Exit, voice, loyalty, neglect, and retaliation: The impact of different maintenance motivations on customers' responses to dissatisfaction

Transcription

1 Exit, voice, loyalty, neglect, and retaliation: The impact of different maintenance motivations on customers' responses to dissatisfaction Abstract Because failure and breakdown are part of social life, the operations of every organization are punctuated by customers dissatisfaction, annoyance and discontent. Consequently, managers are often confronted with deterioration or break-down of their business relationships. Therefore it becomes increasingly important to not only understand how to acquire and maintain ongoing relationships but also how to prevent them from dissolution. The main objective of this paper is to analyze the impact of relationship maintenance motivations on customers' responses to dissatisfaction. A PLS causal model is presented and empirically tested in an industrial context. The results suggest that constraint-based and dedication-based motivations for maintaining a relationship have different impacts on customers' reactions to dissatisfaction. Keywords: EVLN model; retaliation; dissatisfaction; relationship; maintenance Track: Business-to-Business Marketing & Networks

2 1. Introduction Because failure and breakdown are part of social life, the operations of every organization are punctuated by customers' dissatisfaction, annoyance and discontent. Consequently, managers are often confronted with deterioration or break-down of their business relationships. Therefore, it becomes increasingly important to not only understand how to acquire and maintain ongoing relationships, but also how to prevent them from dissolution. The main objective of this paper is to analyze the implications of relationship structure on customers' responses to dissatisfaction. How do these different motivations explain customers' reactions in the face of a relational incident? The first section presents the conceptual framework and the research hypothesis (2). The methodological modalities are then presented (3). Finally, results are exposed (4), discussed (5) and limitations analyzed (5). 2. Conceptual framework and research hypothesis By combining the reactions derived from the EVL framework (1.1) and the propositions concerning customers' motivations for maintaining a relationship (1.2), it is possible to develop an integrative model explaining customers' reactions to dissatisfying incidents. CONSTRAINT-BASED MAINTENANCE - Economic satisfaction Calculative commitment - EXIT NEGLECT Social satisfaction Trust Affective commitment DEDICATION-BASED MAINTENANCE Figure 1 : A theoretical model of customers' reactions to dissatisfaction RETALIATION 2.1. A typology of customers' reactions to dissatisfaction based on the EVL framework VOICE LOYALTY Initially developed by Hirschman (1970), the EVL model states that customers faced with dissatisfaction can either exit (withdraw from the relationship) or voice their discontent, the decision being essentially moderated by the degree of attachment (loyalty) to the firm. The most notable extension of this framework can be attributed to Rusbult, Zembrodt and Gunn (1982) who integrated a fourth option, called neglect (pessimistically waiting for relationship dissolution). However, another line of research (Huefner & Hunt, 2000; Huefner et al., 2002) focuses on retaliation as a possible alternative. The first reference to retaliation can be attributed to Axelrod (1984). Based on this work, some scholars (Grégoire & Fisher, 2006; Huefner & Hunt, 2000; Huefner et al., 2002; Kumar, Scheer, & Steenkamp, 1998) applied these conclusions in a commercial context and define retaliation as "a customer's effort to punish and make a firm pay for the damages it has caused" The different impacts of constraint-based and dedication-based maintenance on customers' reactions to dissatisfaction Although recognizing the cumulative effect of different maintenance motivations on relationship dynamics, few scholars integrate these aspects into a unifying framework (for - 1

3 notable exceptions see Bendapudi & Berry, 1997; Ganesan, 1994; Morgan & Hunt, 1994). The theoretical model developed here relies on one fundamental conjecture based on Bendapudi and Berrys' propositions: while constraint-based motivations (CBM) determines relationship stability and is mainly associated with destructive reactions (exit and neglect), dedication-based maintenance (DBM) promotes relationship development and appears as a sole determinant of constructive modalities (retaliation 1, voice and loyalty) The impact of constraint-based maintenance Derived from an economic perspective, CBM explains continuance of a relationship in terms of the costs and benefits associated with a relationship (Bendapudi & Berry, 1997). In this research, CBM is thought of as a meta-variable, consisting of two interrelated constructs: economic satisfaction and calculative commitment. Economic satisfaction is defined as a "member's positive affective response to the economic rewards that flow from the relationship with its partner" (Geyskens, Steenkamp, & Kumar, 1999). It has been associated with instrumental motivations for maintenance and is, consequently, presented as a determinant of calculative commitment (Palmantier, Dant, Grewal, & Evans, 2006). At the same time, economic satisfaction is also considered as an antecedent of trust (Geyskens, Steenkamp, Scheer, & Kumar, 1996; Gilliland & Bello, 2002). Hypothesis 1: Economic satisfaction is positively associated with (a) calculative commitment and (b) trust. Rooted in the Gesellschaft strain of relationships, calculative commitment is defined as the degree to which a member perceives the need to maintain a relationship given the significant anticipated termination or switching costs associated with leaving (Allen & Meyer, 1990; Anderson & Weitz, 1992; Dwyer, Schurr, & Oh, 1987; Geyskens et al., 1996). In the case of high dependence, exit would appear too difficult to accomplish (Luchak, 2003; Ping, 1993, 1995) and neglect would then represent the best alternative (Bendapudi & Berry, 1997; Ping, 1993; Ping & Dwyer, 1992). Hypothesis 2: Calculative commitment is (a) negatively associated with exit and (b) is positively associated with neglect The effects of dedication-based maintenance DBM focuses on the affective responses and explains maintenance based on an active desire of the parties (Bendapudi & Berry, 1997). DBM is also considered as a supra-construct composed of three interrelated variables: social satisfaction, trust and affective commitment. Social satisfaction is defined as a "member's positive affective response to the noneconomic, psychological aspects of its relationship, in that interactions with its exchange partner are fulfilling, gratifying and easy" (1998). Although social satisfaction gave rise to few studies, it can be considered as the main determinant of affective commitment. At the same time, social satisfaction has also been validated as an antecedent of trust (Geyskens et al., 1999). Hypothesis 3: Social satisfaction is positively associated with (a) affective commitment and (b) trust. Trust, defined as a willingness to rely on an exchange partner in whom one has confidence, has been validated as the principal determinant of commitment (Anderson & Weitz, 1989; 1 Given that the main objective of retaliation is to restore cooperation, it is considered as a constructive reaction to dissatisfaction. In fact, a customer resorting to this strategy does not wish to dissolve its partnership. He just tries to return to a fair level of cooperation by applying some sort of revenge action to its partner. 2

4 Blois, 1999; Mayer, Davis, & Schoorman, 1995; 1993; Moorman, Zaltman, & Deshpande, 1992; Morgan & Hunt, 1994). But recent findings show that while trust appears to influence affective commitment positively (de Ruyter & Wetzels, 1999; Geyskens et al., 1996; Gilliland & Bello, 2002; N'Goala, 2006; 1998), it seems to have a negative impact on calculative commitment (de Ruyter & Wetzels, 1999; Geyskens et al., 1996; Wetzels et al., 1998). Hypothesis 4: Trust is (a) positively associated with affective commitment and (b) negatively associated with calculative commitment. Finally, affective commitment is solely related to constructive reactions. The key argument supporting this conjecture rests on the idea that a customer deliberately maintaining a relationship will tend to resort to positive strategies in the face of relational problems. As a result, he will be very likely to opt for loyalty or voice (Luchak, 2003; Roloff, Soule, & Carey, 2001; Withey & Cooper, 1989) and refuse retaliation (Grégoire & Fisher, 2006). Hypothesis 5: Affective commitment is positively associated with (a) loyalty and (b) voice and (c) is negatively associated with retaliation. 3. Study method 3.1. Research context, data collection and sample An industrial distribution context was chosen to empirically test the research hypothesis. The relationships between a manufacturer (FabIndus 2 ) of industrial materials and its distributors operating in the aeronautic, automotive and industrial markets were studied. The sampling frame was a customized list of 775 independent distributors affiliated with FabIndus. Key informants were carefully selected through a telephone identification process. Of the 775 potential targets, 400 were ultimately contacted and invited to answer the web survey. 174 usable questionnaires were electronically returned, for a response rate of 43.5 percent Measurement modalities All the constructs used in our research were based on prior studies and measured using multiple-items scales. Economic and social satisfaction were measured with the ten-items scale developed by Geyskens and Steenkamp (2000). Trust was measured by the five indicators composing Morgan and Hunt's scale (1994). Calculative commitment was operationalized based on the scale developed by Ganesan (1994) and affective commitment was measured using Bloemer and Odekerken-Schröder's five-items scale (2003). Exit and neglect were each derived using two four-items scales borrowed from Ping (1993). Voice and loyalty were measured using Hibbard, Kumar and Stern's three-items scales (2001). The operationalization of retaliation was based on, and adapted from, Grégoire and Fisher (2006) Model estimation The structural equation model presented in figure 1 was estimated using a partial least square (PLS) procedure. PLS is a non-parametric estimation procedure (Chin, 1998; Lohmöller, 1984; 1966; Wold, 1973; 1975, 1982) that can accommodate small samples and avoids some of the restrictive assumptions imposed by maximum likelihood estimation. Bootstrapping and jackknifing procedures were used to evaluate statistical validity of parameters and model fit. 2 For confidentiality reasons, the names of the manufacturer and its distributors cannot be divulged here. Consequently, an assumed name (FabIndus) is used to refer to this supplier. 3

5 4. Analysis and results 4.1. Measurement model EFA was first used to evaluate the psychometric qualities of each construct of the study. This procedure suggested the elimination of four items (SocSat1, CalcCom4, CalcCom6, Retal3) presenting non-significant loading (< 0.5). After deletion, all remaining constructs had satisfactory results and were integrated in the final measurement model (Table 1). Constructs RELIABILITY CONVERGENT VALIDITY DISCRIMINANT VALVIDITY Cronbach's Composite Communality alpha reliability Cv-communality (3) Higher squared AVE Root AVE correlation Economic satisfaction 0,855 0,897 0,635 0,635 0,635 0,797 0,281 Social satisfaction 0,734 0,832 0,554 0,555 0,554 0,744 0,348 Trust 0,901 0,927 0,717 0,717 0,717 0,847 0,281 Calculative commitment 0,874 0,913 0,725 0,725 0,725 0,851 0,408 Affective commitment 0,904 0,929 0,723 0,723 0,723 0,850 0,408 Exit 0,949 0,963 0,868 0,868 0,868 0,932 0,398 Neglect 0,906 0,933 0,777 0,778 0,777 0,881 0,398 Retaliation 0,870 0,918 0,790 0,790 0,790 0,889 0,315 Loyalty 0,826 0,894 0,737 0,736 0,737 0,858 0,082 Voice 0,890 0,927 0,809 0,809 0,809 0,899 0,160 Table 1: Psychometric qualities of the measurement model Reliability of the constructs is confirmed as the Cronbach's alpha and composite reliability indices are well above the 0.70 threshold. The communality and AVE indexes are also above this 0.70 threshold (except for economic and social satisfaction) demonstrating convergent validity. Discriminant validity is established since the correlations of each construct with its indicators (root AVE) exceed the squared correlation between the construct and any other construct (Fornell & Lacker, 1981) Structural model The proposed model demonstrated a relatively weak predictive power (Table 2) as the average variance explained (average R 2 ) was 0,186. This score was essentially explained by the relative inability of the model to predict loyalty and voice. Redundancy indexes confirmed these results. Finally, the GoF index demonstrated a relatively medium fit (0,369). Constructs R 2 Redundancy (4) Cv-redundancy (5) GoF (6) Trust 0,380 0,272 0,260 - Aff.Com. 0,290 0,210 0,206 - Calc.Com. 0,147 0,107 0,093 - Exit 0,239 0,207 0,202 - Neglect 0,146 0,114 0,102 - Retaliation 0,106 0,084 0,081 - Loyalty 0,024 0,018 0,017 - Voice 0,078 0,063 0,056 - Average 0,186 0,134 0,127 0,369 Table 2: PLS model predictive power and goodness-of-fit Concerning hypothesis validity, results show a strong support (Table 3). Indeed, all but two of the research conjectures (H2 b and H4 b ) are confirmed by the bootstrap simulation. 3 Cv-communality = cross-validated communality (calculated by a blindfolding procedure) 4 Redundancy = R 2 x communality 5 Cv-redundancy = cross-validated redundancy (calculated by a blindfolding procedure) 6 Gof = [(average R 2 ) x (average communality)] 4

6 Hypothesis Original Sample Bootstrapped Sample Standard Error t p Validity H1a Eco.Sat Calc.Com. 0,309 0,311 0,077 4,025 *** Confirmed H1b Eco.Sat Trust 0,386 0,377 0,086 4,502 *** Confirmed H2a Calc.Com. Exit -0,489-0,495 0,060 8,155 *** Confirmed H2b Calc.Com. Neglect -0,382-0,394 0,069 5,522 *** Rejected H3a Soc.Sat Aff.Com. 0,219 0,224 0,093 2,352 0,019 Confirmed H3b Soc.Sat. Trust 0,345 0,363 0,087 3,958 *** Confirmed H4a Trust Aff.Com. 0,393 0,395 0,087 4,510 *** Confirmed H4b Trust Calc.Com. 0,117 0,120 0,073 1,593 0,111 Rejected H5a Aff.Com. Loyalty 0,155 0,168 0,063 2,458 0,014 Confirmed H5b Aff.Com. Voice 0,280 0,289 0,064 4,410 *** Confirmed H5c Aff.Com. Retaliation -0,326-0,335 0,069 4,746 *** Confirmed Table 3: Parameter estimation of the PLS Model by the Bootstrap procedure 5. Discussion As a whole, the results confirm the proposition of a differential effect of maintenance motivations on reactions to dissatisfaction. In fact, while CBM explains relationship stability by inhibiting destructive reactions, DBM encourages relationship development by promoting constructive ones. However some contradictory results deserve an explanation. Although calculative commitment was postulated to promote neglect (H2 b ), a positive association was found between these two constructs. In other words, just as calculative commitment discourages exit, it also inhibits neglect. Consequently, neglect has to be thought of as an alternative form of exit, rather than a mere alternative modality of reaction used in the case of high dependence. In this respect, neglect should be characterized as a manifestation of silent exit (Alajoutsijärvi, Möller, & Tähtinen, 2000; Baxter, 1984, 1985; Freeman, 2001; Freeman & Browne, 2004). This perspective would explain the negative association between calculative commitment and neglect, e.g., a distributor being forced to maintain its relationship due to dependency factors, at least for a period of time. Another surprising result is the non-significant association revealed between trust and calculative commitment (H4 b ). This result suggests that although trust promotes affective commitment, it has no effect on the level of calculative commitment in the relationship. This conclusion implies that DBM and CBM have independent determinants, trust being exclusively related to the former. As a result, this outcome does not contradict our hypothesis. It just underlines the specific nature of DBM and its sole impact on relationship quality. 6. Limitations and directions for future research This study has several limitations. First, it is restricted to a single industry, limiting its external validity. Hence, cross-validation in other contexts is required. Second, the sample is relatively small. Consequently, a non-parametric method of estimation has been employed. As a consequence, the results deserve a more robust method to be fully validated. Third, the cross-sectional nature of this study should be considered with caution, since dynamic phenomena are under analysis. A longitudinal design would be more appropriate. Future research on this topic can follow different paths of investigation. The first one would concentrate on reactions to dissatisfaction. Although numerous studies have been undertaken, there is still a lot of work to be devoted to the analysis of each mode of response. Notably, retaliation deserves greater attention. Another direction can be followed to incorporate other determinants in the actual framework. In this perspective, the propositions of Bendapudi & Berry (1997) represent a good starting point. 5

7 References Alajoutsijärvi, K., Möller, K., & Tähtinen, J. (2000). Beautiful exit: How to leave your business partner. European Journal of Marketing, 34(11/12), Allen, N. J., & Meyer, J. P. (1990). The measurement and antecedents of affective, continuance and normative commitment to the organization. Journal of Occupational Psychology, 63(1), Anderson, E., & Weitz, B. (1989). Determinants of continuity in conventional industrial channel dyads. Marketing Science, 8(4), Anderson, E., & Weitz, B. (1992). The use of pledges to build and sustain commitment in distribution channels. Journal of Marketing Research, 29(1), Axelrod, R. (1984). The evolution of cooperation. New-York: Basic Books. Baxter, L. A. (1984). Trajectories of relationship disengagement. Journal of Social and Personal Relationships, 1(1), Baxter, L. A. (1985). Accomplishing relationship disengagement. In S. Duck (Ed.), Understanding personal relationships An Interdisciplinary approach (pp ). London: SAGE Publications. Bendapudi, N., & Berry, L. L. (1997). Customer's motivations for maintaining relationships with service providers. Journal of Retailing, 73(1), Bloemer, J., & Odekerken-Schröder, G. (2003). Antecedents and consequences of affective commitment. Australasian Marketing Journal, 11(3), Blois, K. J. (1999). Trust in business to business relationships: An evaluation of its status. Journal of Management Studies, 36(2), Chin, W. W. (1998). The partial least square approach for structural equation modelling. In G. A. Marcoulides (Ed.), Modern methods for business research (pp ). London: Lawrence Erlbaum Associates. de Ruyter, K., & Wetzels, M. (1999). Commitment in auditor-client relationships: Antecedents and consequences. Accounting, Organizations and Society, 24(1), Dwyer, F. R., Schurr, P. H., & Oh, S. (1987). Developing buyer-seller relationships. Journal of Marketing, 51(2), Fornell, C., & Lacker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement errors. Journal of Marketing Research, 18(2), Freeman, S. (2001). Conflict management and exit strategies in buyer-relationships in foreign markets: A case study of an Australian citrus fruit exporter. Paper presented at the 17th IMP conference, Oslo - Norway. Freeman, S., & Browne, E. (2004). The influence of national culture on dissolution communication strategies in Western versus Asian business relationships: A theoretical model. Supply chain management: An International Journal, 9(2), Ganesan, S. (1994). Determinants of long-term orientation in buyer-seller relationships. Journal of Marketing, 58(2), Geyskens, I., & Steenkamp, J.-B. E. M. (2000). Economic and social satisfaction: Measurement and relevance to marketing channel relationships. Journal of Retailing, 76(1), Geyskens, I., Steenkamp, J.-B. E. M., & Kumar, N. (1999). A meta-analysis of satisfaction in marketing channel relationships. Journal of Marketing Research, 36(2), Geyskens, I., Steenkamp, J.-B. E. M., Scheer, L. K., & Kumar, N. (1996). The effects of trust and interdependence on relationship commitment: A trans-atlantic study. International Journal of Research in Marketing, 13(4), Gilliland, D. I., & Bello, D. C. (2002). Two sides to attitudinal commitment: The effect of calculative and loyalty commitment on enforcement mechanisms in distribution channels. Journal of the Academy of Marketing Science, 30(1), Grégoire, Y., & Fisher, R. J. (2006). The effects of relationship quality on customer retaliation. Marketing Letters, 17(1), Hibbard, J. D., Kumar, N., & Stern, L. W. (2001). Examining the impact of destructive acts in marketing channel relationships. Journal of Marketing, 38(1), Hirschman, A. O. (1970). Exit Voice and Loyalty: Responses to Decline in Firms, Organizations, and States (Economie humaine ed.). Cambridge MA: Harvard University Press. 6

8 Huefner, J. C., & Hunt, H. K. (2000). Consumer retaliation as a response to dissatisfaction. Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 13(1), Huefner, J. C., Parry, B. L., Payne, C. R., Otto, S. D., Huff, S. D., Swenson, M. J., et al. (2002). Consumer retaliation: Confirmation and extension. Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 15(1), Kumar, N., Scheer, L. K., & Steenkamp, J.-B. E. M. (1998). Interdependence, punitive capability, and the reciprocation of punitive actions in channel relationships. Journal of Marketing Research, 35(2), Lohmöller, J.-B. (1984). LVPLS Program Manual - Version 1.6: Latent variable path analysis with partial least-squares estimation. Paper presented at the Zentralarchiv für Empirische Sozialforschung, Universität Zu Köln. Luchak, A. A. (2003). What kind of voice do loyal employees use? British Journal of Industrial Relations, 41(1), Mayer, R. C., Davis, J. H., & Schoorman, D. F. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3), Moorman, C., Deshpande, R., & Zaltman, G. (1993). Factors affecting trust in market research relationships. Journal of Marketing, 57(1), Moorman, C., Zaltman, G., & Deshpande, R. (1992). Relationships between providers ans users of market research: The dynamics of trust within and between organizations. Journal of Marketing Research, 29(3), Morgan, R. M., & Hunt, S. D. (1994). The commitment-trust theory of relationship marketing. Journal of Marketing, 58(3), N'Goala, G. (2006). How to make customers resist switching to another service provider when a critical incident occurs? The complementary roles of trust and relationship commitment. Paper presented at the 9th International Research Seminar in Service Management, La Londe les Maures - France. Palmantier, R. W., Dant, R. P., Grewal, D., & Evans, K. R. (2006). Factors influencing the effectiveness or relationship marketing: A meta-analysis. Journal of Marketing, 70(4), Ping, R. A. (1993). The effects of satisfaction and structural constraints on retailer exiting, voice, loyalty, opportunism, and neglect. Journal of Retailing, 69(3), Ping, R. A. (1995). Some uninvestigated antecedents of retailer exit intention. Journal of Business Research, 34(3), Ping, R. A., & Dwyer, F. R. (1992). A preliminary model of relationship termination in marketing channels. Advances in Distribution Channel Research, 1, Roloff, M. E., Soule, K. P., & Carey, C. M. (2001). Reasons for remaining in a relationship and responses to relational transgressions. Journal of Social and Personal Relationships, 18(3), Rusbult, C. E., Zembrodt, I. M., & Gunn, L. K. (1982). Exit, voice, loyalty, and neglect: Responses to dissatisfaction in romantic involvements. Journal of Personality and Social Psychology, 43(6), Wetzels, M., de Ruyter, K., & van Birgelen, M. (1998). Marketing service relationships: the role of commitment. Journal of Business and Industrial Marketing, 13(4/5), Withey, M. J., & Cooper, W. H. (1989). Predicting exit, voice, loyalty, and neglect. Administrative Science Quarterly, 34(4), Wold, H. (1966). Estimation of principal components and related models by iterative least squares. In P. R. Krishnaiah (Ed.), Multivariate Analysis (pp ). New York: Academic Press. Wold, H. (1973). Non-linear iterative partial least squares (NIPALS) modelling - Some current developments. In P. R. Krishnaiah (Ed.), Multivariate Analysis - Vol. III (pp ). New York: Academic Press. Wold, H. (1975). Soft modelling by latent variables: The non-linear iterative partial least squares (NIPALS) approach. In J. Gani (Ed.), Perspectives in Probability and Statistics (pp ). London: Academic Press. Wold, H. (1982). Soft modelling: The basic design and some extensions. In K. G. Jöreskog & H. Wold (Eds.), Systems under indirect observation - Part 2 (pp. 1-54). Amsterdam - North Holland. 7