Comment on Becker and Murphy (1993) and an alternative model

 

 

AER's referee report and my refutation.pdf

 

 

JEL Classification:  D60, L15, M37

Keywords:   advertising, advertising and welfare, advertising and market        structure, advertising and game

Abstract

Becker and Murphy (1993)  say that advertisements and the goods advertised are complements in stable metautility functions and so consumers buy advertisements on a particular product. These premises are wrong. Consumers buy advertisements as a whole. Consumers do not know what products will be advertised. Consumers only expect the benefit or the harm of advertisements as a whole. Advertisements on a particular product is given away to consumers and the quantities are controlled by producers of the goods advertised as the traditional approach says. And there is a critical math error in their welfare analysis. Even if we admit their premises, their results about welfare are quite wrong.

I suggest an alternative model about advertising and analyse advertising with the new model, in which the consumers' demand for advertising as a whole is completely incorporated into the theory of rational consumer choice just like the supply of labor, and the firms' advertising is incorporated into the theory of rational firm choice just like other inputs.

Comment on Becker and Murphy (1993) and an alternative model

1.  Introduction

Traditional approach assumes that advertisements on a particular product is given away to consumers and the quantities are controlled by producers of the goods advertised. Becker and Murphy (1993) criticize traditional approach for having no way of determining how consumers make their choices about advertising. They treat advertisements and the goods advertised as complements in stable metautility functions. They insist that their approach readily incorporates the demand for advertising into the theory of rational consumer choice. Becker and Murphy (1993) is quite right and excellent to say that consumers buy advertisements. But consumers cannot buy advertisements on a particular product. And a particular advertising and the product advertised cannot be complements. I explain how consumers buy advertisements as a whole and how firms control advertisements on their products.

Advertising changes the good advertised. The goods unadvertised are totally different from the goods advertised. The traditional approach including Becker and Murphy (1993) overlooks this fact. Instead, they assume that firms' advertising affects consumers' demand curve, presuming that advertised goods are identical with unadvertised goods. All the confusions about advertising stems from overlooking that the goods unadvertised are totally different from the goods advertised. I clearly disclose their confusions.

2.  Comment on Becker and Murphy (QJE, 1993)

1) How do consumers make their choices about advertising?

Becker and Murphy (1993) include advertising as one of the goods that enters the fixed preferences of consumers. They treat advertisements and the goods advertised as complements in stable metautility functions. They say that their approach readily incorporates the demand for advertising into the theory of rational consumer choice. They criticize the dominant model of advertising which assumes both that advertisements are given away to consumers, and that quantities are controlled by producers of the goods advertised. They say this model has no way of determining how consumers make their choices about advertising.

Becker and Murphy (1993) is quite right and excellent to say that consumers buy advertisements. But consumers cannot buy advertisements on a particular product. When consumers subscribe to cable T.V.'s, newspapers, and magazines, they are buying advertisements as a whole. Consumers do not know what products will be advertised. They only expect the benefit or the harm of advertisements as a whole. Consumers make their choices about advertising as a whole, but cannot make their choices about a particular advertising.

Consumers have no way of determining how much they consume advertising on a particular product. Advertisements on a particular product is given away to consumers and the quantities are controlled by producers of the goods advertised. Consumers chose 10 advertisements. But the firm did not advertise. How can consumers buy 10 advertisements? The maximum quantity of advertisements consumers can buy is determined by the firm. The firm, not consumers, decides the quantity of advertisements for maximum profit.

Let's think about a simple example. Some companies give a computer to consumers, who should see a certain amount of advertising by internet. In this case consumers buy advertising by a negative price (a computer). When consumers buy the advertising, they do not know what products will be advertised. Consumers expect the utility of the advertisements as a whole. The important thing is that any firm can advertise its product on the websites of these companies. This mechanism is the same to T.V.'s, newspapers, and magazines.

Becker and Murphy (1993) say that the traditional approach cannot explain how consumers choose among different ads that require time, money, or other scarce resources. Becker and Murphy say about intensity of consumers' interest in advertising. But it is already considered by firms. Some firms advertise their products at news channels. Others advertise their products at sports channels. Computer manufacturers advertise at computer magazines. Firms carefully consider the effectiveness of their advertising. If consumers are not affected by their advertising firms do not advertise.

Consumers choose programs, channels, newspapers, or magazines. Consumers do not choose a particular advertising. A T.V. program is popular with teenagers. Then the program is sponsored (advertised) by the firms which make teenager needs. When teenager try to watch the program they have no idea about what goods will be advertised with that program. Consumers make their choice about advertising as a whole, but cannot make their choices about a particular advertising.

2) Are advertisement and the goods advertised complements?

Becker and Murphy treats advertisements and the goods advertised as complements in stable metautility functions. But a particular advertising and the product advertised cannot be complements. Consumers cannot choose the amount of advertising and the product advertised simultaneously in commodity space. Consumers cannot decide what advertising and how much advertising on a certain product will be seen on T.V.'s, newspapers, and magazines. Consumers can decide their preference on a certain product only after being exposed to advertising. That is a critical difference from complements. If two products are complements, consumers can choose optimal consumption level simultaneously in commodity space. Let's assume that firm A produce product G. How can consumers determine the optimal amount of advertising on the product G and the product G simultaneously? Consumers cannot know where and how much firm A advertises. It is impossible.

Becker and Murphy cite a study by several psychologists (Ehrlich, Guttman, Schobach, and Mills 1957). The study found that people who have recently purchased a new car were more likely to read ads for the same type of car than for other type. They insist this example supports their theory of complements. Naturally, consumers seek out and pay more attention to ads for products they have already purchased. Consumers want to know well their belongings and confirm their choices were well enough. But firms do not advertise for this kind of consumers. If a firm stops business after selling out inventory, the firm will never advertise after selling out inventory. There are advertisements on a particular product, not because consumers wants, but because firms wants. Firms use the input of advertisements to make their product. Advertisements cannot be separated from the goods advertised. Advertising is only a psychological input.

3) math error with indirect utility function

Becker and Murphy's premises are wrong. Consumers cannot buy advertisements on a particular product. Advertisement and the goods advertised are not complements. We can stop here. But there is another critical math error in their welfare analysis. So I continue to refute their paper. Even if we admit their premises, their results about welfare are quite wrong. They have a mistake in math.

Becker and Murphy use indirect utility function V (pp. 956-957). Indirect utility function V is a function of price (p) and income (W). If we assume, as Becker and Murphy insist, that advertising is a good or a bad which is traded in market, then indirect utility function V cannot be a function of A (the quantity of advertising). We should remember the following basic microeconomics.

V(p,W) = U(x^*(p,W))     (x^*: optimal consumption bundle)

If A is the market equilibrium amount of advertising, then dV/dA is always zero. Consumers choose optimal amount of advertising in market. So if consumers have to consume other amount of advertising V should decrease. Their dS/dA is always zero. (S=V+Π equation (9))

This does not mean that advertising is always optimal. In Becker and Murphy's model, advertising can be excessive, optimal, or underproduced as other goods because of externality or monopoly. In case of externality, S=V+Π+E (E: externality) and dS/dA=dE/dA will not be zero. In case of monopoly, dS/dP_A will not be zero (P_A: the price of advertising).

3. alternative paradigm about advertising

1) consumers' demand for advertising

Consumers supply the viewing services to massmedia. Consumer's viewing services are raw material to massmedia. Massmedia pay consumers by programs, articles, news, and so forth. Massmedia produce advertising delivery services with the raw material of consumer's viewing services. Massmedia sell advertising delivery services to firms.

Let's think about an explicit example. Some companies give a computer to consumers, who should see a certain amount of advertising by internet. In this case consumers sell the viewing services at a computer. These companies supply advertising delivery services to the firms which want to advertise their products. This mechanism is the same to T.V.'s, newspapers, and magazines.

2) firms' demand for advertising as an input

The goods are the things perceived by consumers. The goods become different by advertising. Advertising is only an input for production as capital, labor, and parts are. Advertising makes the product different as other parts do. Good parts make good quality product. Good advertising makes good quality product. The goods unadvertised is totally different from the goods advertised like the product being made of good parts is different from the product being made of bad parts. Psychological packing and distance from consumers are as important as physical packing and distance. Advertising is used for psychological packing and transporting just like physical inputs are used for physical packing and transportation. Advertising changes the psychological transactioncost of the goods advertised. Advertising is just a psychological input. It is only an input to production. It is an input just as a deliveryperson or a salesperson is an input.

If a firm wants to make a solid chair it uses more iron. If a firm wants to make friendly goods it uses more advertising. The employed amount of advertising is determined by the goods' inherent attribute firms want to create. Advertisements on a particular product is given away to consumers and the quantities are controlled by producers.

3) confusions about advertising

Firms' advertising changes the goods advertised. The traditional approach including Becker and Murphy (1993) overlooks this fact. Instead, they assume that firms' advertising affects consumers' demand curve, presuming that advertised goods are identical with unadvertised goods. They overlook the effects on product itself which is most important and essential.

(1) advertising and welfare analysis

Firms employ advertising as an input. Firms will choose optimal amount of advertising since they want to maximize profit. There can be neither excessive nor insufficient advertising if massmedia industry is competitive and produces no externality. It is the same when consumers supply the viewing services. There can be neither excessive nor insufficient advertising if massmedia industry is competitive and produces no externality.

If massmedia industry is competitive and produces no externality, we don't have to consider the welfare effect of advertising. We do not consider the welfare effect of a certain market if it is competitive and has no externality, since the market allocates resources efficiently.

The question whether advertising increases consumer surplus is meaningless. Advertising makes the product different as other parts do. The goods unadvertised are totally different from the goods advertised. Some firms use iron as one of inputs. We do not care for the circumstances that these firms should produce without iron or with a different amount of iron, since firms rationally demand iron. Then, why should we care for the case of advertising?

(2) advertising, competition, elasticity of demand, and market share stability

Some  products need a large amount of iron in production. Other products does not need any iron in production. We do not say that the former industries are competitive or that the latter industries have quite elastic demands. We do not say that the market share is more stable in the former industries than the latter. Some products need a large amount of advertising in production. Others do not. Why? We don't know. Firms know. Firms want to maximize profit. Their profit maximization input demands determine the amount of advertising. The only thing that we can know is that the employed amount of advertising is determined by the product's inherent attribute firms want to create, not by market structure, nor by elasticity of demand, nor by market share stability.

(3) advertising and prices of advertised goods

This question is also meaningless since advertising makes the product different as other parts do. The goods unadvertised is totally different from the goods advertised. Why should we compare the prices of totally different products?

(4) advertising as strategy, and advertising signaling the quality of the goods

It is not adequate to say that firms choose the amount of advertising as a strategy in a game. In reality firms choose output itself. Advertising changes the product itself. Advertising is just an input in any case. As long as massmedia industry is competitive, firms advertising choice cannot be a strategic variable just like a certain input demand in the competitive market cannot be a strategic variable. Firms' output might be a strategy in most games. When a firm chooses certain output as its strategy, its input demand is automatically determined. The amount of advertising is automatically determined when a firm chooses certain output as its strategy.

The firms recognized as producing high quality products naturally advertise much and charge a high price. It is just the same as high quality goods are packed luxuriously. Firms cannot use advertising as signaling its quality. Consumers don't know how much money firms spend in advertising. Advertising is only an input in any case.

(5) name brand and generic brand

Physical qualities are the same. But name brand advertises much, occupies most market, and charge a higher price. Can advertising alone make it happen? Then, why cannot generic brand advertise more and increase its market share? The important thing is that name brand is recognized as high quality producer from the very first. Name brand advertises to use consumer's belief that name brand is better. It is just the same as high quality goods are packed luxuriously. Name brand is packing its product luxuriously by advertising. Advertising is only an input in any case.

4. conclusion

Becker and Murphy (1993) have failed in incorporating advertising into the theory of rational consumer choice. A particular advertising and the product advertised cannot be complements.

Consumers supply the viewing services to massmedia. Consumer's viewing services are raw material to massmedia. Massmedia pay consumers by programs, articles, news, and so forth. Massmedia produce advertising delivery services with the raw material of consumer's viewing services. Massmedia sell advertising delivery services to firms. In my model, the demand for advertising as a whole i.e., the supply of the viewing services to massmedia can be completely incorporated into the theory of rational consumer choice just like the supply of labor. And firms' advertising is incorporated into the theory of rational firm choice just like other inputs.

A product is the thing perceived by consumers. Psychological packing and distance from consumers are as important as physical packing and distance. Advertising is used for psychological packing and transporting just like physical inputs are used for physical packing and transportation. Advertising is just a psychological input. It is only an input to production. All the confusions about advertising will be clear unless we overlook this essence of advertising.

5. Conclusion 2:  limited rationality and advertising

Consumers are not super-computers. There are innumerable goods in market. Consumers cannot use complete information about goods, even if they have complete information. Consumers choose goods by familiarity and good image, not by complete or available information. Firms advertise to give their products familiarity and good image. Psychological transporting make products familiar. Psychological packing give products good image. Advertising is just a psychological input. The goods are the things perceived by consumers. The goods become different by advertising. Advertising is only an input for production as capital, labor, and parts are. Advertising makes the product different as other parts do. Advertisements cannot be separated from the goods advertised. Advertising is just a part of goods since consumers' rationality is limited. There have been so many confusions about advertising since researchers separated advertising from goods.

References

Becker, G. S. and Murphy, K. M., 1993. "A Simple Theory of Advertising as a   Good or Bad," The Quarterly Journal of Economics, 108, 941-964.

Caves, R. E. and Greene, D. P., 1996. "Brands' Quality Levels, Prices, and    Advertising Outlays: Empirical Evidence on Signals and Information Costs,"  International Journal of Industrial Organization, 14 pp. 29-52.

Dixit, A. and Norman, V., 1978. "Advertising and Welfare," Bell Journal of    Economics, 9, 1-17.

Dorfman, R. and Steiner, P. O., 1954. "Optimal Advertising and Optimal Quality,"   American Economic Review, 44. 826-836.

Ehrlich, D., Guttman, I., Schobach, P., and Mills, J., 1957. "Postdecision Exposure to  Relevant Information," Journal of Abnormal Psychology, 54. 98-102.

Ehrlich, I. and Fisher, L., 1982. "The Derived Demand for Advertising; A    Theoretical and Empirical Investigation," American Economic Review, 72.   366-388.

Grossman, G. M. and Shapiro, C., 1984. "Informative Advertising with Differentiated  Products," Review of Economic Studies, 51, 43-61.

Hurwitz, M. A. and Caves, R. E., 1988. "Persuasion or Information? Promotion and   the Shares of Brand Name and Generic Pharmaceuticals," Journal of Law   and Economics, 31. 299-320.

Kadiyali, V., 1996. "Entry, Its Deterrence, and Its Accommodation:  A Study of the  U.S.  Photographic Film Industry," Rand Journal of Economics, 27. 452-478.

Kihlstrom, R. E. and Riordan, M. H., 1984. "Advertising as a Signal," Journal of   Political Economy, 92. 427-450.

Klein, B. and Murphy, K. M., 1988. "Vertical Restraints as Contract Enforcement   Mechanisms," Journal of Law and Economics, 31 265-297.

Mas-Colell, A., Whinston, M. D. and Green, J. R., 1995. Microeconomic Theory,   New York: Oxford University Press.

Milgrom, P. and Roberts, J., 1986. "Price and Advertising Signals of Product   Quality," Journal of Political Economy, 94. 796-821.

Nelson, P. 1974. "Advertising as Information," Journal of Political Economy, 82.   729-754.

Shapiro, C. 1982. "Consumer Information, Product Quality, and Seller Reputation,"   Bell Journal of Economics, 13. 20-35.

Tirole, J., 1988. The Theory of Industrial Organization, Cambridge,       MA:MIT Press.