|Copyright Annual Reviews, Inc. 1999|
Broadly construed, the field of organizational behavior is concerned with the study of organizations as complex social systems (Pfeffer 1997, Scott 1997). From a psychological perspective, organizational behavior theory and research examine the antecedents and consequences of human behavior-both individual and collective-within organizational settings (Katz & Kahn 1978, Murnighan 1993, Weick 1979). From its inception, a central concern of the field has been identifying the determinants of intraorganizational cooperation, coordination, and control (Arrow 1974, March & Simon 1958). Within the past few years, there has been a dramatic resurgence of interest among social scientists in exploring the role that trust plays in such processes (Coleman 1990, Fukuyama 1995, Kramer & Tyler 1996, Mayer et al 1995, McAlister 1995, Putnam 1993, Misztal 1996, Seligman 1997, Sitkin & Roth 1993). This burst of scholarly activity has been paralleled by equally earnest efforts to apply emerging trust theory to a variety of important organizational problems (Brown 1994, Carnevale 1995, Shaw 1997, Whitney 1994, Zand 1997).
Despite this intense interest and activity, few attempts have been made to assess the state of this rapidly growing literature or to draw out its links to the psychological literature on trust. A primary aim of this review, accordingly, is to survey some of the prominent themes and emerging perspectives on the nature and functions of trust within organizations. In particular, I summarize progress in conceptualizing trust, noting some of the more influential images of trust found in contemporary organizational research. Second, I describe recent research on the antecedents and consequences of trust in organizations. Finally, I discuss some of the barriers to trust that arise within organizations.
IMAGES OF TRUST IN ORGANIZATIONAL THEORY
Although social scientists have afforded considerable attention to the problem of defining trust (e.g. Barber 1983, Luhmann 1988, Mayer et al 1995), a concise and universally accepted definition has remained elusive. As a consequence, the term trust is used in a variety of distinct, and not always compatible, ways within organizational research. At one end of the spectrum are formulations that highlight social and ethical facets of trust. For example, Hosmer (1995) characterized trust as "the expectation... of ethically justifiable behavior-that is, morally correct decisions and actions based upon ethical principles of analysis" (p. 399). Other conceptions emphasize the strategic and calculative dimensions of trust in organizational settings. Thus, Burt & Knez (1996) defined trust simply as "anticipated cooperation" (p. 70), arguing that the "issue isn't moral... It is office politics" (p. 70).
Despite divergence in such particulars, most trust theorists agree that, whatever else its essential features, trust is fundamentally a psychological state.
Trust as a Psychological State
When conceptualized as a psychological state, trust has been defined in terms of several interrelated cognitive processes and orientations. First and foremost, trust entails a state of perceived vulnerability or risk that is derived from individuals' uncertainty regarding the motives, intentions, and prospective actions of others on whom they depend. For example, Lewis & Weigert (1985) characterized trust as the "undertaking of a risky course of action on the confident expectation that all persons involved in the action will act competently and dutifully" (p. 971). Similarly, Robinson (1996) defined trust as a person's "expectations, assumptions, or beliefs about the likelihood that another's future actions will be beneficial, favorable, or at least not deterimental to one's interests" (p. 576).
Other influential definitions construe trust as a more general attitude or expectancy about other people and the social systems in which they are embedded (Garfinkel 1963, Luhmann 1988). For example, Barber (1983) characterized trust as a set of "socially learned and socially confirmed expectations that people have of each other, of the organizations and institutions in which they live, and of the natural and moral social orders that set the fundamental understandings for their lives" (p. 164-65).
Although acknowledging the importance of these cognitive correlates of trust, other researchers have argued that trust needs to be conceptualized as a more complex, multidimensional psychological state that includes affective and motivational components (Bromiley & Cummings 1996, Kramer et al 1996, Lewis & Weigert 1985, McAlister 1995, Tyler & Degoey 1996b). As Fine and Holyfield (1996) noted along these lines, cognitive models of trust provide a necessary but not sufficient understanding of trust phenomena. They suggest that trust embodies also, aspects of the "world of cultural meanings, emotional responses, and social relations... one not only thinks trust, but feels trust" (p. 25).
Trust as Choice Behavior
Several organizational researchers have argued the usefulness of conceptualizing trust in terms of individuals' choice behavior in various kinds of trust dilemma situations (Arrow 1974, Kreps 1990, Miller 1992). One advantage of conceptualizing trust in terms of choice is that decisions are observable behaviors. Another is that organizational theorists possess a well-developed conceptual armamentarium for pursuing the theoretical and empirical implications of trust-as-choice (March 1994).
Within this literature, two contrasting images of choice have gained particular prominence, one that construes choice in relatively rational, calculative terms and another that affords more weight to the social and relational underpinnings of choice in trust dilemma situations.
TRUST AS RATIONAL CHOICE The rational choice perspective, imported largely from sociological (Coleman 1990), economic (Williamson 1993), and political (Hardin 1992) theory, remains arguably the most influential image of trust within organizational science. From the perspective of rational choice theory, decisions about trust are similar to other forms of risky choice; individuals are presumed to be motivated to make rational, efficient choices (i.e. to maximize expected gains or minimize expected losses from their transactions). Such models posit further, as Schelling (1960) noted, that choice is motivated by a "conscious calculation of advantages, a calculation that in turn is based on an explicit and internally consistent value system" (p. 4).
Hardin's (1992) conception of encapsulated trust articulates many of the essential features of this view. A rational account of trust, he notes, includes two central elements. The first is the knowledge that enables a person to trust another. The second is the incentives of the person who is trusted (the trustee) to honor or fulfill that trust. Individuals can trust someone, Hardin proposes, if they have adequate grounds for believing it will be in that person's interest to be trustworthy "in the relevant way at the relevant time" (p. 153). This notion of trust, he observes, is not predicated on individuals' narrow contemplation of their own interests but is enfolded instead in a sophisticated understanding of the other party's interests. "You can more confidently trust me," Hardin (1991) posits, "if you know that my own interest will induce me to live up to your expectations. Your trust then encapsulates my interests" (p. 189).
CRITIQUES OF RATIONAL CHOICE CONCEPTIONS Given its prominence as a conceptual platform from which much recent organizational theory and research proceeds, it is appropriate to note some of the concerns that have been raised about rational choice perspectives on trust. First, although the approach has proven enormously useful in terms of clarifying how individuals should, from a normative or prescriptive standpoint, make decisions about trust, it's adequacy as a descriptive account of how people actually do make such decisions has been questioned on several grounds. Most notably, a large and robust literature on behavioral decision making suggests that many of the assumptions of rational choice models are empirically untenable. Specifically, the extent to which decisions about trust, or any other risky decision for that matter, are products of conscious calculation and internally consistent value systems is suspect. As March ( 1994) cogently noted in summarizing such research, rational choice models overstate decision makers' cognitive capacities, the degree to which they engage in conscious calculation, and the extent to which they possess stable values and orderly preferences.
From a psychological perspective, another limitation of conceptions of trust grounded in presumptions regarding the rationality of choice is that they are too narrowly cognitive. Such conceptions afford too little role to emotional and social influences on trust decisions. As Granovetter (1985) aptly noted in this regard, such conceptions provide, at best, an undersocialized conception of trust. At a more fundamental level, March & Olsen (1989) take exception to the idea that notions of rational expectation and calculation are even central to the phenomenon of trust. The core idea of trust, they propose, is that it is not based on an expectation of its justification. "When trust is justified by expectations of positive reciprocal consequences, it is simply another version of economic exchange, as is clear from treatments of trust as reputation in repeated games" (p. 27).
RELATIONAL MODELS OF TRUST In response to these limitations and concems, a number of scholars have suggested that an adequate theory of organizational trust must incorporate more systematically the social and relational underpinnings of trust-related choices (Mayer et al 1995, McAlister 1995, Tyler & Kramer 1996). According to these arguments, trust needs to be conceptualized not only as a calculative orientation toward risk, but also a social orientation toward other people and toward society as a whole. The initial impetus for these relational models, it should be noted, was sociological theory and research on the impact of social embeddedness on economic transactions (Granovetter 1985). The development of relational conceptions of trust was further fueled by research implicating a variety of "macro-level" structures, including networks and governance systems, in the emergence and diffusion of trust within and between organizations (Burt & Knez 1995, Coleman 1990, Kollock 1994, Powell 1996).
Recent psychological research has extended this initial work by elaborating on the cognitive, motivational, and affective underpinnings of relational trust (Shapiro et al 1992, Sheppard & Tuckinsky 1996). Within social psychology, attempts to develop systematic frameworks for conceptualizing the nature, determinants, and consequences of relational trust have taken as a point of departure either social identity theory (Brewer 1981, Kramer et al 1996) or the group-value model (Tyler & Degoey 1996b, Tyler & Lind 1992). A common feature of these models is their broader emphasis on social rather than purely instrumental (resource-based) motives driving trust behavior, including consideration of how actors' self-presentational concerns and identity-related needs and motives influence trust-related cognition and choice.
Unresolved Questions and Enduring Tensions
Rational choice and relational perspectives on trust project fundamentally different images of trust and have pushed empirical research in quite different directions. To a large extent, however, the ongoing tensions between these perspectives owe more to their distinct disciplinary origins, than to inherent features of the organizational phenomena they seek to explain.
To reconcile these diverse views of trust, it is helpful to avoid thinking of the disparity between them as reflecting conflict between mutually incompatible models of choice (i.e. that trust is either instrumental and calculative or social and relational). Rather, a more useful approach is to move in the direction of developing a contextualist account that acknowledges the role of both calculative considerations and social inputs in trust judgments and decisions. In other words, what is needed is a conception of organizational trust that incorporates calculative processes as part of the fundamental "arithmetic" of trust, but that also articulates how social and situational factors influence the salience and relative weight afforded to various instrumental and noninstrumental concerns in such calculations.
Hardin (1992) provides one promising way of moving beyond this conceptual impasse. It is useful, he argues, to conceptualize trust as a three-part relation involving properties of a truster, attributes of a trustee, and a specific context or domain over which trust is conferred. From this perspective, strategic, calculative and instrumental considerations would be expected to exert a dominant influence in some organizational contexts (e.g. transactions involving comparative strangers). However, in other contexts (such as those involving members of one's own group), relational considerations might be more salient and exert more influence over how trust is construed. Fully elaborated, a three-part theory of trust would thus afford adequate attention to both the calculative and relational underpinnings of trust.
BASES OF TRUST WITHIN ORGANIZATIONS
Considerable theory and research has focused on identifying the bases of trust within organizations (Creed & Miles 1996, Lewicki & Bunker 1995, Sheppard & Tuckinsky 1996, Mayer et al 1995, Zucker 1986). Research on this question attempts to explicate antecedent conditions that promote the emergence of trust, including psychological, social, and organizational factors that influence individuals' expectations about others' trustworthiness and their willingness to engage in trusting behavior when interacting with them.
Ample evidence exists from both laboratory experiments and field-based research that individuals differ considerably in their general predisposition to trust other people (Gurtman 1992, Sorrentino et al 1995). Research suggests further that the predisposition to trust or distrust others tends to be correlated with other dispositional orientations, including people's beliefs about human nature (PEW 1996, Wrightsman 1991). To explain the origins of such dispositional trust, Rotter (1971, 1980) proposed that people extrapolate from their early trust-related experiences to build up general beliefs about other people. As expectancies are generalized from one social agent to another, he argued, people acquire a kind of diffuse expectancy for trust of others that eventually assumes the form of a relatively stable personality characteristic.
While acknowledging their existence, organizational theorists generally have not evinced much interest in such individual differences, except in so far as they might be reliably measured and used as a basis for screening and selection of more trustworthy employees (Kipnis 1996).
Research on trust development has shown that individuals' perceptions of others' trustworthiness and their willingness to engage in trusting behavior when interacting with them are largely history-dependent processes (Boon & Holmes 1991, Deutsch 1958, Lindskold 1978, Pilisuk & Skolnick 1968, Solomon 1960). According to such models, trust between two or more interdependent actors thickens or thins as a function of their cumulative interaction. Interactional histories give decision makers information that is useful in assessing others dispositions, intentions, and motives. This information, in turn, provides a basis for drawing inferences regarding their trustworthiness and for making predictions about their future behavior. Evidence of the importance of interactional histories in judgments about trust comes from a substantial body of experimental research linking specific patterns of behavioral interaction with changes in trust. For example, a number of studies have demonstrated that reciprocity in exchange relations enhances trust, while the absence or violation of reciprocity erodes it (Deutsch 1958, Lindskold 1978, Pilisuk et al 1971, Pilisuk & Skolnick 1968).
In noting the formative role that interactional histories play in the emergence of trust, these models draw attention to two psychological facets of trust judgments. First, individuals' judgments about others' trustworthiness are anchored, at least in part, on their a priori expectations about others' behavior. Second, and relatedly, those expectations change in response to the extent to which subsequent experience either validates or discredits them. Boyle and Bonacich's (1970) analysis of trust development is representative of such arguments. Individuals' expectations about trustworthy behavior, they posit, tend to change "in the direction of experience and to a degree proportional to the difference between this experience and the initial expectations applied to it" (p.130). According to such models, therefore, interactional histories become a basis for initially calibrating and then updating trust-related expectations. In this regard, history-based trust can be construed as an important form of knowledge-based or personalized trust in organizations (Lewicki & Bunker 1995, Shapiro et al 1992).
While personalized knowledge about other organizational members represents one possible foundation for trust, such knowledge is often hard to obtain. Within most organizations, it is difficult for decision makers to accumulate sufficient knowledge about all of the persons with whom they interact or on whom they depend. The size and degree of social and structural differentiation found within most organizations precludes the sort of repeated interactions and dense social relations required for the development of such personalized trust. As a consequence, "proxies" or substitutes for direct, personalized knowledge are often sought or utilized (Creed & Miles 1996, Zucker 1986). Recent research suggests there are several bases for such presumptive trust in others.
Third Parties as Conduits of Trust
Appreciating both the importance of information regarding others' trustworthiness and the difficulty in obtaining such information, Burt & Knez (1995) argued that third parties in organizations are important conduits of trust because of their ability to diffuse trust-relevant information via gossip. As they demonstrated in a study of trust among managers in a high-tech firm, gossip constitutes a valuable source of "second-hand" knowledge about others. However, the effects of gossip on trust judgments are complex and not always in the service of rational assessment of others' trustworthiness. Part of the problem, Burt & Knez theorized, is that third parties tend to make only partial disclosures about others. In particular, third parties often communicate incomplete and skewed accounts regarding the trustworthiness of a prospective trustee because people prefer to communicate information consistent with what they believe the other party wants to hear. Consequently, when a person has a strong relation to a prospective trustee, third parties tend to convey stories and information that corroborate and strengthen the tie, thereby increasing certainty about the person's trustworthiness. Thus, third parties tend to amplify such trust.
Uzzi's (1997) more recent study of exchange relations among firms in the New York apparel industry provides further evidence of the crucial role third parties play in the development and diffusion of trust. He found that third parties acted as important "go-betweens" in new relationships enabling individuals to "roll over" their expectations from well-established relationships to others in which adequate knowledge or history was not yet available. In explaining how this worked, Uzzi argued that go-betweens transfer expectations and opportunities of existing embedded relationships to newly formed ones, thereby "furnishing a basis for trust and subsequent commitments to be offered and discharged" (p. 48).
Category-based trust refers to trust predicated on information regarding a trustee's membership in a social or organizational category-information which, when salient, often unknowingly influences others' judgments about their trustworthiness. As Brewer (1981) noted, there are a number of reasons why membership in a salient category can provide a basis for presumptive trust. First, shared membership in a given category can serve as a "rule for defining the boundaries of low-risk interpersonal trust that bypasses the need for personal knowledge and the costs of negotiating reciprocity" when interacting with other members of that category (p. 356). Further, because of the cognitive consequences of categorization and ingroup bias, individuals tend to attribute positive characteristics such as honesty, cooperativeness, and trustworthiness to other ingroup members (Brewer 1996). As a consequence, individuals may confer a sort of depersonalized trust on other ingroup members that is predicated simply on awareness of their shared category membership.
Orbell et al ( 1994) investigated the effects of category-based trust on expectations and choice in a trust dilemma situation, using gender as a social category. Consistent with the notion that social perceivers possess category-based expectancies, they found that both male and female judges expected females to cooperate in prisoner's dilemma games more than males. However, females did not actually differ in their cooperation rates. Thus, this expectation was based more on gender categories than on actual gender differences between individual targets. Moreover, Orbell et al found that neither male nor female judges used gender to predict cooperation from particular individuals, or as a criterion for choosing whether to play the game with them. Based on these findings, Orbell et al argued that strong categoric expectations do not necessarily carry over to expectations about particular individuals in particular circumstances, suggesting how readily category-based expectancies are sometimes overridden by target-based expectancies.
Role-based trust represents another important form of presumptive trust found within organizations. As with category-based trust, role-based trust constitutes a form of depersonalized trust because it is predicated on knowledge that a person occupies a particular role in the organization rather than specific knowledge about the person's capabilities, dispositions, motives, and intentions.
Roles can serve as proxies for personalized knowledge about other organizational members in several ways. First, as Barber (1983) noted, strong expectations regarding technically competent role performance are typically aligned with roles in organizations, as well as expectations that role occupants will fulfill the fiduciary responsibilities and obligations associated with the roles they occupy. Thus, to the extent that people within an organization have confidence in the fact that role occupancy signals both an intent to fulfill such obligations and competence in carrying them out, individuals can adopt a sort of presumptive trust based upon knowledge of role relations, even in the absence of personalized knowledge or history of prior interaction.
Such trust develops from and is sustained by people's common knowledge regarding the barriers to entry into organizational roles, their presumptions of the training and socialization processes that role occupants undergo, and their perceptions of various accountability mechanisms intended to ensure role compliance. As numerous scholars (Barber 1983, Dawes 1994, Meyerson et al 1996) have noted, it is not the person in the role that is trusted so much as the system of expertise that produces and maintains role-appropriate behavior of role occupants. As Dawes (1994) suggested in this regard, "We trust engineers because we trust engineering and believe that engineers are trained to apply valid principles of engineering, moreover, we have evidence every day that these principles are valid when we observe airplanes flying" (p. 24).
As with other bases of presumptive trust, roles function to reduce uncertainty regarding role occupant's trust-related intentions and capabilities. Roles thus lessen the perceived need for and costs of negotiating trust when interacting with others. Relatedly, they facilitate unilateral acts of cooperation and coordination, even when other psychological correlates usually associated with trust are missing (Meyerson et al 1996, Weick & Roberts 1993). However, role-based trust also can be quite fragile and produce catastrophic failures of cooperation and coordination, especially during organizational crises or when novel situations arise which blur roles or break down role-based interaction scripts (Mishra 1996, Webb 1996, Weick 1993).
If trust within organizations is largely about individuals' diffuse expectations and depersonalized beliefs regarding other organizational members, then both explicit and tacit understandings regarding transaction norms, interactional routines, and exchange practices provide an important basis for inferring that others in the organization are likely to behave in a trustworthy fashion, even in the absence of individuating knowledge about them. Rules, both formal and informal, capture much of the knowledge members have about such tacit understandings (March 1994). Rule-based trust is predicated not on a conscious calculation of consequences, but rather on shared understandings regarding the system of rules regarding appropriate behavior. As March & Olson (1989) put it, rule-based trust is sustained within an organization "not [by] an explicit contract.. . [but] by socialization into the structure of rules" (p. 27). When reciprocal confidence in members' socialization into and continued adherence to a normative system is high, mutual trust can acquire a taken-for-granted quality.
Fine & Holyfield (1996) provide a nice illustration of how explicit rules and tacit understandings function to create and sustain high levels of mutual trust within an organization. Their study examined the bases of trust in the Minnesota Mycological Society, an organization that consists of amateur mushroom afficionados. This organization provides a rich setting in which to study the bases of trust for several reasons. First, the costs of misplaced trust in this organization can be quite severe: eating a mushroom that someone else in the organization has mistakenly declared safe for consumption can lead to serious illness and even, in rare instances, death. Given such risks, Fine & Holyfield note, credibility is lost only once unless a mistake is reasonable. Consequently, members are likely to be vigilant about assessing and maintaining mutual trust and trustworthiness. Second, because membership in the organization is voluntary, exit is comparatively costless. If doubts about others' trustworthiness become too great, therefore, members will take their trust elsewhere and the organization will die. Thus, the organization's survival depends upon its ability to successfully instill and sustain perceptions of mutual trustworthiness among its members.
Fine & Holyfield identified three important bases of trust within this organization, which they termed awarding trust, managing risk, and transforming trust. One way trust is created, they observed, is to award trust to others even when confidence in them may be lacking. For example, considerable social pressure is exerted on novices to consume dishes prepared by other members at banquets. As Fine & Holyfield put it, there is an insistence on trust. Thus, even if members remain privately anxious, their public behavior connotes high levels of trust. Collectively, these displays constitute a potent form of social proof to members that their individual acts of trust are sensible.
This insistence on trust is adaptive, of course, only if collective trustworthiness is actually in place. Accordingly, a second crucial element in the management of trust within this organization occurs through practices and arrangements that ensure competence and due diligence. This result is achieved partially through meticulous socialization of newcomers to the organization. Novices participate in these socialization processes with appropriate levels of commitment because it helps them manage the risks of mushroom eating and secure a place in the social order of the group. In turn, more seasoned organizational members teach novices out of a sense of obligation, having themselves benefitted from instruction from those who came before them. This repaying of their own instruction constitutes an interesting temporal (transgenerational) variant of depersonalized trust.
Over time, Fine and Holyfield argue, as members acquire knowledge about the organization, the nature of trust is transformed. Early on, the organization is simply a "validator" of trust for new members. Subsequently, however, it becomes an "arena in which trusting relations are enacted and organizational interaction serves as its own reward" (p. 29). As with trust in engineers, this form of trust is not simply trust in the expertise of specific individuals, but more importantly, trust in a system of expertise.
Another way in which rules foster trust is through their effects on individuals' self-perceptions and their shaping of expectations about other organizational members. As March (1994) observed in this regard, organizations function much like "stage managers" by providing "prompts that evoke particular identities in particular situations" (p. 72). Miller (1992) offers an excellent example of this kind of socially constructed and ultimately self-reinforcing dynamic. In discussing the underpinnings of cooperation at Hewlett-Packard, he noted that, "The reality of cooperation is suggested by the open lab stock policy, which not only allows engineers access to all equipment, but encourages them to take it home for personal use" (p. 197).
From a strictly economic perspective, this policy simply reduces monitoring and transaction costs. However, from the standpoint of a rule-based conception of trust-related interactions, its consequences are more subtle and farreaching. As Miller (1992) observes, "the open door symbolizes and demonstrates management's trust in the cooperativeness of the employees" (p. 197). Because such acts are so manifestly predicated on trust in others, they tend to breed trust in turn.
Rule-based practices of this sort can also exert subtle influences, not only on individuals' perceptions of their own honesty and trustworthiness, but also their expectations and beliefs about other organizational members' honesty and trustworthiness. As Miller notes in this regard, by eliminating time clocks and locks on equipment room doors at Hewlett-Packard, the organization builds a "shared expectation among all the players that cooperation will most likely be reciprocated" creating "a shared `common knowledge' in the ability of the players to reach cooperative outcomes" (p. 197). By institutionalizing trust through practices at the macro-organizational (collective) level, trust becomes internalized at the micro-organizational (individual) level. Thus, rulebased trust becomes a potent form of expectational asset (Knez & Camerer 1994) that facilitates spontaneous coordination and cooperation among organizational members.
Cross-cultural studies on trust provide further insights into the complexity and variety of these subtle links between rule-based understandings and expectational assets. Yamagishi & Yamagishi (1994) reviewed survey evidence that Japanese citizens often report lower levels of trust compared with their American counterparts. At first glance, this result is quite surprising. From the perspective of the widely-held view that Japanese society is characterized by close, stable, long-term social relations, one might expect that trust should be stronger within Japanese society. To resolve this anomaly, Yamagishi & Yamagishi proposed an important distinction between generalized trust and assurance. What characterizes Japanese society, they argue, is not generalized trust but rather mutual assurance. This mutual assurance is predicated on the stability of interpersonal and interorganizational relationships within the society. Because of this high degree of perceived stability, social uncertainty in transactions is greatly reduced. As Yamagishi & Yamagishi observe, individuals within such a society can count more strongly on ingroup bias for preferential treatment!
In American society, in contrast, no comparable sense of stability or assurance is readily available. As a result, social uncertainty in transactions tends to be greater. Thus, trust concerns among Americans revolve around reducing such uncertainty, and this is manifested in terms of greater reliance on personalized knowledge, reputational information about others, or both. The results of this study by Yamagishi & Yamagishi provide strong support for this argument.
BENEFITS OF TRUST
The ascension of trust as a major focus of recent organizational research reflects, in no small measure, accumulating evidence of the substantial and varied benefits, both individual and collective, that accrue when trust is in place. Perhaps most influential in this regard have been Putnam's (1993) provocative findings implicating trust as a critical factor in understanding the origins of civic engagment and its role in the development of democratic regimes in Italian communities. Fukuyama's (1995) more recent, panoramic characterization of the role trust plays in societal functioning has provided further impetus for arguments that trust constitutes an important source of social capital within social systems.
Although strictly speaking, Putnam's and Fukuyama's findings pertain to social systems writ large, their organizational implications have not gone unnoticed. Within organizational settings, trust as a form of social capital has been discussed primarily on three levels, including its constructive effects on
1. reducing transaction costs within organizations, 2. increasing spontaneous sociability among organizational members, and 3. facilitating appropriate (i.e. adaptive) forms of deference to organizational authorities.
Trust and Transaction Costs
In the absence of personalized knowledge about others, or adequate grounds for conferring trust on them presumptively, trust within organizations must be either individually negotiated or substitutes for trust located (Barber 1983, Kollock 1994, Sabel 1993, Shapiro 1987, Sitkin 1995, Sitken & Roth 1993). Even when effective, such remedies are often inefficient and costly. Recognition of this problem has led a number of theorists to focus on the role of trust in reducing the costs of both intra- and interorganizational transactions (Bromiley & Cummings 1996, Chiles & McMackin 1996, Creed & Miles 1996, Granovetter 1985, Uzzi 1997, Williamson 1993).
From a psychological perspective, one way in which trust can function to reduce transaction costs is by operating as a social decision heuristic. Social decision heuristics represent behavioral rules of thumb actors use when making decisions about how to respond to various kinds of choice dilemma situations they encounter (Allison & Messick 1990). The utility of such heuristics in trust dilemma situations is suggested by Uzzi's (1997) study, described earlier, of exchange relations among firms in the New York apparel industry. Uzzi found that trust in this setting operated not like the calculated risk of economic models, but more like a heuristic assumption that decision makers adopted "...a predilection to assume the best when interpreting another's motives and actions" (p. 43). As evidence of the heuristic quality of judgment and action in this setting, he noted an absence of formal monitoring or measuring devices for gauging and enforcing reciprocity. Instead, individuals spontaneously and unilaterally engaged in a variety of actions that helped solve others' problems as they arose. In interpreting these findings, Uzzi reasoned that "The heuristic character of trust permits actors to be responsive to stimuli" (p. 44). In this fashion, he noted, trust heuristics facilitate the exchange of a variety of assets that are difficult to put a price on but that mutually enrich and benefit each organization's ability to compete and overcome unexpected problems.
Recent research on the evolution of cooperation within complex social systems provides further evidence of the substantial benefits that accrue, at both the individual and collective level, from heuristic forms of trust behavior (Bendor et al 1991 Kollock 1993, Messick & Liebrand 1995, Parks & Komorita 1997). Viewed in aggregate, the findings from these studies suggest that heuristics that predicate presumptive trust (i.e. that are generous with respect to giving others the "benefit of the doubt" when "noise" or uncertainty regarding their trustworthiness is present) can produce substantial increases in both individual and joint payoffs-at least within the context of ecologies in which reasonable numbers of other trustworthy actors are present.
Trust and Spontaneous Sociability
Fukuyama (1995) argued that one of the most important manifestations of trust as a form of social capital is the spontaneous sociability such trust engenders. When operationalized in behavioral terms, spontaneous sociability refers to the myriad forms of cooperative, altruistic, and extra-role behavior in which members of a social community engage, that enhance collective well-being and further the attainment of collective goals (PEW 1996). Within organizational contexts, spontaneous sociability assumes many forms. Organizational members are expected, for example, to contribute their time and attention towards the achievement of collective goals (Murnighan et al 1994, Olson 1965), they are expected to share useful information with other organizational members (Bonacich & Schneider 1992), and they are expected to exercise responsible restraint when using valuable but limited organizational resources (Messick et al 1983, Tyler & Degoey 1996a).
Several empirical studies document the important role trust plays in people's willingness to engage in such behaviors. In an early study, Messick et al (1983) investigated the hypothesis that trust, operationalized in terms of individuals' expectations of reciprocity (i.e. their belief that if they cooperated, others would do so as well), would influence individuals' willingness to voluntarily reduce their consumption of a rapidly depleting common resource pool. In support of this prediction, they found that as individuals received feedback that collective resources were becoming more scarce, those who expected reciprocal restraint from others were much more likely to exercise restraint themselves. In contrast, those whose expectations of reciprocity were low displayed little self-restraint. Significantly, the behavior of low and high trusters did not diverge when resources were plentiful.
In a subsequent study, Brann and Foddy (1988) investigated the effects of interpersonal trust on individuals' cooperative response in a simulated social dilemma. They varied the rate at which a shared resource pool was declining, so that some study participants received feedback that the pool was experiencing only minimal decline, whereas others were led to believe it was facing depletion. Before participating in this task, participants' levels of interpersonal trust were measured by using Rotter's Interpersonal Trust Scale (Rotter 1980). Based on their responses, they were classified as either high or low trusters. Brann & Foddy found that, although low trusters' consumption behavior was unaffected by feedback about the state of the collective resource, high trusters showed differential response, consuming more when resource deterioration was minimal but less under conditions of increasing scarcity.
Parks, Henager, & Scamahorn (1996) examined how low- and high-trust individuals respond to messages of intent from other participants in a social dilemma. They found that low trusters reacted to a competitive message by decreasing cooperation, but were unaffected by a cooperative message. In contrast, high trusters reacted to the cooperative message by increasing cooperation but were unaffected by the competitive message. In a second study, they assessed differences in low- and high-trusters' response to inconsistent messages from others about cooperative intent. They found that a period of unconditional, message-consistent behavior immediately after stated intent can make low and high trusters responsive to cooperative and competitive messages, respectively. They also found, interestingly, that low trusters reacted particularly strongly to inconsistent messages about cooperative intent, rating an inconsistent person as less trustworthy compared with high-trusters' ratings of the same person.
In another study, Parks & Hulbert ( 1995) investigated the effects of trust on decision makers' responses to fear in social dilemma situations. Specficially, they compared individuals' choices in two different forms of social dilemma-a resource contribution (public goods) version of the dilemma and a resource restraint (commons dilemma) version. They used individuals' responses to Yamagishi's Trust Scale to separate study participants into groups of low and high trusters. They found that high trusters cooperated more than low trusters when fear was present, but cooperated at the same rate when fear was absent. However, the effects of fear within each dilemma were complex. Removing fear from the public goods version, they found, decreased high trusters' cooperation, while removing fear from the commons dilemma version increased low truster's cooperation. In discussing these findings, Parks & Hulbert used prospect theory to argue that low trusters may tend to frame zero payoffs in terms of negative reference points ("I didn't gain anything"), whereas high trusters frame comparable outcomes in terms of positive anchors ("I didn't lose anything").
In sum, the results from these studies demonstrate that trust enhances individuals' willingness to engage in various forms of spontaneous sociability, but in complex and often unexpected ways. In assessing the generalizability of this finding, it should be noted that there also exists evidence that the efficacy of trust for solving large-scale collective-action dilemmas of this sort may be limited. Survey results from a study of citizen community involvement (PEW 1996) show that, although trust is important in predicting civic engagement, trust alone is not always enough. Instead, people's level of participation in their communities depends also on perceptions regarding the efficacy of their actions. Additional evidence of possible limitations of trust comes from laboratory research by Sato (1988). Using a simulated social dilemma, she demonstrated that the effects of trust weaken as group size increases. To explain this finding, Sato argued that strategic considerations, including diminished perceptions of the impact of one's own actions on others, as well as diminished expectations about others' cooperativeness, reduce the perceived efficacy of trust as a collective becomes larger.
Trust and Voluntary Deference
Another important stream of organizational research has examined the relationship between trust and various forms of voluntary deference within hierarchical relationships in organizations. Although hierarchical relationships assume varied forms (e.g. leader-follower, manager-subordinate, and employer-employee), the centrality of trust within such relationships has long been recognized (Arrow 1974, Miller 1992).
From the standpoint of those in positions of authority, trust is crucial for a variety of reasons. First, as Tyler & Degoey (1996b) noted, if organizational authorities have to continually explain and justify their actions, their ability to effectively manage would be greatly diminished. Second, because of the costs and impracticality of monitoring performance, authorities cannot detect and punish every failure to cooperate, nor can they recognize and reward every cooperative act. As a result, efficient organizational performance depends on individuals' feelings of obligation toward the organization, their willingness to comply with its directives and regulations, and their willingness to voluntarily defer to organizational authorities. In addition, when conflict arises, trust is important because it influences acceptance of dispute resolution procedures and outcomes. Research has shown that individuals are more likely to accept outcomes, even if unfavorable, when individuals trust an authority's motives and intentions (Tyler 1994).
Recognizing its importance, researchers have investigated the conditions under which people are likely to attribute trustworthiness to those in positions of authority. Early research on this topic sought to identify specific attributes associated with perceived trustworthiness. For example, Gabarro (1978) found that perceived integrity, motives, consistency, openness, discreteness, functional competence, interpersonal competence, and decision making judgment contributed to attributions of trustworthiness between vice-presidents and presidents. Along similar lines, Butler (1991) found that perceived availability, competence, consistency, fairness, integrity, loyalty, openness, overall trust, promise fulfillment, and receptivity influenced subordinates' judgments of an authority's trustworthiness.
More recent social psychological research has refined and extended our understanding of the factors that influence trustworthiness attributions. The most systematic research on this topic has been conducted by Tyler and his associates (reviewed in Tyler & Degoey 1996b and Tyler & Lind 1992). This research identifies several important components of trustworthiness attributions. These include status recognition, which reflects the extent to which authorities recognize and validate individuals' sense of full-fledged membership in their organization, as well as trust in benevolence, which refers to individuals' beliefs that authorities with whom they deal are well intentioned and honest in their decisions. A third important factor is neutrality, which implies perceived fairness and impartiality in decisions. Another finding from this stream of research is that trust matters more in relationships when some sort of common bond exists between authorities and their subordinates.
Other research by Brockner and his associates has investigated the influence of procedural variables on attributions regarding authorities' trustworthiness. Brockner and Siegel (1996) noted that procedures are important because they communicate information not only about authorities' motivation and intention to behave in a trustworthy fashion, but also their ability to do so, a factor they characterize as procedural competence. In support of this general argument, they report evidence that procedures that are structurally and interactionally fair tend to increase trust, whereas lack of perceived structural and procedural fairness tends to elicit low levels of trust.
More recently, Brockner et al (1997) explored some of the conditions under which trust matters more or less. They argued that, all else being equal, trust matters more to individuals when outcomes are unfavorable. In explaining why, they note that receipt of favorable outcomes does not raise issues of authorities' trustworthiness, because the outcomes themselves constitute evidence that the authorities can be counted on to perform behaviors desired by the trustor. Under these circumstances, "...trust is neither threatened nor critical in determining support for authorities" (p. 560). In contrast, when outcomes are unfavorable, trust becomes more critical and authorities are unlikely to receive much support. Brockner et al tested this general prediction in three different studies and found, consistent with it, that trust was more strongly related to support for an authority when outcomes were relatively unfavorable.
BARRIERS TO TRUST
Although recognizing the importance of trust, and the diverse benefits that flow from it, organizational theorists have been cognizant also of the difficulties that attend the creation and maintenance of trust within organizations (Fox 1974, Sitkin & Roth 1993, Sitkin & Stickel 1996). However desirable trust may be, its purchase-to paraphrase Arrow (1974-is neither easy nor assured. In trying to understand why trust remains such an elusive and ephemeral resource within many organizations, researchers have focused attention on identifying psychological and social factors that impede the development of trust and that contribute to the fragility of ongoing trust relations.
Dynamics of Distrust and Suspicion
There has been widespread recognition among organizational theorists that distrust and suspicion are common and recurring problems within many organizations (Fox 1974, PEW 1998, Sitkin & Roth 1993). Distrust has been defined as a "lack of confidence in the other, a concern that the other may act so as to harm one, that he does not care about one's welfare or intends to act harmfully, or is hostile" (Grovier 1994, p. 240). Suspicion has been viewed as one of the central cognitive components of distrust (Deustch 1958) and has been characterized as a psychological state in which perceivers "actively entertain multiple, possibly rival, hypotheses about the motives or genuineness of a person's behavior" (Fein & Hilton 1994, p. 168).
Fein (1996) argued that suspicion can be triggered by a variety of circumstances, including situations where perceivers have forewarnings that another might be insincere or untrustworthy, in which their expectations have been violated, and when they recognize situational cues or possess contextual information that suggests another might have ulterior motives. Experiments by Fein and his associates (Fein 1996, Hilton et al 1993) have explored two competing predictions about the effects of suspicion on judgment. One possibility, they noted, is that information that makes perceivers suspicious might lead to a state of attributional conservatism, elevating their threshold for accepting behavioral information. In other words, once alerted to the possibility of deception, individuals may be predisposed to avoid a rush to judgment, remaining open to the prospect that there is more to a situation than meets the eye. A second and contrasting possibility is that suspicion triggers more sophisticated attributional analyses "characterized by active, careful consideration of the potential motives and causes that may influence people's behaviors" (Fein 1996, p. 1167). In general, the results of their studies demonstrate that suspicion evokes relatively active, mindful processing of attributionrelevant information, thereby supporting the attributional sophistication argument.
Taken together, the results of Fein et al's studies lend support to an image of social perceivers as "intuitive scientists" attempting to draw reasonable inferences about others' trustworthiness from available social data. Other recent research has highlighted the origins and antecedents of less rational forms of distrust and suspicion that may arise within organizations (Kramer 1998). A number of recent studies, for example, have identified a variety of factors that increase individuals' distrust and suspicion of others, including dispositional and situational factors that influence individuals' self-consciousness and perceptions of being under evaluative scrutiny in organizations (Fenigstein & Vanable 1992, Kramer 1994, Zimbardo et al 1981). As Fein and Hilton (1994) noted along these lines, contextual factors that elevate suspicion regarding an actor's motives can have negative consequences for both the perceiver and the suspected social actor. "Often through no fault of their own," they note, "actors may be the targets of others' suspicions because of the context in which their behavior occurred" (p. 171).
Related psychological research suggests that basic cognitive processes such as social categorization may heighten distrust and suspicion between individuals from different groups within an organization. Evidence for the existence of such category-based distrust was provided initially by ethnographic research and laboratory studies on ingroup bias (reviewed in Brewer 1981, 1996). These studies demonstrated that categorization of individuals into distinct groups often resulted in individuals' evaluating outgroup members as less honest, reliable, open, and trustworthy than members of their own group. These effects were observed, moreover, even when those group boundaries were based on arbitrary, minimal and transient criteria. Recent research by Insko, Schopler, and their associates on the discontinuity effect (reviewed in Insko & Schopler 1997) points to a similar conclusion. Viewed in aggregate, these studies suggest that mere categorization and perceived differentiation may create a climate of presumptive distrust between groups within an organization.
In addition to exploring the antecedents and consequences of distrust and suspicion within organizations, other research has examined the more general problem of distrust in organizations, including both public and private institutions (Brown 1994, Carnevale 1995, Nye et al 1997). As Pfeffer (1997) noted, from the moment we are born and until we die, such institutions exert a profound impact on the quality of our lives. Their ability to do so, however, depends in no small measure upon public trust in them, including trust in both their competence and their integrity (Barber 1983, Brown 1994).
Unfortunately, however important it may be, there is substantial evidence that trust in both public and private institutions has been declining for several decades (Carnevale 1995, Coleman 1990, Nye et al 1997, PEW 1996, 1998). For example, although 75% of Americans said they trusted the federal government in 1964, only 25% expressed comparable levels of trust in 1997. Similarly, trust in universities has fallen from 61 to 30%, medical institutions from 73 to 29%, and journalism from 29 to 14% (Nye 1997). Major private companies fare no better, trust in them having fallen from 55% to 21% over this same period (Nye 1997). Another indicator of the pervasiveness of institutional distrust and suspicion is provided by data regarding the frequency with which many Americans endorse various conspiracy theories and abuses of trust involving public institutions (Butler et al 1995, Goertzel 1994, Harrison & Thomas 1997, Pipes 1997).
Although data regarding their prevalence seem unequivocal, the sources of institutional distrust and suspicion remain much less clear. Researchers have advanced many different explanations for the decline in institutional trust, ranging from historical, economic, organizational, psychological, and sociological factors (for overviews, see Harrison & Thomas 1997, Nye et al 1997). A number of studies highlight the importance of unmet or violated expectancies in explaining why public trust in institutions has eroded. Nye (1997), for example, has noted that the decline of public trust in government can be attributed, at least in part, to its perceived failure to solve a variety of social ills. According to this hypothesis, government promises to remedy urgent social problems (e.g. to eradicate poverty, racial injustice, and catastrophic illnesses) led to heightened expectations that government would solve these problems. As these expectations have gone unfillfulled, so has trust diminished.
Pursuing the effects of violated expectancies on people's trust in instititutions, Zimmer (1972) argued that individuals, when making judgments regarding institutional trustworthiness, tend to overgeneralize from vivid, highly salient events involving institutions and their leaders. To investigate this hypothesis, he examined the impact of Watergate on public perceptions of trust in government. He noted that, before Watergate, public trust in Richard Nixon had been generally quite high (in fact, voters in one survey had rated Nixon as more trustworthy than either George McGovern or Edward Kennedy!). Zimmer theorized that subsequent revelations during the Watergate hearings demonstrated to people that their trust in Nixon had been misplaced. Such revelations would lead, he predicted, to a decrease in public trust, especially among those who had originally voted for Nixon. After all, he reasoned, the sense of violated trust should be especially acute among such individuals, resulting in greater generalized distrust of government. Zimmer found support for this hypothesis: People who had voted for Nixon showed the highest levels of subsequent distrust in government.
In interpreting these results, Zimmer speculated that people may use the behaviors of institutional leaders as reference points for gauging their basic beliefs about the state of society and as reality-testing mechanisms when appraising the trustworthiness of its institutions in general. In other words, people may draw general inferences about institutional trust from the behavior of highly visible role models. As a consequence, he reasoned, the behavior of public leaders while in office "may unknowingly or indirectly define reality in more ways and for more of the public than has been appreciated" (p. 749).
If unmet expectations and general beliefs about institutions do contribute to the erosion of trust, it is instructive to consider the sources of such expectations and beliefs. Cappella and Jamieson (1997) recently reviewed evidence implicating the media in the growing distrust and cynicism of the public towards its institutions. The framing of news, they argue, directly affects the public's mistrust of institutions. In particular, news stories that adopt strategic frames [i.e. frames that emphasize themes of "winning and losing and the self-interest implied by that orientation" and that activate negative actor traits such as those indicative of "artifice, pandering, deceit, staging, and positioning for advantage" (p. 85)] tend to promote greater mistrust and cynicism than more neutral, issue-oriented frames. To investigate this hypothesis, they conducted a series of carefully controlled experiments in which news involving public leaders and institutions was systematically framed in either strategic terms or in more neutral, issue-pertinent terms. The results from these studies support their argument that strategic frames produce greater mistrust and cynicism.
Although there is little doubt that trust in public institutions has declined substantially over the past several decades and remains disquietingly low today, it is important to note that the implications of this trend are far less evident than sometimes assumed. First, as Cappella and Jamieson (1997) have cogently noted, it is not clear whether observed levels of distrust and suspicion reflect unwarranted cynicism about the state of contemporary institutions or just realism. Second, even if real, the tangible (behavioral) implications of such distrust and suspicion are far from clear. As the PEW Report (1996) concluded, based on its survey results, although a "general distrust of others is an obvious social ill . . . its direct relevance to the way people act is unclear" (p. 7). Finally, although often portrayed in the popular press and social science literature largely in negative terms, distrust and suspicion may constitute appropriate and even highly adaptive stances toward institutions. Vigilance and wariness about institutions, some have argued, constitute essential components of healthy and resilient organizations and societies (Barber 1983, March & Olsen 1994, Shapiro 1987). From this perspective, distrust and suspicion may, in a very fundamental sense, constitute potent and important forms of social capital.
Technologies That Undermine Trust
An emerging area of organizational research, and one that is almost certain to become increasingly important over the next few years, concerns the relationship between technology and trust. Enthusiasm over technological remedies to trust-related problems has been considerable, as evidenced by the rapid infusion into the workplace of surveillance systems and other forms of electronic monitoring of employee performance. For example, according to Aiello (1993), over 70,000 U.S. companies purchased surveillance software between 1990 and 1992, at an cost of more than $500 million dollars.
Organizations typically adopt such technological remedies in the hope of enhancing employee trustworthiness (e.g. assuring compliance with regulations and deterring misbehavior). Ironically, there is increasing evidence that such systems can actually undermine trust and may even elicit the very behaviors they are intended to suppress or eliminate. In a recent discussion of this evidence, Cialdini (1996) identified several reasons why monitoring and surveillance can diminish trust within an organization. First, there is evidence that when people think their behavior is under the control of extrinsic motivators, intrinsic motivation may be reduced (Enzle & Anderson 1993). Thus, surveillance may undermine individuals' motivation to engage in the very behaviors such monitoring is intended to induce or ensure. For example, innocent employees who are subjected to compulsory polygraphs, drug testing, and other forms of mass screening designed to deter misbehavior may become less committed to internal standards of honesty and intregity in the workplace.
Monitoring systems can produce other unintended and ironic consequences with respect to trust. Hochschild (1983) described how fear of monitoring adversely impacted organizational trust and customer service among flight attendants at Delta Airlines. Flight attendants came to fear and distrust their passengers because of a policy allowing passengers to write letters of complaint about inflight service. No matter how justified or unjustified the complaint from the flight attendant's perspective, such letters would automatically end up in the attendant's file. The resultant climate of suspicion created by this policy was further intensified because flight attendants feared that "passengers" they were serving might not even really be passengers at all. Instead, they might be supervisors working undercover to monitor their performance. Thus, a system intended to produce more trustworthy (reliable) and friendly service encounters unintendedly produced fearful, suspicious ones.
As Cialdini (1996) also notes, monitoring and surveillance systems communicate to employees that they are not trusted, potentially breeding mistrust and resentment in return. When people feel coerced into complying with a behavior, they may resist the behavior when they think monitoring is imperfect and they can get away with it. Because of psychological reactance, even honest employees may try to cheat or sabotage monitoring systems.
Well-intentioned and highly committed organizational members may also feel impelled to cheat when regulations designed to increase trustworthy performance enforce perverse behaviors. Moore-Ede (1993), for example, described the unintended effects of a regulation requiring long-distance truckers to keep detailed logs of their driving time. This policy was implemented to increase driver and public safety by promoting compliance with laws mandating prudent limits on driving hours. Unfortunately, drivers who tried to adhere to these regulations often found themselves trying to sleep when wide awake and, worse, driving when tired. To circumvent these regulations, therefore, some drivers began keeping two sets of books-one for inspection purposes and one that tracked hours actually worked.
Other evidence suggests that the corrosive effects of surveillance extend to those doing the surveillance. Several studies have shown, for example, that the act of surveillance may increase distrust of surveillants over those they monitor (Kruglanski 1970, Strickland 1958). This result has been explained in terms of self-perception theory. Less obvious, but no less insidious in terms of their consequences, are the behaviors that surveillants don't engage in when surveillance and other substitutes for trust are utilized in organizations. As one executive who had implemented a computer monitoring system called Overview mused, "If I didn't have the Overview, I would walk around and talk to people more . . . I would be more interested in what people are thinking about" (Zuboff 1988, quoted in Kipnis 1996, p. 331). Thus, systems intended to guarantee trust may, ironically, not only make it more difficult for employees to demonstrate their trustworthiness, but also for authorities to learn about the distribution of trust within their organizations.
Breach of the Psychological Contract
Another promising stream of recent organizational research has examined the conditions under which initial trust in organizations unravels. Robinson (1996) examined the relationship between employees' trust in an organization and their perceptions of the extent to which the organization had either fulfilled or breached its psychological contract with them. She defined psychological contracts in terms of employees' beliefs regarding the terms and conditions of their reciprocal exchange relation with their employer (i.e. what they owed the employer and what the employer owed them). Psychological contract breach was characterized, in turn, as a subjective experience based on employes' perceptions that the organization had failed to fulfill its perceived obligations.
To investigate the relationship between trust and psychological contract breach, Robinson conducted a longitudinal study of newly hired managers, measuring their initial trust in the organization, as well as their trust levels at 18 and 30 months on the job. She found that initial trust in an employer was negatively related to subsequent perceptions of psychological contract breach. Specifically, individuals with high initial trust were less likely to perceive the psychological contract had been breached compared with those with low initial trust. She also found that prior trust moderated the relationship between psychological contract breach and subsequent trust, such that employees with low initial trust in their employer reported a greater decline in trust after perceived breach than employees with high initial trust. Finally, Robinson found a negative correlation between psychological contract breach and several important forms of employee contributions to the organization, including job performance, civic virtue (extra-role) behaviors, and intentions to remain with the organization.
The Fragility of Trust
Numerous scholars have noted that trust is easier to destroy than create (Barber 1983, Janoff-Bulman 1992, Meyerson et al 1996). To explain the fragility of trust, Slovic ( 1993) argued that there are a variety of cognitive factors that contribute to asymmetries in the trust-building versus trust-destroying process. First, negative (trust-destroying) events are more visible and noticeable than positive (trust-building) events. Second, trust-destroying events carry more weight in judgment than trust-building events of comparable magnitude. To provide evidence for this asymmetry principle, Slovic evaluated the impact of hypothetical news events on people's trustjudgments. In support of his general thesis, he found that negative events had more impact on trust judgments than positive events. Slovic noted further that asymmetries between trust and distrust may be reinforced by the fact that sources of bad (trust-destroying) news tend to be perceived as more credible than sources of good news.
In addition to these purely cognitive factors, researchers have been interested in how organizational factors, such as structural position in an organization, contribute to asymmetries in judgments regarding trust and distrust. Burt and Knez (1995) examined how network structures and the social dynamics they create differentially affect trust and distrust judgments. In the same study described earlier of managers in a high tech firm, they investigated the influence of third parties on the diffusion of distrust within managers' networks. They found that, although both trust and distrust were amplified by third-party disclosures, distrust was amplified to a greater extent than trust. As a result, judgments about distrust had, as Burt and Knez put it, a "catastrophic" quality to them. In explaining these findings, Burt and Knez posited that third parties are more attentive to negative information and often prefer negative gossip to positive information and gossip. Consequently, indirect connections amplify the distrust associated with weak relations much more than they amplify trust among strong relations.
Another study (Kramer 1996) investigated asymmetries in the construal of trust-enhancing versus trust-decreasing behaviors as a function of individuals' location within a hierarchical relationship. Specifically, this study examined how graduate students and their faculty advisors judged the level of trust in their relationship and the evidence they used in making those judgments. Using an autobiographical narrative methodology, students and faculty described the history of their interactions, recalling those behaviors that enhanced or undermined trust in the relationship. Content analysis of these narratives showed that individuals in low status positions (graduate students) tended to code more of their advisors' behaviors as diagnostic of trustworthiness compared with those in positions of high status (faculty advisors). To explain these findings, Kramer argued that, because of their greater dependence and vulnerability, trust concerns are more salient to individuals in low-status positions. As a consequence, they tend to be more viligant and ruminative about trust-relevant transactions. They also code more transactions as diagnostic of trustworthiness and can more easily recall instances of trust violation.
The empirical patterns observed in these studies are consistent with arguments by other theorists, such as Hardin (1992) and Gambetta (1988), that asymmetries in the accumulation of relevant experience by low and high trusters may differentially impact their opportunities to sample and learn from trust-building versus trust-eroding experiences. As Gambetta (1988) noted in this regard, distrust is very difficult to invalidate through experience, because it either "prevents people from engaging in the appropriate kind of social experiment, or, worse, it leads to behavior which bolsters the validity of distrust itself" (p. 234). Consequently, presumptive distrust tends to become perpetual distrust.
In assessing the state of social psychological theory at the end of the 1970s, Kelley (1980) once observed that the field read "more like a Sears Roebuck catalogue than like a novel" (p. 8). Athough social psychology contained an impressive listing of items of possible interest, he noted, there was still "no story with a plot, development of characters, and so on" (p. 8). Kelley's quip can be applied just as aptly to contemporary research on trust in organizations. Although there has been an impressive proliferation of middle-range theories about trust, an integrative theory of organizational trust continues to elude researchers. Relatedly, while empirical evidence continues to accumulate at a rapid rate, there has been a dearth of studies using overarching concepts and multiple-level measures that might help bridge the increasingly diverse conceptions of trust represented by economic, sociological, and social psychological perspectives.
If the glass remains half empty with respect to trust research, however, it is also half full. Not long ago, Luhmann (1988) was able to lament the existence of "a regretably sparse literature with trust as its main theme" (p. 8). Happily, that lament can no longer be uttered with the same force. Trust has rightly moved from bit player to center stage in contemporary organizational theory and research. Nor does Williamson's (1993) more recent assessment that trust remains a "diffuse and disappointing" (p. 485) concept in the social sciences seem as true today as it did only a few years ago. Recent research has sharpened our understanding of the complexity of trust in organizations and enhanced our appreciation of the myriad and often subtle benefits such trust confers. Future theory and research on trust will undoubtedly add to its stock in the organizational sciences.
Visit the Annual Reviews home page at http://www.AnnualReviews.org.[Reference]