Comments on Papers on Consumer Policy Proposals

E. Scott Maynes, Cornell University
[ to cite ]:
E. Scott Maynes (1979) ,"Comments on Papers on Consumer Policy Proposals", in NA - Advances in Consumer Research Volume 06, eds. William L. Wilkie, Ann Abor, MI : Association for Consumer Research, Pages: 472-476.

Advances in Consumer Research Volume 6, 1979      Pages 472-476

COMMENTS ON PAPERS ON CONSUMER POLICY PROPOSALS

E. Scott Maynes, Cornell University

In my view successful research provides convincing answers to interesting and/or important questions. Part I of these comments considers the three papers presented in this session as successful or unsuccessful research from this viewpoint. In Part II, I turn to the lessons that these papers hold for future research on consumer policy.

I. RITCHEY-CLAXTON ON "ACTIONS TO REDUCE CONSUMER SHOPPING PROBLEMS"

The first task of this paper is to uncover an exhaustive list of shopping problems for various products as consumers see them. In a second stage the research seeks to assess the relative importance of these perceived shopping problems. Next, the researchers ascertain consumers' preferences for alternative corrective actions. Finally the major findings of the study are presented, first, to a small group of top government policymakers and, second, to larger groups of middle-level government officials who are asked to evaluate alternative corrective actions as to their desirability and feasibility.

The ACR paper deals only with Stages 1-3.

The General Approach

"I shot an arrow into the air ..." This phrase graphically conveys the spirit in which most of us conceive and report our research. Or, more cynically, we craftily design our research with an eye to acceptance by a particular journal (lest we perish!). Not so Professors Ritchie and Claxton. The communication of their findings and the discussion of their implications with policymakers are an integral part of this research plan. For this we can only applaud. Unfortunately, our applause does not extend all of their research design and its execution.

A second feature of this research that does merit our approval is the assumption that parties affected by consumer policy--government officials in particular, but also consumer activists and businessmen--should be familiar with consumer shopping problems as consumers themselves perceive them.

Where I part company from Mssrs. Ritchie and Claxton is with their implicit assumption that consumers can pass meaningful judgment on the desirability of various corrective actions. In Table 4, for example, we learn that 75 percent of respondents agree that "expanded complaint bureau" facilities are desirable for dealing with auto repair problems. At what cost to each taxpayer? We learn also that only 34 percent agree that "shopping center information booths to help consumers" are desirable. But what information is assumed to be provided?--(1) platitutidous advice such as "Shopping on credit is more expensive" or (2) relevant local information--for example (an actual example from Minneapolis), the fact that pocket cameras of superior (but equivalent) quality in Consumers Union's judgment can be purchased at prices ranging from $80 to $300, and giving the name of the retailer quoting the lower price? Without specification of the information a booth would provide, the recorded attitudes are meaningless.

Thus, I praise Ritchie and Claxton's plan to report and discuss their research findings with appropriate government officials. Certainly, too, public officials as political animals must be aware of public attitudes toward proposed policies, be these attitudes well or ill-informed. But the consideration of appropriate consumer policy requires more than knowledge of consumer attitudes. It requires economic analysis, taking account of prospective benefits and costs.

PILOT STUDY OR NO: THE NATURE OF THE POPULATION AND THE SAMPLE

Pilot Study or No: The Nature of the Population and the Sample

Research pertaining to an intrinsically uninteresting population (someone's class) or based on a defective sample of an interesting population commands interest only as a pilot study, that is, as an exploration of unknown territory or a demonstration of a new technique or approach.

As consumers of the Ritchie-Claxton paper, you may have already discovered that you are unable to discover the population to which the data in Tables 2-4 allude. Nor will you find any description of the techniques by which this unknown population is sampled. Hence, you will be unable to know whether the fetching data of Table 3 are representative of the respondents only, or of a larger and (perhaps interesting) population. Further research on my part proved inconclusive, but only increased my doubts. In the preliminary version of this paper, the authors wrote:

Participants in the study were obtained through contact with the local headquarters of the Boy Scouts of Canada in each of five cities across Canada. Five cities (Vancouver, Regina, Ottawa, Quebec City, Halifax), were chosen in order to obtain inputs from each of the geographic regions of the country. Within each city the local Boy Scout representatives were asked to provide a total of approximately 60 participants...

Good for the Boy Scouts! But how did they choose? No information for answering this question is provided. Hence, I can only recommend excommunication for Ritchie and Claxton and others like them who fail to provide such essential information.

It would be my judgment that these authors do not break enough new ground to legitimize their research on pilot study grounds.

Consumer Shopping Behavior As Terra Incognita

Judging from their references (the Sentry Insurance Company survey excepted), Ritchie and Claxton have approached the problem of consumer shopping problems as though they were exploring virgin wilderness.

We all know that it is not. Presumably reflective of problems encountered in shopping, the first Better Business Bureau was established in 1912! Consumer economics texts (Gordon's Economics for Consumers and Troelstrup's The Consumer in American Society) began discussing consumer shopping problems normatively in the 1930's. A paper of mine (Maynes, 1973) explored the causes of consumer information and shopping problems in 1973. The excellent Best-Andreasen national study of consumer dissatisfaction and complaints was published in 1976 (Best and Andreasen). Finally, two conference proceedings (Day, 1977 and Hunt, 1977) have focused entirely on consumer satisfaction and dissatisfaction, much of it related to shopping.

"Better" research builds on previous knowledge. I chide Mssrs. Ritchie and Claxton for approaching mapped territory as though it were totally unexplored.

Summing-Up

Comments that began with approval have unfortunately been succeeded by the citation of blemishes that convert this from "successful" to "unsuccessful" research. Its merits are an interesting and useful target, and the feedback of the research findings to officials responsible for formulating and executing corrective policies, a feature highly worthy of evaluation. But apparently all this effort went for naught, lost on a sample that may be representative only of itself. (I would be delighted if information were provided to show that it was a probability sample, satisfactorily executed, or--alternatively--a quota sample that could be shown to be reasonably "representative.") Finally, this research did not build--at least there was no evidence to show it--on past exploration of this area by others--the Better Business Bureaus, the normative consumer economists, and recent consumer behaviorists.

MOSCHIS ON "AN EMPIRICAL ASSESSMENT OF FORMAL CONSUMER EDUCATION"

When 92 percent of Americans support the introduction of more consumer education in American schools--as the Sentry Insurance survey shows [Sentry, 1976, p. 76], then it becomes important to know whether some variant of formal "consumer education" is effective. Hence, Moschis' research poses an appropriate question. Unfortunately, as the remarks below suggest, his research does not provide a "convincing answer" to the question posed for some reasons that are beyond his control and some that are within his control.

The Factors Beyond Moschis' Control

A fundamental problem for this investigation is that no consensus exists as to what constitutes "consumer economics'' or "consumer education." As an example, I as the author of an undergraduate text in Consumer Economics would argue that an understanding of price discrimination is a sine qua non for the consumer who would purchase effectively in a contemporary local consumer market. [Salop (1977) has modeled local markets in which sellers sort our consumers on the basis of their knowledgeability, charging higher prices to the ignorant and lower prices to the knowledgeable. It would be my judgment that this model is highly realistic.] Price discrimination (and especially the point that it may be initiated by the consumer) is a topic that is emphasized in my text. It is omitted or passed over most cursorily in most texts now on the market. [Daniel A. McGowan's Consumer Economics, (Chicago: Rand-McNally, 1978) is a welcome exception.]

Going further, I would assert that most texts are mainly descriptive ("preachy") while the problems of consumers are analytical: How much should I search? How do local retail markets work? How do I assess the quality of products? Given inflation and taxes, what asset should I choose for the investment of my savings? What's more, the profession has not yet agreed on which disciplinary background--economics, psychology, consumer behavior(!), home economics--should be stressed most strongly. Given this undeveloped, non-consensus state of consumer economics and consumer education at the university level, is it strange that Professor Moschis may encounter difficulties in deciding what consumers should know and be tested on? Or that I should criticize one of his tests as "invalid"?

The test that I single out for criticism is that relating to "consumer financial management." In essence, "mastery" of consumer financial management is signaled by the test when a student-respondent can accurately estimate the percent of income an average American family would spend on such major categories as food, clothes, home expenses, automobile expenses, other expenses, and saving. Though this information may be interesting, I doubt that it aids in effective financial management for "my family." My major doubts are confined to this particular test; the others seem defensible. However, they might be more defensible had they been gleaned from a conference of "experts" in this (diverse) field.

The Factors Within Moschis' Control

The Sample. In seeking to test the effectiveness of formal consumer education, Moschis has faced two sampling problems: (1) the selection of consumer education courses or subject matter taught, and (2) the selection of a representative group of students.

Moschis could have selected--and this is what I would have viewed as preferable--a course viewed by some group of "experts" as a "very good" or "model" course. Such a procedure would have the advantage that we would (or could) know the exact contents of the stimulus to which students were exposed. Instead, Moschis chose to select a sample of 13 schools in Wisconsin. In effect, this means that the consumer education materials to which students were exposed are a weighted average of whatever is taught in the "consumer education" courses of these 13 schools.

And there lies the rub. According to Moschis, "some [how many?] of the schools were chosen on a convenience basis and some on a random basis." We have no information on the degree to which either courses or the students enrolled in them are representative of all Wisconsin consumer education courses and students. One can observe that, if school-to-school variation in courses or students is great, a sample of 13 from the hundreds of Wisconsin schools would be frightfully inadequate, even if chosen by probability methods. So I chide Moschis for failing to utilize adequate sampling methods and for failing to provide a detailed description of the methods he did apply. As consumers of this research, we can only hope that this sample is representative! But it would have been appropriate--and is still appropriate --for Moschis to tell the reader in some detail just what consumer education courses students were exposed to, what their content was, and, quantitatively, in what ways students in these courses differed from average students in their schools. We know they were more female; were their other factors operative?

"Significant" Versus "Statistically Significant" Differences. In his analysis Moschis devotes considerable space to asking whether mastery of consumer education materials differs "significantly" by age (younger vs. older adolescents), social class, and sex. Not very interesting questions. But they provide classic examples of "statistically significant" differences that are not substantively important. For example, from Table 1 we learn that older adolescents score 12.07 (out of 30) on "information seeking, while younger adolescents score only 11.17 (out of 30). A statistically significant difference at the .01 level. But does Moschis contend that a less-than-one-point difference on a rather crude 30 point scale is "important" in determining the effectiveness of formal consumer education?

What interests me far more--and what goes virtually undiscussed--are the mean values of these scales. But of even greater interest is whether students with higher scores have become more effective consumers in real life. Here we have no evidence at all.

The Quantification of Formal Consumer Education. Finally, in Table 4, we come to the key question: does exposure to formal consumer education exert a statistically significant effect (the narrow question first) on the five components of consumer skill--consumer knowledge, finance management, economic motivations, information seeking, consumer activism? The answer of the partial correlation coefficients is emphatically "no." But is this answer convincing, even for this sample? It may not be. The formal education variable, as entered in the equations, was defined as "the number of consumer related courses taken." The mean value was 3.3 courses. My complaint is that each course is treated as exerting a uniform effect; should not the first course taken have been given a greater weight?

Because of the unsatisfactory statistical formulation of the narrow question, there is no point in asking the important question: does exposure to formal consumer education courses contribute to more effective consumer behavior?

Summing-Up

For reasons both beyond and within the author's control, this investigation does not, in my judgment, provide a convincing answer to the question posed. But it does provide useful lessons for those to come both with respect to steps to be taken and steps not to be taken. As a profession, we must press on to a convincing answer

GOODWIN, MAHAJAN, AND BHATT ON "CONSUMER ARBITRATION AS AN ALTERNATIVE DISPUTE RESOLUTION MECHANISM"

This is not your usual research paper. It reviews the history and then describes the working of consumer arbitration as a complaint resolution mechanism. Finally data are presented for one "large northeastern metropolitan area" relating to such facets of arbitration as the fraction of cases resulting in an arbitration award, the reasons for non-arbitration, the types of businesses, the dollar amounts, the resolution, and time taken for arbitration.

Fraud in Perspective

Following the presentation of sample data on consumer arbitration in one metropolitan area, the authors declare: "Consumer arbitration appears to have a great potential for becoming a satisfactory dispute-resolving mechanism for both consumers and businessmen." In my judgment neither their discussion nor the data they present support this conjecture. Let me say why.

I start with a proposition that a complaint resolution mechanism that is open only to a tiny fraction (the most persistent and most knowledgeable) complaining consumers is not a "satisfactory" dispute-resolving mechanism even though it treats those who use it fairly. The reason: it may offer no satisfactory avenue for the great mass of consumer complainants. Indeed, if the arbitration mechanism is utilized by a sufficiently small fraction of complainants, it may be appropriate to label it a consumer "fraud" because it gives the appearance of providing justice without providing "satisfactory'' complaint resolution to the large majority of complainants. Underlying this observation is the assumption that the time costs and subjective costs (feelings of awkwardness, inadequacy, annoyance, etc.) are likely to be very great on the part of consumer complainants. If they are sufficiently great, consumers may not take the next step in the consumer redress process.

The critical element missing from this paper is the failure to tell us what fraction of grievances presented to the Better Business Bureaus have been taken to arbitration in this sample area. Bear in mind that the Better Business Bureaus process only a small fraction of voiced grievances--0.3 percent of cases where consumers perceived problems and 0.8 percent of voiced consumer complaints (estimated from data published in (Best and Andreasen, 1976, pp. 46-47).

One further doubt on the author's optimism. In my judgment a median resolution time of 68 days (from Table 6) is "speedy" only when compared with litigation.

Summing up, consumer arbitration may turn out to be either (1) a satisfactory dispute-revolving mechanism (it is devoutly to be wished!), (2) a consumer fraud, or (3) something in between. We are in urgent need of further research.

II. THE FUTURE AND CONSUMER POLICY RESEARCH

The organizer of these meetings, William Wilkie, asked each discussant to discuss the field in general. It is pleasant indeed to do what comes so naturally!

Parochialism and Publications Overload

Consumer policy is endemically multidisciplinary. You can make the argument as well as I. So I will practice parsimony and let you do it.

What appalls me--and I speak not only of the papers in this session but of many of the papers published in the Journal of Consumer Research, the Journal of Consumer Affairs, and the new Journal of Consumer Policy--is the parochialism that I note in the citation and revealed ignorance of adjacent disciplines. Consumer economists cite consumer economists, markets cite marketers, economists-economists, etc. It seems to me that this parochialism has great perils. First, we may omit essential elements and come up with inappropriate policy recommendations. An example: the consumer economists who advocated unit pricing and failed to recognize its information processing defects (Russo, 1977). Second, the limitations of narrowly conceived consumer policy research may become obvious to policymakers and hence our influence on policy may be greatly (and rightly) diminished.

It is appropriate to ask why this parochialism. One answer comes from the narrow, academic guilds into which we are organized and within which rewards are dispensed (Departments of Marketing, Consumer Economics, etc.). "Outsiders"--the adjacent disciplines--tend to be unrecognized, uncared for, and denigrated.

But a second answer comes from a concept that has occupied students of consumer behavior much in recent years--information overload and, by extension, publication overload. This Conference, responding to the publish-or-perish imperative of the academic reward system, will contribute at least 600 pages more, certainly to be published and perhaps to be read. (Little wonder we tend to underread the literature of our adjacent disciplines!)

(I find it ironic and instructive that one of the proponents of the "Information overload" concept, Jacob Jacoby, did not discover and cite the articulator of an economic theory of information overload. I refer to Staffan Linder and his Harried Leisure Class (Columbia Press, 1970) in which he notes (Ch. 6) how increased affluence applied to fixed time will necessarily result in less search and increasingly sloppy decisions!)

We in consumer research have no monopoly on the publication overload problem; it is a scourge of all academia. The solution? Two steps, panaceas by no means, come to mind. The first is to make publication more costly, perhaps by charging a stiff review fee for submitted manuscriptions. The second is to reward discriminating reading more, by counting, assessing, and by shifting the rewards in favor of discriminating reading, for example, by taking account in the making of salary and promotion decisions of the quantity and quality of review activities.

Information is so central to our profession that reform should start with us. More particularly, can we realistically expect policymakers to read the 600 or more pages that this Conference will produce?

Pilot Studies vs. Substantive Research

For empirical research novelty justifies the use of data taken from an intrinsically uninteresting, unrepresentative population, or a badly biased sample of an interesting population. In short, there is a place for pilot studies.

But substantive studies will deserve our attention only when they are based on adequate samples of interesting populations. As consumers of research, we must insist that researchers tell us enough, in the article we are reading, to enable us to make an informed judgment ourselves regarding the character of the population and the adequacy of the sample as executed. We should ask reviewers to dismiss as unacceptable (or only acceptable after revisions) papers that do not meet these standards.

A MATTER OF PROPER CONNECTIONS: THE CONCEPTION, EXECUTION AND FOLLOW-UP TO CONSUMER POLICY RESEARCH

Ask and answer an uninteresting or unimportant question and you have wasted resources. So it is a matter of some importance to pose a question that is worth answering.

Ask an important question and answer it unrealistically and you will also have wasted resources since your "answer" will prove impracticable.

Finally, answer an important question realistically and you may still fail, because the relevant decision-makers may fail to notice or to use your answer.

How may these pitfalls be avoided? One possible means is to touch base with consumer policy people of various roles before, during, and after the undertaking of research.

Whether any groups or individuals can take a sufficiently broad view of a field so as to be able to identify the "proper" questions I am not sure. But many have tried. For those contemplating consumer policy research I would call your attention to several papers that seek to map the consumer policy field and its gaps.

A recent NSF Project (Consumer Affairs Institute, 1976) brought together an altogether remarkably assemblage of consumer advocates, consumer "bureaucrats," consumer affairs people from business, and consumer researchers to identify, through organized brainstorming sessions, important consumer research problems. A second NSF-sponsored Conference (Denney and Lund, 1978) included researchers, businessmen, consumer advocates, and academics and ex post sought their reactions to the consumer policy research that was reported at the Conference. A recent paper by a consumer economist (Shepard, 1978) has provided a series of taxonomies of consumer policy research that should help to identify gaps. Finally, responding to a call from the United Nations Centre on Transnational Corporations I, as one student of consumer policy, sought to organize the field and pose the most important issues (Maynes, 1978).

But one would be fatuous to expect that any group or any individual would raise either all the appropriate questions and certainly not the outlines of appropriate answers. Is it not precisely the original question or answer that is unlikely to be anticipated? By the same token, it can be argued that practitioners of consumer policy or consumer affairs are so close that they tend to be preoccupied with the immediate. Consider a possible policy to deal with the problem of consumer complaints: change the law so that complainants whose grievances are judged genuine are entitled to be compensated for the time they spend in seeking consumer redress. A plausible proposal, I suggest. But never have I heard it voiced by a consumerist!

As a final comment, I would like to reiterate my praise of Mssrs. Ritchey and Claxton for incorporating the feedback of their research results to consumer policy-makers. This is indeed a feature that all of us should seek to incorporate into our research programs!

REFERENCES

Arthur Best and Alan R. Andreasen, Talking Back to Business: Varied and Unvaried Consumer Complaints, Working Paper (Washington Center for the Study of Responsive Law, 1976).

Consumer Affairs Institute, Toward the Development of a Comprehensive Consumer Research Program (Washington: National Science Foundation, 1976).

Ralph L. Day (ed.), Consumer Satisfaction, Dissatisfaction and Complaining Behavior (Bloomington, Ind.: Indiana University, School of Business, 1977).

William Michael Denney and Robert T. Lund, Research For Consumer Policy, Proceedings of a Conference (Cambridge: MIT Center for Policy Alternatives, 1978).

H. Keith Hunt (ed.), Conceptualization and Measurement of Consumer Satisfaction and Dissatisfaction (Marketing Science Institute, 1977).

E. Scott Maynes, "Consumer Protection: The Issues" (Unpublished Paper prepared for the United Nations Centre on Transnational Corporations, 1978).

E. Scott Maynes, "Consumerism: Origins and Research Implications," in Eleanor B. Sheldon (ed.), Family Economic Behavior: Problems and Prospects (Philadelphia: Lippincott, 1973).

Daniel A. McGowan, Consumer Economics (Chicago: Rand McNally, 1978).

J. Edward Russo, "The Value of Unit Price Information," Journal of Marketing Research, 14 (May, 1977), 193-201.

Steve Salop, "The Noisy Monopolist" Price Dispersion, and Price Discrimination," Review of Economic Studies, 1977, Forthcoming.

Sentry Insurance Company, Consumer at the Crossroads (Cambridge: Marketing Science Institute, 1977).

Lawrence Shepard, "Toward a Framework for Consumer Policy Analysis," Journal of Consumer Affairs, 12 (Summer, 1978, 1-11.

Hans B. Thorelli and Sarah V., Consumer Information Systems and Public Policy (Thorelli: Ballinger, 1977).

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