Consumer Behavior and Acceptable Risk: Some Research Issues

M. Venkatesan, University of Oregon
ABSTRACT - It is clear that the growing risk assessment literature is very relevant to the problems posed in the hazard management area for consumer products and services. However, there are some research issues that must be investigated now so that these researchers can proceed with a programmatic research emphasis. This paper has identified five such research issues.
[ to cite ]:
M. Venkatesan (1981) ,"Consumer Behavior and Acceptable Risk: Some Research Issues", in NA - Advances in Consumer Research Volume 08, eds. Kent B. Monroe, Ann Abor, MI : Association for Consumer Research, Pages: 503-505.

Advances in Consumer Research Volume 8, 1981      Pages 503-505

CONSUMER BEHAVIOR AND ACCEPTABLE RISK: SOME RESEARCH ISSUES

M. Venkatesan, University of Oregon

[Portions of this paper are based on Fischhoff (1980) preliminary draft. Therefore, please do not quote from this paper.]

ABSTRACT -

It is clear that the growing risk assessment literature is very relevant to the problems posed in the hazard management area for consumer products and services. However, there are some research issues that must be investigated now so that these researchers can proceed with a programmatic research emphasis. This paper has identified five such research issues.

INTRODUCTION

Product safety as an issue has reappeared vigorously with the recent disclosures and linking of "toxic shock syndrome" to the use of a certain brand of tampon. While the number of deaths reported is considerably smaller than a day's carnage on the highways, that brand of tampon is already off the market. The safety question relating to the use of saccharin is heating up. The recent legislation specifying minimum standards for the infant formula points up to another product area where lack of safety has produced tragic results, though not in vast numbers.

The recent Titan II mishap amply demonstrates that no technology is foolproof. Thus, technological risks are inherent in products developed for consumption by consumers. The right to safety has been made very explicit in the 1960s and the Delaney Amendments, the Product Safety Commission, and the like are the result of the prominence of the safety issue. Consumers are not unaware that perfect safety is impossible, and they are also cognizant of the risk-benefits relationships. Engel et al. (1978) pointed out that one of the important roles of consumer research is in establishing the level of safety that mist be considered in decisions concerning the products.

As Rethans (1980) has pointed out, the assessment and management of hazards arising from technology have been of major concern to general public and to a number of governmental agencies. In order to help society cope with the technological risk, risk assessments have been undertaken in many areas. For example, the U.S. Nuclear Regulatory Commission made an assessment of the rink associated with nuclear power in their Reactor Safety Study. A number of other studies relate to hazards emanating from chemicals, other new technologies, etc. In the consumer arena, risk assessments for products are generally not made public. The only major visible effort is by the Product Safety Commission, which is concerned with technical facts of hazard assessment.

It appears from the newly emerging risk assessment literature that risk assessment has two components: one dealing with risk estimation and the other dealing with risk evaluation. As Rethans (1980) points out, the social facet of risk evaluation pertains to the importance attached to these hazards by society, and thus risk evaluation inevitably involves the judging of the "acceptability of the risk."

While some notions of perceived risk included the "physical risk" inherent in the product, no empirical work had been undertaken to determine the "acceptable level of risk" with respect to products utilizing the "perceived risk" concept. The first study which has looked at the hazard management area for the products is by Rethans (1980). Since his paper has dealt with his study extensively, I will not detail his study here. Suffice to point out that his study makes a contribution to the important area of risk evaluation by consumers, its determinants and how to deal with product risks. The studies that are planned by Jacoby and his associates as part of their NSF-funded studies, are aimed at answering the question: "Do consumers actually consider health and safety information when making purchase, usage, and disposal decisions regarding technologically innovative products?" Thus, very little empirical evidence is currently available to make policy recommendations, nor is there enough evidence to make suggestions that will aid management decisions with respect to management of hazards relating to products. Therefore, this paper attempts to point out the relevance of this area to researchers in the consumer behavior area and to provide some guidelines regarding some of the issues that are emerging in this area.

Approaches to Acceptable Risk

It would appear that the greatest relevance is in the area of investigating the determinants of risk perception by consumers so as to aid in the evaluation of which hazards, and in what form, the consumers are willing to tolerate. Since there is no such thing as a completely risk-free product, consumers may be said to have an "acceptable level of risk." Some (Fischhoff, et al. 1980) contend that the term acceptable level of risk is a misnomer inasmuch as what we have at best is "that level of risk accepted according to a particular version of a particular kind of rationality" (p. 9). There are five approaches which appear to be currently utilized to determine the "acceptable level." They are: (1) formal analysis, (2) nonintervention, (3) procedural approaches, (4) professional standards, and (5) comparative approaches.

In the formal analysis approach, such familiar analyses as cost-benefit analysis, risk-benefit analysis, decision analysis and value-aspect assessment are included. The nonintervention approach is characterized by the free market which sets prices on alternative hazard management option, and individuals choose according to price and personal definition of acceptance risk. In other words, this is an extreme position, as this approach opposes any intervention into dealings between buyers and sellers, which is conducted under the protection of a watchful and responsive legal system.

The common approach that is currently in vogue is the procedural approach. This approach relies on the political and economic forces molding the standards, by their interaction with one another and the varied feedback provided by their environment. The professional standards are provided by the technical experts most knowledgeable in a field. Finally, the comparative approaches look at the way risks have been balanced in the past or balanced at present, and they should guide the risk management decisions for the present and for future action.

As is evident, some of these approaches require some centralized decision-making body. Each approach has its pros and cons, and its limitations. None of these approaches incorporate the subjective approach or the psychological elements of the risk assessment process that are critical in the evaluation and management of hazards relating to consumer products. Slovic et al. (1980(b)) paper examines the nature of perception of risks by individuals, which has direct relevance to researchers in consumer product safety area. They advocate a method of "expressed preferences" to get directly to the problem of perception of risk and in my judgment such a method is well suited to the study of acceptable risk for consumer products.

Slovic's (1980(b)) study attempted to ascertain what people meant when they said that a particular technology was risky. They found that the risk judgements of lay people related only moderately to the annual death rates and the risk perceptions were not closely related to their own fatality estimates, In other words, their study indicated that risk perception did not mean annual fatalities or frequency of deaths. It revealed that individual's perception of "disaster potential" played an important role in arriving at risk judgments. In addition, an array of qualitative considerations were associated with risk perception by individuals. The characteristics that are associated with risk perceptions in their study are: (1) voluntariness of risk, (2) immediacy of effect, (3) knowledge about risk, (4) to what extent are the risks known to science, (5) control over risk, (6) newness/ familiar risk, (7) chronic or catastrophic nature of the risk, (8) common hazard or a dread hazard, and (9) severity of consequences.

Some Issues to Consider

Much of the individual decisions regarding choice of products relate to voluntariness of the risk. Where solutions are mandated, as in the example of air bags for automobiles, or standards of food additives, etc. such situations characterize the involuntariness of the risks on the part of consumers, While the Slavic (l1980(b)) study did not find much difference in the evaluation of risk with respect to voluntariness, one would expect that in the context of consumer behavior, this consideration might play an important role. This is a major area which needs investigation by consumer researchers before proceeding to investigate other characteristics. The rationale is provided by Lorance (1976) who asserts that what is acceptable (by consumers) definitely depends on the degree to which we, as consumers, are free to opt for or decline the risk. Where individual's options are limited, such as the example where a governmental agency determines the acceptable level of risk and preempts the individual's voluntary decision-making, the involuntariness might greatly affect the perception of risk and its acceptable levels.

The second aspect which needs the attention of consumer researchers is the perception of "disaster potential." For example, in the recent incidences of toxic shock syndrome, the ailment rate is three in 100,000. However, preliminary surveys indicate that consumers may not have been concerned with average fatality per year (as in the Slavic study), but the potential catastrophic effect such products may have on the health and safety of consumers. Experts point out that not only for this product but for a variety of consumer products, testing before introduction of such products would not have been able to point out the possible hazard. That is, hazard information is unlikely to emerge before the products are marketed. This may also point up to the desirability of post-marketing surveillance in order to aid risk evaluation.

A related problem in the context of consumers is that they must rely on inferences based on what they remember hearing or observing about the hazards. Slavic (1979) points out that one heuristic that has special relevance for risk perception is called "availability," Thus consumers can imagine or recall instances that occur frequently and numerous factors (such as recent incidences) affecting availability might distort their judgements.

Another factor that relates to consumers is the "overconfidence" factor. Many issues relating to products result in injuries rather than being fatal. In addition, their own experience as consumers without facing any hazards tend to strengthen their views that the hazards cannot be serious, particularly when an individual consumer's own hazard experience with respect to any of the risks associated with a product may be overwhelmingly benign. Thus, such overconfidence might affect their risk perception and the level of acceptable risk for a variety of products.

To aid consumers in risk perceptions, more and more warning labels are advocated for products with the hope that such information may he utilized by the consumer both for making purchase and use decisions. However, as Slovic et al. (1980(a)) point out, merely mentioning possible adverse consequences (no matter how rare) could enhance their perceived likelihood and make them appear more frightening. A recent study on a hypothetical Patient Package Insert by Keown (1980) lends some support to this notion. This is an area where considerable research is needed to help public policy makers and governmental agencies charged with investigating and recommending warning labels to products.

Another problem peculiar to consumer research may be the perception by a large number of consumers that risks associated with products are managed by competent professionals (both in the business organization and in the governmental agencies) that they need not be concerned about them. In addition, lay people seem to believe that our technological progress will solve the hazard problems. Such complacency on the part of consumers might well affect the efficiency of educational programs and the use of labels provided with the products in aiding consumer decision-making.

There are a number of other issues that should receive the attention of consumer researchers concerned with the perception of risk by consumers and how they handle such risks. Suffice to point out that the new discipline of "risk assessment" is very much relevant to researchers in the consumer behavior area, and we have a long way to go before we will be in a position to make recommendations for policy-makers.

REFERENCES

Engel, James F. et al. (1978), Consumer Behavior, Hinsdale, Ill.: The Dryden Press (3rd Edition).

Fischhoff, Baruch, Slavic, Paul, Lichtenstein, Sarah (1978) "How Safe is Safe Enough? A Psychometric Study of Attitudes Towards Technological Risks and Benefits," Policy Sciences, 9, 127-152.

Fischhoff, Baruch, Slavic, Paul and Lichtenstein, Sarah, (1979), "Weighing the Risks. Which Risks are Acceptable?" Environment, 21, 4, 17-38.

Fischhoff, Baruch, Lichtenstein, Sarah, Slovic, Paul, Derby, Stephen L., and Keeney, Ralph L. (1980), Approaches to Acceptable Risk (Preliminary Draft).

Keown, Charles Jr. (1980), Attitudes of Laypeople. Physicians and Pharmacists Towards Prescription Drug Side Effects, Doctoral Dissertation, University of Oregon, College of Business Administration.

Lowrance, W. (1976), Of Acceptable Risk, Los Altos, California: William Kaufman.

Rethans, Arno (1979), An Investigation of Consumer Perceptions of Product Hazards, Doctoral Dissertation, University of Oregon, College of Business Administration.

Slovic, Paul, Fischhoff, Baruch, and Lichtenstein, Sarah (1979), "Rating the Risks," Environment, 21, 3, 14-39.

Slovic, Paul, Fischhoff, Baruch, and Lichtenstein, Sarah (1980(a)), "Informing People About Risk," Paper Presented at the Conference on Product Labeling and Health Risks, Banbury Center, Cold Spring Harbor, New York, May 21-23.

Slovic, Paul, Fischhoff, Baruch, and Lichtenstein, Sarah (1980(b)), "Perceived Risk," in R. C. Schwing and W. A. Albers, Jr., (Eds.) Societal Risk Assessment: How Safe is Safe Enough? New York: Plenum Press.

----------------------------------------