\

Shankar Ganesan

Determinants of Ijong-Term Orientation in Buyer-Seller Relationships
Marketing managers must know the time orientation of a customer to select and use marketing tools that correspond to the time horizons of the customer. Insufficient understanding of a customer's time orientation can lead to problems, such as attempting a relationship marketing when transaction marketing is more appropriate. The author suggests that long-term orientation in a buyer/seller relationship is a function of two main factors: mutual dependence and the extent to which they trust one another. Dependence and trust are related to environmental uncertainty, transaction-specific investments, reputation, and satisfaction in a buyer/seller relationship. The framework presented here is tested with 124 retail buyers and 52 vendors supplying to those retailers. The results indicate that trust and dependence play key roles in determining the long-term orientation of both retail buyers and their vendors. The results also indicate that both similarities and differences exist across retailers and vendors with respect to the effects of several variables on long-term orientation, dependence, and trust.

ith increasing frequency, manufacturers and suppliers are reporting a change in the nature of buyer-seller relationships. It is not unusual to read that companies such as General Motors, Xerox, Black & Decker, Neiman-Marcus, and others are looking to their suppliers to help them achieve a stronger competitive position. Such a strong competitive position can be achieved only by developing a competitive advantage that can be sustained over long periods of time. A sustainable competitive advantage can be created by such factors as good merchandise, customer service, and efficient distribution systems. However, most firms overlook the sustainable competitive advantage that can be created through long-term relationships with their suppliers. For example, in a retailing environment, retailers (buyers) with long-term relationships can achieve a competitive advantage by receiving merchandise in short supply, information on new and best-selling products and competitive activity, best allowable prices, and advertising and markdown allowances. Similarly, vendors (sellers) with long-term relationships can achieve a competitive advantage by obtaining infonnation on best-selling products and competitive activity, better cooperative advertising, and special displays for their merchandise. A recent study (Noordewier, John, and Nevin 1990) indicates that relational elements (such as longterm orientation) enhance the performance outcomes in buyer-seller relationships. Anderson and Weitz (1992) refer
Shankar Ganesan is an Assistant Professor of Marketing at State University of New York in Albany. He is currently a visiting Assistant Professor of Marketing at Virginia Polytechnic Institute and State University in Blacksburg. The author thanks the Center for Retailing Education and Research at the University of Florida for sponsoring this project. He also thanks David Brinberg, Bart Weitz, and two anonymous JM reviewers for their helpful comments on drafts of this article.

W

to such a long-term orientation in a relationship as ' 'commitment" and indicate that mutual commitment results in independent channel members working together to serve customer needs better and increase mutual profitability. My first purpose is to examine the antecedents of longterm orientation of both retailers and vendors in an ongoing channel relationship. Vendors need to know the time orientation (short- or long-term) of a retailer to select and use marketing tools with time characteristics that correspond to the time horizons of the retailer. Insufficient understanding of a retailer's time orientation can lead to problems, such as attempting a relationship marketing when transaction marketing is more suitable. Previous studies of long-term orientation in channel relationships (Anderson and Narus 1990; Anderson and Weitz 1989, 1992) have concentrated mainly on the importance of transaction-specific investments (TSIs) in determining long-term orientation. The focus has been on developing long-term relationships through creating dependence and locking in customers by getting them to invest in transaction-specific assets. Though TSIs and dependence play an important role in affecting long-term orientation, this study indicates that they are not sufficient to explain it—trust is also a necessary ingredient. The reason that TSIs and dependence are not enough is that they both focus on present or existing conditions. This can result in the party with high dependence not getting, or perceiving that they will not get, a fair share of the pie of resources in the future. Firms with high dependence and specific assets will seek constantly to escape from this dependence. However, with trust, the focus is on future conditions. Trust is necessary for the perception of a fair division of the pie of resources in the future. Second, I explicitly identify the major dimensions of trust and investigate their effect on long-term orientation.
Determinants of Ljong-Term Orientation / 1

Journal of Marketing Vol. 58 (April 1994), 1-19

FIGURE 1 Model of Retailer's and Vendor's Long-Term Orientation

si-

Environmental Diversity Dependence of retailer on vendor

82-

Environmental Volatility

83-

/Transaction specific' ( investments by \ ^ ^ retailer

Perception of vendor^s dependenece on retailer

Retailei's long-termA orientation J

84-

Perception of specific investments by vendor ^

Vendor's credibility (trust) 1)3

85-

Reputation of the vendor

86-

Retailei's experience with the vendor | 6

Vendort benevolence (trust) r\4

87 •

Satisfaction with previous outcomes

So far, researchers in marketing (Anderson and Narus 1990; Anderson and Weitz 1989; Moorman, Deshpande, and Zaltman 1992, 1993) have treated trust as a unidimensional construct. However, research in interpersonal trust has shown that trust is a multidimensional construct. Such a multidimensional approach provides greater diagnosticity with respect to the effect of trust on long- and short-term orientation. For example, this study indicates that one dimension of trust, credibility, has a significant effect on longterm orientation, whereas the other dimensions do not. I also develop scales for the multiple dimensions of trust and explicitly test the validity of a multidimensional model of trust. Furthermore, I investigate various antecedents of trust in a buyer-seller relationship. For example, I look at the importance of TSIs as a signal of trust. Third, the empirical research reported here assesses the retailer and vendor perspectives separately on long-term orientation and discusses the commonalties and differences across the two perspectives. Finally, the channel context examined in this study is quite different from those studied in the past because the re-

tailing environment reflects a wide variety of dependence levels between the channel members. For example, retailers cany certain product lines for which they are highly dependent on their vendors, whereas in other product lines, retailers have several alternatives, resulting in low dependence on their vendors.

Conceptual Framework
The conceptual model specifying the antecedents of a retailer's (vendor's)' long-term orientation is presented in Figure 1.

Long-Term Orientation
A retailer's long-term orientation is the perception of interdependence of outcomes in which both a vendor's outcomes and joint outcomes are expected to benefit the re'TO make the hypotheses easier to read, henceforth the term retailer (vendor) will be replaced by retailer throughout the article. Thus, the a priori assumption is that retailers and vendors would behave in a similar manner.

2 / Journai of Mariceting, Aprii 1994

tailer in the long run (Kelley and Thibaut 1978). Retailers with a short-term orientation are concerned only with the options and outcomes of the current period, whereas retailers with a long-term orientation focus on achieving future goals and are concerned with both current and future outcomes. The difference between short- and long-term orientations also can be explained by the nature of interfirm exchange adopted by channel members. Firms with a shortterm orientation rely on the efficiencies of market exchanges to maximize their profits in a transaction, whereas firms with a long-term orientation rely on relational exchanges to maximize their profits over a series of transactions. Relational exchanges obtain efficiencies through joint synergies resulting from investment in and exploitation of idiosyncratic assets and risk sharing. Both orientations have the ultimate objective of maximizing the outcomes obtained by channel members and do not imply any altruistic motives on the part of channel members. Researchers in the area of organizational theory and economics (Williamson 1979) have suggested that governance structures can be arrayed on a continuum of relationalism, which ranges from market to relational exchanges (Macneil 1980). Recently, in the marketing literature, researchers (Dwyer, Schurr, and Oh 1987; Kaufmann and Stem 1988; Noordewier, John, and Nevin 1990) have attempted to define the characteristics of relationalism. One characteristic proposed by Noordewier, John, and Nevin (1990) is "Expectations of continuity of a relationship," which captures the probability of a futtire interaction between a retailer and vendor. This definition suggests that at one end (discrete transaction) there is a low probability of future interaction and at the other (relational transaction) there is high probability of future interaction. In contrast, the long-term orientation proposed here goes beyond mere probability to capture both the desire and utility of a buyer toward having a long-term relationship. This long-term orientation includes the element of future interaction proposed by Noordewier, John, and Nevin (1990) but also focuses on the desire of the parties toward a long-term relationship. It refers to an orientation toward a specific vendor, not a generalized orientation toward all vendors. Finally, a retailer's long-term orientation is distinct from the longevity of a relationship. Though duration of an existing relationship is likely to affect a retailer's long-term orientation, longevity in itself is not sufficient to capture the retailer's long-term orientation (Kelley 1983). In other words, a retailer's long-term orientation in an existing relationship rather than the length of the relationship seems to be a better indicator of closeness in relationships. Here, long- and short-term orientations are conceptualized as polar opposites on a single dimension. r/ie Effects of Trust and Dependence on LongTerm Orientation Trust and long-term orientation. Trust is the willingness to rely on an exchange partner in whom one has confidence (Moorman, Zaltman, and Deshpand6 1992). An important aspect of this definition is the notion of trust as a belief.

a sentiment, or an expectation about an exchange partner that results from the partner's expertise, reliability, and intentionality. The definition of trust proposed here reflects two distinct components: (1) credibility, which is based on the extent to which the retailer believes that the vendor has the required expertise to perform the job effectively and reliably and (2) benevolence, which is based on the extent to which the retailer believes that the vendor has intentions and motives beneficial to the retailer when new conditions arise, conditions for which a commitment was not made. Trust based on a partner's expertise and reliability focuses on the objective credibility of an exchange partner: an expectancy held by an individual that the partner's word or written statement can be relied on (Lindskold 1978). This dimension encompasses the consistency, stability, and control over the pattem of behavior exhibited. Benevolence focuses on the motives and intentions of the exchange partner. This dimension includes the qualities, intentions, and characteristics attributed to the partner rather than its specific behaviors (Rempel, Holmes, and Zanna 1985). Vendors who are concerned with the outcomes of a retailer along with their own will be trusted to a greater extent than vendors who are interested solely in their own welfare. It should be noted that the second type of trust can exist even when the objective credibility of vendors is less than perfect: Vendors may not act the way they say they will because of competing demands or situations beyond their control. In other words, vendors will be trusted if their actions are perceived as benevolent by the retailer (Lindskold 1978). In this definition of trust, the focus is on an individual's (retail buyer's) trust of the other individual (vendor's representative). A retailer's trust in a vendor affects the long-term orientation of a retailer in three ways: (1) it reduces the perception of risk associated with opportunistic behaviors by the vendor, (2) it increases the confidence of the retailer that short-term inequities will be resolved over a long period, and (3) it reduces the transaction costs in an exchange relationship. According to WiUiamson (1975, 1981), an understanding of transaction costs is central to the study of organizations. Transaction costs include the costs of reaching an agreement satisfactory to both sides, adapting the agreement to unanticipated contingencies, and enforcing its terms. Because of bounded rationality and the costs of writing, negotiating, and implementing a contract, a comprehensive contract involving a long-term relationship is not possible. At best, only incomplete contracting can be achieved. In such incomplete contracts, the hazards of opportunistic behavior are greater because the termination of the relationship cannot be achieved easily. The hazards of opportunistic behavior in long-term relationships can be mitigated or removed if there is trust between the two parties. Incomplete contracting in a trusting relationship means that the two parties agree to adapt to unanticipated contingencies in a mutually profitable manner. In such trusting relationships, retailers and vendors are likely to respond to inequities through solutions over the long run instead of short-term opportunistic behaviors. This suggests that trust reduces the

Determinants of Ljong-Term Orientation / 3

risk of opportunistic behavior in a long-term exchange relationship. Thus, when trust exists, retailers and vendors believe that long-term idiosyncratic investments can be made with limited risk because both parties will refrain from using their power to renege on contracts or use a shift in circumstances to obtain profits in their favor. Furthermore, trusting relationships are likely to have lower transaction costs because incomplete contracts are sufficient for running the exchange relationship. A retailer's trust in a vendor reduces the risk of opportunistic behaviors by the vendor, lowers transaction costs in the exchange relationship, and thus increases the likelihood of a long-term orientation toward the vendor. Credibility and benevolence will have similar effects on long-term orientation because both specific trusting behaviors and intentions are required to mitigate the perception of risk due to opportunistic behaviors. Hp Trust in a vendor's credibility and benevolence is positively related to retailer's long-term orientation. Dependence and long-term orientation. Dependence of a retailer on a vendor refers to a retailer's need to maintain the channel relationship to achieve desired goals (Frazier 1983). Heide and John (1988) indicate that dependence of a retailer on a vendor is increased when (1) outcomes obtained by the retailer from the vendor are important and highly valued and the magnitude of the exchange is high (sales and profit approach suggested by El-Ansary and Stern 1972), (2) outcomes obtained by the retailer exceed outcomes available to the retailer from the best alternative vendor (role performance and comparison level of outcome approaches suggested by Anderson and Narus 1984 and Frazier 1983), and (3) retailer has few alternative sources or potential sources of exchange (concentration of exchange approach suggested by Pfeffer and Salancik 1978). A retailer highly dependent on a vendor has several options to deal with the asymmetrical dependence (Emerson 1962): (1) maintaining status quo, (2) partial or complete withdrawal from the relationship, (3) formation of coalitions, (4) extension of the power network (similar to the notions of offsetting investments), and (5) status enhancement of the more powerful party. The first option is to maintain status quo such that the asymmetrical dependence relationship continues. Research in channel relationships indicates that asymmetrical dependence in an exchange situation makes one party susceptible to the power and influence of the other party (Beier and Stem 1969). The more powerful party is in a position to create more favorable terms of trade and divert the profit from the less powerful party. Anderson and Weitz (1989) point out that asymmetrical dependence relationships are less stable and likely to break up over time. Maintaining status quo is not a profitable action for the retailer over the long run. The second alternative involves replacing the current vendor. This assumes that the retailer has a reasonable alternative. Oftentimes, lack of alternatives is the primary cause of dependency. Therefore, dissolving the relationship is not a viable solution in many situations. Similarly, the third al4 / Journai of i\/lari(eting, Aprii 1994

ternative, formation of coalitions, is also not an option because of legal and economic restrictions. The fourth and fifth altematives seem to be the most feasible solutions to manage dependence. These altematives involve making TSIs in the relationship with the vendor and enhancing the status of the vendor in their relationship (making the vendor feel important). There are two types of TSIs that the retailer can make; they can be explained best by considering the simple vendor/retailer/customer (consumers who buy the specific vendor's products) link. The first one involves specific investments in the vendor/retailer relationship. This would include investing in special displays for the vendor's products and training salespeople to sell such products. However, organizational economics literature (Williamson 1979) has shown clearly that one-sided specific investments are open to exploitation because these specific assets cannot be safeguarded. One safeguard suggested by Heide and John (1988) deals with the investments in the retailer/customer relationship. Specifically, the retailer will reduce its dependence on the vendor by engaging in bonding behaviors with the vendor's customers for the vendor's lines. This includes developing personal relationships with the vendor's customers and creating an identity with those customers separate from the vendor's line. Developing a unique identity could involve personahzed services to their customers such as informing the customers of new products and designs. Conceptually, the actions of the retailer are designed to create exit barriers for the customers such that the customers would not go easily to another retailer to buy the vendor's products. The customers would incur switching costs such as lack of service and up-to-date information. Thus, a retailer's best approach to reduce the asymmetrical dependence involves making specific investments in the vendor and the vendor's customers. Such specific investments cannot be made unless the retailer has a long-term orientation toward the vendor. However, when retailers perceive the vendor as being dependent on them, they have little motivation to develop a strong, cooperative long-term relationship. The net benefits (benefits minus costs) provided by vendors who are highly dependent on the retailer are either marginally greater than or equal to the net benefits offered by altemative vendors. In such relationships, the costs of maintaining long-term relationships through long-term contracts exceed the benefits provided by them. This reduces a retailer's desire to initiate and develop a long-term relationship with such vendors. In many cases, even the long-term benefits of cooperation with marginal vendors are difficult to identify, resulting in very little incentive for the retailer to engage in a long-term relationship. H2: Dependence of a retailer on a vendor is positively related to the retailer's long-term orientation. H3: Perceived dependence of a vendor on a retailer is negatively related to the retailer's long-term orientation. Satisfaction and long-term orientation. Satisfaction of a channel member with past outcomes reflects a positive affectiye state based on the outcomes obtained from the relationship. Previous research in channel relationships has indi-

cated that satisfaction of a channel member is instrumental in increased morale, cooperation between channel members, few terminations of relationships, and reduced litigation (Hunt and Nevin 1974; Lusch 1976). Though numerous studies have investigated the determinants of channel member satisfaction (see Gaski 1984 for a review), few studies have investigated the impact of channel member satisfaction on their orientation toward a longterm relationship. However, studies in close relationships have found a significant relationship between satisfaction with outcomes and commitment to a relationship (Rusbult et al. 1991). Furthermore, research in the area of family decision making has indicated that individuals who are dissatisfied because of poor outcomes in the past are likely to focus on short-term gains and thus have a short-term orientation (Corfmann and Lehmann 1987). H^: A retailer's satisfaction with past outcomes is positively related to the retailer's long-term orientation. Antecedents of Trust Reputation and trust. Retailers and vendors provide signals of their future actions through their behaviors in other channel relationships. By making sacrifices and showing their concern for the other channel member, retailers and vendors develop a reputation for fairness within the industry (Anderson and Weitz 1992). Retailers who perceive the vendor to possess a reputation for fairness are more likely to trust the vendor. Vendors who have a reputation for terminating relationships and seeking high profits provide a signal to the retailer that the vendor is interested solely in its own interests, rather than being concerned with mutual welfare of the dyad. Such a negative reputation is likely to reduce the trust between the channel members. Reputation for fairness is likely to have a positive effect on a vendor's credibility but not on benevolence. Reputation for fairness is built on the edifice of reliable and consistent behavior over time. Such a reputation of effective performance is easily transferable across firms and enhances the credibility of the vendor. In contrast, benevolence is based on caring and making sacrifices for the channel partner. Such motives can be realized only through actual interaction, not word-of-mouth. H5: Reputation of a vendor is positively related to the retailer's perception of vendor's credibility. Satisfaction and trust. Research in equity and social exchange theory suggests that (in)equitable outcomes affect behaviors in subsequent periods (Adams 1965; Kelly and Thibaut 1978). In a continuing relationship, satisfaction with past outcomes indicates equity in the exchange. Such equitable outcomes provide confidence that parties are not being taken advantage of in a relationship and that both parties are concerned about the other's welfare in the relationship. In contrast, when individuals find themselves participating In inequitable relationships, they feel angry and resentful. Such feelings toward the other party due to dissatisfaction result in suspicion and mistrust of the channel partner. A retailer who perceives inequity in a relationship due to poor outcomes will be dissatisfied and likely to view the

vendor as untrustworthy and exploitative. Though the retailer still could continue such a relationship there is likely to be a reduced level of trust in the vendor. A retailer's satisfaction with outcomes will increase its perception of a vendor's benevolence and credibility. Satisfaction affects a vendor's benevolence because it indicates the vendor's concern for equitable outcomes and welfare of the retailer. Satisfaction with outcomes is also likely to be related to a vendor's credibility because it indicates effective performance of channel functions through greater reliability and expertise. Hg: A retailer's satisfaction with past outcomes is positively related to the retailer's perception of a vendor's benevolence and credibility. Experience and trust. Experience with a channel partner breeds trust. Scanzoni (1979) and Dwyer, Schurr, and Oh (1987) argue that as the experience with the vendor increases, the dyad is more likely to have passed through critical shakeout periods in the relationship. Such periods provide both parties with a greater understanding of each other and their idiosyncrasies. Thus, experience with the vendor is likely to increase a retailer's trust in the vendor's credibility and benevolence. H^: A retailer's experience with a vendor is positively related to the retailer's perception of the vendor's benevolence and credibility. Perception of vendor's specific investments and trust. A retailer's perception of a vendor's specific investments in a relationship provides a signal that the vendor can be trusted. Vendor TSIs include investments in people, lasting assets, and procedures. Examples of such investments are training the retailer's salespeople in merchandising, developing displays, providing dedicated electronic linkups for inventory control and ordering, and offering information on new products. These investments communicate strong commitment to the relationship. Such idiosyncratic investments are not hollow promises, because of the economic consequences that the other party will incur if the relationship ends. A retailer's perception of vendor TSIs increases both the vendor's credibility and benevolence. TSIs offer tangible evidence that the vendor can be believed, it cares for the relationship, and it is willing to make sacrifices through such investments. Hg: A retailer's perception of vendor TSIs is positively related to the retailer's perception of the vendor's benevolence and credibility. Antecedents of Dependence Uncertainty and dependence. Decision-making uncertainty is the degree to which an individual or organization cannot anticipate or accurately predict the environment (Pfeffer and Salancik 1978). Because different facets of uncertainty have opposite effects on channel structure and channel member behavior (Balakrishnan and Wemerfelt 1986), two dimensions of environmental uncertainty, volatility and diversity, are examined. These dimensions were found to

Determinants of Long-Term Orientation / 5

have strong effects on the decision processes of organizations (Leblebici and Salancik 1981). Environmental volatility refers to the extent to which market and demand changes are rapid. High volatility in a retail industry would reflect rapid fluctuations in customer demand and the inability to predict trends and future outcomes in specific markets (Klein, Frazier, and Roth 1990). Under conditions of high volatility, writing contracts that cover all unanticipated contingencies is difficult and costly. In such circumstances, vendors can take advantage of retailers by interpreting unspecified clauses in their contract to their own benefit. To avoid such opportunistic behaviors, retailers are likely to develop long-term relationships that permit sequential, adaptive decision making (John and Weitz 1988). These relationships are likely to increase the dependence of the retailer on the vendor. Environmental diversity reflects the extent to which there is uncertainty in the environment (cf. Aldrich 1979), for example, a retail market that includes many users of products, vendors, and competitors. A firm facing such a wide variety of market segments would have difficulty obtaining information and formulating effective strategic programs for each element of the market. Such diversity within a market will encourage a retailer to develop relationships with multiple channel partners capable of handling the demands of specialized markets. This will encourage the retailer to develop channel structures that are relatively flexible and temporary and thus reduce the dependence of a retailer on a vendor. H,: Environmental volatility is positively related to a retailer's dependence on a vendor. H,g: Environmental diversity is negatively related to a retailer's dependence on a vendor. TSIs and dependence. Transaction-specific assets are investments in durable assets that are highly specialized to the exchange relationship and not easily redeployable and have little salvage value in other relationships (Williamson 1981). Transaction-specific assets can create exchange difficulties for the investing party (Heide and John 1988); for example, an opportunistic exchange partner could appropriate some fraction of the value of the immobile assets because the investing party can no longer rely on the threat to switch to another supplier or vendor. In other words, immobile transaction-specific assets prevent credible threats because of switching costs. Therefore, the extent of dependence is a function of the magnitude of the specific assets. A retailer who has invested in specific assets has created barriers to exit from such a relationship. These relationships result in the retailer's inability to replace the vendor easily. Thus, retailer TSIs increase the dependence of the retailer on the vendor but decrease the perception of the vendor's dependence on the retailer. In contrast, a retailer's perception of vendor TSIs are likely to decrease the retailer's dependence on the vendor but increase the perception of the vendor's dependence on the retailer. H,,: Retailer TSIs are positively related to a retailer's dependence on a vendor and negatively related to the retailer's perception of the vendor's dependence on the retailer. 6 / Journal of Marketing, April 1994

H12: A retailer's perception of vendor TSIs are negatively related to a retailer's dependence on a vendor and positively related to the retailer's perception of the vendor's dependence on a retailer.

Method
The data used to test the model were collected in a mail survey of retail buyers and their vendors from six regional department store chains with annual sales ranging from $200 million to $800 million in 1990. The retail buyers and vendors dealt with a wide variety of products such as men's and women's clothing and accessories, silverware, perfumes, and jewelry. Every retail buyer taking part in the survey was solely responsible for the purchase and profitability of the products purchased. Most of the vendors were senior sales representatives or sales managers and served as liaisons for the vendor organizations. Therefore, key informant bias (Phillips 1981) is not a major problem in this study. The data were obtained in two separate surveys. In the first, retail buyers selected a vendor and answered questions dealing with that specific vendor. The questions focused on antecedent factors and characteristics of the specific relationship with the vendor. Once a response was obtained from the retailer buyer, a second questionnaire was sent to the vendor selected by the retail buyer with a cover letter indicating that a department store had offered their name as a supplier of merchandise and requested information on their relationship with the specific retailer. The survey sent to the vendor focused on the same antecedent factors and characteristics as the retailer questionnaire. Pretest Pretests involved exploratory interviews with retail buyers from two department store chains. The exploratory interviews indicated considerable interest in the issue at hand and revealed substantial diversity (short- and long-term) in buyer/vendor relationships. The pretests also involved sending a draft of the questionnaire to 14 retail buyers in two department stores and a follow-up interview for their feedback on the appropriateness and answerability of the questions. Finally, the pretests revealed retail buyers' proclivity toward selecting vendors with whom they had long-term relationships. To overcome this selection bias, buyers in the retailer survey were assigned randomly to conditions in which they were asked to select a vendor on the basis of two criteria: (1) they have a long-term (short-term) relationship with this vendor and (2) the vendor's product is very (moderately) important to their organization. This resulted in a 2 X 2 matrix with two factors: duration of the relationship and the importance of the product. The random assignment of the retail buyers to one of these four cells overcame the selection bias of the retail buyers and resulted in a diverse group of retailer/vendor relationships dealing with a wide variety of product categories. A few vendors selling merchandise to the two department stores were interviewed on telephone to obtain their feedback on the issue of buyer/vendor relationships. All the

vendor representatives interviewed showed considerable interest in participating in the survey. Respondent Soiicitation and Response Rate Eight retail department store chains were sent a letter of solicitation that included a brief description of the study, its purpose, and a sample copy of the questionnaire to be completed by retail buyers. Five of the firms agreed to participate in the study. They then were sent 30 questionnaires to be distributed throughout the buying department by a coordinator in each department store, resulting in a final sample of 150 retail buyers. A total of 124 retail buyers responded to the retailer survey (response rate of 83%). All the buyers provided the name and address of the vendor representative (key informant) who was most knowledgeable about the relationship. Coded surveys were mailed to all the vendor representatives with a cover letter explaining the purpose of the study and the confidentiality of the responses. A total of 52 vendor representatives responded to the survey (response rate of 42%). Out of the 124 retail respondents, 48 (39%) retail buyers responded to short-term relationships and 76 (61%) responded to long-term relationships with their vendors. Of the vendors, 21 (40%) responded to shortterm relationships and 31 (60%) responded to long-term relationships. A comparison of the percentages of short- and long-term relationships across retailers and vendors showed no significant difference. All retail buyers had dealt with the vendor for at least four months (one retail season), and 89% of them had dealt with the vendor for a minimum of one year. The minimum experience of the vendor with a retailer was one year. Deveiopment of Measures The first step involved developing multiple items for each construct. In developing the items, parallel wording was used for both the retailer and vendor and self-reported and perceived measures^ for all the constructs. For example, an item used to measure a retailer's long-term orientation, "Any concessions we make to help this resource will even out in the long run," was modified to "Any concessions we make to help this retailer will even out in the long-run." Finally, items representing the same constructs across retailers and vendors were checked for a common set of core items. Items idiosyncratic to the retailers and vendors also were included in the measures along with the common set of items. These items refiect the unique orientation associated with the roles of the buyer and seller. Item analysis and exploratory factor analysis were used to clarify the scales. Items showing high factor loadings and those not loading on multiple factors were retained. The resulting set of items was tested for unidimensionality and convergent and discriminant validity using a confirmatory factor analytic procedure (LISREL 7.16). The goodness of fit indices (GFI) and t-values associated with individual items were used to identify the final set of items repre^Though we use the terms self-reported and perceived to distinguish the retailer's viewpoint from retailer's perception of vendor's viewpoint, both measures reflect a retailer's perception.

senting the construct. Once unidimensionality was established, internal consistency was calculated using Cronbach's alpha. Measures All the measures used in the study are reported in Appendix A, as well as the means, standard deviations, and Cronbach's alpha for the measures. The correlation matrix for the measures is indicated in Table 1. Retailer's and vendor's long-term orientation.^ This scale captures the orientation of the retailers and vendors toward developing long-term relationships. Because extant scales for this construct were not available, items were generated through interviews with retail buyers and vendors. The seven Likert items representing a retailer's longterm orientation capture the focus of a retailer on long-term goals, long-run concessions, long-run profitability of the relationship, and concern for both own and vendor outcomes. A similar scale was developed to assess a vendor's longterm orientation toward the retailer. Retailer's dependence and perception of vendor's dependence. A retailer's dependence on the vendor was assessed through two measures. First, retail buyers were asked, "What percentage of the volume in this category is accounted for by this resource?" Second, they were asked to respond to eight items focusing on the extent to which the vendor was important to them on factors such as sales and product lines. A retailer's perception of a vendor's dependence was measured by a set of three items. A percentage of volume measure used for a retailer's dependence was not used on the basis of the suggestions of retail buyers during prestudy interviews. Vendor's dependence and perception of retailer's dependence. Similar to the retailer's dependence, a vendor's dependence is based on a set of items and a question on the percentage of volume in a product category accounted for by the retailer. Vendors were asked, "What percentage of the volume in this category is accounted for by this retailer?" The vendors also were asked to respond to a set of six items focusing on the extent to which the retailer was important to them on factors such as sales and product lines. These scores were standardized and added to the standardized volume measure to obtain an indicator of vendor dependence on the retailer, with higher scores indicating higher levels of vendor dependence. In contrast to the multiple measures used to assess retailer dependence, a single measure was used for vendor dependence."* Vendor perception of retailer dependence was measured by a set of three items focusing
'No objective time frame would be appropriate for defining the long- or short-term orientation because of the wide variety of products. Thus, we allowed the retailer or vendor to report what they considered a long-term orientation. ""Because of sample size limitations, a LISREL model was not utilized for the vendor side data. The LISREL procedure allows for multiple indicators for a single latent construct. Because an ordinary least squares regression procedure was used to test the hypotheses on the vendor side, multiple indicators of a construct were collapsed (through standardization) to form a single indicator.

Determinants of Long-Term Orientation / 7

TABLE 1 Means, Standard Deviations, and Correlations for Retaiier's and Vendor's Side
RLTO/ RD0V1/ RD0V2/ VDOR/ VCRED/ VBEN/ VLTO VD0R1 VD0R2 RDOV RCRED RBEN (1) (2) (3) (4) (5) (6) 1. Retailer's long-term orientation / Vendor's long-term orientation 2. Retailer's dependence on vendor (Measure 1) / Vendor's dependence on retaiier 3. Retaiier's dependence on vendor (Measure 2) / Vendor's dependence on retaiier 4. Vendor's dependence on retailer / Retailer's dependence on vendor 5. Vendor's credibility / Retailer's credibiiity 6. Vendor's benevolence / Retailer's benevolence 7. Environmentai diversity / Environmental diversity 8. Environmentai volatility / Environmentai volatility 9. Retailer transaction specific investments (TSI) / vendor TSI 10. Retailer's perception of vendor TSI / Vendor's perception of retailer TSi 11. Reputation of the vendor/ Reputation of the retailer 12. Retailer's experience with the vendor / Vendor's experience with the retailer 13. Retailer's satisfaction / Vendor's satisfaction 1.0 .17 .12 .49* REDV/ REVOL/ VEDV VEVOL (7) (8) TSIR/ TSIV (9) TSIV/ TSIR (10) VREP/ RREP (11) REXP/ VEXP (12) RSAT/ VSAT (13) .47*

.52**

.29

-.07

-.48**

.4**

.6*

.53"

-.10

.43

1.0

.21

.61**

-.16

.25

-.32*

-.29*

.21

.10

-.09

.14

.03

.20

.44

1.0

.08

-.08

.10

-.23

-.15

.08

.13

-.04

.36

-.16

.37

.39

.17

1.0

.24

.43"

-.23

-.54**

.39"

.33*

.11

.05

.24

.57 .58 .07 -.12 .19

.17 .24 -.17 -.06 .40

.00 .05 -.15 -.08 .32

.38 .47 .16 .01 .33

1.0 .63 .03 -.08 .09.

.50** 1.0 .10 -.05 .28

.11 .08 1.0 .32 -.06

-.23 -.24 .36" 1.0 -.05

.14 .23 -.10

.54" .38** -.09

.45** -.16 .27 -.07 -.15 .25 .05 -.28* .02 -.04

.51* .39* -.01 -.35* .03

- . 5 1 * * -.34* 1.0 .55**

.54

.22

.06

.45

.46

.49

.12

-.04

.28

1.0

.70**

.07

.40**

.45

-.14

-.02 .17

.27

.52 -.01

.26 .07

.04 -.04

-.25 .04

.13 .05

.38 .01

1.0 -.01

.15 1.0

.38 -.14

-.06

.12

.20

.53

.08

-.10

.18

.29

.36

.02

.43

.43

-0.12

1.0

1. The lower triangular matrix (in bold) provides the correiations for the retailer's side data and the upper triangular matrix indicates the correlations for the vendor's side. For example, the correlation between vendor's credibiiity (as perceived by the retailer) and retailer's dependence on the vendor (RDOV, measure 1) is .17, which is in row 5, column 2. Similarly, the correlation between retailer's credibility (as perceived by thevendor)andvendor'sdependenceontheretaiier-.16, whichisin row2, column5. 2. For the retailer's side, correlations greater than .18 are significant at .05 level and correlations greater than .24 are significant at .01 level The total number of respondents used in the listwise correlation matrix for the retailer's side is 120. 3. For the vendor's side, a painwise correlati9n matrix with 52 respondents was used. Correlations significant at .05 ievel are indicated by and correlations significant at .01 level are indicated by **.

on the extent to which the vendor was important to the retailer. Retailer's trust in vendor's representative. The two dimensions of a retailer's trust in a vendor's representative, vendor credibility and benevolence, were assessed through seven and five items respectively. All items representing a retailer's trust in the vendor's representative were subjected to exploratory factor analysis with a oblimin rotation. The items that loaded highly (greater than .40) on the two factors were analyzed further through a confirmatory factor analytic procedure using LISREL VII (Joreskog and Sorbom

1988). Though the overall chi-square test of the two-dimensional model was statistically significant (x^(53) = 89.27), the GFI of .90 and the root mean square residual of .047 suggested a good model fit. A unidimensional and three-dimensional model of trust also were tested. A chi-square difference test of both models with a two-dimensional model of trust suggested acceptance of the two-dimensional model of trust. Vendor's trust in a retail buyer. The two dimensions of

a vendor's trust in a retail buyer, retailer credibility and benevolence, were assessed through four and three items

8 / Journal of Marketing, April 1994

respectively. Confirmatory factor analysis suggested the acceptance of the two-dimensional model of the vendor's trust in a retailer. Retailer's and vendor's perception of environmental uncertainty. Separate measures for the two dimensions of environmental uncertainty were developed. Environmental volatility was assessed through five items for the retailer's side and three items for the vendor's side. Environmental diversity was assessed through two items for both sides. These items are adapted from Klein, Frazier, and Roth (1990). Retailer TSIs and perception of vendor's investments. The retailer's investments in their relationship with the vendor and their perception of the vendor's investment were assessed by four and three items respectively. The items address a range of investments, including trained salespeople, displays, and custom-made products. The items used to measure investments by the retailer and vendor are adapted from those used by Anderson and Weitz (1992). Vendor TSIs and perception of retailer's investments. The vendor's investments in its relationship with the retailer and its perception ofthe retailer's investment were assessed by four and three items respectively. Similar to the retailer's specific investments, these scales assess the extent of investments made by the vendor in displays, training programs, and exclusive products and the vendor's perception of the retailer's specific investments. Retailer's perception of vendor's reputation and vendor's perception of retailer's reputation. A retailer's perception of a vendor's reputation in the marketplace was assessed with four items that elicited the vendor's reputation for fairness, concern, and honesty. This scale has been used by Anderson and Weitz (1992) to measure the reputation of manufacturers and distributors. A vendor's perception of a retailer's reputation also was assessed with the same four items. Retailer's and vendor's satisfaction. A retailer's and vendor's satisfaction with the outcomes was measured using a four-item semantic differential scale. Retailers and vendors were asked to respond to the following: "Describe your feelings with respect to negotiation outcomes with this resource (retailer) in the past one year." The anchors were pleaseddispleased, sad-happy, contented-disgusted, and satisfieddissatisfied. Retailer's and vendor's experience. A retailer's and vendor's experience with each other was obtained by asking the following question: "How many years' experience do you have in buying (selling) merchandise from (to) this particular vendor (retailer)?"

structural coefficients. The effects of such error were modeled explicitly using a strategy suggested by Hayduk (1987). This involved constraining the theta delta and theta epsilon matrices to predetermined values corresponding to a priori determined levels of error variance. These estimates for the error variance are based on reliabilities obtained through coefficient alpha for each measure except the retailer's experience with the vendor, which was a singleitem measure. For retailer's experience, a reliability of .95 was chosen. The dependence of the retailer on the vendor was measured using two distinct measures. The dependence measure using a multiple item scale with a reliability of .94 was used to define the latent variable metric (i.e., the path in the lambda Y matrix was constrained to equal 1.0). The dependence measure using the volume percentage was the second measure. The residuals of the two latent dependence constructs (retailer dependence on vendor and perception of vendor dependence on retailer) and the latent trust constructs (vendor credibility and benevolence) are assumed to be correlated because of correlated errors. All the latent independent variables are assumed to be correlated. Model fit. The overall model fit was good. The chisquare was nonsignificant (x^(31) = 39.95, /? > .1), the GFI was large (.96), and the standardized residuals were generally small and nonsignificant. In Table 2,1 present the standardized path coefficients. Long-term orientation. All the factors hypothesized to affect a retailer's long-term orientation were significant except one: A retailer's perception of vendor benevolence was not found to be significantly related to the retailer's longterm orientation (P54 = .112, n.s.). However, a retailer's perception of vendor credibility was found to be significantly related to the retailer's long-term orientation ((353 = .297, p < .01). Thus, the results provide partial support for H,. The dependence of the retailer on vendor is positively related to long-term orientation 0 5 1 = .419, p < .01) and the retailer's perception of the vendor's dependence on retailer is negatively related to the retailer's long-term orientation (P52 = -.266, p < .1). The results support Hj and Hj. A retailer's satisfaction with past outcomes is significantly related to the retailer's long-term orientation (^57 = .323, p < .01) supporting H^. As hypothesized, the results indicate that a retailer's trust in a vendor, the extent of a retailer's and vendor's dependence in a relationship, and a retailer's satisfaction with the outcomes in a relationship play a significant role in determining a retailer's long-term orientation toward a vendor. These three variables explain 75.2% of the variance associated with a retailer's long-term orientation. Trust. The effect of a vendor's reputation on the vendor's credibility was significant (735 = .442, p < .01), but a vendor's reputation was not significantly related to the vendor's benevolence (745 = -.116, n.s.). These results offer support to Hj. They suggest that reputation for fairness can create credibility but not benevolence.

Analysis and Results
Retailer Perspective
The model in Figure 1 was evaluated using LISREL 7.16 (Joreskog and Sorbom 1988) with sample covariance matrix as the input matrix. All the measures are subject to measurement error that can introduce bias in the estimates of the

Determinants of Long-Term Orientation / 9

TABLE 2 Results of the LISREL Analysis for the Retailer's side structural model Parameters1 Estimates 7II 7I2 7I4 723 724 734 735 736 737 744 745 746 747 757 -.324 .119 .397 .257 .256 .588 .433 .442 -.049 -.135 .755 -.116 .039 .082 .323 .419 -.266 .297 .112 .36 .55 .49 .55 .75 T-Value -1.700 .735 2.977 1.723 2.038 4.105 3.286 3.740 -.590 -1.232 4.919 -.863 .418 .661 3.857 3.372 -1.685 2.671 .756 Measurement model Parameters ^x11 A,x22 Jlx33 3Oj(66
5LX77

Estimates .75 .838 .869 .818 .903 .975 .968 .864 .529 .844 .946 .885 .961 .209 .345 .509 .289 -.313 .398 .54 .521 .49

T-Value _

_ _ 4.565 _ _

P51 P52 P53 P54
R2T|1 R2TI2 R2TI3 R2TI4

Xy^^ A.y21 A,y32 A,y43 >.y54 Xy65 \i;12 \(>34 (t)12 ())17 . 10)

tailer TSIs had a significant positive effect on a retailer's dependence on the vendor (7I3 =.397, p < .01), and a retailer's perception of vendor's dependence (723 = .256, p < .05) providing partial support for H,,. Perception of vendor TSIs had a significant positive effect on retailer's dependence on the vendor (7I4 = .257, p < .1) and perception of vendor TSIs (724 = .588, p < .01), offering partial support to Hj2. These results offer mixed support for Hj, and H,2. Uncertainty and investment variables explain 36% ofthe variance in a retailer's dependence on the vendor and 55% of the variance in a retailer's perception ofthe vendor's dependence on the retailer. As predicted, retailer TSIs increase the retailer's dependence on the vendor and perceived vendor TSIs increase the vendor's dependence on the retailer. These results suggest that TSIs create barriers to exit in a relationship and therefore create dependence on the other party. In addition, the results suggest that retailer TSIs increase the vendor's dependence, and vendor TSIs increase the retailer's dependence. Though these results offer no support for the predicted direction of these relationships, they make intuitive sense. Any vendor TSI (even if not matched by reciprocal investments by the retailer) is likely to improve the outcomes obtained by the retailer in that relationship compared with the outcomes provided by alternate vendors in the marketplace. This is likely to increase the retailer's dependence. All the independent variables were tested for their indirect effects on a retailer's long-term orientation. None of the indirect effects were significant, suggesting that the effects of the independent variables on long-term orientation were mediated through the dependence and trust constructs. Vendor Perspective

Perceptions of specific investments by vendor affect vendor credibility (734 = .433, p < .01) as well as vendor benevolence (744 = .755, p < .01), providing full support to Hg. Specific investments made by the vendor provide a powerful signal to the retailer about the vendor's credibility and desire to make sacrifices by investing in assets that are not easily redeployable elsewhere. Contrary to previous studies (e.g., Anderson and Weitz 1989), a retailer's experience with the vendor did not have a significant effect on vendor credibility or benevolence. This suggests that creating trust could be related to actual behaviors within the specific relationship rather than length of the relationship. Thus, H^ is not supported. Satisfaction with past outcomes does not have a significant effect on a vendor's benevolence or on credibility, offering no support for Hg. Though surprising, the result may suggest an egocentric bias on the part of buyers. Buyers may view the outcomes and satisfaction as being due to their own efforts rather than the vendor's inputs. Dependence. The support for various hypotheses of uncertainty and specific investments on dependence is mixed. Environmental diversity had a significant negative effect on a retailer's dependence on the vendor as predicted by H,Q (7II = -.324, p < .1). However, environmental volatility did not have a significant impact on a retailer's dependence on the vendor (7I2 = .119, n.s.) as hypothesized in Hj,. Re10 / Journal of Marketing, April 1994

The model in Figure 1 for the vendor was tested through multiple regression procedure instead of LISREL. The vendor side data has a sample size of 52 respondents.' When a LISREL model is fitted with a small sample size, parameter estimates can be unstable (Fomell 1983). Furthermore, this problem can be compounded by a large number of observed variables (13) and a complex model. In view of this, standard ordinary least squares regression analysis was applied to the data. The regression limitation of one indicator per construct was handled by creating an additive index (after standardization of each indicator) for vendor dependence on retailer variable. Because the objective of this study is to compare and contrast a retailer's perspective with a vendor's perspective, all the hypothesized effects were tested for the vendor's side. The results are provided in Table 3. Long-term orientation. A vendor's long-term orientation was found to be significantly related to the vendor's per'TO test out if the sample of 52 provides adequate power, the power associated with all the results of vendor data set (Table 6) was assessed. For vendor's long-term orientation as the dependent variable (VLTO), with N= 52, a = .05, f2 = .35 or pop. R^ = .26 (assumption of large effect size), number of variables = 5, the power was .90. Cohen (1988) argues that this is an adequate power to make reasonable inferences about covariation. The assumption of large effect size is based on the R^ obtained from the retailer side data and previous research in this area (Anderson and Weitz 1992). Similarly, with a large effect size, the power for all the other dependent variables was .80 or greater.

TABLE 3

Results of the Regression Anaiysis for the Vendor's Side
Vendor's iong-term orientation (VLTO) Vendor's dependence on the retailer (DEPVOR) Perception of retailer's dependence on the vendor (DEPROV) Retailer's credibility (RCRED) Retailer's benevolence RBEN) Environmental diversity (EDIV) Environmental volatility (EVOL) Vendor's TSI (TSIV) Vendor's perception of retailer's TSI (TSIR) Reputation of the retailer (REPUT) Vendor's experience with the retailer (VENEXP) Satisfaction with past outcomes (SATOUT) F (dfn. dfd)
R2

Vendor's dependence on retailer (DEPVOR)

Perception of retailer's dependence on vendor (DEPROV)

Retailer's credibiiity (RCRED)

Retaiier's benevoience (RBEN)

.208 (2.189) .3022 (4.316) .3842 (5.971) -.218 (2.13) -.311 (3.95) -.167 (.772) .076 (.134) .046 (.068) .2981 (3.687) .167 (1.148) .3342 (4.233) .112 (.475) -.154 (1.623) .3102 (4.338) 7.143 (5.38) .484 .416 2.441 (4,39) 5.093 (2.49) .172 .138 .3082 (5.583) 7.3273 (4.44) .40 .35 .253 (1.789) -.075 (.157) .08 (.345) .3122 (4.216) 2.4891 (4.44) .185 .110

Adjusted R2 2 p < .05

.20 .12

3 p < .01 Note: Regression coefficients are standardized coefficients with F-values within parentheses.

ception of a retailer's dependence on the vendor (h = .302, p < .05) and not related to the vendor's dependence on the retailer (b = .16, n.s.). Though a vendor's perception of a retailer's dependence was significantly related to the vendor's long-term orientation, the result was opposite the predicted direction. Taken together, these results do not support Hj and Hj. A vendor's perception of a retailer's credibility was found to he significantly related to the vendor's long-term orientation (b = .384, p < .05) whereas a retailer's benevolence was not significantly related to the vendor's longterm orientation (b = -.22, n.s.). These results provide partial support for H, and are similar to those found in the retailer's perspective. Finally, satisfaction with outcomes was significantly related to a vendor's long-term orientation (b = .31, p < .05), supporting H^. Overall, a retailer's credibility and dependence on the vendor and satisfaction explained 42% of the variance associated with the vendor's long-term orientation.

Trust. The effect of a retailer's reputation on the retailer's credibility and retailer's benevolence was not significant (b = .11, n.s., and b = -.07, n.s.). Thus, Hj is not supported. These results suggest that, unlike a retailer's perspective, vendors do not value the retailer's reputation in the marketplace. As found in the retailer's perspective, a vendor's experience with the retailer did not significantly influence the retailer's credibility or benevolence, providing no support for H7. Satisfaction with past outcomes has a significant effect on both retailer credibility (b = .308, p < .05) and retailer benevolence (b = .312, p < .05), supporting Hg. This suggests that vendors may be tuned much more closely to the retailer's behavior in a relationship rather than using external cues to develop their trust in a retailer. Finally, perception of specific investments by a retailer affects vendor credibility (b = .334, p < .05) but not vendor benevolence ( b = .253, n.s.), offering partial support to Hg.

Determinants of Long-Term Orientation /11

Similar to the retailer's perspective, a retailer's investments send a powerful signal to the vendor regarding their trustworthiness and intentions. Dependence. As found in the retailer's perspective, the support for the hypotheses related to uncertainty and specific investments on dependence is mixed. Environmental diversity had a significant negative effect on a vendor's dependence on the retailer as predicted by HJQ (b = -.31, p < .1). However, environmental volatility did not have a significant impact on a vendor's dependence on the retailer (b = - .17, n.s.). Thus, H^ was not supported. Vendor TSIs did not affect a vendor's dependence on the retailer or a vendor's perception of a retailer's dependence on the vendor, offering no support for H,,. However, a vendor's perception of retailer TSIs had a significant positive effect on a vendor's perception of the retailer's dependence (b = .30, p < .1) but not on the vendor's dependence on the retailer. This provides mixed support for Hjj. Uncertainty and investment variable explain 20% of the variance associated with a vendor's dependence on the retailer and 17% on a vendor's perception of a retailer's dependence.

Discussion
The results from this study provide substantial support for the framework of Figure 1. Many of the factors that influence retailers also influence the vendors in their formation of long-term orientation, trust, and dependence. However, some variables have different effects across the two perspectives. These results challenge the traditional notion that all channel members behave in a similar fashion. If the results from this study are robust, it suggests that channel research should consider carefully the implications of channel member roles on various relationship factors. A summary of the similarities and differences across the two perspectives is provided in Table 4. Long-Term Orientation Both retailer and vendor perspectives indicate that longterm orientation is determined by the extent to which a retailer or vendor trusts the channel partner. In previous studies of long-term orientation in channel relationships (Anderson and Weitz 1989, 1992; Noordewier, John, and Nevin 1990), researchers have concentrated mainly on developing long-term relationships through dependence and locking in channel partners by getting them to invest in transactionspecific assets. Though the results in this study indicate that dependence (and indirectly TSIs) plays a role in determining long-term orientation, it is not sufficient to explain it. An element of trust is necessary for parties to have a longterm orientation. The reason that trust is such a necessary ingredient for long-term orientation is that it shifts the focus to future conditions. With trust, both parties believe that even under unanticipated contingencies, the pie of rewards will be divided in an equitable and fair manner. In contrast, longterm relationships bound by dependence and TSIs signify a forced collaboration. Both parties constantly search to reduce this dependence. Thus, I suggest that locking in chan-

nel partners through TSIs and dependence is not enough to develop a long-term relationship: Actions that foster a climate of trust are needed. Though the results suggest that trust is important in developing a long-term orientation, only one dimension of trust (credibility) is a significant predictor. Credibility is the belief that the other party would make commitments that are reliable and has representatives who perform the selling/ buying functions effectively. However, both retailer and vendor viewpoints indicate that the perception of benevolence is not a significant determinant of long-term orientation. Though surprising, this result suggests that retail buyers and vendors are trained to focus on objective evidence of reliability rather than motives of their channel partners. The results show that the effect of dependence on longterm orientation is affected by the channel role. Retailers are likely to have a long-term orientation with a vendor on whom they are dependent; that is, one who provides or is likely to provide critical and important resources. The approach taken by retailers is consistent with Pfeffer and Salancik (1978), who suggest increasing the control over critical resources to reduce dependence. One way to gain control over important and critical vendors is to be long-term oriented and improve the overall profitability of both parties through investments in the relationship. Such an approach eventually would reduce the asymmetry in dependence and increase mutual dependence. In contrast, retailers are likely to have a short-term orientation with vendors who are dependent on them. Retailers view a vendor's dependence on them as a liability for the vendor and use it to their own advantage. Vendors, however, consider the extent of a retailer's dependence as a key predictor of long-term orientation rather than the extent of their own dependence. In other words, vendors are likely to develop a long-term relationship with a retailer only if the retailer is highly dependent on them. This suggests that the vendors are more cautious in their effort to manage dependence. This is not surprising given the retail environment, in which power retailers such as WalMart and Kmart and category killers such as Toys 'R' Us have squeezed out small vendors. In summary, the results indicate that retailers are more likely to take a proactive approach toward managing their dependence, whereas vendors are more likely to take a reactive approach. Trust In previous studies on trust (Anderson and Weitz 1989), researchers have found that antecedents such as goal congruence, negative reputation, and length of the relationship affect trust. However, no researcher has investigated the effect of TSIs on trust. The results from this study indicate that retailers and vendors who perceived idiosyncratic investments by their channel partners believed their partners to be trustworthy. This suggests that TSIs in a relationship can provide strong signals regarding the channel member's trustworthiness. The value of such investments is enhanced when parties make one-sided commitments. However, though TSIs affected the perception of credibility, they did not affect the perception of benevolence.

12 / Journal of Marketing, April 1994

TABLE 4 Summary of Results
Nature of relationship A. Sinnilaritles 1. Vendor's (retailer's) credibility on retailer's (vendor's) longterm orientation 2. Vendor's (retailer's) benevolence on retailer's (vendor's) long-term orientation 3. Satisfaction with past outcomes and retailer's (vendor's) long-term orientation 4. Environmental diversity and dependence of a retailer on vendor (vendor on a retailer) 5. Environmental volatility and dependence of a retailer on vendor (vendor on a retailer) 6. Retailer's (vendor's) perception of transaction specific investments by a vendor (retailer) and perception of vendor's dependence on the retailer (retailer^s dependence on the vendor) 7. Retailer's (vendor's) perception of specific investments by the vendor (retailer) and vendor's (retailer's) credibility 8. Retailer's (vendor's) experience with the vendor (retailer) and vendor's (retailer's) credibility 9. Reputation of the vendor (retailer) and vendor's (retailer's) benevolence 10. Retailer's (vendor's) experience with the vendor (retailer) and vendor's (retailer's) benevolence B. Differences 1. Dependence of a retailer (vendor) on vendor (retailer) and retailer's (vendor's) long-term orientation 2. Perception of vendor's (retailer's) dependence on a retailer (vendor) and retailer's (vendor's) long-term orientation 3. Transaction specific investments by a retailer (vendor) and dependence of a retailer on a vendor (vendor on a retailer) 4. Transaction specific investments by a retailer (vendor) and perception of vendor's (retailer's) dependence on the retailer (vendor) 5. Retailer's (vendor's) perception of transaction specific investments by a vendor (retailer) and dependence of a retailer on a vendor (vendor on a retailer) 6. Reputation of the vendor (retailer) and vendor's (retailer's) credibility 7. Satisfaction with past outcomes and retailer's (vendor's) credibility 8. Perception of specific investments by the vendor (retailer) and vendor's (retailer's) benevolence 9. Satisfaction with past outcomes and retailer's (vendor's) benevolence positive, significant negative, significant positive, significant positive, significant positive, significant positive, significant nonsignificant positive, significant nonsignificant nonsignificant positive, significant nonsignificant nonsignificant HJ. partsup H3. partsup H,,, partsup H,,. notsup positive, significant nonsignificant positive, significant negative, significant nonsignificant positive, significant positive, significant nonsignificant positive, significant negative, significant nonsignificant positive, significant H,, supa H,. notsup H4. sup H,o. sup Hg. notsup H12. sup Retailer's side Vendor's side Support/ Nonsupport

positive, significant nonsignificant nonsignificant nonsignificant

positive, significant nonsignificant nonsignificant nonsignificant

Hg. sup HJ, notsup H5. sup Hg. notsup

nonsignificant

H,2. notsup

nonsignificant positive, significant nonsignificant positive, significant

H5. partsup Hg. partsup Hg. partsup Hg. partsup

°Sup indicates that a hypothesis is supported for both the retailer and vendor, notsup, indicates that a hypothesis is supported for neither the vendor nor the retailer, and partsup, indicates that a hypothesis is supported either for the retailer but not the vendor or for the vendor but not the retailer.

Determinants of Long-Term Orientation / 1 3

From a vendor's perspective, satisfaction with outcomes creates trust, whereas retailers rely, to a certain extent, on the reputation of the vendors in the marketplace. The effect of satisfaction and reputation show the divergence across the two perspectives. Reputation of the vendor had a significant effect on its credibility hut reputation of the retailer did not. This suggests that retailers and vendors could he influenced hy a different set of cues regarding the trustworthiness of their partners. Retailers could rely on external cues, such as reputation for fairness in the marketplace, whereas vendors could rely more on internal cues, such as actual interactions with the retailers. Satisfaction with past outcomes has a significant effect on trustworthiness of their partners for vendors hut has no effect for retailers. This suggests that vendors may he more conscious of their outcomes and resulting satisfaction from a relationship than retailers. The effect of satisfaction on trust for the vendors also could reflect a more cautious approach of the vendors. A vendor's perception of the retailer's trustworthiness could he related to the extent to which outcomes refiect equity in the exchange.

Dependence
Both perspectives suggest that uncertainty—in the form of diversity—reduces dependence of the channel memher. When retailers and vendors have a diverse market with many products, customers, and competitors, they are more inclined to reduce their dependence on a particular retailer or vendor and deal with multiple entities on a short-term hasis. Environmental volatility did not affect dependence for either retailers or vendors. The effect of specific investments on dependence is different across the two perspectives. From a retailer's viewpoint, specific investments hy retailers and vendors significantly increased the dependence of the retailer and the vendor on each other. This result supports the finding of Anderson and Weitz (1992) that specific investments by a manufacturer and distributor act as potent pledges in the channel relationship and have a positive effect on their commitment to the relationship. Because greater commitment implies greater dependence of channel members on their partners, I believe the results from this study are consistent with their study. For vendors, however, only the perceptions of retailer TSIs had a significant effect on a retailer's dependence on the vendor. This suggests a role bias in terms of how retailers and vendors perceive the specific investments of each other. Manageriai impiications From a managerial perspective, it is important to know the orientation of the customer (retailer) toward a vendor. Insufficient understanding of a customer's orientation can lead to problems, such as attempting to form a long-term relationship with a customer when a short-term relationship is more appropriate (Jackson 1985). Customers with a longterm orientation are not likely to change vendors, and their accounts are likely to be attractive to the vendor. The vendors will benefit from concentrating their sales and marketing efforts and developing a unique strategy for such an ac14 / Journal of Marketing, April 1994

count. In contrast, short-term oriented customers are likely to switch their patronage from one vendor to another and often will buy on the basis of short-term inducements such as price. The first benefit of the analysis of customer's orientation in relation to time is that it enables the vendor to select and use marketing tools with time characteristics that correspond to the time horizons of the customer. For example, a short-term tool such as price could be most appropriate for a customer with a short-term orientation, whereas technological investments and technical capabilities could be appropriate for long-term oriented customers. A second benefit of analyzing a customer's time orientation is that the vendors can use the insight from the analysis to change the customer's position along the spectrum of long- to short-term orientation. Though a variety of factors (e.g., product category, competitors, regulations) constrain the vendor's ability to change the customer's position on the time horizon, the vendor's actions can infiuence the customer's orientation. A major determinant of short- and longterm orientation is the extent to which customers trust the vendor. A vendor can affect the customer's time orientation by engaging in trust-enhancing hehaviors. Specifically, the vendor's ahihty to provide promised benefits reliably and effectively can change a customer's orientation. A second major determinant of a customer's orientation is the extent to which the customer is locked in with the vendor. Previous research has shown that vendor TSIs often induce substantial reciprocatory investment actions by customers and thus create exit barriers to both parties. For example, VF corporation, a manufacturer of jeans and lingerie, worked with WalMart to develop an artificial intelligent software program that automatically adjusted the assortment models. If a particular WalMart outlet sold, for example, more large sizes than the norm, VF's software changed the assortment accordingly. Such dedicated systems create mutual dependence, which shifts the focus of a customer (in this case, WalMart) toward a long-term orientation. Therefore, the key for managers is to understand the customer's time orientation and either develop a strategy suitable to that time orientation or change the customer's time orientation through TSIs and trust-enhancing actions. Limitations The results and implications of this study are somewhat constrained by the research method employed. Though the tests of the models yield several results that are consistent with the hypotheses, the cross-sectional design used limits the ability to rule out altemative causal inferences. For example, the model is predicated on the assumption that trust and dependence affect long-term orientation. Though the results support this general hypothesis, it is also conceivable that a reverse sequence of events is operating. Specifically, the long-term orientation of a channel member affects the trust and dependence in a relationship. Typically, most studies have conceptualized trust as a determinant rather than a consequence of relationship quality. For example, Anderson and Narus (1990) view it as a determinant of cooperation and level of conflict in a working relationship. Anderson and Weitz (1989) view trust as a deter-

minant of perceived continuity of relationship. Finally, Dwyer, Schurr, and Oh (1987) suggest that it is a key determinant of commitment in a relationship. Most studies in channels of distribution have viewed dependence as a determinant of a firm's behavior and strategic decisions. Pfeffer and Salancik (1978) argue that because dependence is created by trade partners who provide critical and important resources, it is important to manage and control such resources. Heide and John (1988) suggest different ways of balancing dependence in conventional channels. Thus, similar to trust, most studies have viewed dependence as a determinant of long- or short-term relationships rather than a consequence of such relationships. Furthermore, I examine only one dimension of longterm relationships, that is, the long-term orientation. In particular, the study does not include measures of relational norms (Dwyer, Schurr, and Oh 1987; Kaufmann and Stem 1988; Macneil 1980). An interesting extension of the conceptual framework would include several dimensions of longterm relationships. Another limitation of the study is the small sample of vendors and potential self-selection bias. Though the proportion of short- and long-term relationships for vendors is similar to that of retailers, the small sample size affects the stability of the parameter estimates. Therefore, the results obtained from the vendor data should be interpreted with caution. I utilize a single source for retailers and vendors for all measures. As a result, the study does not eliminate the possibility of a common method bias resulting from the use of a single source. Using multiple sources in the data collection would improve the reliability of the measures and increase the confidence in the results. Finally, the sample of retailers from department store chains may limit variation on some dimensions and reflect considerable market power on the part of retailers. Further studies should attempt to collect data from retailer/vendor relationships that include smaller retailers and their vendors. Directions for Further Research One interesting issue that should be investigated is the decision to develop a long-term orientation toward a retailer or vendor. The results from this study suggest that developing a long-term orientation requires substantial sacrifices, such as increased specific investments in the relationship. When should a retailer or vendor develop such a long-term

orientation? What type of channel partner should be selected for such a long-term commitment? Addressing these issues will help retailers and vendors search for channel partners who provide a competitive advantage in an increasingly competitive environment. A second issue is identifying the advantages of developing a long-term orientation. Does such a long-term orientation in a retailer-vendor relationship result in a strategic advantage? Another related issue is "first mover" advantages in such long-term relationships: Do retailers and vendors who develop a long-term orientation with their partners first in a category obtain substantial benefits from such a move? This study indicates that retailer and vendor credibility plays an important role in the development of a long-term orientation. There are several issues regarding the implementation of trust. How does one get channel members to trust other parties? What skills are necessary to develop credibility in retailer/vendor relationships? How should retailers and vendors identify channel partners who participate in trust-creating behaviors? Another issue is the examination of asymmetry in longterm orientation. Recent research (Buchanan 1992) has indicated that the degree of asymmetry in dependence is likely to affect retailers' profitability. Similarly, the degree of asymmetry in long-term orientation could lead to mismatches between retailers and vendors. Such mismatches can result in heightened conflict, dissatisfaction, and eventual termination of the relationship. How should parties interested in a long-term orientation signal to their channel partners their commitment without being taken advantage of? Such issues can be addressed better by longitudinal studies that highlight evolving dynamics in a relationship not apparent in cross-sectional studies. This study focuses on dependence and trust as the key to long-term orientation. Other approaches such as social exchange theory that capture the importance of alternatives (Anderson and Narus 1984) also could affect the long-term orientation. Another issue that should be investigated is the perceptual convergence between a retailer's and vendor's longterm orientation. As found in this study, retailers and vendors differ in terms of the importance attached to various constructs. Further research should involve looking at the consequences of perceptual convergence or divergence related to constructs such as long-term orientation. They can provide insights into the success of channel relationships.

APPENDIX A Measures Used for the Retailer's and Vendor's Side
Retailer side measures A. Retailer's iong-term orientation 1. We believe that over the long run our relationship with this resource will be profitable. 2. Maintaining a long-term relationship with this resource is important to us. 3. We focus on long-term goals in this relationship. 4. We are willing to make sacrifices to help this resource from time to time. Vendor side measures A. Vendor's long-term orientation 1. We believe that over the long run our relationship with the retailer will be profitable. 2. Maintaining a long-term relationship with this retailer is important to us. 3. We focus on long-term goals in this relationship. 4. We are willing to make sacrifices to help this retailer from time to time.

Determinants of Long-Term Orientation / 1 5

APPENDIX A (Continued)
Retailer side measures A. Retaiier's long-term orientation 5. We are only concerned with our outcomes in this relationship. (R) 6. We expect this resource to be working with us for a long time. 7. Any concessions we make to help out this resource will even out in the long run. Cronbach's alpha: .94 Mean: 5.49, Standard Deviation: .93 B. Dependence of retaiier on vendor Measure 1 1. If our relationship was discontinued with this resource, we would have difficulty in making up the sales volume in our trading area. 2. This resource is crucial to our future performance. 3. It would be difficult for us to replace this resource. 4. We are dependent on this resource. 5. We do not have a good alternative to this resource. 6. This resource is important to our business. 7. This resource's product lines are essential to round out our product offering. 8. If our relationship was discontinued, we would have difficulty replacing this resource. Cronbach's alpha: .94 Mean: 4.62, Standard Deviation: 1.29 Measure 2 1. What percentage of the volume in this category is accounted for by this resource? % Mean: 26.32, Standard Deviation: 24.06 C. Perception of vendor's dependence on the retailer 1. We are important to this vendor. 2. We are a major outlet for this vendor in our trading area. 3. We are not a major outlet for this resource. (R) Cronbach's alpha: .71 Mean: 5.52, Standard Deviation: 1.09 C. Perception of retaiier's dependence on the vendor 1. We are important to this retailer. 2. We are a major supplier to this retailer in our product category. 3. If we discontinued supplying to this retailer, this retailer would have difficulty making up the sales volume in our product category. Cronbach's alpha: .76 Mean: 5.54, Standard Deviation: 1.17 D. Retaiier's trust in vendor (vendor's credibiiity) 1. This resource's representative has been frank in dealing with us. 2. Promises made by this resource's representative are reliable. 3. This resource's representative is knowledgeable regarding his/her products. 4. This resource's representative does not make false claims. 5. This resource's representative is not open in dealing with us. (R) 6. If problems such as shipment delays arise, the resource's representative is honest about the problems. 7. This resource's representative has problems answering our questions. Cronbach's alpha: .90 Mean: 5.76, Standard Deviation: .91 D. Vendor's trust in retaiier (retaiier's credibiiity) 1. The buyer representing this retailer has been frank in dealing with me. 2. Promises made by the buyer representing this retailer are reliable. 3. The buyer representing the retailer is knowledgeable about the product. 4. The buyer representing this retailer has problems understanding our position. (R) Cronbach's alpha: .80 Mean: 5.73, Standard Deviation: 1.18 B. Dependence of vendor on retailer Measure 1 1. If our relationship was discontinued with this retailer, we would have difficulty in making up the sales volume in this trading area. 2. This retailer is crucial to our future performance. 3. It would be difficult for us to replace this retailer in this trading area. 4. We are dependent on this retailer for sales in this region. 5. We do not have a good alternative to this retailer. 6. This retailer generates high sales volume for us. Cronbach's alpha: .85 Mean: 4.84, Standard Deviation: 1.32 Measure 2 1. What percentage of the volume in this category is accounted for by this retailer ? % Mean: 14.91, Standard Deviation: 16.01 Vendor side measures A. Vendor's long-term orientation 5. We share our long-term goals with this retailer. 6. We would like to develop a long-term relationship with this retailer. Cronbach's alpha: .82 Mean: 6.26, Standard Deviation: .69

16 / Journai of Mariteting, Aprii 1994

APPENDIX A (Continued)
Retailer side measures E. Retaiier's trust in vendor (vendor's benevoience) 1. This resource's representative has made sacrifices for us in the past. 2. This resource's representative cares for us. 3. In times of shortages, this resource's representative has gone out on a iimb for us. 4. This resource's representative is like a friend. 5. We feel the resource's representative has been on our side. Cronbach's alpha: .88 Mean: 4.76, Standard Deviation: 1.14 F. Environmental diversity How would you describe the market for the product you buy from this vendor compared with other products in general? 1. Few new products—Many new products 2. Few new competitors—Many new competitors. Cronbach's alpha: .57 Mean: 4.12, Standard Deviation: 1.39 G. Environmental volatility How would you describe the market for the product you buy from this vendor compared with other products in general? 1. Unpredictable demand—Predictable demand (R) 2. Sales forecasts are accurate—Sales forecasts are inaccurate 3. Stable industry volume—^Volatile industry volume 4. Easy to monitor trends—Difficult to monitor trends 5. Complex—Simple (R) Cronbach's alpha: .72 Mean: 3.38, Standard Deviation: .95 H. Retailer TSIs 1. We have made significant investments in displays, trained salespeople, etc. dedicated to our relationship with this vendor. 2. If we switched to a competing resource, we would lose a lot of the investment we have made in this resource. 3. We have invested substantially in personnel dedicated to this resource. 4. If we decided to stop working with this resource, we would be wasting a lot of knowledge regarding their method of operation. Cronbach's alpha: .76 Mean: 3.17, Standard Deviation: 1.49 i. Perception of vendor TSis 1. This resource has gone out of their way to link us with their business. 2. This vendor has tailored its merchandise and procedures to meet the specific needs of our company. 3. It would be difficult for this resource to recoup its investment in us if they switched to another retailer as an outlet for their products. Cronbach's alpha: .67 Mean: 4.51, Standard Deviation: 1.43 H. Vendor TSIs 1. We have made significant investments in displays, training salespeople etc. dedicated to our relationship with this retailer. 2. If we switched to a competing retailer, we would lose a lot of the investment we have made in this retailer. 3. We have invested substantially in personnel dedicated to this retailer. 4. If we decided to stop working with this retailer, we would be wasting a lot of knowledge regarding their method of operation. Cronbach's alpha: .71 Mean: 4.31, Standard Deviation: 1.31 i. Perception of retailer TSIs 1. This retailer has gone out of their way to link us with their business. 2. This retailer has changed their floor displays to meet the specific needs of our products. 3. This retailer has made significant investments in training their sales staff to handle questions about our products. Cronbach's alpha: .66 Mean: 4.47, Standard Deviation: 1.25 F. Environmentai diversity How would you describe the market for the product you sell to this retailer compared with other products in general? 1. Few new products—Many new products 2. Few new competitors—Many new competitors. Cronbach's alpha: .56 Mean: 4.32, Standard Deviation: 1.42 G. Environmental volatility How would you describe the market for the product you sell to this retailer compared with other products in general? 1. Unpredictable demand—Predictable demand (R) 2. Sales forecasts are accurate—Sales forecasts are inaccurate 3. Stable market shares—^Volatile market shares Cronbach's alpha: .56 Mean: 3.18, Standard Deviation: 1.12 Vendor side measures E. Vendor's trust in retaiier (retaiier's benevoience) 1. The buyer representing this retailer has made sacrifices for us in the past. 2. The buyer representing this retailer cares for my welfare. 3. In times of delivery problems, the buyer representing this retailer has been very understanding. Cronbach's alpha: .76 Mean: 4.62, Standard Deviation: 1.55

Determinants of Ljong-Term Orientation / 1 7

APPENDIX A (Continued)
Retailer side measures J. Reputation of the vendor 1. This resource has a reputation for being honest. 2. This resource has a reputation for being concerned about the retailers. 3. This resource has a bad reputation in the market. (R) 4. Most retailers think that this resource has a reputation for being fair. Cronbach's alpha: .82 Mean: 5.29, Standard Deviation: 1.02 K. Satisfaction witii past outcomes Describe your feelings with respect to the outcomes with this resource in the past one year: 1. 2. 3. 4. Pleased—Displeased (R) Sad—Happy Contented—Disgusted (R) Dissatisfied—Satisfied Vendor side measures J. Reputation of the retaiier 1. This retailer has a reputation for being honest. 2. This retailer has a reputation for being concerned about their suppliers. 3. This retailer has a bad reputation in the market. (R) 4. Most suppliers would like to deal with this retailer. Cronbach's alpha: .75 Mean: 5.85, Standard Deviation: 1.04 K. Satisfaction with past outcomes Describe your feelings with respect to the outcomes with this retailer in the past one year: 1. 2. 3. 4. Pleased—Displeased (R) Sad—Happy Contented—Disgusted (R) Dissatisfied—Satisfied

Cronbach's alpha: .94 Mean: 5.17, Standard Deviation: 1.25 L. Retaiier's experience with the vendor 1. How many years' experience do you have in buying merchandise from this particular vendor: years Mean: 2.87, Standard Deviation: 3.49

Cronbach's alpha: .94 Mean: 5.45, Standard Deviation: 1.34 L. Vendor's experience with the retaiier 1. How many years' experience do you have in selling merchandise to this particular retailer: years Mean: 4.24, Standard Deviation: 5.38

1. All items used for assessing constructs A, B, C, D, E, H, I, and J are Likert items anchored by strongly disagree (1) and strongly agree (7). 2. All items used for assessing constructs F, G, and K are semantic differential items anchored by 1 and 7. 3. R indicates reverse worded

REFERENCES
Adams, Stacy J. (1965), "Inequity in Social Exchange," in Advances in Experimental Social Psychology, Vol. 2, Leonard Berkowitz, ed. New York: Academic Press, 267-99. Aldrich, Howard (1979), Organizations and Environments. Engelwood Cliffs, NJ: Prentice-Hall, Inc. Anderson, Erin and Barton A. Weitz (1989), "Determinants of Continuity in Conventional Industrial Channel Dyads," Marketing Science, 8 (Fall), 310-23. and (1992), ' 'The Use of Pledges to Build and Sustain Commitment in Distribution Channels," Journal of Marketing Research, 29 (February), 18-34. Anderson, James C. and James A. Narus (1984), "A Model of the Distributor's Perspective of Distributor-Manufacturer Working Relationships," Journal of Marketing, 48 (Fall), 62-74. and (1990), "A Model of Distributor Firm and Manufacturer Firm Working Partnerships," Journal of Marketing, 54 (January), 42-58. Balakrishnan, Srinivasan and Birger Wemerfelt (1986), "Technical Change, Competition, and Vertical Integration," Strategic Management Journal, 7 (July-August), 347-59. Beier, Fredrick and Louis W. Stem (1969), "Power in the Channel of Distribution," in Distribution Channels: Behavioral Dimensions, Louis W. Stern, ed. Boston: Houghton-Mifflin Company. Buchanan, Lauranne (1992), "Vertical Trade Relationships: The Role of Dependence and Symmetry in Attaining Organizational Goals," Journal of Marketing Research, 29 (February), 65-75. Corfman, Kim P. and Donald R. Lehmann (1987), "Models of Cooperative Group Decision Making and Relative Influence: An Experimental Investigation of Family Purchase Decisions," Journal of Consumer Research, 14 (June), 1-13. Dwyer, Robert F., Paul H: Schurr, and Sejo Oh (1987), "Developing Buyer-Seller Relationships," Journal of Marketing, 51 (April), 11-27. El-Ansary, Adel I. and Louis W. Stem (1972), "Power Measurement in the Distribution Channel," Journal of Marketing Research, 9 (February), 47-52. Emerson, Richard M. (1962), "Power-Dependence Relations," American Sociological Review, 27 (February), 32-33. Fornell, Claes (1983), "Issues in the Application of Covariance Structure Analysis: A Comment," Journal of Consumer Research, 9 (March), 443-48. Frazier, Gary L.(1983), "On the Measurement of Interfirm Power in Channels of Distribution," Journal of Marketing Research, 20 (May), 158-66. Gaski, John (1984), "The Theory of Power and Conflict in Channels of Distribution," Journal of Marketing, 48 (Summer), 9 29. Hayduk, Leslie A. (1987), Structural Equation Modeling with LISREL: Essential and Advances. Baltimore: Johns Hopkins University Press. Heide, Jan B. and George John (1988), "The Role of Dependence Balancing in Safeguarding Transaction-Specific Assets in Conventional Channels," Journal of Marketing, 52 (January), 2035.

18 / Journai of Mariteting, Aprii 1994

Hunt, Shelby and John R. Nevin (1974), "Power in a Channel of Distribution: Sources and Consequences," Journal of Marketing Research, 11 (May), 186-93. Jackson, Barbara Bund (1985), Winning and Keeping Industrial Customers: The Dynamics of Customer Relationships. Lexington, MA: Lexington Books. John, George and Barton A. Weitz (1988), "Forward Integration into Distribution: Empirical Test of Transaction Cost Analysis," Journal of Law, Economics, and Organization, 4 (Fall), 121-39. Joreskog, Karl G. and Dag Sorbom (1988), LISREL VII: Analysis of Linear Structural Relationships by Maximum Likelihood. Mooresville, D^I: Scientific Software. Kaufmann, Patrick J. and Louis W. Stem (1988), "Relational Exchange Norms, Perceptions of Unfairness, and Retained Hostility in Commercial Litigation," Journal of Conflict Resolution, 32 (September), 534-52. Kelly, Harold H. (1983), "Love and Commitment," in Close Relationships, H. H. Kelley et al., eds. New York: W. H. Freeman, 265-314. and John W. Thibaut (1978), Interpersonal Relations: A Theory of Interdependence. New York: John Wiley & Sons, Inc. Klein, Saul, Gary L. Frazier, and Victor J. Roth (1990), "A Transaction Cost Analysis Model of Channel Integration in International Markets," Journal of Marketing Research, 27 (May), 196-208. Leblebici, Huseyin and Gerald Salancik (1981), "Effects of Environmental Uncertainty on Information and Decision Processes in Banks," Administrative Science Quarterly, 26 (December), 578-96. Lindskold, Svenn (1978), "Trust Development, the GRIT Proposal, and the Effects of Conciliatory Acts on Conflict and Cooperation," Psychological Bulletin, 85 (4), 772-93. Lusch, Robert (1976), "Sources of Power: Their Impact on IntraChannel Conflict," Journal of Marketing Research, 13 (August), 382-90. Macneil, Ian (1980), The New Social Contract, An Inquiry into Modem Contractual Relations. New Haven, CT: Yale University Press.

Moorman, Christine, Gerald Zaltman, and Rohit Deshpande (1992), "Relationships Between Providers and Users of Market Research: The Dynamics of Trust Within and Between Organizations," Journal of Marketing Research, 29 (August), 314-28. , , and (1993), "Factors Affecting Trust in Market Research Relationships," Journal of Marketing, 57 (January), 81-101. Noordewier, Thomas G., George John, and John R. Nevin (1990), "Performance Outcomes of Purchasing Arrangements in Industrial Buyer-Vendor Relationships," Journal of Marketing, 54 (October), 80-94. Pfeffer, Jeffrey and Gerald R. Salancik (1978), The External Control of Organizations: A Resource Dependence Perspective. New York: Harper Row. Phillips, Lynn W. (1981), "Assessing Measurement Error in Key Informant Reports: A Methodological Note on Organizational Analysis in Marketing," Journal of Marketing Research, 18 (November), 395^15. Rempel, John K., John G. Holmes, and Mark P. Zanna (1985), "Trust in Close Relationships," Journal of Personality and Social Psychology, 49 (1), 95-112. Rusbult, Caryl E., Julie Verette, Gregory A. Whitney, Linda F. Slovik, and Isaac Lipkus (1991), "Accommodation Processes in Close Relationships: Theory and Preliminary Empirical Evidence," Journal of Personality and Social Psychology, 60(1), 53-78. Scanzoni, John (1979), "Social Exchange and Behavioral Interdependence," in Social Exchange in Developing Relationships, Robert L. Burgess and Ted L. Huston, eds. New York: Academic Press. Williamson, Oliver E. (1975), Markets and Hierarchies, Analysis and Anti-trust Implications. New York: The Free Press. (1979) "Transaction-Cost Economics: The Governance of Contractual Relations," Journal of Law and Economics, 22 (October), 3-61. - (1981), "The Economics of Organization: The Transaction Cost Approach," American Journal of Sociology, 87 (3), 548-77.

Determinants of Long-Term Orientation / 1 9