Trust and the Decision to Outsource: Affective Responses and Cognitive Processes

Many of the various forms of cooperative strategy that firms are pursuing in today’s economy entail the placing of important business functions in the hands of a partner. This paper examines the role of trust in the decision by a producer to place the marketing function in the hands of another entity, namely a cooperative. Although others have studied the effect of what may be termed general trust on inter-organizational relationships, few have examined the antecedents of that trust. We propose a model in which affective responses and cognitive processes are precursors to a sense of general trust, which, in turn, influences the outsourcing decision. These affective responses and cognitive processes have both direct and indirect (mediated) effects on the decision to place an important function in the hands of another entity. Perceptions of partner expertise in the business function at hand and the perceived need for the focal firm to maintain control over that function are also considered in the model. The model is tested in a somewhat novel context: the decision of cotton producers to outsource the marketing of their cotton fiber. Using survey data gathered from the actual decision-maker, and structural equations modeling, we find that the inclusion of affective responses and cognitive processes in our model produces a richer explanation of the outsourcing decision. The differences between the effects of affective responses and cognitive processes have potentially important implications for managers engaged in cooperative strategies and for the scholars who study them. Corresponding author: Tel: + 1-801-422-4362 Fax: + 1-801-422-0539 Email: [email protected] Other contact information: J.L. Morrow, Jr. Phone: (205) 226-4827, Fax: (205) 226-3080, [email protected] M. Hansen and J.L. Morrow / The International Food and Agribusiness Management Review Vol 6 Iss 3 2003 41 Strategic alliances appear to be an increasingly important business model for conducting economic exchange. The strategic alliance form of organization is widely used throughout the food and agribusiness industry. The agricultural marketing cooperative may be viewed as one of the oldest, most well established forms of strategic alliance (Staatz, 1983; Vitaliano, 1983). Grocery retailers also use the cooperative form of organization to gain purchasing and distribution efficiencies. (Progressive Grocer, 2000). One important source of economic benefit to be gained from the strategic alliance model is outsourcing (Murray & Kotabe, 1999; Quinn & Hilmer, 1994; Venkatraman, 1997). Outsourcing through a strategic alliance typically involves the placing of a business function in the hands of a partner. Agricultural marketing cooperatives have been facilitating the transfer of the marketing function from the producer to the management of the cooperative, in varying degrees, for decades. More recently, a similar transfer has been observed in many other industries. For example, e-businesses often enter into alliances with logistics and shipping firms. In these alliances, the e-business places all the necessary logistics and shipping functions in the hands of their partners. The economic benefit of these alliances stems primarily from efficiency advantages of specialization. These advantages are revealed not only in the lowered costs of logistics and shipping but also in the increased ability of the e-business to focus on core competencies such as product development, marketing, financing, etc. Similarly, agricultural marketing cooperatives have allowed producers to focus their efforts on production. Entering into a strategic alliance is a significant decision for a firm. Scholars have argued that trust plays an important role in the adoption and management of the strategic alliance model (Barney & Hansen, 1994; Dyer, 1996, 1997; Gulati, 1995; Sako, 1992). Placing a business function, especially a critical one, in the hands of a partner is also a significant decision. In fact, some have argued that this decision may be a critical determinant in economic performance (Quinn & Hilmer, 1994; Venkatraman, 1997). It stands to reason that if trust were important in the initial decision to enter an alliance, it would also play a role in subsequent outsourcing decisions. The purpose of this paper is to develop and test arguments as to what the role of trust is in outsourcing decisions. Specifically, we examine the affective responses and cognitive processes that lead to a sense of general trust. We also examine how the relationship between trust and outsourcing is mediated by perceptions of competency and the need to maintain control over a business function. The research on trust in organizations, including strategic alliances, may be divided into two broad categories: 1) the development of trust, and 2) the effect of trust. This paper is an attempt to link these two broad categories. Several models of the development of trust have been offered by scholars: repeated ties (Gulati, 1995), affectand cognition-based trust (McAllister, 1995, high initial trust (cognition) (McKnight, Cummings, & Chervany, 1998); conditional, unconditional, distrust (Jones & George, 1998), exchange framework (Whitener, Brodt, Korsgaard, & Werner, 1998), weak-, semi-strong-, and strong form (Barney & Hansen, 1994). The definitions offered in the development of trust literature are usually specific to the M. Hansen and J.L. Morrow / The International Food and Agribusiness Management Review Vol 6 Iss 3 2003 42 model being developed. However, a common thread running through each one of these models, to varying degrees, is that of trust resulting from affective responses and cognitive processes. Whereas the definitions in the trust development literature tend to be relatively specific, the definition of trust used in much of the effect of trust literature is more general. For example, Brockner, Siegel, Daly, Tyler, & Martin (1997) use a working definition of trust offered by Mayer, Davis, and Schoorman (1995: 712): “The willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the trustor, irrespective of the ability to monitor or control the other party.” Brockner, et al., (1997) did elaborate on this definition so as to include the elements of risk, competency, and motivation. In the end, however, the notion of trust is very general in nature. For example in casual conversation, if a person is asked whether or not another individual can be trusted, the person asking and the person responding both understand what is meant by ‘trust’. In fact, Brockner, et al., (1997: 563) measured trust by asking simple, straightforward questions such as “I trust the management to treat me fairly.” We refer to such trust as general trust. Regardless of whether the definition of trust used is general or specific, the notion of an expectation about the behavior of another party seems to be a constant in the effect of trust literature (e.g., Barney & Hansen, 1994; Brockner, et al., 1997; Dyer, 1997; Sako, 1992; Zaheer, McEvily, & Perrone, 1998). Much of this research has focused on the cost savings associated with trusting relationships (Dyer, 1996,1997; Parkhe, 1993, Sako, 1992). As such, trust is generally viewed as an important element in minimizing transaction costs. However, some have also argued that trust may be an important element in expanding the set of potential revenue opportunities associated with alliances (Barney & Hansen, 1994; Dyer & Singh, 1998; Hansen, Hoskisson, & Barney, 1999). Outsourcing is one of the vehicles by which alliances may offer the dual benefits of cost reduction and revenue enhancement. We argue that general trust is an important element in the decision to outsource. However, we also argue that there are significant antecedents to general trust that must be considered in any robust explanation of the role of trust in the outsourcing decision. Nature of Outsourcing Outsourcing is largely a matter of degree for most organizations. For example, a firm could outsource its janitorial function and, it could also outsource key elements of its R&D function. Both would be outsourcing, and yet, the strategic relevance of the two decisions would likely range from trivial to critical, respectively. Certainly a company faces less risk in outsourcing janitorial services than it does in outsourcing an R&D function. Thus, the significance of the decision to outsource varies depending on what is being outsourced. This is not to say that decisions such as the janitorial outsourcing decision are not important, as such a decision may present significant cost savings opportunities. Although trust is likely a factor in both M. Hansen and J.L. Morrow / The International Food and Agribusiness Management Review Vol 6 Iss 3 2003 43 strategically relevant and strategically irrelevant decisions, the focus in this paper is on outsourcing decisions that are strategically relevant. Researchers have concluded that outsourcing definitely has its upside and its downside. Quinn (2000) argues that outsourcing is a strategic necessity in today’s innovation-dependent economy. He argues that no single firm can innovate as efficiently or as effectively alone as it could with the collaboration of other firms through outsourcing. Bettis, Bradley, and Hamel (1992), on the other hand, warn that outsourcing may also lead to industrial decline if done without the proper care. They point out that firms stand to lose important competencies over time that may be critical to strategic competitiveness. Once a firm has lost these competencies, the firm may be unable to regain these competencies to a degree that would confer competitive advantage. There is broad agreement in the literature that outsourcing done correctly has the potential to enhance a firm’s competitive position, but that outsourcing done incorrectly will almost certainly be detrimental (Bettis, et al., 1992; Murray & Kotabe, 1999; Quinn, 2000; Quinn & Hilmer, 1994). A key element in an outsourcing strategy is deciding which activities to outsource. Indeed the main problem with outsourcing noted by Bettis, et al., (1992) was that firms too often choose to outsource activities that are key to competitiveness. Quinn and Hilmer (1994: 43) suggest that firms follow a dual approach to outsourcing: 1) concentrate on core competencies and achieve definable preeminence, and 2) strategically outsource other activities for which the firms have neither a critical need nor special capability. This logic suggests that firms must understand the relevant marketplace and firm-specific capabilities in order to determine which firm competencies are or may become preeminent. Firms must then exercise discipline in deciding which activities to outsource and which activities to nurture and protect. Further discipline is required as firms choose suppliers to whom activities are to be outsourced. Given the potential benefits and pitfalls of outsourcing, the decision to outsource is one that deserves careful consideration. The decision is made even more salient by the fact that most supplier markets are imperfect (Quinn & Hilmer, 1994). This imperfection arises, in part, from the uncertainty associated with turning over an important business function to another entity. Assessments of the trustworthiness of the managers of the other entity play a crucial role in attenuating these uncertainty concerns. Trust helps facilitate exchanges in uncertain circumstances because of the expectation that supplier-firms will not exploit vulnerabilities. While we recognize that trust alone is not a sufficient condition, we argue that it is certainly a necessary condition of the outsourcing decision. As such, the development of trust in the outsourcing context merits explanation. M. Hansen and J.L. Morrow / The International Food and Agribusiness Management Review Vol 6 Iss 3 2003 44 The Development of General Trust While many scholars may agree on a fundamental definition of trust and the conditions necessary for trust to arise, a variety of conceptualizations of trust have been offered, particularly in terms of the nature of trust and its dimensionality. For example, among the different forms of trust suggested by Rousseau et. al. (1998) are deterrence-based trust, calculus-based trust, and relational trust. Sako (1992) categorized trust as being of three basic types: contractual, competence, and goodwill. Other conceptualizations view trust as having conditional and unconditional states (Jones & George, 1998) or weak, strong and semi-strong forms (Barney & Hansen, 1994). Several scholars have conceptualized trust as having cognitive and affective dimensions (e.g., Cummings & Bromiley, 1996; Lewis & Weigert, 1985; McAllister, 1995; Zaheer, McEvily, & Perrone, 1998). Empirically, McAllister (1995) found that interpersonal trust among members in an organization was both affect-based and cognition-based. The Cognitive and Affective Bases of Trust What are the antecedents of general trust? While Jones and George (1998) conceptualized trust as consisting of three states: conditional, unconditional, and distrust, they described both cognitive processes and affective influences in arriving at the different states of trust. Thus, trust develops from a process, or pattern, of thinking and feeling on the part of the trustor regarding the potential object of trust. Others have described this process and the resulting cognitive and affective trust (Cummings & Bromiley, 1996; Lewis & Weigert, 1985; McAllister, 1995; Zaheer, McEvily, & Perrone, 1998). Cognitive processes. McAllister (1995) argued that interpersonal trust is cognitionbased because individuals choose who they will trust and base this decision on what they believe are “good reasons” (Lewis & Weigert, 1985). The choice to trust and the search for “good reasons” suggest a cognitive process by which one determines that an individual, group or organization is trustworthy. Therefore, a key to understanding general trust is the recognition of the process by which individuals arrive at some assessment of the trustworthiness of another individual, group, or organization. In other words, cognitive processes are descriptive of how one develops “good reasons” that others are trustworthy. Thus, we define cognitive processes as a series of careful, methodical thought routines that culminates in a general belief that an individual, group or organization is trustworthy. This careful, methodical process involves the consideration of “empirical evidence” (Jones & George, 1998). That portion of general trust resulting from cognitive processes is therefore not an instantaneous phenomena, it develops only after an individual is able to cognitively process and assess the available evidence. The cognitive process base of general trust seems consistent with what Jones and George (1998) termed unconditional trust. Their unconditional trust state is reached only after an individual is able to develop confidence in others that is “backed up by empirical evidence” (1998: 537). Jones and George also argued that M. Hansen and J.L. Morrow / The International Food and Agribusiness Management Review Vol 6 Iss 3 2003 45 trust is based on attitudes, which can be viewed as “(1) the knowledge structures containing the specific thoughts and feelings people have about other people, groups, or organizations and (2) the means [process] through which they define and structure their interactions with others” (1998: 532). Thus, attitudes, which partially determine trust, may be developed as individuals engage in a cognitive process to assess the trustworthiness of others. Calculus-based trust as described by Rousseau, et al., (1998) is also consistent with a cognitive process base of general trust. They argued that calculus-based trust is the result of a rational choice that is made based on “credible information” concerning the intentions or competence of others. Thus, a rational, cognitive-based process is needed for individuals to gain “proof sources” that provide this credible evidence concerning the trustworthiness of others (Rousseau et. al., 1998). Finally, a cognitive-based process is also crucial to Barney and Hansen’s (1994) typology of semi-strong and strong form trustworthiness. In both of these forms of trustworthiness, one party assesses the trustworthiness of another party. The party weighs the evidence embedded in both the attributes of the transaction and the characteristics of the other party(s) to the transaction. In the case of semi-strong form trustworthiness, the cognitive process results in the determination that the costs of opportunistic behavior by a partner would outweigh the benefit of such behavior to the other partner, and, therefore, the other party may be trusted. Strong form trustworthiness suggests that a party to a transaction has come to the conclusion, after weighing the evidence of partner characteristics, that the other partner’s own moral development would likely prevent such opportunistic behavior. A sense of general trust develops as managers engage in the cognitive process of evaluating the available information about the trustworthiness of the managers of supplier-firms. While it is certainly the case that in some instances such cognitive processes result in lower general trust or perhaps even distrust, we argue that increasing levels of cognitive processes will result in increasing levels of general trust. Thus, H1: Cognitive processes will have a positive effect on the general trust that managers have for the managers of supplier-firms. Affective responses. Lewis and Weigert (1985) argued that trust also consists of an emotional base that is distinct from, but complementary to, its cognitive base. Thus, we define affective responses as the response an individual experiences based on one’s instincts, intuitions or feelings that culminate in a general belief that an individual, group or organization is trustworthy. While cognitive processes refer to how one develops “good reasons” that others may be trusted, affective responses refer to the “emotional bonds” that may result in trust between parties. Affective responses may lead to general trust because these emotional bonds may eventually provide the basis for trust (Lewis & Weigert, 1985; McAllister, 1995). Jones and George explained that the development of trust begins with one party “suspend[ing] belief that the other party may not be trustworthy” (1998: 536). This suspension M. Hansen and J.L. Morrow / The International Food and Agribusiness Management Review Vol 6 Iss 3 2003 46 takes place because of the initial absence of evidence to cognitively evaluate. We argue that during this suspension the parties to an exchange are relying on affective responses (e.g., instincts, intuitions, and feelings) to arrive at some determination of the trustworthiness of the other party(s). Affective responses in the very early stages of an exchange relationship may be likened to a “gut” or “knee-jerk” reaction based on the intuitive feelings generated by the experience with the other party. As the relationship progresses, these initial instincts, intuitions and feelings influence the cognitive processes by which perceptions of trustworthiness continue to develop. The affective states experienced during interactions with others influence the level or state of trust held among parties (Jones & George, 1998). While Jones and George argued that moods and emotions impact all types of trust, we suggest that the notion of affective responses is consistent with their conditional trust construct. They defined conditional trust as a state in which the attitudes of parties toward each other are favorable, thus they are willing to continue future interactions, as long as both parties exhibit appropriate behavior. Jones and George note that “sufficient positive affect and a relative lack of negative affect” act to reinforce the attitudes that lead to conditional trust (1998: 536). This suggests that conditional trust has a strong affect-based component. Affective responses are also conceptually similar to what Rousseau and her colleagues (1998) termed relational trust. Indeed, they acknowledged that because relational trust has a large emotional component, scholars often refer to this form of trust as affective trust. An element of affective trust is also present in Barney and Hansen’s (1994) typology of trustworthiness. The affective states experienced in dealing with a partner would certainly influence perceptions about the trustworthiness of that partner. Positive affect would serve to bolster perceptions that another partner possessed the type of character that would prevent opportunistic behavior (strong form trustworthiness). Negative affect, on the other hand, would likely cause partners to insist on contractual safeguards (semi-strong form trustworthiness). General trust may vary in the degree to which it is based on cognitive processes and affective responses. As suggested above, the balance between these two antecedents of general trust may shift depending on the experiences of the partners. For example, the trust that one has for a fellow airline passenger who is a complete stranger is largely a matter of affective response to initial appearances and impressions. However, as one visits with the fellow airline passenger and learns of accomplishments and experiences cognitive processes begin to influence the sense of trust felt. Likewise, the general trust that one has for colleagues at work may be affect-based because these individuals seem like “nice” people and everyone “gets along.” However, if one is considering pursuing a business venture with colleagues from work where the potential for opportunistic behavior is high, then the general trust that one has for colleagues is likely to be more informed by cognitive processes. The intensity of the affective response experienced by a person is dependent upon factors that are both internal and external to that person. Internal factors such as M. Hansen and J.L. Morrow / The International Food and Agribusiness Management Review Vol 6 Iss 3 2003 47 attitudes, likes and dislikes that have developed over time, and memories of past experiences strongly influence the affective response a person will have to any given situation. These factors lead to a reaction with little or no cognitive effort or process. External factors such as the appearance, language, and mannerisms of the party with whom the focal person is interacting also influence the affective response. The setting of the interaction may also influence the affective response of a person. Lighting, sound, and smell can play a role in determining the intensity of affective response. Managers who have a more intense affective response are more likely to develop higher levels of general trust. Of course, the affective response managers experience may also be negative, in which case low levels of general trust would likely result. However, we argue that overall, managers experiencing more intense affective responses will develop higher levels of general trust. Thus, we propose, H2: The affective response of managers will have a positive effect on the general trust felt for the mangers of supplier-firms. The Effect of General Trust on Outsourcing The primary effect of general trust on the outsourcing decision lies in the expectation that supplier-firm managers can and will perform the business function in question in a manner consistent with the interests of the focal firm. The role of trust in facilitating economic exchange is well documented (Barney & Hansen, 1994; Rousseau et. al., 1998; Williamson, 1993). A sense of trust for another economic actor reduces the inherent uncertainty surrounding any transaction. General trust as defined here, and trust as used in common language, entails an expectation as to both the competence and the character of another economic actor. If one were to lack confidence in the competence of another, one would probably have a low level of general trust, even though one might have confidence in the character of the other person. Thus, if a person holds a high level of general trust for another person it is assumed that the person has confidence in the other person’s competence and character. This confidence therefore reduces both ability-to-perform uncertainty and behavioral uncertainty (Williamson, 1993). The uncertainty reducing nature of general trust leads managers to favor supplierfirm managers who are judged to be trustworthy. Again, we do not argue that trust is the only factor influencing the outsourcing decision. Rather, we argue that general trust matters in the decision and that managers will tend to choose supplier firms whose management is trusted. Thus, H3: General trust will be positively associated with the decision to outsource. Expertise and The Need for Control In addition to the influence of general trust, two other important factors influence the decision to outsource, namely the perceived expertise of the supplier firm and M. Hansen and J.L. Morrow / The International Food and Agribusiness Management Review Vol 6 Iss 3 2003 48 the need for control on the part of the focal firm. As managers decide which activities to outsource these two factors create a tension. Managers may recognize that a supplier firm has more expertise in a certain business function than the focal firm. On the other hand, the managers may feel compelled to maintain control of that particular business function. The resolution of this tension may largely determine the success or failure of not only the outsourcing strategy but of the firm (Bettis, et al., 1992). Central to the outsourcing decision is the issue of whether or not an outside supplier can perform an activity better than the focal firm can. Indeed, Quinn and Hilmer (1994) would argue that identifying the best-in-world supplier of an activity, be it internal or external, is the first step in a successful outsourcing strategy. The routines used in making this important first-step decision likely vary from firm to firm. These routines may range from casual observation to intense study and research. Regardless of how the conclusion is reached, the perception that an outside supplier possesses more expertise than the focal firm in a particular business function bears a strong influence on the outsourcing decision. Managers who arrive at this conclusion would likely opt to outsource in the absence of any tension due to the need to control a function. As managers develop perceptions of supplier-firm expertise, cognitive processes and affective responses shape those perceptions. These cognitive processes and affective responses are likely very similar to those involved in developing general trust. We propose that general trust for managers of a potential outside supplier influences perceptions of expertise. In other words, managers of the focal firm develop some level of general trust for managers of a potential outside supplier that, in turn, affects the perception of the expertise of those outside suppliers. General trust felt for managers of outside suppliers will have a positive effect on the development of perceptions of expertise. Thus, we propose: H4: General trust is positively associated with perceptions of supplierfirm expertise. Obviously, managers who are convinced that supplier firms possess superior expertise would be inclined to choose an outsourcing strategy. However, the influence of general trust and the tension created by the need to maintain control of a business function may make the correct outsourcing decision less obvious. Owens Corning recently outsourced its expense management function to VIN.net on an ASP basis (Kearney, 2000). An important consideration for Owens Corning was the potential loss of control. Owens Corning determined that any loss of control in outsourcing was more than offset by the cost savings provided by VIN.net’s system. We suspect that for every outsourcing decision that gets written-up in a trade journal like Owens Corning’s did, there are many more decisions not to outsource that never become publicly known. The proportion of these decisions not to outsource that result from a need to control concern outweighing the benefit of supplier-firm expertise remains an empirical question beyond the scope of this paper. However, it seems reasonable to assume that many managers decide against M. Hansen and J.L. Morrow / The International Food and Agribusiness Management Review Vol 6 Iss 3 2003 49 outsourcing because the expected costs of loss of control exceed the expected benefits of putting a business function into the hands of an outside supplier. The expertise-control tension begs an intriguing question: Does control of a secondrate business function ever trump having that business function in the hands of the best-in-world supplier? The arguments of Bettis, et al., (1992) would suggest that firms should avoid losing competencies even if other suppliers may have an advantage. Quinn and Hilmer (1994) and Quinn (2000) suggest that firms should strive to be best-in-world in their own core competencies and outsource other activities to best-in-world suppliers. It is important to note that in the eight years between Bettis, et al., (1992) and Quinn (2000) rapid innovation has become increasingly relevant to strategic success. Therefore, it would appear that having business functions in the hands of best-in-world suppliers would generally offset any loss of control disadvantages in today’s innovation-intensive environment. Thus, we argue that: H5: Perceptions of supplier-firm expertise in a business function will have a stronger influence on the outsourcing decision than the focal firm’s need for control. The outsourcing decision model developed here is depicted in Figure 1. This paper focuses primarily on the antecedents of general trust and its effect on the outsourcing decision. Certainly there are other important relationship possibilities and influences on the decision to outsource. For example, the effects of cognitive processes and affective responses on perceptions of supplier-firm expertise are especially interesting. However, by linking the development of trust with the effect of trust on an important management decision this model bridges a gap in the trust literature.