Advances in Consumer Research Volume 20, 1993 Pages 109-112
UNETHICAL SELLER PRACTICES: A NEGLECTED ISSUE IN CONSUMER SATISFACTION AND DISSATISFACTION RESEARCH
Alan R. Andreasen, University of Connecticut
The research literature on dissatisfaction and complaining behavior has largely ignored instances of unethical behavior on the part of marketers. Three biases have led to this outcome. Research has focused on outcomes rather than processes. It has largely ignored possible ethical problems in interactions with sellers after purchase. And, economic costs have been emphasized over emotional and psychological costs. Policing systems do not appear likely to correct unethical seller behaviors and thus better monitoring is needed as is new research exploring the impact of the three sources of past bias.
The 1980s have seen a very rapid acceleration in interest in the topic of consumer satisfaction and dissatisfaction on the part of both marketers and public policymakers. Marketers have become increasingly concerned, in part, because of their growing emphasis on quality as a key determinant of future profitability and their realization that service intangibles play an extremely important role in determining that quality. For their part, public policymakers are increasingly concerned because of a growing realization that lack of oversight has been a major contributor to the ethical excesses in business performance in the 1980's.
This rapidly growing interest has led to a significant outpouring of research. A recently review cited over 900 entries on consumer satisfaction and dissatisfaction and complaining behavior through 1990, the majority of these completed since 1982 (Perkins 1991). This literature has covered a wide range of topics including papers on the process of post-purchased evaluation, alternatives to measuring dissatisfaction and dissatisfaction, determinants of complaint behavior, characteristics of complainers, use of third party interveners, and the responses of the business community. The literature has been particularly responsive to the needs of commercial marketers. It has, however, been much less responsive to the needs of public policymakers.
Public policymakers look to such research to provide two kinds of input (Andreasen and Manning 1980). First, they expect the literature to help them detect problems that are in need of public intervention. Second, they need help in priority-setting. Given limited financial and political resources, regulators must make careful strategic judgments as to where to intervene to achieve maximum ameliorative impact. These judgments require four types of information. They need to know:
1. The frequency of market problems in need of potential intervention;
2. The seriousness of these problems, which is, in turn, a function of:
a. The economic and social cost of each instance; and
b. The characteristics of those who are primarily affected (e.g. whether they are in some ways disadvantaged);
3. The availability and potential costs (to all parties) of possible solutions, including seller or industry action; and
4. The likelihood that proposed solutions will have the desired impact.
While a certain amount of the existing published research addresses these issues, it is seriously deficient in failing to address a major concern of public policymakers, namely, the existence and impact of unethical marketing behavior. In my judgment, this gap is attributable to three major biases in the research literature on consumer satisfaction and dissatisfaction. The first bias is found in the almost universal tendency to focus consumer satisfaction and dissatisfaction research on the ends of the purchase process and not on the process itself. We study how products and services turned out in the post-purchase evaluation phase. We investigate whether prior expectations are met by the product or service and whether there is a "disconfirmation" experience. A typical research protocol asks consumers first to think about a situation in which a purchase turned out to be unsatisfactory and then to tell the researcher what they did about it.
The trouble with this way of framing the issue is that it tends to minimize the chance that researchers would detect consumer exposure to unethical marketer behavior. It generates comments about outcomes; i.e. was the product or service satisfactory? It does not prompt consumers to think about C or comment on C cases where there might be seller misconduct in the process of creating the transaction. Such misconduct could include the following:
- Misrepresentation and deception
- Excessively high pressure selling
- Unreasonable use of fear tactics
- Demeaning references to race, age, and sex
In all of these cases, these unethical practices involve the selling process, not the performance of the goods and services per se.
A second major bias in the research literature also minimizes the detection of cases of unethical behavior. This is the tendency to measure consumer satisfaction in terms of what I have called initial evaluations (Andreasen 1977). Researchers ask how the purchase turned out but do not investigate interactions with sellers after the purchase takes place. In our own research in the mid-70s (Andreasen and Best 1977), we found that consumers had problems with purchases about 20 percent of the time. When things went wrong, they voiced a complaint to the seller about 40 percent of the time. This means that perhaps eight percent of all purchases involve some kind of post-purchase interaction with the seller. These post-purchase encounters also present many opportunities for unethical marketer behavior. For example:
- Refusal to honor warranties
- Illegal use of threats to secure payment (e.g. threatening to call employers, advise neighbors, secure court orders, and so on)
- Denial of responsibility
- Disparagement of buyer and his or her judgment, integrity, truthfulness.
The third bias in the literature would make it difficult to calibrate the impact of unethical marketer behavior even if it were detected. This bias is the emphasis on economic losses. A typical measure used to assess the impact of dissatisfaction is to take the cost of the purchase or the cost of repairing some problem (sometimes including the economic cost of lost time to the consumer). The difficulty here is that the cost of unethical behavior to consumers is very often emotional and psychological. Take for example these hypothetical, but not atypical, cases:
- An elderly person is contacted by a door-to-door salesperson selling hearing aids. In order to convince the consumer he or she needs the product, the seller pretends not to hear the consumer or uses a rigged "testing device" to prove that the consumer needs help. Certainly, the consumer will have wasted money on an unneeded product. But, surely, the cost in personal self-worth may well be enormous, especially for an elderly person who felt heretofore that he or she "was not really that old."
- A poor minority household is sold a set of encyclopedias by a salesperson who shames them into believing that, without the seller's product, their children are destined to live in poverty like them. When the head of the household loses his or her job, the seller insists on other collateral and, when payments are still not made, threatens to take the buyer to court, pointing out that "people like you never win in the courts, it's a white man's institution." Again, there are certainly economic costs. But the social and psychological costs of this personally damaging and racist behavior are great and may, indeed, last a lifetime.
These biases in the CS/D literature significantly reduce the possibility of public policymakers becoming aware of the incidence and nature of the ethical problems that exist in the marketplace. Because the orientations and protocols of present research designs tend to ignore ethical problems and to neglect emotional and psychological costs, they also diminish the likelihood of finding that some groups bear inordinate impacts of unethical behavior. This was the finding in the late 60s and 70s (e.g. Caplovitz 1964; Andreasen 1975). It is undoubtedly true today. One of the frequent findings of research, including my own, is that the poor, the elderly and minority groups often report fewer problems and voice them about as often as non-disadvantaged groups. I suspect that this may well be attributable to the way we did the studies!
Does it matter that we do not detect or measure unethical behavior very well? Even if such biases do exist, is it possible that the marketplace already addresses and resolves these problems without the need for outside intervention? If the answer to this question is in the negative, the issue then becomes one of how to develop a signalling system that does not have the biases I have just described. First, let us consider how consumer satisfaction and dissatisfaction is dealt with by the market already.
1. Self-correcting Systems: In an ideal self-correcting system, signals reporting problems experienced by consumers are sent back to sellers who take note of them and correct their own behavior (while also compensating parties already injured). A basic problem with self-correcting systems is that they cannot attempt to correct a problem if it is not detected by consumers. This can happen for one of three reasons:
a. Problems are invisible even to experts given the present state of detection technology;
b. Problems are made invisible by the deceptive practices of sellers;
c. Detection of problems requires a level of sophistication beyond the reach of most consumers (÷lander, 1977)
If a problem is detected, the system could still fail if the consumer chooses not to send a signal. This can happen for a number of reasons:
a. Consumers are not "certain" the problem really exists;
b. Consumers attribute responsibility for the problem fully or partially to themselves and/or to other conditions and not to the seller;
c. Consumers believe the economic, social or psychological costs of sending the signal exceed the potential gain;
d. Consumers believe the size of the potential gain from complaining and/or the probability of obtaining it do not justify action.
It would seem highly unlikely that a system that relies on consumers to detect unethical behavior and send signals to sellers about them would be effective for a number of reasons:
1. Very often, unethical behavior is designed to insure that consumers do not notice that they are being wronged. Indeed, the definition of deception requires that the consumer be deceived.
2. Unethical practices are often directed against groups that may have more difficulty realizing that they are being subjected to such practices. These groups would include the elderly, the undereducated, and those who do not speak English.
3. Part of the unethical practice itself may be designed to make the buyer believe that, if something is wrong, it is probably the buyer's fault and should not be acted upon.
4. Consumers subjected to unethical practices may know that the seller is unethical and expect that:
a. The seller will not respond even if the buyer did send a signal;
b. Sending a signal may subject the consumer to even more psychological and emotional pain through harassment, name-calling and so on.
5. Assuming that unethical practices are more often directed against the disadvantaged, these consumers may decide that they should not jeopardize their limited access to market alternatives and credit by speaking up.
Assuming that consumers do detect the unethical behavior and do decide to send a signal, they have two options that Albert Hirschman refers to as "exiting" and "voicing" (Hirschman 1970). Exiting behavior is Adam Smith's classic "Invisible Hand." It is the action of individual consumers who simply choose to "exit" a particular buyer-seller relationship and take their patronage elsewhere. In theory, the cumulative weight of dozens of exit behaviors signals to the seller the existence of a problem which the seller then, in an ideal system, sets about correcting, sometimes after further primary research to ascertain the exact nature and cause of the problem. Exit signals are effective in a great many situations. However, they are likely to prove unsatisfactory in the case of unethical behavior, even in the unlikely case that the seller wishes to correct the problem, under one or more of the following conditions:
1. Even though buyers vow never to buy a product or service or to patronize an unethical seller, purchase cycles are sufficiently long that exit signals do not emerge until substantial economic or personal injury has been done (both to buyers and to sellers);
2. Even though buyers exit quickly and repurchase cycles are short, competition is so dynamic that exit signals are masked by the "noise" of temporary in-switching by other consumers;
3. The consumer has no real option for changing behavior (i.e. sending a market signal).
The alternative method of signalling displeasure is by voicing a complaint directly to the seller. Again ideally, the seller will take note of this signal and correct an unsatisfactory market condition. One would think that sellers would actively seek out voicing (complaints) in that it (a) provides early warning in the case of products or services with long repurchase cycles; (b) provides a signal that will be distinct from other marketplace noise; and (c) provides detail as to the exact nature of the problem, thus often obviating the need for further research.
Of course, the most serious defect in self-correcting systems is that they will not work if sellers choose not to respond even when a signal is detected. There are a number of cases in which a seller will not respond in the case of unethical behavior:
1. The seller is truly unethical and ignores all signals of dissatisfaction of any kind;
2. The seller believes that the costs of correcting the problem are greater than the potential lost business (the Pinto case);
3. If the signal is a voicing signal, the seller believes that there will be no lost sales because customers will not take their patronage elsewhere;
4. Those consumers who do complain are of limited interest to the seller;
5. Buyers who are ignored are unlikely to cause the seller other problems such as bad publicity;
6. The seller's business does not involve repeat purchases and so he or she does not care if customers are unhappy.
Often unethical sellers are involved in one-time scams like the classic Holland Furnace case. Further, because the victims of unethical practices may be customers disadvantaged with respect to income, age, race and/or language, sellers may have little to fear either because the buyers have few alternatives or because loss of their patronage may be of little consequence.
2. Amplified Systems: If markets are not self-correcting through the actions of individual consumers protecting themselves, they may still not require intervention by public regulators in one of two cases. One possibility is that elites will police the market for them. In such cases, sellers respond to a small number of prosperous and educated consumers because these consumers either are very vocal in their dissatisfaction or represent a significant loss to the seller if they exit. While this group could police unethical behavior on behalf of all consumers, they are less likely to be victims or to be aware of the victimization of the disadvantaged.
The second possibility is for buyers to amplify their voices by taking their dissatisfaction to others. The most common forms of amplification available to dissatisfied consumers are:
1. Engaging in negative word-of-mouth that results in other consumers exiting or voicing (Richins, 1983) or even going to the extreme of undertaking consumer boycotts (Smith 1990);
2. Creating publicity for the problem in the media (e.g. in consumer "help" columns or feature articles);
3. Enlisting the help of influential individuals such as celebrities or politicians to whom the seller and/or the media are likely to pay attention;
4. Enlisting the help of formal or informal associations such as the Consumer Federation of America, industry-specific consumer groups such as the Airline Passengers Association, or nonconsumer groups such as church clubs and labor unions to make the seller aware of the problem;
5. (Finally) Resorting to a formal or informal complaint proceeding such as an industry arbitration system, small claims or other court or a federal, state or local regulatory agency to coerce the seller into paying attention.
To the extent that these amplification mechanisms do not distort the original consumer interest, they can be very helpful. Amplifiers can command more resources and expertise to investigate and document problems, to bring them to the attention of sellers, and to press for change or individual resolution. However, if those subjected to unethical behavior are primarily disadvantaged consumers, then the literature suggests that they are unlikely to know about or have the assertiveness to use these amplification alternatives.
3. External Monitoring: If, as seems likely, the marketplace will not be self-correcting and consumers faced with unethical behavior will not be protected by either "elite policing" or amplification systems, then public policymakers such as the FTC and the FDA will need to intervene. The question still remains: how will they detect and evaluate unethical marketer behavior? There are a number of options ranging from the informal to the scientifically rigorous:
1. Reliance on direct "mailbag" complaints, assuming that faint signals of unethical behavior represent the tip of a large iceberg;
2. Reliance on the investigative "research" of amplification agencies such as Consumers Union or individuals such as Ralph Nader;
3. Requiring industries to keep specific records of complaints, as in the airline industry;
4. Funding of a systematic Dissatisfaction Tracking System.
The latter has been explored in the past by government agencies in the U.S. through the work of Day and his colleagues (e.g. Day and Bodur 1978) and through the Technical Assistance Research Program, Inc. (1979). There have been periodic studies in Canada and Great Britain and a more serious continuing effort just begun in Sweden (Fornell 1992).
Clearly, a major challenge facing the consumer research community is to begin a series of studies of unethical marketer behavior that overcomes the biases outlined earlier. First, consumer satisfaction and dissatisfaction research needs to be undertaken that focuses on both outcomes and processes that result in purchases of goods and services. Second, studies should be undertaken that identify purchase problems, ask consumers what they did about these, and, for those who did contacted sellers, ask whether this experience involved any unethical behavior on the part of the sellers. Finally, in both these types of studies, when unethical behavior is reported, researchers need to investigate both the economic and noneconomic costs incurred.
Finally, further research should be undertaken of other behaviors that may reflect responses to unethical selling practices. These include studies of consumer boycotts, use of third party complaint handling agencies, and grudgeholding (Hunt and Hunt 1990).
Andreasen, Alan R. (1975), The Disadvantaged Consumer. New York: The Free Press.
Andreasen, Alan R. (1977) "A Taxonomy of Consumer Satisfaction/ Dissatisfaction Measures," Journal of Consumer Affairs, 11, 2, 11-24.
Andreasen, Alan R. and Arthur Best (1977) "Consumers Complain C Does Business Respond?" Harvard Business Review, 55, 4 (July-August), 93-101.
Andreasen, Alan R. and Jean Manning (1980) "Information Needs for Consumer Protection Planning," Zeitschrift fur Verbraucherpolitik/Journal of Consumer Policy, 4, 1, 115-126.
Caplovitz, David (1967). The Poor Pay More. New York: The Free Press.
Day, Ralph L. and M. Bodur (1978), "Consumer Response to Dissatisfaction with Services and Intangibles," in Advances in Consumer Research, 3, 263-268.
Fornell, Claes (1992), "A National Customer Satisfaction Barometer: The Swedish Experience," Journal of Marketing, 56, 1 (January), 6-21.
Hirschman, Albert O. (1970), Exit, Voice and Loyalty: Responses to Declines in Firms, Organizations, and States. Cambridge, MA: Harvard University Press.
Hunt, H. David and H. Keith Hunt (1990), "Consumer Grudgeholding: Further Conceptualization and Analysis," Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 3, 117-122.
+lander, Folke (1977), "Consumer Satisfaction: A Skeptic's View," in H. Keith Hunt (ed), Conceptualization and Measurement of Consumer Satisfaction and Dissatisfaction. Cambridge, MA: Marketing Science Institute, 409-452.
Perkins, Debra S. (1991), "A Consumer Satisfaction Dissatisfaction and Complaining Behavior Bibliography: 1982-1990," Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 4, 194-228.
Richins, Marsha L. (1983), "Negative Word-of-Mouth by Dissatisfied Consumers: A Pilot Study," Journal of Marketing, 47, (Winter), 58-78.
Smith, N. Craig (1990), Morality and the Market: Consumer Pressure for Corporate Accountability, London: Routledge.
Technical Assistance Research Program (TARP) (1979), Consumer Complaint Handling in America: Final Report. Washington, DC: U.S. Department of Health, Education, and Welfare.