretive lenses for social facts, including
other institutions. Without downplaying the difficulties of combining ideas from
theories with very different behavioral assumptions, the choice-within-constraints
framework needs a concept like that of cognitive institutions.
INSTITUTIONAL CHANGE
New institutionalists need to account both for how institutions emerge and change
and for the timing of emergence and change. Early work on institutions by Davis
& North (1971), North (1981), Williamson (1975, 1985), and others asserted that
efficientinstitutions wouldemergeand changeas the need arose and thatthetiming
of their emergence or change would be economically optimal. This view has now
largely been discredited by economists and sociologists. In a now famous critique,
Granavotter (1985) attacked Williamson’s early accounts of institutional change
as bad functionalism. What made Williamson’s arguments bad functionalism was
that they did not do a sufficient job of tying the function of institutional forms to
the mechanisms by which they came to predominate. Even North’s (1993, p. 12)
view has changed; as he puts it, “there is ...little evidence to support the view ...
that the necessary institutions will be the automatic outcome of getting the prices
right.”
The challenge for institutionalists is to create a richer theory of the origin and
changeof institutions. One of the biggest barriers to creating a richer theoryisalack
of empirical work, particularly on private institutions. Ostrom’s (1990) discussion
of the emergence of institutions to govern water in the Los Angeles Basin, and
Ingram & Inman’s (1996) study of the emergence of an institution to promote
tourism at Niagara Falls are some of the few accounts of the origin of private
institutions. Yet, precisely how actors overcome the second-order collective action
problem inherent in the creation of institutions is central to any theory of the origin
of institutions. Empirical studies of change are equally scarce. Part of the problem
is that most private institutions are not well documented, making it difficult to do
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?
540 INGRAM ¥CLAY
careful, detailed empirical work. Ostrom’s and Ingram & Inman’s work suggests
that studies can be done; more studies are badly needed.
Thelimited empirical workthathas been donehasfocusedon public institutions.
North(1993) summarizes the current understandings,beginning with the emphasis
on organizations as the primary source of institutional change: “... competition
forces organizations to continually invest in skills and knowledge to survive. The
kinds of skills and knowledge that individuals and their organizations acquire will
shape evolving perceptions about opportunities and hence choices that will incre-
mentally alter institutions” (p. 17). The organizational engine of change operates
within the existing institutional framework, which affects efforts to change insti-
tutions because existing organizations have a stake in the status quo (North 1990).
Particularlyrelevantpartsof the existinginstitutionalframeworkare the beliefsand
shared understandings of individuals, which frame proposed institutional changes.
North’s summary points to two key questions that must be answered for a
theory of institutional change to advance. First, if organizations are the motor of
institutional change, then the new institutionalism needs a theory of organizational
action and change. Empirical work suggests that organizations are inertial, that
they pursue noneconomic interests, and that they are only boundedly rational.
Organizations are notoriously poor filters of changing interests because they are
rife with inertia (Hannan & Freeman 1984), so much so that they tend to reflect the
conditions of their founding, even decades later (Stinchcombe 1965). Having been
founded to exploit a given institutional framework and with limited capacity to
change, existing organizations are favored by stability of institutions. Rather than
the source of pressure to change institutions, existing organizations are a source
of resistance to institutional change. And although it is true that entrepreneurs are
drivento developnewskills and knowledge, theytendto exploittheseopportunities
through new types of organizations (Ingram 1998).
Effortsto effect institutional change may be motivated by a varietyof factors, of
whicheconomicinterestisonlyone.Carruthers(1990)pointstotheroleof noneco-
nomicinterests for institutional changein a criticism ofNorth & Weingast’s (1989)
account of the Glorious Revolution. This instance of institutional change, among
the most celebrated in terms of the gains in economic efficiency it created, was,
according to Carruthers, largely the product of religious competition, determined
by “popery, not property.” The prominence of noneconomic interests does not de-
cline when competing organizations are the engines of institutional change. Orga-
nizations, like individuals, pursue political ideologies, even at substantial material
expense (Simons & Ingram 1997). Organizational efforts to change institutions of-
ten reflect a mix of economic and noneconomic interests. For example, hotel chain
entrepreneurs in the United States of the 1920s changed institutions surrounding
trainingby establishing universityhotel schools, partly to generate theprofessional
managers their organizations required, but also in the hope that university educa-
tion would improve the social status of their occupation (Ingram 1998). In other
instances, organizations seek to affect institutions for purely ideological purposes,
but economic results occur as well, as when the Women’s Christian Temperance
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?NEW INSTITUTIONALISM 541
Union lobbied for prohibition legislation that forced breweries out of business in
the jurisdictions in which it was passed (Wade et al 1998).
Finally, the bounded rationality that the new institutionalism attributes to indi-
vidualsdoesnotconvenientlydisappear when those individuals form organizations
(Cyert & March 1963). As North (1993) notes, the complexity of the institutional
framework presents a particular strain on actors that might manage to be more or
less rational when making simple decisions. Consequently, even if inertial orga-
nizations pursue their varied interests by trying to change institutions, there is no
reason to believe that they will always attempt the right changes. It is not uncom-
mon to find that organizational efforts to change institutions in their favor backfire
(Ingram & Inman 1996, Wade et al 1998).
Thesecond key questionthatemergesfrom North’ssummary is, like theneedfor
a better organizational theory, one that sociologists are already working to answer.
It concerns the role of beliefs and shared understandings, which were also identi-
fied as relevant to the operation of institutions. Beliefs and shared understandings
feature prominently in efforts of institutional change as they define the perceived
legitimacy of institutional arrangements, forming the frames through which pro-
posed changes can be related to broader societal interests (Fligstein 1997, Rao
1998). This framing may determine what institutional alternatives are accepted
or rejected. This is seen in the efforts of South American coffee producers to
present US acquiescence to the International Coffee Organization, a treaty-based
cartel that raised prices to consumers, as necessary to fight communism (Bates
1997). It is also apparent in the efforts of hoteliers in Niagara Falls, Ontario, and
NiagaraFalls,NewYork,to frame laws that aimed tomaketheFallsmore attractive
for tourists, as acts of nationalism in a United States-Canadian rivalry (Ingram &
Inman 1996). These examples make it clear that the process of institutional change
goes far beyond the weighing of competing interests. Instead, it appears to be as
much a comparison of rhetorical claims, the outcome of which depends on deep
social values of the audience.
This argument contributes to a broader recognition that, in institutional change
as in the operation of institutions, centralized actions often rely on decentralized
ones. Knight (1992) offers a theory of institutional change that is notable in that
it focuses on social norms as the foundation for change in other types of insti-
tutions and it focuses on the distributional consequences of institutions. Often,
new institutionalists are sufficiently taken by the public-good nature of many in-
stitutions that they characterize them as cooperative arrangements that benefit all.
In contrast, Knight emphasizes that institutional change always has distributional
consequences and that bargaining between interested parties is the process that de-
termines institutional outcomes. The role of such bargaining seems clear when we
consider change in centralized institutions, like a law that would favor some finan-
cial organizations but disfavor others (Haveman & Rao 1997), but Knight argues
thatbargainingalsooccurs overdecentralized institutions and that relativebargain-
ing power determines which social norms arise and persist. Knight & Ensminger
(1998) illustrate the argument with a case study of changing norms among the
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?
542 INGRAM ¥CLAY
Galole Orma of the Tana River District, Kenya. The norm of clan exogamy, for
example, is declining among the Orma. This change occurred after a redistribution
of bargaining power from those who the old norm favored—the clan elite who
hope to build alliances through marriage ties—to those it harmed—young men
and women who want more freedom to choose marriage partners.
Hopcroft (1998) also examines the role of institutional interdependence for in-
stitutional change, arguing that changes in state laws and ideologies flowed from
localinstitutionsin rural villages. In doingso,Hopcraft challenges the emphasis on
public institutions to explain economic development in the Western world (North
& Thomas 1973), observing that, in preindustrial England, there were local differ-
ences in economic action and performance, although regions shared the same um-
brella of state institutions. Hopcraft attends particularly to the differences between
the champion (regular open-field) system of central England and the noncham-
pion systems of eastern and southwestern England, arguing that the nonchampion
systems, which provided for more individual property rights, resulted in lower
transaction costs and higher productivity. Beyond that, these private-centralized
institutions formed the groundwork for subsequent institutional change. Enclosure
occurred early and easily in the nonchampion regions, whereas it was late and dif-
ficult in champion areas, often forced by an act of Parliament. The nonchampion
regions adopted wage labor, while feudal rules persisted elsewhere. People from
the nonchampion regions also spearheaded efforts to change state institutions, for
example, campaigning against the granting of royal trading monopolies and sup-
porting the parliamentary cause in the Civil War of the 1640s. Hopcroft goes as
far as to attribute ideological change, in the form of a shift from a communitarian
culture to one of puritanical, civic-minded individualism, to the advantages of the
nonchampion system.
CONCLUSION
The potential of the choice-within-constraints new institutionalism is apparent
in the scope of the preceding sections. This is a theory that can be applied to
explain a substantial portion of the performance and change of all social systems,
including states, organizations, and groups. This comprehensiveness is also what
is most exciting about the empirical analyses conducted within the paradigm. The
new institutionalism has been applied to explain the differential performance of
economies throughout history, the structures of parliaments, the rise and fall of
particular organizations and industries in the twentieth century, and even changes
in marriage norms of African tribes. Impressively, these phenomena are tackled
withoutfalling back on a trivialtheoryofaction—the behavioral assumptions of the
new institutionalism recognize the constraints on human information-processing
capability and the range of interests pursued by individuals, organizations, and
states. Indeed, the recent trend is to give an increasing role to the influence of
ideologies, beliefs, and shared understandings on behavior.
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?NEW INSTITUTIONALISM 543
At the same time, there remain many weaknesses in the new institutionalism.
Sociologists are particularly well positioned to address the key weaknesses in the
theory. In the behavioral assumptions, there is a need for a theory of the origin
of preferences. Preferences, no doubt, arise partly through processes of social
interaction, which are deep within sociological territory.
Our categorization scheme for institutional forms allowed us to identify and
describe three. This approach indicated that there were interdependencies among
the forms across the dimensions of categorization: public institutions rely on pri-
vate institutions, and centralized institutions rely on decentralized institutions.
These interdependencies are also reciprocal. Sociological theory can be useful
in unpacking the interdependencies. In particular, embeddedness theory and cog-
nitive new institutionalism can contribute to understanding private-decentralized
institutions, which are the foundation of the institutional framework.
Regardingthe question of institutionalchange,theemerginganswers emphasize
the role of competing organizations as an engine for change and the importance of
framing institutional alternatives for legitimacy. Sociology’s organizational theory
can provide the necessary underpinnings for a change theory that is driven by
organizations. At the same time, cognitive new institutionalists in sociology have
begun to analyze the processes of cultural entrepreneurship through which an
institutional alternative comes to be accepted.
ACKNOWLEDGMENTS
Helpful comments on an earlier draft were provided by Frank Dobbin, Guy
Holburn, Stan Li, Michael Lounsbury, John Meyer, Victor Nee, Peter Roberts,
Jeff Robinson, Tim Rowley, Sampsa Samila, and Tal Simons.
Visit the Annual Reviews home page at www.AnnualReviews.org
LITERATURE CITED
Barnett WP, Carroll GR. 1993. How institu-
tionalconstraints affected the organization of
early American telephony. J. Law. Econ. Or-
gan. 9:98–126
BarzelY.1989. The EconomicAnalysisof Prop-
erty Rights. New York: Cambridge Univ.
Press
Bates RH. 1997. Open Economy Politics: The
PoliticalEconomy of the WorldCoffeeTrade.
Princeton, NJ: Princeton Univ. Press
BaumJAC,OliverC.1992.Institutional embed-
dedness and the dynamics of organizational
populations. Am. Sociol. Rev. 57:540–59
Bernstein L. 1992. Opting out of the legal
system: extralegal contractual relations in
the diamond industry. J. Leg. Stud. 21:115–
57
Bogue AG. 1958. The Iowa claims clubs: sym-
bol and substance. Mississippi Val. Hist. Rev.
45:231–53
Carroll GR, Teo ACY. 1998. How regulation
and globalization affected organizational le-
gitimation and competition among commer-
cial banks in Singapore, 1840–1994. Pre-
sented at Annu. Meet. Acad. Manage., 58th,
San Diego
Carruthers BG. 1990. Politics, popery, and
property: a comment on North and Weingast.
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?
544 INGRAM ¥CLAY
J. Econ. Hist. 50:693–98
ClayK. 1997a. Tradewithout law:private-order
institutions in Mexican California. J. Law
Econ. Organ. 13:202–31
Clay K. 1997b. Trade, institutions, and credit.
Explor. Econ. Hist. 34:495–521
Clay K, Wright G. 1999. Order without law:
property rights during the California gold
rush. Dept. Econ., Stanford Univ. Unpub-
lished paper
Coase RH. 1937. The nature of the firm. Eco-
nomica 4:86–405
Coase RH. 1998. The new institutional eco-
nomics. Am. Econ. Rev. 88:72–74
ColemanJ. 1990. FoundationsofSocial Theory.
Cambridge, MA: Harvard Univ. Press
Cyert RM, March JG. 1963. A Behavioral The-
ory of the Firm. Englewood Cliffs, NJ: Pren-
tice Hall
Davis LE, North DC. 1971. Institutional
Change and American Economic Growth.
London: Cambridge Univ. Press
DennenRT.1976. Cattlemen’sassociations and
property rights in land in the American west.
Explor. Econ. Hist. 14:423–36
DiMaggio PJ. 1990. Cultural aspects of eco-
nomic organization. In Beyond the Market
Place,ed. R Friedland,A Robertson, pp.113–
36. New York: Aldine de Gruyter
Dobbin F, Dowd TJ. 1997. How policy
shapes competition: early railroad foundings
in Massachusetts. Admin. Sci. Q. 42:501–
29
Dobbin F, Sutton JR. 1998. The strength of a
weak state: the rights revolution and the rise
of human resource management divisions.
Am. J. Sociol. 104:441–76
Ellickson RC. 1991. Order Without Law. Cam-
bridge, MA: Harvard Univ. Press
Ensminger J. 1992. Making a Market: The In-
stitutional Transformation of an African So-
ciety. New York: Cambridge Univ. Press
Ficker SK. 1999. Institutional constraints and
the strategy of the firm: an American rail-
road corporation in nineteenth-century Mex-
ico. Presented at Conf. Inst. Mark. Comp.-
Hist. Perspect., Palo Alto, CA
Fligstein N. 1990. The Transformation of Cor-
porate Control. Cambridge, MA: Harvard
Univ. Press
Fligstein N. 1997. Fields, power, and social
skill: a critical analysis of the new institu-
tionalisms. Presented at Conf. Power Organ.,
Hamburg, Germany
Fligstein N, Mara-Drita I. 1996. How to make
a market: reflections on the attempt to create
a single market in the European Union. Am.
J. Sociol. 102:1–33
Friedland R, Alford R. 1991. Bringing soci-
ety back in: symbols, practices, and institu-
tional contradictions. In The New Institution-
alism in Organizational Analysis, ed. WW
Powell, PJ DiMaggio, pp. 232–66. Chicago:
Univ. Chicago Press
Ghoshal S, Moran P. 1996. Bad for practice: a
critique of the transaction cost theory. Acad.
Manage. Rev. 21:13–47
Granovetter MS. 1985. Economic action and
social structure: the problem of embedded-
ness. Am. J. Sociol. 91:481–510
Greif A. 1989. Reputation and coalitions in
medieval trade: evidence on the Maghribi
traders. J. Econ. Hist. 49:857–82
Greif A. 1994. Cultural beliefs and the orga-
nization of society. J. Polit. Econ. 102:912–
50
Greif A, Milgrom P, Weingast BR. 1994. Coor-
dination, commitment, and enforcement: the
case of the merchant guild. J. Polit. Econ.
102:715–76
HannanMT,FreemanJ.1984.Structural inertia
and organizational change. Am. Sociol. Rev.
49:149–64
Haveman HA, Rao H. 1997. Institutional and
organizationalcoevolutioninthethrift indus-
try. Am. J. Sociol. 102:1606–51
HeckathornDD. 1990. Collective sanctions and
compliance norms: a formal theory of group-
mediated social control. Am. Sociol. Rev.
55:366–84
Homans GC. 1950. The Human Group. New
York: Harcourt Brace Jovanovich
Hopcroft RL. 1998. The importance of the lo-
cal rural institutions and economic change
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?NEW INSTITUTIONALISM 545
in preindustrial England. In The New
Institutionalism in Sociology, ed. MC Brin-
ton, V Nee, pp. 277–304. New York: Russell
Sage Found.
Ingram P. 1998. Changing the rules: interests,
organizations, and institutional change in the
U.S. hospitality industry. In The New Insti-
tutionalism in Sociology, ed. MC Brinton, V
Nee, pp. 258–76. New York: Russell Sage
Found.
Ingram P, Inman C. 1996. Institutions, inter-
group rivalry, and the evolution of hotel pop-
ulations around Niagara Falls. Admin. Sci. Q.
41:629–58
Ingram P, Simons T. 2000. State formation,
ideological competition, and the ecology of
Israeli workers’ cooperatives, 1920–1992.
Admin. Sci. Q. 45(1):25–53
Johnson RN, Libecap GD. 1982. Contracting
problems and regulation: the case of the fish-
ery. Am. Econ. Rev. 72:1005–22
Knight J. 1992. Institutions and Social Conflict.
New York: Cambridge Univ. Press
Knight J, Ensminger J. 1998. Conflict over
changingsocial norms: bargaining, ideology,
and enforcement. In The New Institutional-
ism in Sociology, ed. MC Brinton, V Nee,
pp. 105–26. New York: Russell Sage Found.
Macaulay S. 1963. Non-contractual relations
in business: a preliminary study.Am. Sociol.
Rev. 28:55–67
Maurer N, Haber S. 1999. Institutional change
and productivity growth: access to credit and
liquidity constraints in the Mexican cotton
textile manufacture, 1878–1913. Presented
at Conf. Inst. Mark. Compar.-Hist. Perspect.,
Palo Alto, CA
McFaul M. 1995. State power, institutional
change, and the politics of privatization in
Russia. World Polit. 47:210–43
Meyer JW, Boli J, Thomas GM, Ramirez FO.
1997.Worldsociety and the nation-state. Am.
J. Sociol. 103:144–81
Meyer JW, Rowan B. 1977. Institutionalized
organizations: formal structure as myth and
ceremony. Am. J. Sociol. 83:340–63
Nee V. 1992. Organizational dynamics of mar-
ket transitions: hybrid forms, property rights,
and mixed economy in China. Admin. Sci. Q.
37:1–27
Nee V. 1998. Sources of the new institutional-
ism. In The New Institutionalism in Sociol-
ogy, ed. MC Brinton, V Nee, pp. 1–16. New
York: Russell Sage Found.
Nee V, Ingram P. 1998. Embeddedness and
beyond: institutions, exchange, and social
structure. In The New Institutionalism in So-
ciology, ed. MC Brinton, V Nee, pp. 19–45.
New York: Russell Sage Found.
North DC. 1981. Structure and Change in Eco-
nomic History. New York: Norton
North DC. 1990. Institutions, Institutional
Change and Economic Performance. New
York: Cambridge Univ. Press
North DC. 1993. Institutions and credible com-
mitment. J. Inst. Theor. Econ. 149:11–23
North DC, Thomas RP. 1973. The Rise of the
Western World. London: Cambridge Univ.
Press
North DC, Weingast BR. 1989. Constitutions
and credible commitments: the evolution of
the institutions of public choice in 17th cen-
tury England. J. Econ. Hist. 49:803–32
OstromE. 1990. Governing the Commons. New
York: Cambridge Univ. Press
Poole KT, Rosenthal H. 1996. Congress: A
Political-Economic History of Roll Call Vot-
ing. New York: Oxford Univ. Press
Powell WW, DiMaggio PJ, eds. 1991. The New
Institutionalism in Organizational Analysis.
Chicago: Univ. Chicago Press
Rao H. 1998. Caveat emptor: the constructions
of nonprofit consumer watchdog organiza-
tions. Am. J. Sociol. 103:912–61
Reid JP. 1980. Law for the Elephant. San
Marino, CA: Huntingt. Libr.
Rueschemeyer D, Evans PB. 1985. The state
and economic transformation: toward an
analysis of the conditions underlying ef-
fective intervention. In Bringing the State
Back In, ed. PB Evans, D Rueschemeyer,
T Skocpol, pp. 44–77. London: Cambridge
Univ. Press
Schelling TC. 1960. The Strategy of Conflict.
P1: FQK/FQC/FGD P2: FLW
June 3, 2000 16:55 Annual Reviews AR105-22
?
546 INGRAM ¥CLAY
Cambridge, MA: Harvard Univ. Press
Scott WR. 1995. Institutions and Organiza-
tions. Beverly Hills, CA: Sage
Simon HA. 1957. Models of Man. New York:
Wiley & Sons
Simons T, Ingram P. 1997. Organization and
ideology: kibbutzim and hired labor, 1951–
1965. Admin. Sci. Q. 42:784–813
Spicer A. 1997. Building markets from scratch:
institutions and agency in the Russian mass
privatization program. Presented at Annu.
Meet. Acad. Manage., 57th, Boston
Stark D, Bruszt L. 1998. Postsocialist Path-
ways. New York: Cambridge Univ. Press
StinchcombeAL.1965.Socialstructureand or-
ganizations. In Handbook of Organizations,
ed. JG March, pp. 142–93. Chicago: Rand
McNally
Stinchcombe AL. 1997. On the virtues of the
oldinstitutionalism. Annu. Rev.Sociol. 23:1–
18
Stone A, Levy B, Paredes R. 1996. Public in-
stitutions and private transactions: a compar-
ative analysis of the legal and regulatory en-
vironment for business transactions in Brazil
and Chile. In Empirical Studies in Institu-
tional Change, ed. LJ Alston, T Eggertsson,
DC North, pp. 95–128. New York: Cam-
bridge Univ. Press
Ullmann-Margalit E. 1977. The Emergence of
Norms. Oxford, UK: Clarendon
UmbeckJR.1981.ATheoryofPropertyRights,
withApplication to theCalifornia Gold Rush.
Ames, IA: Iowa State Univ. Press
Uzzi B. 1996. The sources and consequences
of embeddedness for the economic perfor-
mance of organizations: the network effect.
Am. Sociol. Rev. 61:674–98
Uzzi B. 1999. Embeddedness in the making
of financial capital: how social relations and
networks benefit firms seeking financing.
Am. Sociol. Rev. 64:481–505
Veblen TB. 1899. The Theory of the Leisure
Class: An Economic Study of Institutions.
New York: Macmillan
Wade JB, Swaminathan A, Saxon MS. 1998.
Normative and resource flow consequences
of local regulations in the American brewing
industry,1845–1918. Admin. Sci. Q. 43:905–
35
Weingast BR. 1993. Constitutions as gover-
nance structures: the political foundations
of secure markets. J. Inst. Theor. Econ.
149:286–311
Wholey DR, Christianson JB, Sanchez SM.
1992. Organizational size and failure among
health maintenance organizations. Am. So-
ciol. Rev. 57:829–42
Wiggins SN, Libecap GD. 1985. Oil field uni-
tization: contractual failure in the presence
of imperfect information. Am. Econ. Rev.
75:368–85
Williamson OE. 1975. Markets and Hierar-
chies: Analysis and Antitrust Implications.
New York: Free Press
Williamson OE. 1985. The Economic Institu-
tions of Capitalism. New York: Free Press.
Williamson OE. 1994. Transaction cost eco-
nomics and organization theory. In The
Handbook of Economic Sociology, ed. NJ
Smelser, R Swedberg, pp. 77–107. Prince-
ton, NJ: Princeton Univ. Press
