Number Eight

1' Academy of Management Review

1998, Vol. 23, No. 3, 393-404.

INTRODUCTION TO SPECIAL TOPIC FORUM

NOT SO DIFFERENT AFTER ALL: A CROSS- DISCIPLINE VIEW OF TRUST

DENISE M. ROUSSEAU

Carnegie Mellon University

SIM B. SITKIN

Duke University

RONALD S. BURT

University of Chicago

COLIN CAMERER

California Institute of Technology

Our task is to adopt a multidisciplinary view of trust within and between firms, in an effort to

synthesize and give insight into a fundamental

construct of organizational science. We seek to

identify the shared understandings of trust

across disciplines, while recognizing that the

divergent meanings scholars bring to the study

of trust also can add value. Disciplinary differences characterizing tradi-

tional treatments of trust suggest that inherent

conflicts and divergent assumptions are at work (Fichman, 1997). Economists tend to view trust as

either calculative (Williamson, 1993) or institu-

tional (North, 1990). Psychologists commonly frame their assessments of trust in terms of at-

tributes of trustors and trustees and focus upon

a host of internal cognitions that personal at-

tributes yield (Rotter, 1967; Tyler, 1990; see Deutsch, 1962, for an example of more calcula-

tive framing by a psychologist). Sociologists of- ten find trust in socially embedded properties of relationships among people (Granovetter, 1985)

or institutions (Zucker, 1986). These different assumptions are manifest in

our divergent use of language. To some scholars

the term "contract" refers to a legal means for avoiding risk where trust is not particularly high

(Smitka, 1994; Williamson, 1975); to others the

word signals a basis for trust resulting from

sharing and mutuality (Macauley, 1963; Rous-

seau, 1995). Notwithstanding our differences in

emphasis and approach, if our disciplines are to

communicate, we must assume that others are

seeking some common meanings, just as we are.

Without that assumption, we will not make a

good-faith effort to understand one another, re-

ferred to by Sabel as the "act of charity" (1993: 111) necessary to produce mutual intelligibility

out of a mix of ideas and terms. As a result, our

disciplines will continue to work at cross-

purposes and will remain fragmented.

A phenomenon as complex as trust requires

theory and research methodology that reflect

trust's many facets and levels. The features of

trust that characterize the set of papers compos- ing this special issue include

* multilevel trust (individual, group, firm, and institutional),

* trust within and between organizations, * multidisciplinary trust, * the multiple causal roles of trust (trust as a

cause, outcome, and moderator), * trust as impacted by organizational change,

and * new, emerging forms of trust.

This body of work suggests that trust may be a "meso" concept, integrating microlevel psycho- logical processes and group dynamics with ma-

crolevel institutional arrangements (House, Rousseau, & Thomas-Hunt, 1995). In effect, to study trust within and between firms is to ride

We thank Paul Goodman and Bill McEvily for their helpful

comments, Carole McCoy for patient word processing, and

Cathy Senderling for her wonderful editing.

393

This content downloaded from 155.247.166.234 on Tue, 16 Jan 2018 20:20:21 UTC All use subject to http://about.jstor.org/terms

394 Academy of Management Review July

the organizational elevator up and down a va- riety of conceptual levels. Our ride in this ele- vator will focus on the interconnected hallways

of the contemporary firm, rather than its isolated corner offices.

It may appear that we believe trust research

reflects casual scholarship because efforts to

date have focused more on charting the territory than on probing its depths. However, work on trust, as evidenced by this special issue and a very insightful recent book by Kramer and Tyler (1996), is beginning to take a more well-defined and focused form. As the authors of this special topic forum demonstrate, trust is at once related

to dispositions, decisions, behaviors, social net- works, and institutions. However, despite the complex (one might say "multiplex") character of trust, this special issue attempts to promote the accumulation rather than the fragmentation of theory and research on trust.

Our goals in this Introduction are to (1) gauge whether there are common elements underlying trust as it is viewed across disciplines and dif- ferent levels of analysis, (2) discuss views schol-

ars hold regarding the dynamics of trust (i.e., whether it is static or has phases), and (3) clarify the multiple ways in which organizational re- searchers model trust (as cause, effect, or mod- erator), with the ultimate goal of creating a more cumulative body of knowledge on trust in and

between organizations. We begin by investigat- ing a series of assumptions about trust.

TESTING ASSUMPTIONS: WHAT DO WE KNOW ABOUT TRUST?

To date, we have had no universally accepted scholarly definition of trust. There is agreement that trust is important in a number of ways: it enables cooperative behavior (Gambetta, 1988); promotes adaptive organizational forms, such

as network relations (Miles & Snow, 1992); re- duces harmful conflict; decreases transaction costs; facilitates rapid formulation of ad hoc

work groups (Meyerson, Weick, & Kramer, 1996); and promotes effective responses to crisis. Sometimes, scholars use the term "trust" when they mean other things, which has been prob- lematic (Sitkin & Roth, 1993). In a very influential early line of "trust" research, Deutsch (1962), for instance, uses the term when referring to coop- eration within groups. However, cooperation may result from a variety of reasons unrelated to

trust, such as coercion (e.g., court-ordered com- pliance). This blurring of the distinction be- tween trust and cooperation has led to a fuzzi- ness in the treatment of behavior-based trust

and the construct of trust itself.

To advance our understanding of trust, we suggest that the concept of trust requires clear boundaries to usefully inform research and the- ory. We explore how trust's boundaries are con-

stituted by examining the articles that make up this special topic forum in relation to bound- aries proposed in previous treatments of trust. We structure our analysis around four key ques-

tions about the state of our knowledge regard- ing trust. We begin with a basic question.

1. Do Scholars Fundamentally Agree or Disagree on the Meaning of Trust?

To answer this question, let us examine how

trust is defined in the articles in this special topic forum. Regardless of the underlying disci- pline of the authors-from psychology/micro-or-

ganizational behavior (e.g., Lewicki, McAllister, & Bies; Mishra & Spreitzer) to strategy/econom- ics (e.g., Bhattacharya, Devinney, & Pillutla)- confident expectations and a willingness to be vulnerable are critical components of all defini- tions of trust reflected in the articles. The most

frequently cited definition in this special issue

is "willingness to be vulnerable," proposed by Mayer, Davis, and Schoorman (1995).

This contemporary definition plays the central role in defining trust for McKnight, Cummings, and Chervany; Mishra and Spreitzer; and Jones and George. Other authors say the same thing but with different words: "willingness to rely" on another (Doney, Cannon, & Mullen) and "confi- dent, positive expectations" (Lewicki et al.). "Positive expectations" of others defines trust

for Hagen and Choe, Elangovan and Shapiro, and Das and Teng, whereas trust is a positive attitude toward others for authors Whitener, Brodt, Korsgaard, and Werner. Even Bigley and Pearce, who make a case for the absence of an

overarching definition of trust, characterize it in much the same way as other authors-as "vul- nerability," "perception," and "preconscious ex-

pectation." Our evidence from this contempo- rary, cross-disciplinary collection of scholarly writing suggests, therefore, that a widely held definition of trust is as follows:

This content downloaded from 155.247.166.234 on Tue, 16 Jan 2018 20:20:21 UTC All use subject to http://about.jstor.org/terms

1998 Rousseau, Sitkin, Burt, and Camerer 395

Trust is a psychological state compris-

ing the intention to accept vulnerabil-

ity based upon positive expectations

of the intentions or behavior of an-

other.

Note that identification of common meaning

does not imply that all operationalizations of trust reflect the same thing. For example, clear evidence exists that interorganizational and in-

terpersonal trust are different, because the focal object differs (Zaheer, McEvily, & Perrone, in press). However, the composition of trust (i.e., the fundamental elements of its definition) are com-

parable across research and theory focusing on parties both inside and outside firms and inves- tigating trust relations from different disciplin- ary vantage points. We did not expect this re- sult, but it is encouraging to observe.

Across disciplines, there is agreement on the conditions that must exist for trust to arise. Risk is one condition considered essential in psychologi- cal, sociological, and economic conceptualiza- tions of trust (Coleman, 1990; Rotter, 1967; William- son, 1993). Risk is the perceived probability of loss, as interpreted by a decision maker (Chiles & Mc- Mackin, 1996; MacCrimmon & Wehrung, 1986). The path-dependent connection between trust and risk taking arises from a reciprocal relationship: risk creates an opportunity for trust, which leads to risk taking. Moreover, risk taking buttresses a sense of trust when the expected behavior mate-

rializes (Coleman, 1990; Das & Teng, this issue). Trust would not be needed if actions could be undertaken with complete certainty and no risk (Lewis & Weigert, 1985). Uncertainty regarding whether the other intends to and will act appro-

priately is the source of risk. The second necessary condition of trust is in-

terdependence, where the interests of one party

cannot be achieved without reliance upon an- other. Although both risk and interdependence are required for trust to emerge, the nature of risk and trust changes as interdependence in-

creases (Sheppard & Sherman, this issue). De- grees of interdependence actually alter the form trust may take, with the nature of trust a firm places in temporary workers being quite distinct from trust associated with its veteran, core em-

ployees. Authors of several articles in this spe- cial issue provide frameworks linking forms of trust to the context of the relationship (Lewicki,

McAllister, & Bies; Sheppard & Sherman). In-

deed, the time may be right for an overarching view of trust across forms of interdependence.

To answer our initial question, scholars do appear to agree fundamentally on the meaning

of trust. Trust, as the willingness to be vulnera- ble under conditions of risk and interdepen-

dence, is a psychological state that researchers in various disciplines interpret in terms of "per-

ceived probabilities" (Bhattacharya et al., this

issue), "confidence," and "positive expectations" (e.g., Jones & George, Hagen & Choe, Das & Teng, all this issue)-all variations on the same theme. Trust is not a behavior (e.g., cooperation), or a choice (e.g., taking a risk), but an underlying

psychological condition that can cause or result from such actions. Regardless of the discipline

of the researcher, we share the root assumptions that trust is psychological and important to or-

ganizational life. Finally, because risk and interdependence

are necessary conditions for trust, variations in

these factors over the course of a relationship between parties can alter both the level and, potentially, the form that trust takes. The poten- tial for trust to change gives rise to a second

question.

2. Do Researchers View Trust Statically?

A focus on static and stable phenomena is characteristic of normal science, which values

precision and control. Indeed, equilibrium seek- ing is an underlying assumption in such fields as economics. Given this emphasis, it is not

surprising that scholars often have treated trust as static. Social psychologists often see trust as

either/or, where one person either completely trusts or completely distrusts another (Gabarro, 1990, cited in Lewicki et al., this issue). This static, all-or-nothing view is linked to the pre- dominance in early trust research of laboratory studies focusing on highly structured games, such as the Prisoner's Dilemma game (e.g., Ax- elrod, 1984). Under such conditions, the level of trust reflects a single point, rather than a distri- bution along an intra- or interpersonal contin- uum. However, the fact that trust changes over time-developing, building, declining, and even resurfacing in long-standing relation- ships-is evidence from comparative and histor- ical research upon trust in organizations (Miles & Creed, 1995) and in the broader society (Fukuyama, 1995).

This content downloaded from 155.247.166.234 on Tue, 16 Jan 2018 20:20:21 UTC All use subject to http://about.jstor.org/terms

396 Academy of Management Review July

But the question remains whether scholars pursuing precision and control have focused un-

duly upon trust as a static rather than dynamic phenomenon. To answer the question, we exam-

ine the articles in this special topic forum in

terms of three phases of trust: (1) building (where

trust is formed or reformed), (2) stability (where trust already exists), and (3) dissolution (where trust declines). These phases of trust character- ize the ebb and flow of relationships. Character- izing this issue's articles in terms of the phase

upon which they focus helps address how com-

prehensively scholars have viewed the develop- ment of trust.

The building phase is addressed in several articles: in the emergence of trust in new organ- izational settings (McKnight et al.) and new or-

ganizational relationships (Das & Teng), or in

the context of an existing relationship between workers and managers (Whitener et al.) where trust may be created or enhanced. The function- ing of trust under stable conditions is central to the broad treatments given to institutional fac-

tors associated with trust by Hagen and Choe (trust in Japanese society) and Sheppard and Sherman (trust across different relational forms). Declining trust is the focal point in articles on

downsizing (Mishra & Spreitzer) and betrayal (Elangovan & Shapiro), in which the authors ex- plore the effects of reduced trust.

The articles here do not overemphasize stabil-

ity but address a variety of trust's phases, at least raising the possibility of a balance in

scholarship on trust and its dynamics. A related

question thus arises. Do scholars recognize that trust is dynamic but focus only upon a particular phase? In 4 out of the 12 articles, the authors consider multiple phases of trust. Processes of

trust building and decline are examined by Jones and George, as well Bigley and Pearce. Bhattacharya and his colleagues examine how firm-based relationships begin and the condi-

tions under which they achieve stability. Fi- nally, Lewicki and his colleagues, by focusing simultaneously on antecedents of trust and dis-

trust (which they conceptualize as two distinct constructs), run the gamut-from factors build- ing trust or promoting distrust to conditions that

achieve an equilibrium in each and the dynam- ics that can move individuals and firms through distinct levels of trust and distrust.

In answer to the question, scholars do, at

times, focus on multiple phases. However, the

tendency of the authors in the majority of arti- cles here is to focus upon building, stability, or

decline and to specify conceptual frameworks within a particular phase. Such an emphasis on phase-specific trust may be necessary at this

point in the development of trust scholarship. Given the dynamic nature of trust, the next question is related to trust's role in the causal frameworks researchers employ.

3. Does the Status of Trust As a Cause, Effect,

or Interaction Vary Across Disciplines?

Theorists and researchers of trust may model

the concept as an independent variable (cause), dependent variable (effect), or interaction vari- able (a moderating condition for a causal rela- tionship). When economic outcomes are of inter-

est, researchers often conceptualize trust as a potential cause in choice scenarios framed

around social dilemmas. High trust, perhaps based on previous experiences with a partner in a repeated game, tends to result in the decision

to cooperate, which can lead to access to eco- nomic gains, as in the classic Prisoner's Di- lemma (Axelrod, 1984; Miller, 1992). Trust, thus, is conceptualized as an independent variable.

Similarly, transaction cost economists view trust as a cause of reduced opportunism among transacting parties, which results in lower

transaction costs (Williamson, 1975). Trust has also long been found to be an important predic-

tor of successful negotiations and conflict- management efforts (Deutsch, 1958), and it has a

direct effect on disputants' responses to media- tors attempting to settle disputes (Ross & Wie- land, 1996). We note that several articles in this special topic forum consider trust as a cause. Elangovan and Shapiro and Jones and George

focus exclusively on trust as an independent variable.

In contrast, trust can be the result of deep

dependence and identity formation, as has been

the case historically in Japanese firms (Ouchi, 1981). Trust as the result of institutional arrange- ments is characteristic of a sociological per- spective (e.g., Zucker, 1986). Third-party relations have been found to impact trust, where existing social structures shape a person's reputation based upon a third party's ability to tell stories that corroborate one's trustworthiness (or lack of it; Burt & Knez, 1996). Further, trust can also be seen as the result of attributes of the other party,

This content downloaded from 155.247.166.234 on Tue, 16 Jan 2018 20:20:21 UTC All use subject to http://about.jstor.org/terms

1998 Rousseau, Sitkin, Burt, and Camerer 397

such as that party's competence, concern, open- ness, and reliability (Mishra, 1996). In this spe- cial issue trust is conceptualized exclusively as a result in seven articles (Bhattacharya et al.; Doney et al.; Hagen & Choe; Lewicki et al.;

McKnight et al.; Sheppard & Sherman; Whitener et al.). Note that these authors run the gamut of disciplines-from psychology and organization- al behavior to economics and marketing.

The moderating role of trust in shaping causal relationships is characteristic of studies of inter-

personal behavior in organizations and social settings, as found in both micro-organizational behavior and social psychology (e.g., Robinson & Rousseau, 1994). Mishra and Spreitzer (this issue) use trust as a cause and moderator in

their modeling of reactions to downsizing,

whereas Bigley and Pearce (this issue), along with Das and Teng (this issue), systematically model trust in all three roles.

Our authors model the role of trust in diverse ways, regardless of their disciplinary base. The economic focus of Das and Teng and Bhatta- charya and his colleagues does not constrain either article from treating trust as an effect, despite the trend for economic models to use

trust to explain human choice (Miller, 1992). Sim- ilarly, with their micro-organizational behavior perspectives, Mishra and Spreitzer, Bigley and Pearce, Elangovan and Shapiro, and Jones and

George view trust, at least in part, as a cause of the behavior of interest-not simply an outcome (whereas Lewicki et al., Sheppard and Sherman, and Whitener et al. do follow in the traditional use of trust as a micro-organizational behavior individual response). Das and Teng model the moderating effects of trust from an economic

viewpoint, whereas Bigley and Pearce and Mishra and Spreitzer do the same from the van- tage point of micro-organizational behavior.

We conclude, therefore, that disciplinary dif- ferences do not account for the status given to trust as an independent, dependent, or modera- tor variable. Rather, the function of trust in the causal frameworks researchers model appears here to reflect richer and more complex cross- disciplinary views.

4. Do Disciplinary Differences Exist in the Levels of Analysis in Trust Research?

Our last question concerns the level of analy- sis at which scholars conceptualize trust and

related phenomena. Our call for papers empha-

sized the desire for multilevel perspectives on trust in and between organizations-reflecting

the array of entities, individuals, dyads, groups, networks, firms, and interfirm alliances in which trust and related processes play a role. It is commonly assumed that disciplines occupy

different turf in organizational science: psychol- ogists, the individual and occasionally the group; sociologists, the group and society; and

economists, the individual or the larger firm.

Strikingly, the great majority of articles in this special issue include, often exclusively, a focus on the individual-whether as trustor (e.g., Doney et al.; Lewicki et al.; McKnight et al.; Mishra & Spreitzer) or trustee (e.g., Elangovan &

Shapiro; Whitener et al.). Analysis at the indi- vidual level tends to characterize conceptualiza-

tions of trust within firms, particularly the will- ingness of subordinates to trust their bosses (Whitener et al.). In five articles the authors ad- dress firm-level trust: Das and Teng, exclusively at the firm level; Hagen and Choe and Sheppard

and Shapiro, in combination with dyadic anal- ysis; Bhattacharya and colleagues, in their anal- ysis of the behavior of firms and agents; and Bigley and Pearce, who address trust at multiple levels. Although micro-organizational behavior researchers predominate at the individual level

and economists at the firm level, neither turf is "sacred" to a particular field. Scholars from a variety of backgrounds address phenomena at any given level.

This special topic forum provides numerous examples of articles adopting multiple levels of analysis, as well as a simultaneous consider-

ation of trust within and between firms. Why might trust require a multilevel analysis, re- gardless of the disciplinary focus of the scholar?

First, we know that reputation matters, partic- ularly the historical trustworthiness of parties in previous interactions with others (Burt & Knez, 1996), and it is the social context (e.g., networks) that makes reputational effects possible. Social norms shape both the behaviors parties engage in, as well as their beliefs regarding the inten- tions of others (Sitkin & Stickel, 1996; Whitener et

al., this issue). Institutions promote or constrain trust relations (Fukuyama, 1995; Hagen & Choe, this issue; Smitka, 1994). Thus, microlevel trust

relations are constrained and enhanced by macro processes (Sitkin, 1995). Conversely, broader forms of trust, particularly between

This content downloaded from 155.247.166.234 on Tue, 16 Jan 2018 20:20:21 UTC All use subject to http://about.jstor.org/terms

398 Academy of Management Review July

firms, can be influenced by microlevel arrange-

ments-in particular, how individuals repre- senting each firm relate to each other (Fichman

& Goodman, 1996; Zaheer et al., in press). Thus, multilevel processes underlie trust, and schol- ars across a variety of disciplines have em- braced this multilevel view.

In sum, what do we know about trust? We find

that trust is a psychological state composed of the psychological experiences of individuals,

dyads, and firms. There is a common underlying definition of trust across scholars from different

disciplines, and this basic definition applies across trust's levels of analysis and develop-

mental phases. Scholars tend to view trust dy- namically but focus on specific phases in devel- oping their conceptual frameworks. Some are interested in trust's beginning, others in its end,

and still others in trust as an ongoing and stable phenomenon.

Scholars will continue to debate the kinds of

questions that a cross-disciplinary view rais- es-whether contracts are a basis of trust or a

means of reducing risk and whether the lan- guage of the marketplace can be used to de-

scribe personal relationships-and they will puzzle over the question of why one party to a relationship is willing to trust first (is it self- interest or good faith?). The broad convergence of issues and interests in this special topic fo- rum suggests that research on trust from the

perspective of any one discipline is likely to inform several others.

ISSUES AND IMPLICATIONS

The case for integrating trust across disci-

plines rests on the common psychological basis

upon which all formulations of trust rest. Pulling

together disparate research on trust requires at- tention to the context in which trust is studied. One thing is apparent: scholars operationalize trust differently, depending on the focus and phase of trust they study. Trust in one-time transactions typically derives from the calculus

of gains and losses, weighed by perceived risks. However, in ongoing relationships the question is not so much "How much do I trust?" but "In

what areas and in what ways do I trust?" (Lewicki et al.). We acknowledge that some on- going relations remain predominantly transac- tional, with a focus on gains and losses (Wil- liamson, 1975). However, both history and the

nature of the interaction between the parties can shape the form that trust takes.

Scholars who differentiate trust from distrust (e.g., Lewicki et al., this issue; Sitkin & Roth, 1993), in effect, imply that trust has a "band- width," where it can vary in scope as well as

degree. Trust takes different forms in different relationships-from a calculated weighing of

perceived gains and losses to an emotional re- sponse based on interpersonal attachment and identification. Market-based exchanges may emphasize calculus more, whereas communal relationships might emphasize identification.

However, even in the same context, the scope of

trust may vary, depending on the relationship's history, stage of development, and cues in the immediate setting. The recognition of variations in the scope of trust is evident in the writings of

Williamson (1993), who asked where "calcula- tive" trust ends and "people" trust begins.

The bandwidth of trust varies in the same relationship over time. Moreover, broad and nar- row bandwidths characterize different types of

relationships. Where a trustor believes in the positive intentions of the trustee across a broad range of situations, bandwidth is great. In con-

trast, bandwidth is narrow when trust's range is limited to specific conditions only (Sitkin & Roth, 1993). Lewicki et al.'s differentiation of trust and distrust as separate concepts is an implicit rec- ognition of variations in bandwidth across rela-

tionships where trust (expectations of positive intentions) and distrust (expectations of nega- tive intentions) can exist simultaneously. To un- derstand how bandwidth functions, we must consider the different forms of trust included in the bandwidth.

Different Forms of Trust

Deterrence-based trust emphasizes utilitarian considerations that enable one party to believe that another will be trustworthy, because the costly sanctions in place for breach of trust ex-

ceeds any potential benefits from opportunistic behavior (Ring & Van de Ven, 1992, 1994; Sha- piro, Sheppard, & Cheraskin, 1992). Asset speci- ficity effects in the form of switching costs to parties, as in transaction cost economics, are

examples of deterrence-based trust. The ques- tion then becomes whether sanctions foster or

substitute for trust, particularly in interfirm sit- uations (Hagen & Choe, this issue).

This content downloaded from 155.247.166.234 on Tue, 16 Jan 2018 20:20:21 UTC All use subject to http://about.jstor.org/terms

1998 Rousseau, Sitkin, Burt, and Camerer 399

Some have raised the issue that deterrence-

based trust is not trust at all (Sitkin & Roth, 1993). It is clear that sanctions can foster or obstruct

cooperation, which is a behavior. However, al- though trust promotes cooperation, cooperation

can occur for other reasons as well (e.g., coer-

cion or fear of loss). As trust at its narrowest

bandwidth, deterrence-based trust has the qual- ities of low distrust/low trust (Lewicki et al., this issue), where costs of breaching trust are high

and the involvement between the parties is lim-

ited in the first place.

In a sense, trust is not a control mechanism

but a substitute for control, reflecting a positive

attitude about another's motives. Control comes

into play only when adequate trust is not

present. For example, a detailed legal contract is one mechanism for controlling behavior.

However, Macaulay (1963) observes that de-

tailed contracts can get in the way of creating

an effective exchange relationship: in effect,

people do not need to develop trust when their

exchange is highly structured and easily moni- tored. Although detailed contracts promote lim- ited cooperation based upon deterrence, most

firms that form alliances do so because of a

social network of prior alliances, which makes

detailed contracts less necessary (Kogut, Shan, & Walker, 1993).

There is an apparent incompatibility between

strict controls and positive expectations about

the intentions of another party. Some controls actually appear to signal the absence of trust and, therefore, can hamper its emergence, per- haps by limiting the degree of interdependence

that develops between the parties. Moreover,

belief in the absence of "negative intentions" is not the same as beliefs in the presence of posi-

tive ones-the latter being a necessary condi- tion of the generally accepted definition of trust. Deterrence-based trust, therefore, may not be trust at all but may be closer to low levels of distrust.

Calculus-based trust, however, is based on ra- tional choice-characteristic of interactions

based upon economic exchange. Trust emerges when the trustor perceives that the trustee in-

tends to perform an action that is beneficial. The perceived positive intentions in calculus-based trust derive not only from the existence of deter- rence but also because of credible information

regarding the intentions or competence of an-

other (Barber, 1983). For instance, credible infor-

mation about the trustee may be provided by others (reputation) or by certification (e.g., a di-

ploma). Such "proof sources" signal that the

trustee's claims of trustworthiness are true (Doney et al., this issue). In Lewicki et al.'s (this

issue) framework, this would incorporate the

high trust/high distrust condition. Here, parties

trust but verify under conditions where willing-

ness to trust is limited to specific exchanges

(e.g., financial but not personal). Opportunities are pursued and risks continually monitored. The range of calculus-based trust is often lim-

ited to situations where evidence of failure to

perform can be obtained in the short term. Risk

may entail short-term performance losses but not threaten the trustor's broader interests.

Relational trust derives from repeated interac-

tions over time between trustor and trustee. In-

formation available to the trustor from within the relationship itself forms the basis of rela-

tional trust. Reliability and dependability in

previous interactions with the trustor give rise to

positive expectations about the trustee's inten- tions. Emotion enters into the relationship be-

tween the parties, because frequent, longer- term interaction leads to the formation of

attachments based upon reciprocated interper- sonal care and concern (McAllister, 1995). (For this reason, scholars often refer to this form of trust as "affective trust" [McAllister, 1995] and as "identity-based trust" at its broadest scope

[Coleman, 1990]). Repeated interactions can vary considerably

in the resources exchanged and in the scope of interdependence between the parties (from re-

peated employment of contract labor from the same agency to career advancement of full-time employees in a firm). Repeated cycles of ex- change, risk taking, and successful fulfillment of expectations strengthen the willingness of trusting parties to rely upon each other and ex-

pand the resources brought into the exchange. Thus, an exchange can evolve from an arm's- length transaction into a relationship: from a

"fair day's work for a fair day's pay" arrange- ment to a high-performance employment rela-

tionship characterized by mutual loyalty and broad support. Citizenship behavior from em- ployees and organizational support from em- ployers are characteristic of high levels of rela-

tional trust based upon experience within the relationship (Eisenberger, Huntington, Hutchin-

son, & Sowa, 1986; Organ, 1990).

This content downloaded from 155.247.166.234 on Tue, 16 Jan 2018 20:20:21 UTC All use subject to http://about.jstor.org/terms

400 Academy of Management Review July

In Lewicki et al.'s (this issue) framework, rela-

tional trust entails not only beliefs in the posi- tive intentions of the trustee but also in the ab-

sence of negative intentions, giving rise to the

condition of high trust/low distrust. Interdepen- dence between the parties to relational trust is likely to increase over time as new opportunities and initiatives are pursued. A dynamic of rela- tional trust is its potential for expansion or con- traction, where experiences over time can esca-

late positive beliefs regarding the intentions of the other or, conversely, exacerbate negative be-

liefs (Lewicki et al., this issue; Sitkin & Roth, 1993).

Whereas calculus-based trust is relatively cir- cumscribed, limited as it is to discrete ex-

changes reinforced by the existence of deter- rents, relational trust involves a broader array of resource exchange (including socioemotional

support, as well as concrete resources) and en- tails a greater level of faith in the intentions of the other party. Exchanges based on calculus-

based trust are likely to be terminated once vi- olation occurs, but exchanges characterized by relational trust often are more resilient. Unmet expectations can be survived when relational

trust exists, particularly if parties make an effort

to restore a sense of good faith and fair dealing to their interactions.

There is a fine line between the existence of a good-faith relationship between parties and the

emergence of a shared identity. Two parties may continually cooperate and share informa-

tion and assets while still believing that the other party is "them" instead of "us." However, there is a tendency for repeated interactions to

create expanded resources, including shared in- formation, status, and concern. These expanded resources can, in turn, give rise to a psycholog-

ical identity (Gaertner, Dovidio, & Bachman, 1996). Employees may come to characterize themselves in relationship to their teammates or firm as "we," and may derive psychic benefits from being part of a successful enterprise. Iden- tity-based trust is relational trust at its broadest.

Institution-based trust can ease the way to formulating both calculus-based and relational

trust. Ex ante deterrents may promote trust, be- cause one's confidence that reputation matters permits relationships to form in the first place. Institutional factors can act as broad supports for the critical mass of trust that sustains further

risk taking and trust behavior (e.g., Gulati, 1995;

Ring & Van De Ven, 1992; Sitkin, 1995). These supports can exist at the organizational level, in the form of teamwork culture (Miles & Creed,

1995; Whitener et al., this issue), and at the so- cietal level, through such cultural supports as legal systems that protect individual rights and property (Fukuyama, 1995).

One example where both organizational and societal factors affect trust can be found in com-

parative research on Hungarian firms under

communism (Pearce & Branyicki, in press). The

absence of within-firm procedures for promoting consistent employee treatment, coupled with the

autocratic nature of the Hungarian government, undermined trust and good-faith relations be- tween supervisors and subordinates. In con-

trast, the U.S. firms studied enjoyed the benefits of standardized human resource practices within firms (e.g., performance reviews and

compensation) and legal protections from the federal government. Not surprisingly, trust be- tween supervisors and subordinates was con- siderably higher in American firms than in Hun- garian ones.

Whether institutional trust is a control or a

form of trust support is a fundamental issue (Shapiro, 1987). Control, as manifested in laws and reputational sanctions, acts as a deterrent from opportunism. These mechanisms, however, can serve as a springboard for the creation of

trust. Hagen and Choe (this issue) argue that, in Japan, societal controls function as both a

means of controlling behavior and as a basis for supporting the development of trust. Conceptu- alizing trust and distrust as distinct offers in- sights into how institutions promote trust in so-

cieties where the anticipation of positive motives can be fostered without high levels of monitoring, because a legal system makes ex-

pectations of harm low-probability events. Institutional controls can also undermine

trust, particularly where legal mechanisms give rise to rigidity in responses to conflict and sub- stitute high levels of formalization for more flex- ible conflict management (Sitkin & Bies, 1994).

Scholars have variously interpreted the tension between institutional mechanisms creating im- personal forms of trust and less standardized (but more flexible) interpersonal trust. Based on research in North American organizations, Zucker (1986) views institutional mechanisms as

reducing the opportunity for creating interper- sonal trust. In contrast, based upon comparative

This content downloaded from 155.247.166.234 on Tue, 16 Jan 2018 20:20:21 UTC All use subject to http://about.jstor.org/terms

1998 Rousseau, Sitkin, Burt, and Camerer 401

research in Hungary and the United States,

Pearce and Brzenksky (in press) see a minimum

level of institutional trust as sine qua non for the

emergence of interpersonal trust. The possibility of a step function characteriz-

ing the role institutional trust plays in shaping

interpersonal trust remains a subject for future

research. Nonetheless, a variety of institutional

factors, including legal forms, social networks, and societal norms regarding conflict manage-

ment and cooperation, are likely to interact in creating a context for interpersonal and interor- ganizational trust.

The variations we observe in the bandwidth of

trust across relationships suggest there may be a tension between acting out of self-interest (agency) and acting out of the interests of a

broader collective (community). Sabel (1993) sug- gests that we need to move away from thinking

of trust as rational self-interest toward a shared sense of community with a common fate. Beliefs

can arise in communities that lead to avoidance

of exploitation, where trusting others is a condi- tion of membership. Such shared understand- ings between individuals or between firms can

arise out of interactions and from shared or com-

mon knowledge. Indeed, some societies have penalties for put-

ting oneself ahead of community interests (Hearn, 1904, as cited by Hagen & Choe, this issue). High power distance societies, such as

Japan, may build obligations into societal roles, creating codes of conduct that reinforce collec- tive behavior through relational sanctions. In low power distance societies, such as North America, mechanisms that support repeated in-

teractions, including stable employment, net- work ties, and laws protecting property rights of

individuals and firms, may also enable trust

(Nooteboom, Berger, & Noorderhaven, 1997). Our

discussion of bandwidth suggests that institu-

tional mechanisms can play a critical role in shaping the mix of trust and distrust that exists.

In Figure 1 we model the three basic forms of

trust (calculative, relational, and institutional) with respect to the issue of bandwidth. (Note that we conclude that deterrence is not trust and exclude it from the model.)

The various forms trust can take-and the

possibility that trust in a particular situation can mix several forms together-account for some of the apparent confusion among scholars. Conceptualizing trust in only one form in a given relationship risks missing the rich diver-

sity of trust in organizational settings. Recogniz- ing that, in a given relationship, trust has a bandwidth (which may exist to different degrees between the same parties, depending on the task or setting) introduces the idea that experi- ences over the life of a relationship may lead to

pendulum swings. The interests of each party

separately and their mutual concerns might be met to a limited degree at any single point in time-but to a large degree over the life of the relationship.

Is Trust in Transition?

The form and context of organizations are in transition. We observe in society a move toward small-scale relations (Miles & Creed, 1995; Miles

& Snow, 1992). In this era of more flexible forms

FIGURE 1 A Model of Trust

Ins l tttional trust Calculative trust

Institutional trust

Early Middle Later

Developmental time

This content downloaded from 155.247.166.234 on Tue, 16 Jan 2018 20:20:21 UTC All use subject to http://about.jstor.org/terms

402 Academy of Management Review July

of organizing, the breaking up of large firms, the

rise of independent contractors, and shifts in the

prevalent forms of trust are likely. In fact, we

may currently observe a shift from institutional

trust to individual and network-based trust (Sax-

enian, 1996). Even highly competitive industries,

such as investment banking, manifest evidence

of individual and network-based trust mecha-

nisms, despite high turnover and loose ties to

one another. In a recent study researchers ob-

served friendship or advice relationships or ties

of around 2 percent among investment bankers,

whereas individual ratings of investment

banker trustworthiness display inter-rater agree-

ment of .8 (Burt, 1998).

In more fluid work settings, trust may be par-

ticularly important for the ability of workers to

self-organize. Where trust is present, it can pro-

mote a critical mass of trust-related behaviors,

such as the cooperation needed to create higher-

unit trustworthiness (Whitener et al., this issue). In a knowledge-based economy, a trustee's com-

petence, ability, and expertise become increas-

ingly important as an indicator of his or her

ability to act as anticipated.

In this issue McKnight et al. and Sheppard

and Sherman raise the issue of whether new

organizational forms imply new forms of trust, characterized by changes in the nature of vul- nerabilities and risks or changes in bandwidth.

These shifts raise the possibility that violations

of trust may take new and different forms. In a

highly interdependent marketplace, violations

of calculus-based trust may be signs of irration-

ality, whereas breaches of relational trust may be indicators of unpredictability at the middle

level and of bad faith and lack of caring at the

highest level (Sheppard & Sherman, this issue).

In sum, the impact of contemporary organiza-

tional changes on trust leads us to conclude that

context is critical to understanding trust. Acon-

textual research will be limited in its ability to

represent the true functioning of trust. Future

empirical research on trust needs to address

whether a particular context has given rise to a

single form of trust (what Lewicki et al. refer to as "uniplex") or to broader multiplex forms. New organizational forms built around the manage-

ment of interdependence will provide a catalyst

for innovative research on trust and its band-

width into the next millennium.

CONCLUSION

Our disciplines are intellectually dense at dif-

ferent points in the network of constructs (Bigley

& Pearce, this issue). We may be divided by

jargon and tribal identification, but we are part of the same network of ideas. Despite the com-

mon concern regarding our different disciplin-

ary lenses (i.e., "blinders"), we observe consid- erable overlap and synthesis in contemporary

scholarship on trust. But we have a confession for the reader: by collaborating across disci- plines to identify and develop multilevel, multi-

disciplinary views of trust, we have stacked the deck.

This special topic forum has been designed to

reinforce integration and cumulation of insights.

The scholars who have contributed to this spe-

cial issue have undertaken the challenge to be

consciously integrative in their approaches to fundamental problems surrounding trust in or- ganizational settings. We applaud AMR's will- ingness to provide the context and the incen-

tives to support a creative synthesis across disciplines, and we deeply appreciate the schol-

arship and creativity of the authors who have

stepped up to the challenge this special issue offered. Enjoy the product of their efforts. This special topic forum will have achieved its pur- pose if it acts as a pointer to further the cumu- lative nature of research on trust in and between

organizations.

REFERENCES

Axelrod, R. 1984. The evolution of cooperation. New York:

Basic Books.

Barber, B. 1983. The logic and limits of trust. New Brunswick,

NJ: Rutgers University Press.

Burt, R. S. 1998. The network structure of representation.

Working paper, Graduate School of Business, University

of Chicago, IL.

Burt, R. S., & Knez, M. 1996. Trust and third-party gossip. In

R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations:

Frontiers of theory and research: 68-89. Thousand Oaks,

CA: Sage.

Chiles, T. H., & McMackin, J. F. 1996. Integrating variable risk preference, trust, and transaction cost economics. Acad-

emy of Management Review, 21: 73-99.

Coleman, J. S. 1990. Foundations of social theory. Cambridge, MA: Belknap Press.

Deutsch, M. 1958. Trust and suspicion. Journal of Conflict Resolution, 2: 265-279.

Deutsch, M. 1962. Cooperation and trust: Some theoretical

This content downloaded from 155.247.166.234 on Tue, 16 Jan 2018 20:20:21 UTC All use subject to http://about.jstor.org/terms

1998 Rousseau, Sitkin, Burt, and Camerer 403

notes. Nebraska symposium on motivation: 275-320. Lin- coln: Nebraska University Press.

Eisenberger, R., Huntington, R., Hutchinson, S., & Sowa, D.

1986. Perceived organizational support. Journal of Ap- plied Psychology, 71: 500-507.

Fichman, M. 1997. A multilevel analysis of trust in interor-

ganizational customer-supplier ties. Paper presented at

the annual meeting of the Academy of Management, Boston.

Fichman, M., & Goodman, P. S. 1996. Customer supplier ties in interorganizational relations. In B. M. Staw & L. L.

Cummings (Eds.), Research in organizational behavior, vol. 18: 286-329. Greenwich, CT: JAI Press.

Fukuyama, F. 1995. Trust: The social virtues and the creation of prosperity. New York: Free Press.

Gabarro, J. J. 1990. The development of working relation- ships. In J. Gallagher, R. E. Kraut, & C. Egido (Eds.), Intellectual teamwork: Social and technological founda- tions of cooperative work: 79-110. Hillsdale, NJ: Law- rence Erlbaum Associates.

Gaertner, S. L., Dovidio, J. F., & Bachman, B. A. 1996. Revising the contact hypothesis: The induction of a common

group identity. International Journal of Intercultural Re- lations, 20: 271-290.

Gambetta, D. 1988. Trust: Making and breaking cooperative relations. New York: Basil Blackwell.

Granovetter, M. S. 1985. Economic action and social struc- ture. American Journal of Sociology, 91: 481-510.

Gulati, R. 1995. Does familiarity breed trust? The implica- tions of repeated ties for contractual choice in alliances.

Academy of Management Journal, 38: 85-112.

Earn, L. 1904. Japan: An interpretation. Rutland, VT: Tuttle.

House, R., Rousseau, D. M., & Thomas-Hunt, M. 1995. The meso paradigm: A framework for the integration of mi-

cro and macro organizational behavior. In L. L. Cum-

mings & B. M. Staw (Eds.), Research in organizational behavior, vol. 17: 71-114. Greenwich, CT: JAI Press.

Kogut, B., Shan, W., & Walker, G. 1993. Competitive cooper- ation in biotechnology: Learning through networks? In N. Nohria & R. Eccles (Eds.), Networks and organizations:

Structure, form, and action: 348-365. Boston: Harvard Business School Press.

Kramer, R. M., & Tyler, T. R. 1996. Trust in organizations:

Frontiers of theory and research. Thousand Oaks, CA: Sage.

Lewis, J. D., & Weigert, A. 1985. Trust as a social reality. Social Forces, 63: 967-985.

Macauley, S. 1963. Non-contractual relations in business: A

preliminary study. American Sociological Review, 28: 55-67.

MacCrimmon, K. R., & Wehrung, D. A. 1986. Taking risks: The management of uncertainty. New York: Free Press.

Mayer, R. C., Davis, J. H., & Schoorman, D. 1995. An integra- tive model of organizational trust. Academy of Manage- ment Review, 20: 709-734.

McAllister, D. J. 1995. Affect- and cognition-based trust as

foundations for interpersonal cooperation in organiza-

tions. Academy of Management Journal, 38: 24-59.

Meyerson, D., Weick, K. E., & Kramer, R. M. 1996. Swift trust

and temporary groups. In R. M. Kramer & T. R. Tyler

(Eds.), Trust in organizations: Frontiers of theory and research: 166-195. Thousand Oaks, CA: Sage.

Miles, R. E., & Creed, W. E. D. 1995. Organizational forms and

managerial philosophies: A descriptive and analytical

review. In B. M. Staw & L. L. Cummings (Eds.), Research

in organizational behavior, vol. 17: 333-372. Greenwich,

CT: JAI Press.

Miles, R. E., & Snow, C. C. 1992. Causes of failure in network

organizations. California Management Review, Sum- mer: 93-72.

Miller, G. J. 1992. Managerial dilemmas: The political econ-

omy of hierarchy. New York: Cambridge University Press.

Mishra, A. K. 1996. Organizational responses to crisis: The

centrality of trust. In R. M. Kramer & T. R. Tyler (Eds.),

Trust in organizations: Frontiers of theory and research:

261-287. Thousand Oaks, CA: Sage.

Nooteboom, B., Berger, H., & Noorderhaven, N. G. 1997. Effects

of trust and governance on relational risk. Academy of Management Journal, 40: 308-338.

North, D. C. 1990. Institutions, institutional change, and eco-

nomic performance. New York: Cambridge University Press.

Organ, D. W. 1990. The motivational basis of organizational citizenship behavior. In L. L. Cummings & B. M. Staw

(Eds.), Research in organizational behavior, vol. 12: 43-

72. Greenwich, CT: JAI Press.

Ouchi, W. G. 1981. Theory Z. New York: Avon.

Pearce, J., & Branyicki, I. In press. Insufficient bureaucracy:

Trust and commitment in particularistic organizations.

Organizational Science.

Ring, P. S., & Van de Ven, A. 1992. Structuring cooperative

relationships between organizations. Strategic Manage-

ment Journal, 13: 483-498.

Ring, P. S., & Van de Ven, A. H. 1994. Developmental process

of cooperative interorganizational relationships. Acad- emy of Management Review, 19: 90-118.

Robinson, S. L., & Rousseau, D. M. 1994. Violating the psy-

chological contract: Not the exception but the norm. Jour-

nal of Organizational Behavior, 15: 245-259.

Rousseau, D. M. 1995. Psychological contracts in organiza-

tions: Understanding written and unwritten agreements.

Thousand Oaks, CA: Sage.

Ross, W. H., & Wieland, C. 1996. Effects of interpersonal trust

and time pressure on managerial mediation strategy in

a simulated organizational dispute. Journal of Applied Psychology, 81: 228-248.

Rotter, J. B. 1967. A new scale for the measurement of inter-

personal trust. Journal of Personality, 35: 615-665.

Sabel, C. F. 1993. Studied trust: Building new forms of coop- eration in a volatile economy. In R. Swedberg (Ed.), Ex-

This content downloaded from 155.247.166.234 on Tue, 16 Jan 2018 20:20:21 UTC All use subject to http://about.jstor.org/terms

404 Academy of Management Review July

plorations in Economic Sociology: 104-144. New York: Russell Sage Foundation.

Saxenian, A. 1996. Beyond boundaries: Open labor markets and learning in Silicon Valley. In M. B. Arthur & D. M. Rousseau (Eds.), The boundaryless career: A new organ- izational principle for a new organizational era: 23-39. New York: Oxford University Press.

Shapiro, D., Sheppard, B. H., & Cheraskin, L. 1992. Business on a handshake. Negotiation Journal, 8: 365-377.

Shapiro, S. P. 1987. The social control of impersonal trust. American Journal of Sociology 93: 623-658.

Sitkin, S. B. 1995. On the positive effect of legalization on

trust. In R. J. Bies, R. J. Lewicki, & B. H. Sheppard (Eds.), Research on negotiation in organizations, vol. 5: 185-217. Greenwich, CT: JAI Press.

Sitkin, S. B., & Bies, R. J. (Eds.). 1994. The legalistic organiza- tion. Thousand Oaks, CA: Sage.

Sitkin, S. B., & Roth, N. L. 1993. Explaining the limited effec- tiveness of legalistic "remedies" for trust/distrust. Or- ganizational Science, 4: 367-392.

Sitkin, S. B., & Stickel, D. 1996. The road to hell: The dynamics of distrust in an era of quality. In R. M. Kramer & T. R.

Tyler (Eds.), Trust in organizations: 196-215. Thousand

Oaks, CA: Sage.

Smitka, M. 1994. Contracting without contracts: How the Jap- anese manage organizational transactions. In S. B. Sit-

kin & R. J. Bies (Eds.), The legalistic organization: 91-108. Thousand Oaks, CA: Sage.

Snow, C. C., Miles, R. E., & Coleman, H. J., Jr. 1992. Managing 21st century network organizations. Organizational Dy- namics, 20: 5-20.

Tyler, T. R. 1990. Why people obey the law. New Haven, CT: Yale University Press.

Williamson, 0. E. 1975. Markets and hierarchies. New York: Free Press.

Williamson, 0. E. 1993. Calculativeness, trust and economic organization. Journal of Law and Economics, 30: 131-145.

Zaheer, A., McEvily, B., & Perrone, V. In press. Does trust matter? Exploring the effects of interorganizational and interper- sonal trust on performance. Organizational Science.

Zucker, L. G. 1986. Production of trust: Institutional sources of economic structure 1840-1920. In B. M. Staw & L. L. Cum- mings (Eds.), Research in organizational behavior, vol. 8:

53-111. Greenwich, CT: JAI Press.

Denise M. Rousseau is the H. J. Heinz II Professor of Organizational Behavior and Public Policy at Carnegie Mellon University. Her research addresses the psychologi- cal contracts in organizations and changing employment relationships.

Sim B. Sitkin is an associate professor of management at the Fuqua School of Business at Duke University. His research examines the processes by which organizations and their members become more or less capable of change and innovation, focusing on the effect of formal and informal organizational control systems on trust, risk taking, learning, and innovation.

Ronald S. Burt is the Hobart W. Williams Professor of Sociology and Strategy at the University of Chicago. His general interest is the social structure of competition. His recent published research concerns social capital effects on manager performance, the gender of social capital, personality and social capital, the social context of reputation and trust between managers, and the rhetoric of distrust.

Colin Camerer is a professor at Cal Tech and teaches economics and psychology. He does research in behavioral economics, using data, mostly from experiments, to show how limited rationality in economic decisions, games, and markets affects outcomes. He is writing a book on game theory experiments.

This content downloaded from 155.247.166.234 on Tue, 16 Jan 2018 20:20:21 UTC All use subject to http://about.jstor.org/terms