November 20, 2002
DRAFT
Commitment and Exchange: The Emergence of Trust Networks under
Uncertainty
By Karen S. Cook, Stanford University, Eric R.W. Rice, U.C.L.A., and Alexandra
Gerbasi, Stanford University
Introduction
When uncertainty and risk are associated with economic and social transactions
relatively closed trust networks often emerge to facilitate various types of informal
cooperation. In Russia, for example, “blat” was an extensive form of informal exchange that
emerged to provide scarce resources and services or favors (Ledeneva, Lomnitz etc.). Similar
exchange systems have been documented in a number of different contexts. Moving from a
social system in which the dominant mode of interaction is closed groups or networks to
more open networks such as those required to support the transition to a market economy and
democratic institutions may be difficult for a number of reasons. We draw on research on
exchange networks to explore some of these reasons. Our main claims are that uncertainty
and risk (such as that created by corruption and dishonesty) lead to commitment and the
formation of trust networks that tend toward closure. One disadvantage of such closed
networks of exchange is that they tend to limit access to opportunities outside the network.
Those in power in networks of exchange are less likely to form commitments and have more
reason to break out of these networks when they incur opportunity costs. Power asymmetries
impede commitment and under some circumstances reduce trust. Reputation systems and
1
third-party mediators (or guarantors) emerge under certain conditions to facilitate the move
from closed trust networks, often involving only family members and close associates, to
more open networks such as those required for the operation of market economies (see
Radaev, 2002, on the move from affect - based trust to reputation - based trust networks)
involving transactions with strangers. We draw on research (primarily experimental) on
exchange networks to offer insights into these network processes. We include discussion of
various examples of these processes based on evidence from survey and field research in a
wide variety of contexts, especially Eastern Europe.
Uncertainty, Trust and Forms of Exchange
Under conditions of uncertainty, trust networks are created to provide a more secure
transaction environment, especially when there is no reliable contract law or enforcement or
when the issues in exchange cannot be well handled with explicit contracts. The uncertainty
can arise from a number of sources and lead to different types of network solutions. Under
high uncertainty and high risk, transactions are likely to occur primarily among parties who
know each other well and form relatively closed associations or groups (e.g. families or
informal membership associations), in which the group boundaries are clear and membership
is easily determined (e.g. it is easy to detect who is in and who is out of the group). Insiders
are included in trades and outsiders are excluded. Optimum conditions of trade rarely exist
under these conditions. Corruption and a high potential for exploitation in the larger society
may lead to such closed-association systems of trade. As will be argued below, a major
difficulty with a quasi-closed system of trade is that it restricts the market for both “buyers”
and “sellers.”
2
Under lower threat of exploitation and levels of corruption closed-association trade
may give way to the formation of rudimentary reputation systems that enable individuals to
trade across membership boundaries and to establish indirect network ties to facilitate a
broader range of exchanges. These systems may emerge as transitional phases in the move
toward more open networks of exchange. We examine the emergence of trust networks under
varying conditions, but first we review the research on commitment between exchange
partners under uncertainty and the emergence of trust networks.
Within exchange theory (see Molm and Cook, 1995) a variety of types of exchange
have been investigated. The most common form of exchange is dyadic, restricted exchange in
which two parties engage in the exchange of valued goods or services for mutual benefit (see
also Ekeh 1974 and Blau 1964). The term “restricted” refers to the fact that the exchange is
isolated to the dyad. Emerson (1972) expanded the work on restricted exchange to focus on
the linkages between connected sets of exchange relations. Two exchange relations for
Emerson were viewed as connected to the extent that exchange in one relation affected the
frequency or level of exchange in the other relation. Two exchange relations A:B and B:C are
positively connected at B in an A-B-C network if exchange in one relation increases the
probability or frequency of exchange in the other relation. The connection is negative if
exchange in one relation decreases the frequency or probability of exchange in the other
relation. (Logically, the third category is a null relation, which is equivalent to no connection.
In Emerson’s work null relations are not important since they imply that the relations are not
connected and thus do not form a network link). Connected exchange relations form networks
of exchange, which may include different types of connections (i.e. mixed networks include
both positive and negative connections).
3
Dyadic exchange can be either negotiated or non-negotiated (see Molm and Cook,
1995). For Molm non-negotiated exchange involves the reciprocal giving of goods or
services in the hope that reciprocity will apply and the recipient will return the favor. (See
also Blau on the distinction between social and economic exchange). A different version of
reciprocal exchange is generalized exchange in which the reciprocity is generalized and not
particularized (i.e. between two connected actors). In generalized exchange goods or services
are transferred to one party in the hope that some other party may reciprocate, but the one
who reciprocates is not the person who is the direct recipient of the goods or services
(Takahashi 2000, Takahashi and Yamagishi 1999). A standard example in anthropology is
the Kula Ring (Malinowski).
Recent research by Molm, Takahashi and Peterson (2000) reports findings
differentiating two forms of exchange: negotiated and reciprocal dyadic exchange. Classical
exchange theorists (such as Blau 1964) proposed that trust is more likely to develop between
partners when exchange occurs without explicit negotiations or binding agreements (cf.
Macauly 1963). Under uncertainty and risk, exchange partners have greater opportunity to
demonstrate their trustworthiness by acts of reciprocal giving in the absence of negotiated
agreements. Molm et.al. (2000: 1398) demonstrate that reciprocal exchanges produce higher
levels of trust and stronger feelings of affective attachment and/or commitment than do
negotiated exchanges. The initial act of giving in a reciprocal exchange acts as a “signal” of
the actor’s trustworthiness to the recipient and creates the foundation for reciprocity of
exchange and eventually trust. The behavioral commitments that form also reduce the
inequality in the exchange and affect is, to some extent, dependent upon this reduction in
inequality as well as the signaling of trustworthiness. In an interesting paradox Molm claims
4
that these findings indicate that the mechanisms that were created to reduce risk in
transactions (negotiations and strictly binding agreements) have the “unintended consequence
of reducing trust in the relationship” since trust is not required if the agreements are binding.
This conclusion, however, is based on the fact that in an experimental setting the
experimenter serves as the “contract enforcer” and subjects are not allowed to renege on their
commitments. (For an experimental study in which this is not the case see Rice 2002,
discussed below.)
Under environmental uncertainty and conditions of high potential risk (as could be
created by corruption and widespread dishonesty) it can be argued that exchange systems are
more likely to be set up as negotiated exchanges than as reciprocal exchanges which require
more confidence and trust. In some ways, Molm’s findings regarding differences in perceived
trustworthiness of one’s partners under negotiated and reciprocal exchange regimes, suggest a
“Catch 22” exists. Molm, Takashi and Peterson report that one’s most frequent exchange
partner is rated as trustworthier under reciprocal exchange than under negotiated exchange.
The investigators reason that reciprocal exchange creates more uncertainty (since it is not
negotiated) and thus when it is successful and one’s initial “gift” of a service or valuable
resource (to initiate an exchange) is reciprocated, then a more reliable signal has been offered
that one is a trustworthy partner. Negotiated exchange does not provide an opportunity for
such a signal (unless the exchange can be reneged on as in Rice’s {2002} experimental
condition, referred to below as the “non-binding” exchange condition).
Commitment to Exchange Relations: Causes and Consequences
5
Commitment tends to occur more readily among power equals than among power
unequals in exchange networks (Cook and Emerson 1978). This fact has been supported by
recent research indicating that positive and frequent exchange among power equals creates
commitment and positive emotions toward the exchange relation (see Lawler and Yoon,
1998). The focus of most of the recent research within social exchange theory on the concept
of commitment, however, has linked commitment to social uncertainty. Cook and Emerson
(1984) define the degree of uncertainty as “the subjective probability of concluding a
satisfactory transaction with any partner” (Cook and Emerson 1984: 13). They found that
greater uncertainty led to higher levels of commitment with particular exchange partners
within an exchange opportunity structure. Commitment between exchange partners reduces
the uncertainty of finding a partner for trade and insures a higher frequency of exchange.
Commitment as defined by Cook and Emerson (1978) is behavioral. It refers to the decision
to continue to exchange with a particular partner (or set of partners in larger networks) to the
exclusion of alternatives that might be more profitable. Behavioral commitment in an
exchange opportunity structure creates relatively enduring exchange relations (rather than
spot markets). While affective commitment might emerge as a result of such ongoing
exchange it is treated as a separate factor to be explained. In this paper we use the term
commitment to imply behavioral commitment (not affective commitment). Behavioral
commitment implies an ongoing exchange relation that typically provides information about
the relative trustworthiness of the partners to the exchange. High levels of trustworthiness are
assumed to facilitate the emergence of a trust relation. Trust networks are connected
exchange relations formed in this manner. While commitment is expected to emerge under
6
conditions of uncertainty to provide the security for exchange in the future, such networks of
exchange are most likely to become trust networks under conditions of risk.
Recently research within exchange theory has conceptualized social uncertainty as the
probability of suffering from acts of opportunism imposed by one’s exchange partners
(Kollock 1994; Rice 2002; Yamagishi et al. 1998). Opportunism creates a different source of
uncertainty than the concern over the location of an exchange partner, the primary source of
uncertainty in Cook and Emerson (1978), involving the risk of exploitation. (Exploitation is
hard to block in networks of exchange unless the networks become closed and those who
cheat can be excluded from further interactions with those in the network). Social
uncertainty, created by the risk of exploitation, has also been shown to promote commitment.
Kollock (1994), Rice (2002) and Yamagishi et al. (1998) examined behavioral commitments
in environments that allow actors to cheat one another in their exchanges. Securing
commitments to specific relations is often the most viable solution to the problem of
uncertainty in these environments. If actors within a given opportunity structure prove
themselves to be trustworthy exchange partners, continued exchange with those partners
provides a safe haven from opportunistic exchangers. Such commitments, however, have the
drawback of incurring sizable opportunity costs in the form of exchange opportunities
foregone in favor of the relative safety of committed relations (Yamagishi and Yamagishi,
1994). This is one of the main dilemmas facing individuals in settings in which untrustworthy
behavior is common or where opportunism and corruption of some sort is the norm. And if
the equilibrium in the society is relatively closed trust networks, it may be difficult to break
out of this pattern of exchange to generate a more open network of exchange among relative
strangers. These trust networks among close kin or ethnic group members may actually serve
7
to reduce trust in outsiders or make it difficult to create since transactions occur rarely across
group boundaries (Rose-Ackerman 2000:536). (Later we discuss the possible effects of
reputation systems for this transition from closed to more open networks of trade and mutual
benefit.)
In Kollock’s (1994) initial study connecting opportunistic uncertainty and
commitment, actors exchanged in two different environments. In one environment (low
uncertainty) the true value of the goods being exchanged was known, while in the other (high
uncertainty) environment the true value of goods was withheld until the end of the
negotiations. He found that actors had a greater tendency to form commitments in the higher
uncertainty environment. Moreover, actors were willing to forgo potentially more profitable
exchanges with untested partners in favor of continuing to transact with known partners who
have demonstrated their trustworthiness in previous transactions (i.e. they did not
misrepresent the value of their goods).
Yamagishi, Cook and Watabe (1998) further explored the connections between
uncertainty and commitment, deviating from Kollock’s experimental design but coming to
similar conclusions. In their experiment, actors are faced with the decision of remaining with
a given partner or entering a pool of unknown potential partners. They employed several
modifications of this basic design, but in each instance the expected value of exchange
outside the existing relation was higher than the returns from the current relation. They found
that actors were willing to incur sizeable opportunity costs to reduce the risks associated with
opportunism. Moreover, they found that uncertainty in either the form of an uncertain
probability of loss or an unknown size of loss was able to promote commitments between
exchange partners.
8
In both the Kollock (1994) and Yamagishi et al. (1998) studies, exchange occurs
among actors in environments that allow for the potential for opportunism, but in which
actors are guaranteed finding an exchange partner on every round. In Rice’s (2002)
experiment actors exchange in two different environments: one that allows actors to renege
on their negotiated exchange rates (high uncertainty) and one where negotiations are binding
(low uncertainty). Exchange, however, also occurs within two network structures that vary in
degree of power inequality - a complete network in which all actors can always find a
transaction partner, and a T-shaped network, where one actor connecting three others at the
center has greater access to alternatives for trade than those on the periphery on the network.
Uncertainty promoted commitment in the complete network (power-balanced), but not in the
T-shaped network (a monopoly structure with maximum power inequality). Commitments,
he argued, are more viable solutions to uncertainty in networks in which power differences
are minimal. In networks that include a large power difference among the actors, the
structural pull away from commitment to explore alternatives is sufficiently intense as to
undermine the propensity to form commitments. Power is determined in such structures by
access to alternatives. Giving up alternatives reduces power.
Whereas Kollock and Yamagishi and his collaborators suggested that actors would
incur opportunity costs to avoid potentially opportunistic partners, Rice (2002) suggests that
such tendencies can be muted by power inequality in the network structure. This finding is
consistent with a number of studies in which the effects of power inequalities on the
formation of commitments have been investigated (see also Cook and Emerson, 1978, 1984;
Lawler and Yoon, 1996; Molm, Takahasi and Petersen, 2000). It is generally the case that
behavioral commitment is inversely related to the level of the power inequality in the
9
network. Thus, commitment is higher among power equals than among power unequals, all
other things being equal. Commitment is more common in horizontal than in vertical
relationships.
Rice (2002) explores other effects of commitment in exchange networks. In particular
he investigates how commitment affects the distribution of resources across relations and
within networks as a whole. He argues that commitments, when they occur, will reduce the
use of power in unequal power networks, resulting in a more egalitarian distribution of
resources across positions in a network. In networks where power between actors is unequal,
power-advantaged actors have relatively better opportunities for exchange than their powerdisadvantaged partners. These superior alternatives are the basis of power-advantaged actor’s
power. If, as uncertainty increases, power-advantaged actors form commitments with powerdisadvantaged actors, they erode the very base of their power. Forming commitments entails
ignoring potential opportunities. Alternative relations are the basis of structural power and as
these relations atrophy, the use of power and the unequal distribution of resources will be
reduced. For this reason increasing power inequality tends to lower the propensity for
commitment even under uncertainty. With increased risk of loss, however, even poweradvantaged actors seek committed partners for exchange.
Research results on exchange under social uncertainty thus indicate a strong tendency
for actors to incur opportunity costs by forming commitments to achieve the relative safety or
certainty of ongoing exchange with proven trustworthy partners (Kollock 1994; Rice 2002;
Yamagishi et al 1998). In addition to these opportunity costs, Rice (2002) argues that
commitments may also have unintended negative consequences at the macro level of
exchange. Actors tend to invest less heavily in their exchange relations under higher levels of
10
uncertainty. Moreover, acts of defection in exchange, while producing individual gain, result
in a collective loss, an outcome common in prisoner’s dilemma games. Both processes
reduce the overall collective gains to exchange in the network as a whole. So while there is a
socially positive aspect to uncertainty, in so far as commitments may increase feelings of
solidarity (e.g. Lawler and Yoon 1998) and resources are exchanged more equally across
relations (Rice 2002), there is the attendant drawback of reduced aggregate levels of
exchange productivity and efficiency.
As Powell and Smith-Doerr (1994:392) put it, “ties that bind may also turn into ties
that blind. When repeat trading becomes extensive it can turn inward, leading to parochialism
or inertia.” Marsden (1983) points out that networks may restrict access, in part through
brokerage arrangements and in part because they structure the flow of goods, resources, and
information, sometimes in less than optimal ways. Two examples provided by Powell and
Smith-Doerr (1994:392) of the potential negative effects of network arrangements involving
repeat trading include Powell’s (1985) study in which the “ossification of an editor’s
networks” is defined as the major factor in the decline of the list of products available from
the publishing house. The second example they provide is of the embeddedness of an
industry locked into a particular production network that created inertia and made the Swiss
watch making industry vulnerable and not responsive to the digital technology revolution
(Glasmeier 1991, as cited in Powell and Smith-Doerr 1994).
In another example provided by Henry Farrell, in the Italian packaging tool industry,
concentration in the industry has limited the contact between suppliers and end-producers in
ways that are counterproductive. Concentration limits the alternatives for the suppliers and
they become vulnerable to exploitation by the end-producers. A committed relation forms
11
between one particular supplier and end-producer further constraining the market and placing
the financial security of the supplier under the control of one specific end-producer. In a
contrasting case study of French machine producers, Farrell indicates how long-term
commitments are avoided as the end-producers strive to keep suppliers from depending too
heavily on them. Essentially the French producers, though powerful in the network, refrain
from long-term commitments and maintain the conditions that support market competition
for access to suppliers and, hence, responsiveness to changing economic factors.
Commitments can also have unintended consequences. As Mizruchi and Stearns
(2001) discover in their analysis of the use of social networks to complete deal between
commercial bankers and their corporate customers. They hypothesize that high uncertainty
leads the bankers to rely on colleagues with whom they have strong ties for advice and for
support of their deals. Their findings reveal that the tendency of bankers to rely on their
approval networks and on those they trust actually lead them to be less successful in closing
deals. This lower success rate in closing deals is viewed as the unintended consequence of
“purposive action” by Mizruchi and Stearns. Uncertainty, they argue (2001:667), creates
“conditions that trigger a desire for the familiar, and bankers respond to this by turning to
those with whom they are close.” The bankers in high risk environments turn to those they
trust even when seeking advice from a broader range of contacts (perhaps less trusted) might
make them more successful in the long run (especially with complex deals).
Under some conditions, then, networks can constrain the ability of the actors involved
to adapt to rapid economic or environmental change. These conditions have not yet been
spelled out in any systematic way, however. Our paper is an exploration of some of these
12
conditions in environments that are in flux, politically, socially and environmentally
including the dramatic cases of economic transition in Eastern Europe and elsewhere.
Commitment and the Formation of Trust Networks under Uncertainty
Exchanges are often “embedded” in networks of ongoing social relations. Uzzi
(1996) argues that “embeddedness” has profound behavioral consequences, affecting the
shape of exchange relations and the success of economic ventures. “A key behavioral
consequence of embeddedness is that it becomes separate from the narrow economic goals
that originally constituted the exchange and generates outcomes that are independent of the
narrow economic interest of the relationship” (Uzzi 1996: p.681). In related research Lawler
and Yoon (1996; 1998) and Lawler, Yoon and Thye (2000) argue that as exchange relations
emerge actors develop feelings of relational cohesion directed toward the ongoing exchange
relation. These feelings of cohesion result in a wide variety of behaviors which extend
beyond the “economic” interests of the relationship, such as gift-giving, forming new joint
ventures across old ties, and remaining in a relationship despite the presence of new,
potentially more profitable partnerships. While these are clearly positive aspects of these
commitment relations, there is also a downside when these relations become “locked-in” and
limit the range of exchange relations or the exploration of new opportunities.
A number of studies have begun to document the relationship between uncertainty
and the emergence of trust-based networks. Guseva and Rona-Tas (2001) compare the credit
card markets of post-Soviet Russia and the United States. They are concerned with how
credit lenders in each country manage the uncertainties of lending. In the United States, they
argue, credit lending is a highly rationalized process that converts the uncertainty of
13
defaulting debtors to manageable risk. Lenders take advantage of highly routinized systems
of scoring potential debtors, through the use of credit histories and other easily accessed
personal information. This system allows creditors in the United States to be open to any
individuals who meet these impersonal criteria.
In Russia, creditors must reduce uncertainties through personal ties and commitments.
Defaulting is an enormous problem in Russia, aggravated by the fact that credit information
such as that used by American lenders has, until quite recently, been unavailable. To
overcome these uncertainties Russian banks seeking to establish credit card markets must use
and stretch existing personal ties. Loan officers make idiosyncratic decisions about potential
debtors, based largely on connections to the bank, or known customers of the bank. In this
way defaulting debtors cannot easily disappear, as they can be tracked through these ties.
Viewed through the lens of recent theorizing on the connections between uncertainty and
commitments, these different strategies seem quite reasonable. As discussed earlier,
exchange theorists have repeatedly shown that as uncertainty increases, commitment to
specific relations likewise increases (Kollock 1994; Yamagishi et al. 1998; Cook and
Emerson 1984; Rice 2002). In the case of credit card markets, it is clear that the United
States presents an environment of relatively low uncertainty, compared to the high-levels of
uncertainty present in Russia. Exchange theory argues therefore that commitments will be
greater in Russia, which is exactly the case. Lending is facilitated by existing commitments
to the banks or the bank’s known customers. While such theoretical confluence is interesting,
it is in generating new insights that one can see the value of examining this situation through
the lens of exchange theory.
14
Rice (2002) argues that network structure will intervene in the process of commitment
formation. This insight suggests that sociologists ought to ask how different shaped networks
of potential debtors and lenders in Russia affect the use of commitments to procure credit?
Rice also argues that uncertainty, while promoting commitment, simultaneously reduces the
overall level of exchange in networks. This is another outcome observed in the Russian credit
card market, but one that is largely ignored by Guseva and Rona-Tas (2001). It is this aspect
of the problem that is addressed to some extent by Radaev (2002) on the emergence of
reputation systems in Russia. Finally, Yamagishi and his collaborators (Yamagishi et al.
1998) argue that uncertainty can stem from either the probability of loss or the size of loss.
Another question that should be raised in this context is how the size of loss, and not just the
potential for loss, relates to the behaviors observed in the Russian versus the American credit
card markets.
Exchange theory tends to focus on commitments as an outcome, not as a social
mechanism. In the case of the Russian credit card market, existing commitments provide a
mechanism through which network structures are expanded and changed. This raises the
issue of how interpersonal commitments may in turn create opportunities for network
expansion and/or reduction. Russian banks, for example, expand their trust networks by
issuing credit cards to families and friends of top bank executives (see also Ledeneva 1998).
As Guseva and Rona-Tas (2001:638) note: “Here the borrower-creditor relationship is
intermingled with close social bonds that serve as an additional guarantee and a channel of
information.” The social bond serves as a “bond” to reduce the risk of unrepaid credit.
Despite the fact that trust networks in which credit can be extended allow economic
transactions beyond direct exchanges, there is a limit to the extent to which such networks
15
will allow for movement to free trade. In fact, they may serve to hinder the development of
institutions that might serve as the basis for free trade (i.e. trade among strangers – what
Guseva and Rona-Tas view as the U.S. credit market).
Another way in which trust networks can be expanded to enlarge the number of those
in the market, according to Guseva and Rona-Tas, is to stretch the network by extending
credit to those indirectly tied to one another. “Trust is transitive,” they argue. But, because
this extends the risk involved, it is not used as a strategy beyond one-degree of separation. A
bank employee reveals the unwritten rule of credit extension: “there should not be more than
one person separating a bank official from an applicant” (Guseva and Rona-Tas 2001:639).
In the end person-to-person interviews are most often used in Russia to determine whether to
grant credit to an applicant. This requires the interviewer to develop the “art” of assessment
of the trustworthiness of the other. These security officers who grant or deny credit are
described by Guseva and Rona-Tas (2001:639) as lie detectors: “We have to stare the
applicants intently in the eyes, trying to guess whether they are telling the truth and whether
they should be issued a card.” The networks of trust in which the applicants are embedded are
used by the banks for security as well as for information should the applicant default. In the
U.S. this is done through the provision of the names of “credit references” on the application.
Applicants in the U.S. usually list a friend, co-worker or family member.
In his study of emerging markets for non-state businesses in Russia, Radaev (2002)
investigates the mechanisms and institutional arrangements that help actors cope with the
uncertainty and opportunism common in such an environment. Two features of the situation
are significant. Under uncertainty actors turn to interpersonal ties involving trust and greater
16
certainty to produce some security in the context of high levels of opportunism. This is the
behavior that is documented also by Guseva and Rona-Tas (2001) discussed above.
In documenting the uncertainty of business relations in Russia, Radaev (2002) surveys
indicated how important honesty and trustworthiness were in business partners. This result is
driven by the fact that there are frequent infringements of business contracts creating both
risk and high levels of uncertainty. Half of the respondents admitted that contract
infringements were quite frequent in Russian business in general and a third of the
respondents had had a high level of personal experience with such infringements. This degree
of opportunism creates barriers to the formation of reciprocal trust relations. There is
widespread distrust of newcomers to the market but established reliable partners are viewed
as more trustworthy.
In this climate commitment is clearly the most predictable response to uncertainty as
in the case of Kollock’s (1994) rubber markets and the credit card market discussed by
Guseva and Rona-Tas (2001). Another reason for the uncertainty is that the existing
institutions lack credibility and legitimacy. The courts do not effectively manage dispute
resolution and existing institutions do not secure business contracts. Banks can default, even
large ones assumed to be solvent. To cope with this fact the business community creates
closed business networks with reputation systems that define insiders and outsiders. This
system is based on information obtained from third parties, but more importantly on common
face-to-face meetings between potential partners.
In a completely different environment, McMillan and Woodruff (1999) find a similar
process in the transition from a planned economy to a market-based economy in Vietnam.
Here the market began as a result of small entrepreneurs using their ongoing relationships to
17
secure agreements. These social relations take the place of the non-existent contract law in
what remains of the planned economy. Cheating is easy in Vietnam because of the lack of
contract law and enforcement. According to McMillan, “What is striking about Vietnam is
that the entrepreneur’s incentive not to cheat a contract partner is not that the partner will sue
but that he’ll stop dealing altogether” (Quoted from Stanford Business 2000). Hardin (2002)
identifies this condition as the primary basis for trustworthiness. Since the courts can not be
trusted to resolve legal disputes (see also Montinola 2002) they create their own reputational
system promoted by gossip and meetings in teahouses where information is exchanged about
the credibility of various trading partners. “They try to avoid disputes by checking their
customers’ financial backgrounds and personalities with others who have done business with
them.” These informal exchanges, these investigators argue, create a business ethic that
supports a rudimentary market. Here we see the transition between closed trust networks and
the beginnings of a market economy through ties that are brokered in teahouses. Reputation
systems are essential in the formation of credit information that can be used to extend beyond
the reach of personal (and often closed) networks of exchange.
In a 1993 survey conducted by Radaev the emerging networks of entrepreneurs in
Russia primarily included personal acquaintances (42%), friends and their relatives (23%)
and relatives (17%). This fact reflects the reality in the credit card market in Russia (Guseva
and Rona-Tas 2001). Only a small percentage (11%) of the business contacts in 1993 were
new or relatively new acquaintances. More recently, however, there is a move away from
affect-based relations and trust to reputation-based trust as the networks formed purely on the
basis of acquaintance, kin ties, or friendship have tended to fall apart due to their inefficiency.
The relatively closed business networks that have emerged to replace the older “familial” and
18
friendship ties provide better information about the trustworthiness of the partners and their
competence. Within exchange theory the formation of commitment under uncertainty and
trust networks (see Cook and Hardin 2001) in the face of uncertainty provide theoretical
support for the evidence provided by Radaev (2002) and others on the recent emergence of
business networks in Russia. This argument is also consistent with Rice’s (2002) argument
that commitments can have negative aggregate level consequences for productivity and
efficiency in exchange systems.
Research in the exchange theory tradition on topics such as trust, strategic action,
commitment and reputational networks all have potential applications in the analysis of the
emergence of exchange networks in countries with transitional economies as well as in other
types of economies as evidenced by the work of many economic sociologists (e.g. Uzzi,
Granovetter, etc.). Moving from closed groups to more open networks of trade mirrors some
of the processes identified by Emerson (1972) as important for study from an exchange
perspective contrasting group-level exchange systems (productive exchange in corporate
groups) with network-level exchange. In addition, the return to the study of the significant
differences between social processes (e.g. power, justice and commitment) involved in
different types of exchange - negotiated, reciprocal and generalized exchange (Molm 1988;
Molm1990; Molm 1994, etc.) - has the potential to provide new insights into a variety of
emergent forms of exchange under different circumstances. For example, under uncertainty
negotiated, binding exchange may be more likely to emerge before reciprocal (most often,
non-binding) exchange will flourish, in part because it involves greater degrees of
uncertainty. Reciprocal exchange, as Molm and her coauthors have documented, generally
19
requires more trust since the terms of exchange are not simultaneously negotiated and
opportunism is possible (Molm et al 1999; Molm et al 2000).
Internet Trade, Markets and Reputation Systems
Yamagishi (forthcoming) demonstrates that the role of reputation systems varies
depending upon whether the social system is closed or open. In closed communities
reputation systems can be effective because negative reputations force exclusion (see also
Cook and Hardin, 2001). In more open societies negative reputation systems are less effective
because they are limited in the extent to which they are transferred to all those in the network
(or social system). Information flows only across those actors that are linked directly or
indirectly. In addition, actors can alter their identities in ways that make it easier to reenter
trade networks without being recognized as having had a negative reputation. The focus of
Yamagishi’s experimental study is an Internet trading network in which both the level of
honesty in the trades and the price can be tracked. Positive - reputation systems are more
effective than negative - reputation systems in the open networks. Actors are rated on their
honesty and thus accumulate reputation points that are published in the network during each
transaction.
Reputation systems are investigated by Yamagishi (2002, unpublished) as solutions to
the “lemons” problem in Internet markets, that is, the tendency for the goods on the market to
be low quality especially when there is asymmetric information. Typically, the lemons
market (Akerloff) emerges when the buyers have much less valid or reliable information
about the quality of the goods on the market than do the sellers (as is often the case in the
used car market). Buyers must rely on the reputation of the sellers to determine how much
20
confidence they have in the seller (i.e. how trustworthy the seller is). In this world of trade
Yamagishi demonstrates that the lemons problem is exacerbated when actors can change their
identities and reenter the market with a new identity (thus canceling their negative
reputation). They also demonstrate that positive reputations affect behavior differently since
actors develop an investment in their reputations and want to protect them. They lose this
positive reputation and their investment in it if they alter their identities. “The negative
reputation system that is designed to illuminate dishonest traders is particularly vulnerable to
identity changes, whereas the positive reputation system designed to illuminate honest traders
is not so vulnerable to identity changes”. The results concerning the quality of goods
produced under the different reputation systems (with and without the possibility of changing
identities in the exchange network) are presented in Figure 1 below from Yamagishi
(forthcoming). The findings replicate those for the “lemons” market in terms of the low
quality of the goods (Experiment 1 – Control Condition) on the market without reputations,
but clearly demonstrate the superior role of positive reputations systems when identities can
be easily changes as would be the case in Internet trading. (Compare the two conditions in
Experiment 3 – negative and positive reputation conditions.)
Figure 1 About Here
The negative reputation system, however, as Yamagishi suggests, may work only
under the condition that someone who develops a negative reputation can be excluded from
trade effectively by the group acting as a whole. This requires a closed network (see also
Greif for a discussion of the difference between closed associations and open trade networks
among the Maghribi). Cook and Hardin (2001) also suggest that norms of exclusion work
21
only in groups and are not effective in open networks. Norms of exclusion function to
eliminate those who have “cheated” from the system of trade (Hardin 1995, 2002, chapter 5).
Kollock (1999) views the auction houses that have been created for trade on the
Internet as a laboratory for the study of the management of risk when there is little or no
access to external enforcement mechanisms. He studies the emergence of endogenous
solutions to the problems of risky trade (in cases in which no guarantees, warranties, or other
third-party enforcement mechanisms exist). The reputation systems that emerge to manage
this risk are the primary focus of his work. Yamagishi explores the differences documented
by Kollock between negative and positive reputation systems. Kollock (1999: 111) notes that
“a particularly disturbing strategy for fraud…in a number of online markets is the person who
works hard at establishing a trustworthy reputation, and then sets up a whole series of
significant trades and defaults on all of them, disappearing into a new identity after the fact.”
In a quantitative study of auctions on eBay, Kollock (1999b) provides evidence that at least
for some high-value goods, the price buyers paid was positively and significantly affected by
the seller’s reputation.
In a simulation study of the effects of positive reputation systems (similar in character
to the positive reputations studied experimentally by Yamagishi (forthcoming), Whitmeyer
(2000:196) finds that the effects of different types of positive reputation systems depend to a
large extent on the proportion of cooperators (as opposed to non-cooperators) in the
population. He examines the effects of reputation systems on general confidence gains in the
society. If the proportion of non-cooperators is low, any type of positive reputation system
will work because the non-cooperators will be more easily detected, especially if it is hard to
get a positive reputation. (Whitmeyer varied the degree to which it was easy or hard to get a
22
positive reputation in his simulation.) If the proportion of non-cooperators is high, then a
tough reputation system will mean lost opportunities for cooperation because some potential
cooperators will not be detected.
Uncertainty and the Emergence of Trust Networks for Services
The focus of our argument above is that in the economic arena, especially during a
major economic or political transition; trust networks emerge as a result of commitment
formation between exchange partners under conditions of uncertainty. Uncertainty can result
from the general lack of institutional support (and backing) for contracts and for enforcement
of the terms of trade, but also from corruption or the potential for exploitation that goes
undetected or even unpunished. (Add reference to paper by Montinola on corruption in the
courts in the Philippines.)
Illegal forms of behavior also result in risk and uncertainty and can lead to the
formation of similar trust networks. This is the type of trust built up in closed associations
such as the Mafia (Gambetta 1992), which are highly exclusionary networks. In the Mafia the
network includes only those who are members of the association and strong norms exist that
determine the nature of the acceptable behavior as well as the rules of trade. Only insiders
can participate, outsiders are excluded from the perquisites of “membership.” Where
governments have failed or where general institutions are corrupt then alternative
organizations like the Mafia may take over the economic arena and subsequently make it
difficult to put into place general market mechanisms. In such a situation risk and uncertainty
for those outside the organization may remain high since it is in the interest of the Mafia to
23
mediate economic transactions through its “trust networks.” Creating mechanisms to break
down the control of the Mafia (as the case of Russia makes clear) may be very problematic.
In a less dramatic case, Carol Heimer (2001) studies the trust network that emerged
during the early sixties to help women who needed access to abortion services that were
illegal at the time. Before the 1973 U.S. Supreme Court decision that legalized abortion, a
Chicago-based feminist abortion service referred to as “Jane” provided help to about 11,000
women. The trust networks built by this informal organization also served to protect the
identities of the physicians who provided these services. The risk was high for both parties,
since the women’s health was at stake and the physicians’ license to practice (and much more
recently their lives) were at stake. Under such high stakes it is interesting to investigate trust
networks formed and were maintained. Two features of the situation were critical. First, the
clients were vulnerable. The services they needed could not be obtained on the open market
(or in appropriate organizations) since the service was not legal and normal information
channels did not provide information on the reputations of the medical providers. Second, it
was also in the interest of the practitioners to establish trust relations with the clients since the
stakes were high if they were caught performing illegal abortions. They faced threats to jobs
and reputations in addition to possible prison terms. Of course, the most important feature of
the network was to provide information not only on availability of the service, but also to
provide information on the quality of the service. In this way one could assume that
incompetent practitioners had been eliminated from the network (though this was clearly not
always the case since many women died as a result of receiving an abortion during this time
period). This reputational information was collected and provided as a critical service to those
who needed it by this third party organization, “Jane,” committed ideologically to women’s
24
rights in the area of fertility decisions. Later we discuss other third party forms of mediated
information relevant to trustworthiness and securing the possibility of engaging in
interactions with strangers especially those involving risk and uncertainty.
Under some circumstances trust networks emerge for the provision of a broad range
of services such as healthcare. These networks often fill the void left by institutions (or
simply fill the information void even when institutions work fairly well). In addition to
networks used to obtain care, side payments can be required to get high quality services in
certain situations. Hungary is an example (Kornai, 199x). Rose-Ackerman (2001), among
others, comments on the medical arena as an area in which corruption exists in some of the
post-socialist societies - along with university admissions in Poland and Slovakia and
customs officials in a number of countries (Miller, Grodeland, and Kosheckhkina 1999).
Rose-Ackerman examines general issues of trust and honesty in the post-socialist societies.
The kinds of benefits that are obtained from contacts and bribes, she suggests, is an important
topic for further research given that there are rather large differences across countries in the
perceived incidence of corruption. The system of side-payments for medical care, for
example, is maintained in some countries both by professionals who seek bribes and by
clients who pay them in order to obtain individualized benefits or better service. This leads
to what Rose-Ackerman (2001: 424) calls a “self-sustaining system of corruption.” People
pay the extra payments because others do. What they continue to do is based to a large extent
on what they think others are doing, in a common collective action problem. Public opinion,
Rose-Ackerman notes, is against corruption, but the system is maintained at the level of
individual behavior because individuals benefit from the system even though it may be
collectively irrational (even against collective opinion). Solving this problem will be difficult
25
since it requires that individual behavior be coordinated on a different solution entirely to the
problem of obtaining benefits.
Transitions from Closed to Open Networks: Possible Solutions and Limitations
While we have discussed factors that help explain the emergence of relatively closed
trust networks under uncertainty, a problem for transitional economies is to understand the
nature of the changes that occur when societies shift from one type of economy to another
(i.e. from a planned economy to a market economy). Relatively closed trade associations or
closed trust networks may create problems for the shift to a market economy that requires
open networks to facilitate trades among strangers. For example, consumer credit remains
limited in the Russian economy primarily because there is no good way to secure these
transactions until more general institutions are built up that can provide the kind of
“insurance” that will make transactions with strangers involving loans and extensive credit
possible. Without these institutions, or in the face of weak and unreliable institutions (not to
mention untrustworthy institutions) markets are limited by the reach of actors’ ties to one
another in the society, since trust networks can provide the security for trade that cannot be
offered by institutional safeguards. The social embeddedness of these more limited networks
for the distribution of goods and services serves to facilitate exchange, but restricts the
development of completely open markets of trade. Investment in particular social relations of
assurance (as Yamagishi argues) can work to the detriment of the development of
“generalized trust”. His comparisons between the U.S. and Japan, in this respect, are telling
(see also, Cook et. al. 2002, unpublished; and Yamagishi, Cook and Watabe, 1998).
26
For quite different reasons than in Eastern and Central Europe there are extensive
closed networks of trade in Japan, a society in which what has been called a more
“collectivist” culture exists. Generalized trust is lower in Japan than in the United States,
reflecting to some extent this cultural difference, but also a difference in the standard mode of
exchange. It is only relatively recently (in the past two decades) that credit was widely
available to Japanese citizens or that one could order a plane ticket, for example, through the
mail. Many of these transactions were required to be face-to-face. As in the Russian example
above, merchants and bankers viewed themselves as less likely to be exploited in a face-toface transaction. (Recall that in Russia Radaev reports that credit must be obtained through a
personal interview.) Also, because general trust was low in both cases, it was not considered
wise to use the mails to transfer valuable goods.
Lincoln, Gerlach and Takahashi (1992) indicate in their study of Japanese businesses
how important the “keiretsu criteria” of trust and long-term relationships are for business.
The keiretsu networks are both horizontal and vertical linkages between firms that were
believed to provide Japan an economic advantage during the Japanese economic miracle.
Keiretsu networks were fairly stable sets of regularized exchange relations that reduced the
market to a structured set of trading partnerships (Cook and Hardin 2001). To some extent
such networks reduce transaction costs and risk. As Scharpf (1994:27) puts it, “when such
ongoing relations do exist, the reliability of actors’ expectations, and their trust in each
other’s commitments, may be raised far above the level that would be reasonable even among
well-socialized strangers.” While such networks offer some advantages to limit competition
and provide security of trade under some economic conditions, the downside is that they can
constrain economic growth and restrict opportunity.
27
The use of personal ties for business purposes is also a strong tradition in China where
there are many overlapping networks involving different types of social relations. Kin-based
ties form the main avenue of entry into the economy for Chinese families. Reputation
systems exist in the form of “quanxi contacts” that provide assurance of trustworthiness when
a new deal is to be consummated. The quanxi contact is the intermediary or third party
guarantor. Because such quanxi relations are built up over time, however, and embody longterm commitments (some even transferred intergenerationally), they cannot be built up over
night. In addition, they are embedded in what Hamilton and Fei (1992) call “relational
codes”. Such codes clearly restrict entry into the marketplace.
Other factors can also limit entry into the economic marketplace for societies that do
not yet have a market economy. Brown (2003) describes the vanilla trade in Madagascar in
villages in which distrust prevails primarily among kin. Distrust is so pervasive that it invades
families since there are few non-kin available as trading partners. Closeness of relation does
not seem to modify the effect since even husbands and wives do not often trust each other.
Wives typically maintain the books and keep information from their husbands as a way of
limiting their extra-marital activities. In such an environment the eventual establishment of
trust networks with non-kin for trade would seem unlikely, given that kinship does not even
secure trustworthiness.
One key to the problem of trust and the transition from socialism identified by RoseAckerman (2002) is that there may be a contest between the existing trust networks (based on
what she terms reciprocal trust) that emerged as a coping mechanism under socialism and the
necessary “trust in rules” or confidence in new institutions that will act in a fair and impartial
manner. Personal links may undermine reform efforts, she concludes (2002:559). “Russians
28
and Central and Eastern Europeans established dense networks of informal connections to
cope with the difficulties of life under socialism and some of these practices have continued
as ways of coping with the present (Rose 1999:10; Ledeneva 1998). One question raised by
the transition is whether the legacy of these informal connections is helping or hindering the
process of institutionalizing democracy and the market.”
In an interesting comparison between Russia and Central and Eastern Europe, RoseAckerman (2002:565) reports that initial research indicates that in the Central European
economies more reliance is placed on the courts as arbiters in the case of contract failures.
This makes market deals among strangers possible at an acceptable level of risk. Reciprocal
trust can then emerge among initial strangers in repeat transactions. In contrast, where there is
less confidence in legal enforcement such as in Russia and in the Ukraine (where the courts
are viewed as corrupt and open to bribes) “economic actors are reluctant to deal with
outsiders.” In this case, as Rose-Ackerman points out, “both buyers and sellers are locked
into mutually reinforcing relationships that may limit disputes, but also limit competition and
entry.” The research question that is posed in this analysis is precisely when and under what
conditions does reciprocal trust among close kin and friends undermine or enhance the
establishment of “one-sided trust in the reliability of institutions.” It requires insiders to
interact with outsiders on the basis of standard norms of contractual obligations (ideally
backed by law) as if they were members of the same “group”. Whether this can be
accomplished is an important question and one that goes to the heart of the matter in many
countries undergoing a political/economic and major social transition.
The interesting paradox that has been revealed in some of the recent work on different
forms of exchange and their implications for commitment and trust under uncertainty (e.g.
29
Molm, etc.) is that the very procedures that are put in place to make transactions more
reliable and to increase confidence in the market may undermine the basis for trust. As Molm
concludes (2000:1425), when the shadow of the future is short and exploitation is profitable
then the risk inherent in reciprocal exchange may outweigh the benefits. In such situations
what she calls “assurance structures,” which are mechanisms for creating negotiations that
are binding and enforceable, may actually decrease trust - or at a minimum fail to provide the
conditions for building trust.
In contrast, Molm argues that reciprocal trust relations may have a positive benefit if
they lead to generalized trust relations. This is the central dilemma. They may lead to more
general trust or may simply reinforce trust for those within the network created by reciprocal
exchange relations. Molm claims (2000:1425) that “through numerous experiences with
specific others who behave in a trustworthy manner under conditions of risk, we may come to
expect that others, with whom we have no direct experience, will also be worthy of our trust.”
But this is an empirical question and in order to answer it we should vary the level of
uncertainty and risk involved in the situation. Only then could we draw inferences about
situations like those faced in transitional economies. While it is agreed that an environment in
which generalized trust is high (see Fukuyama, Yamagishi and Yamigishi, etc.) may result in
advantages - since individuals and firms are able to explore new relations and take advantage
of new opportunities in social, business, and political arenas - this depends centrally upon the
nature and level of the risks involved. Only under certain conditions is it likely that the kind
of particularized trust that emerges in trust networks will lead to trust of those outside one’s
direct experience (as well as indirect experience through reputations obtained from
trustworthy contacts within the network) or to assessments of the generalized trustworthiness
30
of strangers. This move is complex and may rely on the kinds of institutions that arise to
manage defaulting.
We have focused on uncertainty and its effects on the emergence of trust networks
and subsequently the transitions from trust networks to open networks of exchange that move
beyond direct personal bonds. We have not dealt explicitly with situations in which the level
of distrust is so high that it is difficult to imagine how one would get to real markets. As
Rose-Ackerman (1999:436) puts it, widespread distrust in institutions, as exists in Russia and
some of the Eastern European countries, leads to a focus on interpersonal distrust. The only
“counterweight here,” she claims, “is the creation of exclusive trusting networks operating
inside or outside the formal institutional framework.” In situations of high levels of distrust,
however, as we have indicated above, the move from these closed trust networks to open
networks of trade may be difficult.
The move to reputation systems may be one of the intermediate steps that might work
to extend the network. The information provided from the closed trust networks might then
be useful in providing reputations that are credible since it could include both positive
information about trustworthiness and negative information about defaulting behavior. This
step might foster the extension of trade beyond the bounds of direct personal ties. A potential
difficulty, however, is that if the distrust of outsiders is intense, the reputational system that
evolves may simply reinforce the divide between those in the trust network and those outside
of it, making any extension of trade across this divide difficult. This implies that beyond the
nature of the reputational system that develops (as Yamagishi argues), the degree of distrust
and its distribution across groups in the society is critical. An interesting aside here is that it
may be precisely in this kind of situation that Putnam’s panacea might work. Associations
31
that crosscut the major cleavages in the society, if they exist (perhaps in the form of sports
teams or other interest based clubs not derived from ethnic group identity), may serve to build
bridges across this divide.
Another key factor is the proportion of potential cooperators in the society and, as
Whitmeyer demonstrates, when that proportion is low any reputational system that helps us
locate the ones who are cooperative will be useful. Investigating the different paths from
closed associations to open networks of trade would be a useful research project in the
various countries now undergoing tremendous economic change. From Hungary and Poland
to South Africa and Vietnam, the study of uncertainty and the emergence of trust networks as
one possible path to open markets (or as a hindrance to the development of open markets) is
an important next step in our research agenda. It will also be important to study
experimentally the features of reputation systems or that may help make open network
trading feasible, especially in environments in which some cheating is likely to occur.
Third parties may be important intermediaries in the move from closed to open
networks for trade and other business transactions. As in the case of Vietnam, discussed
above, teahouses link individuals unknown to each through indirect ties and the transfer of
relevant information about trustworthiness. Or, as in the case of Russia, discussed by Radaev,
rudimentary credit associations are beginning to emerge that facilitate face-to-face meetings
in which reputations for trustworthiness can be transmitted in much the same way as in the
Vietnamese teahouses. These informal modes of cooperation may have the externality of
extending the credit available in segments of the society by providing information about the
trustworthiness of those that can be indirectly linked in networks that extend beyond the trust
networks that generally work on the basis of relatively direct ties.
32
Reputational systems are used to extend credit and make transactions possible that are
unlikely without credible information about trustworthiness. In much the same way that
rudimentary credit associations are emerging in Russia to “certify” customers as trustworthy,
professional associations often emerge to “certify” the competence and trustworthiness of
various professionals so that their clients or consumers (often unable to judge for themselves)
can be assured that they will be treated appropriately. These associations can do more than
certify, they may also regulate those who are members. They can sanction those who do not
live up to the reputation of the profession and even exclude from membership those who
violate the professional ethics or relevant codes of conduct. The full-fledged emergence of
credit associations backed eventually by fiduciary and contract law might take this form in
Russia and Eastern Europe. It would involve third party accreditation and confidence that the
information being provided was accurate. Whether any of these reputational mechanisms
work to “guarantee” trustworthiness and under what conditions they succeed in extending the
reach of trust networks to foster markets needs further empirical investigation. Without this
step in a climate of dishonesty and exploitation trust networks are likely to remain the
dominant mode of exchange. Restricted network exchange, in this context, is one mechanism
for avoiding the risk of entering and re-entering the “market for lemons.”
33
References
Axelrod, Robert. 1984. The Evolution of Cooperation. New York: Basic Books.
Blau, P.M. 1986. Exchange and Power in Social Life, 2nd printing. New Brunswick, NJ:
Transaction Books.
Bearman, Peter. 1997. "Generalized Exchange." American Journal of Sociology 102:13831415.
Brown, Margaret L. 2003. “Compensating for Distrust among Kin.” In R. Hardin (ed.)
Distrust. New York: Russell Sage Foundation Press.
Cook, Karen S., Emerson, Richard M. 1978. "Power, Equity and Commitment in Exchange
Networks." American Sociological Review 43:721-739.
Cook, Karen S., Emerson, Richard M. 1984. "Exchange Networks and the Analysis of
Complex Organizations." Research in the Sociology of Organizations 3:1-30.
Cook, Karen S., Richard Emerson, Mary R. Gillmore, and Toshio Yamagishi. 1983. "The
Distribution of Power in Exchange Networks: Theory and Experimental Results."
American Journal of Sociology 89:275-305.
Cook, Karen S., and Mary R. Gillmore. 1984. "Power, Dependence and Coalitions." Pp. 2758 in Advances in Group Processes, edited by Edward J. Lawler. Greenwich, CT: JAI
Press.
Cook, Karen S., and Russell Hardin. 2001. "Norms of Cooperativeness and Networks of
Trust." Pp. 327-347 in Social Norms, edited by M. Hechter and K-D. Opp. New York:
Russell Sage Foundation.
34
Cook, Karen S., and Toshio Yamagishi. 1992. "Power in Exchange Networks: A PowerDependence Formulation." Social Networks 14:245-265.
Ekeh, Peter. 1974. Social Exchange Theory: The Two Traditions. Cambridge: Harvard
University Press.
Emerson, Richard. 1962. "Power-Dependence Relations." American Sociological Review
27:31-41.
Emerson, Richard. 1972. "Exchange Theory Part I: A Psychological Basis for Social
Exchange." Pp. 38-57 in Sociological Theories in Progress, edited by Joseph Berger,
Morris Zelditch Jr., and B. Anderson. Boston: Houghton Mifflin.
Emerson, Richard. 1972. "Exchange Theory, Part II: Exchange Relations and Networks." Pp.
58-87 in Sociological Theories in Progress, edited by Joseph Berger, Morris Zelditch
Jr., and B. Anderson. Boston: Houghton Mifflin.
Farrell, Henry. (2003). “Trust, Distrust, and Power.” In R. Hardin (ed.), Distrust, New York:
Russell Sage Foundation Press.
Friedkin, Noah E. 1993. "An Expected Value Model of Social Exchange Outcomes." Pp.
163-193 in Advances in Group Processes, edited by Edward J. Lawler. Greenwich,
CT: JAI Press.
Guseva, Alya, and Akos Rona-Tas. 2001. "Uncertainty, Risk, and Trust: Russian and
American Credit Card Markets Compared." American Sociological Review 66:623646.
Hamilton, Gary, and Xiaotong Fei. 1992. From the Soil: The Foundation of Chinese Society.
Berkeley, CA: University of California Press.
35
Heckathorn, Douglas D. 1984. "A Formal Theory of Social Exchange: Process and
Outcome." Current Perspectives in Social Theory 5:145-180.
Heckathorn, Douglas D. 1985. "Power and Trust in Social Exchange." Pp. 143-167 in
Advances in Group Processes, edited by Edward J. Lawler. Greenwich, CT: JAI
Press.
Heimer, Carol. 2001. Trust, Vulnerability and Uncertainty. Ppxx in K.S. Cook (ed.) Trust in
Society. New York City, New York: Russell Sage Foundation.
Homans, G. C. 1974. Social Behavior and Its Elementary Forms. New York: Harcourt, Brace
and World.
Kollock, Peter. 1994. "The Emergence of Exchange Structures: An Experimental Study of
Uncertainty, Commitment, and Trust." American Journal of Sociology 100:313-45.
Kollock, Peter. 1999. “Trust in Online Communities.” In E Lawler (ed.) Advances in Group
Processes Vol. Xx (?)
Lawler, Edward J., and Jeongkoo Yoon. 1993. "Power and the Emergence of Commitment
Behavior in Negotiated Exchange." American Sociological Review 58:465-481.
Lawler, Edward J.; Yoon, Jeongkoo. 1996. "Commitment in Exchange Relations: Test of a
Theory of Relational Cohesion." American Sociological Review 61:89-108.
Lawler, Edward J.; Yoon, Jeongkoo. 1998. "Network Structure and Emotion in Exchange
Relations." American Sociological Review 63:871-894.
Lawler, Edward J., Jeongkoo Yoon, and Shane R. Thye. 2000. "Emotion and Group
Cohesion in Productive Exchange." American Journal of Sociology 106:616-657.
Leik, Robert K. 1992. "New Directions for Network Exchange Theory: Strategic
Manipulation of Network Linkages." Social Networks 14:309-323
36
Lincoln, James R., Michael Gerlach, and Peggy Takahashi. 1992. “Keiretsu Networks in the
Japanese Economy: A Dyad Analysis of Intercorporate Ties,” American Sociological
Review 57: 561-585.
Macy, Michael W, and John Skvoretz. 1998. "The Evolution of Trust and Cooperation
between Strangers: A Computational Model." American Sociological Review
63:638-660.
Markovsky, Barry, David Willer, and Travis Patton. 1988. "Power Relations in
Exchange Networks." American Sociological Review 5:101-117.
Markovsky, Barry, John Skvoretz, David Willer, Michael J. Lovaglia, and Jeffrey Erger.
1993. "The Seeds of Weak Power: An Extension of Network Exchange Theory."
American Sociological Review 58:197-209.
Mizruchi, Mark S. and Linda Brewster Stearns. 2001. “Getting Deals Done: The Use of
Social Networks in Bank Decision-Making.” American Sociological Review 66
(October): 47-671.
Molm, Linda. 1997. Coercive Power in Social Exchange. Cambridge: Cambridge University
Press.
Molm, Linda. 1997. "Risk and Power Use: Constraints on the Use of Coercion in Exchange."
American Sociological Review 62:113-133.
Molm, Linda, and Karen S. Cook. 1995. "Social Exchange and Exchange Networks." Pp.
209-235 in Sociological Perspectives on Social Psychology, edited by Karen S. Cook,
Gary Alan Fine, and James S. House. Boston: Allyn and Bacon.
Molm, Linda, Gretchen Peterson, and N. Takahashi. 1999. "Power in Negotiated and
Reciprocal Exchange." American Sociological Review 64:876-890.
37
Molm, Linda, Theron M. Quist, and Phillip A. Wiseley. 1994. "Imbalanced Structures, Unfair
Strategies: Power and Justice in Social Exchange." American Sociological Review
49:98-121.
Molm, Linda, N. Takahashi, and Gretchen Peterson. 2000. "Risk and trust in social exchange:
An experimental test of a classical proposition." American Journal of Sociology
105:1396-1427.
Montgomery, J. 1996. "The Structure of Social Exchange Networks: A Game-Theoretic
Reformulation of Blau's Model." Sociological Methodology 26:193-225.
Radaev, Vadim. 2002. "Entrepreneurial Strategies and the Structure of Transaction Costs in
Russian Business." Problems of Economic Transition 44:57-84.
Rice, Eric R.W. 2002. The Effect of Social Uncertainty in Networks of Social Exchange.
Unpublished Ph.D. dissertation.
Rose-Ackerman, Susan. 2001. “Trust and Honesty in Post-Socialist Societies.” Kyklos, Vol.
54: 415-444.
Takahashi, N. 2000. "The Emergence of Generalized Exchange." American Journal of
Sociology 105:1105-1134.
Takahashi, N., and Toshio Yamagishi. 1996. "Social Relational Foundations of Altruistic
Behavior." Japanese Journal of Experimental Social Psychology 36:1-11.
Takahashi, N., and Toshio Yamagishi. 1999. "Voluntary Formation of a Generalized
Exchange System: An Experimental Study of Discriminating Altruism." Japanese
Journal of Psychology 70:9-16.
38
Uzzi, Brian. 1996. "The Sources and Consequences of Embeddedness for the Economic
Performance of Organizations: The Network Effect." American Sociological Review
61:674--698.
Whitmeyer, Joseph. 2000. “Effects of Positive Reputation Systems” Social Science Research
29, 188-207.
Yamagishi, Toshio, Karen S. Cook, and M Watabe. 1998. "Uncertainty, Trust and
Commitment Formation in the United States and Japan." American Journal of
Sociology 104:165-194.
Yamagishi, Toshio and Midori Yamagishi. 1994. “Trust and Commitment in the United
States and Japan.” Motivation and Emotion Vol. 18 (2) 129-166.
Yamagishi, Toshio. 2002. “The Role of Reputation in Open and Closed Societies: An
Experimental Study of Internet Trade” (unpublished).
39
90
Expt 1,
Control
Expt 1, ID
Average Quality Level
80
70
Expt 1,
Reputation
Expt 2
60
50
40
Expt3,
Positive
Expt3,
Negative
30
20
10
0
1
2
3
4 5 6
Tim e Block
7
8
9
Figure 1: Reputation System Effects on Average Quality Level of Goods Exchanged in the
positive reputation condition and the negative reputation condition in Experiment 3,
compared to the levels in Experiments 1 and 2. (Yamagishi, forthcoming)
40