Exit, Voice, and Loyalty
Exit, Voice, and Loyalty
To extend the analysis of his fellow economists beyond their traditional focus on simple market exchanges, Albert O. Hirschman wrote Exit, Voice, and Loyalty (1970), a seminal work that examines how several different kinds of human behavior might be invoked when consumers confront a decline in firms, organizations, and states.
In Hirschman’s exit, voice, and loyalty (EVL) model, consumers dissatisfied with a product in the marketplace, for example, might complain to the producer about declining quality (voice) or patiently wait for the product to improve (loyalty) instead of purchasing a rival good or service (exit). In politics, voters dissatisfied with their party’s policies may vote for another (exit), work in party caucuses to change those policies (voice), or hope that the party platform will be revised (loyalty). And in an abusive interpersonal situation, one can leave (exit), complain (voice), or avoid confrontation in the hope that the storm will pass (loyalty). Indeed, Hirschman’s analysis of the interrelations among various behaviors an individual might engage in when confronting declining quality in one’s personal, political, economic, and social relationships was significant to all of the social sciences precisely because of its comprehensiveness.
If very different types of behaviors can effectively substitute for each other, one cannot study each independently without missing something vital. Voice behaviors like consumer complaints or writing letters to the city council, for example, cannot be understood without accounting for opportunities to exit through purchasing a substitute product or moving to another city. Hirschman further demonstrated that opportunities to exercise one type of behavior often influence—in sometimes very surprising ways—the effectiveness of other behaviors. The threat of exit, for example, may encourage declining firms, organizations, and states to be more or less attentive to voice depending on who is exiting, when the exiting occurs, and how it bears on the interests of those making decisions about their goods, services, or policies.
Despite its sweeping and immediately recognized import, Hirschman’s analysis is nonetheless incomplete. First, the EVL model does not give equal attention to the three types of behaviors; loyalty only makes an appearance two-thirds of the way through the analysis. In addition, loyalty is discussed almost exclusively as a brake on the exercise of exit and voice rather than as an independent mode of response to dissatisfaction. Second, exit, voice, and loyalty fail to encompass all of the possible behaviors that one might observe when consumers, voters, or family members confront a decline in the quality of their relationships. They may well do nothing. Indeed, Hirschman discussed how consumers might neither exit nor engage in voice behaviors when institutions do not provide producers with incentives to respond to consumer demand. However, he essentially treated such situations as pathologies rather than incorporating them into the model in a systematic manner, recognizing thereby that not responding to decline is potentially as interesting as exit, voice, or loyalty. Third, Hirschman did not develop a general model that communicated how consumers, voters, employees, or family members select among exit, voice, or loyalty behaviors.
In the early 1980s scholars addressed these weaknesses in extensions of Hirschman’s EVL model. First, in their 1983 analysis, Caryl Rusbult and I. M. Zembrodt recast Hirschman’s threefold typology as a two-dimensional space. The first distinguishes active and passive responses to dissatisfaction while the second distinguishes constructive and destructive responses. Thus, voice includes all active responses to dissatisfaction that are constructive with respect to the relationship giving rise to that dissatisfaction. In contrast, exit behaviors, while also active, are destructive to the relationship. Loyalty behaviors are passive, constructive responses to dissatisfaction. A fourth type of response not addressed by Hirschman except as pathological behavior is neglect; that is, passively responding to dissatisfaction by allowing the relationship to further deteriorate. In another significant extension of Hirschman’s analysis, in 1981 Daniel Farrell and Rusbult developed a parsimonious, three-variable model explaining how individuals select among the responses to dissatisfaction. When one has many alternatives to a current job, a product or service, or a romantic partner, one is more likely to respond to dissatisfaction in an active manner through voice or exit rather than through the passive behaviors associated with loyalty or neglect. But when one is highly invested in and/or has had a high level of prior satisfaction with the relationship, one will more likely respond to dissatisfaction with constructive voice or loyalty behaviors and suppress negative behaviors characteristic of exit and neglect responses.
A number of social scientists continue to employ Hirschman’s original threefold typology. By the mid-1980s, however, many others were employing its descendent, the exit, voice, loyalty, and neglect (EVLN) model. This extension of the original highlighted two dimensions defining four distinct responses to dissatisfaction, including the previously neglected category of neglect. Selection among the four responses was further hypothesized to be function of alternatives, prior satisfaction, and investments. The EVLN model was strongly supported in empirical work on private and public sector employment behaviors, romantic relationships, and political behavior in metropolitan settings. All of these studies examine a wide range of responses to dissatisfaction within a common theoretical framework linking what had been previously understood as disparate and distinct behaviors.
There have been fewer applications of either the original version of Hirschman’s model or its EVLN extension in studies of traditional market economics, which was the main focus of Exit, Voice, and Loyalty. To some extent, this reflects some long recognized normative biases in the disciplines. In the same way that some political scientists might regard exiting a political jurisdiction as illegitimate to the point of constituting treason, some economists view voice, loyalty, or neglect as a failure on the part of consumers to exercise due diligence or as evidence of imprudence. But perhaps a more telling explanation, and one applicable to all social sciences, is that simultaneously accounting for exit, voice, loyalty, and neglect behaviors is very hard to do.
SEE ALSO Economics; Political Science; Sociology
Farrell, Daniel, and Caryl E. Rusbult. 1981. Exchange Variables as Predictors of Job Satisfaction, Job Commitment, and Turnover: The Impact of Rewards, Costs, Alternatives, and Investments. Organizational Behavior and Human Performance 27: 78–95.
Hirschman, Albert O. 1970. Exit, Voice, and Loyalty. Cambridge, MA: Harvard University Press.
Lyons, William E., and David Lowery. 1989. Citizen Responses to Dissatisfaction in Urban Communities: A Partial Test of a General Model. Journal of Politics 51: 842–868.
Rusbult, Caryl E., and David Lowery. 1985. When Bureaucrats Get the Blues: Responses to Dissatisfaction among Federal Employees. Journal of Applied Social Psychology 15: 80–103.
Rusbult, Caryl E., and I. M. Zembrodt. 1983. Responses to Dissatisfaction in Romantic Involvements: A Multidimensional Scaling Analysis. Journal of Experimental Social Psychology 19: 274–293.
Rusbult, Caryl E., Daniel Farrell, G. Rogers, and A. G. Mainous III. 1988. Impact of Exchange Variables on Exit, Voice, Loyalty, and Neglect: An Integrative Model of Responses to Declining Job Satisfaction. Academy of Management Journal 31: 599–627.
Rusbult, Caryl E., Dennis J. Johnson, and G. D. Morrow. 1986. Determinants and Consequences of Exit, Voice, Loyalty, and Neglect: Responses to Dissatisfaction in Adult Romantic Involvements. Human Relations 39: 45–63.
Withy, Michael, and William H. Cooper. 1989. Predicting Exit, Voice, Loyalty, and Neglect. Administrative Science Quarterly 34: 521–539.