Discussion of the Three Papers on the Consumption of Consumer Goods

Jagdish N. Sheth, University of Illinois
ABSTRACT - All the three papers relate to issues of social marketing in some way or the other. The theoretical paper in a radical theory of consumption tries to apply some of the concepts of Karl Marx. The paper on failures and successes in bringing about less consuming life style changes is an empirical study. The final paper investigates demographic and attitudinal factors related to energy conservation.
[ to cite ]:
Jagdish N. Sheth (1982) ,"Discussion of the Three Papers on the Consumption of Consumer Goods", in NA - Advances in Consumer Research Volume 09, eds. Andrew Mitchell, Ann Abor, MI : Association for Consumer Research, Pages: 313-314.

Advances in Consumer Research Volume 9, 1982      Pages 313-314

DISCUSSION OF THE THREE PAPERS ON THE CONSUMPTION OF CONSUMER GOODS

Jagdish N. Sheth, University of Illinois

ABSTRACT -

All the three papers relate to issues of social marketing in some way or the other. The theoretical paper in a radical theory of consumption tries to apply some of the concepts of Karl Marx. The paper on failures and successes in bringing about less consuming life style changes is an empirical study. The final paper investigates demographic and attitudinal factors related to energy conservation.

HOW RADICAL IS RADICAL THEORY?

The radical theory of consumption proposed by Dholakia is a thought-provoking paper. There is an excellent review of necessary criteria for a theory to be labeled radical in any discipline. The author then proposes to develop his own radical theory of consumption based on some of the concepts of Karl Marx. The basic proposition of the theory seems to be the allocation of human energy between domestic production and industrial production which generates income and wealth with which to engage in consumption. Reciprocally, consumption of domestic and industrial production generates a set of work ethics and values which influence the choice among leisure, wage work and household activities.

Unfortunately, the theory of consumption is not as radical as the author thinks. With regard to all the three major aspects of a radical theory, there already exists numerous theories and hypotheses in the literature. For example, on the issue of closing the loop, several scholars such as Becker and Lipset have proposed significant theories related to allocation of time between wage work and leisure work. Similarly, on the issue of emphasizing the context, we already have such radical theories as Veblen's conspicuous consumption and Reisman's other directed society. Finally, exposing the ideological aspects of consumption (subverting the hegemony) is strongly entrenched in the traditional literature on welfare economics, and in the more recent literature on consumerism and consumer's bill of rights.

In short, the radical theory of consumption is not as radical as the author thinks. In my opinion, this is a direct consequence of the confusion between a theory of consumer behavior and consumption behavior. The author seems to be not aware of the vast literature in consumption economics and behavioral economics.

I also think it is less appropriate to utilize Karl Marx's propositions in consumption behavior for several reasons. First, they clearly presume that we live in a homogeneous society. While this may have been true in his era, it is certainly not true of the contemporary consumption society. Second, Karl Marx presumes that the quality of production and consumption are comparable. We know that in a highly industrialized society, this is simply not true. Again, it may have been very true in his era which was dominated by the Agrarian economic activities. Finally, Karl Marx also presumes that the magnitudes of production and consumption are equal at a point in time. We already know that in the modern industrial states this is simply not true. We can actually create inequality between production and consumption by providing time and place utilities in addition to the form utility in our production functions.

In fact, we can easily assert that the classical functions of marketing such as collecting, sorting and grading came into existence to accommodate the above three problems inherent in any Production-conSumption system.

I believe a more fundamental issue relates to the choices a society makes between make, barter or buy decisions. In other words, what are the underlying factors which account for the distribution of production and consumption activities in terms of make, barter and buy choices? Adam Smith would argue that the degree of specialization and economy of scale concepts may go a long way to explain these choices a society makes rather than any ideological Concepts.

LOWERING LIFE STYLES IS NOT EASY

The second paper by Gronhaug and Ogaard is an interesting study of the conscious efforts carried out by a group of_ Norwegians who belong to a voluntary organization which believes in self actualization. The authors discover that this group of consumers were more able to intervene and self control such activities as daily diet, energy conservation, growing vegetables, picking berries and performing physical exercises. On the other hand, they were less able to control social engagements, drinking, smoking, vacationing and leisure activities. Beyond this empirical finding, the paper generates very few additional insights despite a lot of statistical analysis including factor analysis.

In some ways, the statistical analysis of the data resembles the use of a cannon to shoot a mouse. In fact, it would appear that the authors fall into the trap of making causal inferences out of the data which are at best correlational. This is particularly true when they assert that intentions precede outcomes based on a moderate correlation between intended and successful efforts in bringing about life style changes.

In my opinion, it is less useful to collect survey research data to measure voluntary efforts at conservation and simplified life styles. First of all, volunteer groups such as the Norwegian group are not representative of the population even if they resemble the national demographics Second, the self reports on various activities are likely to be highly subjective and biased. Finally, even if one can correctly measure these activities, it is very difficult to estimate magnitudes of behavioral change. Instead, it is much more useful to conduct field experiments or observational research

CLAIMING ENERGY CONSERVATION IS NO CONSERVATION

The third paper by Semenik, Belk and Painter is an empirical study on the determinants of energy conservation After a very extensive review of studies on energy conservation, the authors find that most studies are inconsistent and contradictory. Consequently, they fail to pinpoint what demographic and psychographic factors influence or motivate people to conserve energy. The authors carry out their own study on a sample of households in Salt Lake City. They carry out extensive multivariate analysis of the data to conclude that beliefs about gasoline shortage and attitudes toward potential conservation methods discriminate much better than either self reported gasoline consumption behavior or household demographics.

Unfortunately, this study suffers from numerous methodological problems. For example, it is totally inappropriate to create groups out of continuous dependent measures in order to perform discriminant analysis. The groups must be naturally defined and must rem>in invariant. Otherwise, the resultant classification tables are likely to generate arbitrary chi squared values which can be easily changed with each redefinition of the group. The authors make further mistakes of removing the most typical energy conservers by creating two groups from the extreme ends of the energy conservation distribution. Thus, their results of discriminant analysis are more appropriate for the minority of extreme groups of conservers and non-conservers but not at all appropriate for the middle majority who are neither conservers nor non-conservers.

Second, the authors seem to imply that a multivariate analysis of variance somehow generates more significance in the data than univariate analysis of variance. Statistically, it is impossible to create greater significance out of the data than what they actually contain except of course, the illusion of greater significance generated by artifacts of analysis.

Third, the authors seem to be oblivious to issues of multicollinearity. There is not a single reference to this serious problem in interpretation of discriminant weights even though many of the independent variables seem to be correlated.

Finally, it is not a good model or theory when only eight out of 32 independent variables reach statistical significance. There is no explanation or discussion as to why the vast majority of variables failed to emerge as significant discriminators.

However, the biggest problem with this study is the definition of energy conservers and non-conservers. First of all, it is based on self-reports and therefore suffers from the same problems of subjectivity and bias as found in the previous paper. More importantly, the two self report items which determine a person will be classified as energy conservers or non-conservers may have nothing to do with energy conservation. For example, "consuming less gasoline than five years ago" may very well be due to changes in work place, life cycle, divorce, marriage and a host of other factors. Similarly, "estimated mpg of the auto being driven" can be merely a function of changes in automobiles, age of the automobile and host of factors beyond the control of the individual. In short, defining energy conservation based on such self reports which contain many alternative explanations is no conservation at all.

Once again, this study along with the previous study on simplifying life styles clearly indicates the urgency of discouraging survey research and simultaneously encouraging field experiments or observation research which will generate hard data.

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