Economics, Information and Consumer Behavior

ABSTRACT - This paper summarizes important assumptions and implications about consumer and seller behavior that are contained in the economic theory of information. The emphasis is on interesting, testable hypotheses that have escaped the attention of consumer researchers. Some attention is also given to findings from consumer research that are relevant to the economic theory of information.

Citation:

John E. Calfee and Gary T. Ford (1988) ,"Economics, Information and Consumer Behavior", in NA - Advances in Consumer Research Volume 15, eds. Micheal J. Houston, Provo, UT : Association for Consumer Research, Pages: 234-238.

Advances in Consumer Research Volume 15, 1988      Pages 234-238

ECONOMICS, INFORMATION AND CONSUMER BEHAVIOR

John E. Calfee, University of Maryland

Gary T. Ford, The American University

[This research was partially supported by a Summer Research Grant from the Kogod College of Business Administration.]

ABSTRACT -

This paper summarizes important assumptions and implications about consumer and seller behavior that are contained in the economic theory of information. The emphasis is on interesting, testable hypotheses that have escaped the attention of consumer researchers. Some attention is also given to findings from consumer research that are relevant to the economic theory of information.

INTRODUCTION

This paper seeks to demonstrate that the economic theory of information is in many respects a sophisticated theory of consumer and seller behavior, a theory that involves much more than search by consumers and advertising by sellers. We attempt to sketch the outlines of that theory, paying particular attention to those aspects of the economics of information that have been neglected by consumer researchers. Our discussion is intended to be provocative rather than definitive, and the emphasis is on testable, but mostly untested, hypotheses about consumer and seller behavior.

Consumer researchers, of course, have long been interested in the economics of information. But the focus has been primarily on consumer search, price dispersion, the relation between price and quality, and the informational content of advertising. Meanwhile, another large body of economic literature has dealt primarily with market signaling and other theoretical solutions- to the problems of cost and verification. As is demonstrated below, this literature, although not yet the focus of much consumer research, also contains assumptions and predictions about how consumers deal with the difficult problems of extracting useful information from biased sources, i.e., sellers, and how sellers attempt to provide valuable information to consumers.

This paper is divided into four sections: a brief review of those elements of the economics of information that are particularly interesting for consumer research; implications for consumer behavior; implications for seller behavior (mainly advertising); and concluding comments. Where possible we briefly note the empirical literature from consumer research relevant to the topic.

OVERVIEW OF THE ECONOMICS OF INFORMATION

The economic theory of information assumes, like most mainstream economics, that both buyers and sellers attempt to maximize utility, subject to constraints on resources, time and so on. From this perspective, information is simply another commodity that may be produced, marketed, purchased and consumed like other more tangible commodities. This view was the starting point for Stigler's classic article that launched information as a topic in economics. He argued that consumers will gather product information up to the point where the cost of doing so exceeds the value of further information, and that this behavior helps determine such market phenomena as price dispersion (Stigler, 1961). As important as the Stigler article was, it soon became apparent that the peculiar economic problems associated with producing and disseminating the commodity "information" require more than a simple marginal costCmarginal value model.

The first problem is that a large proportion of consumer information is distributed free by sellers in the form of advertising. Although this " free" information reduces buyers' search costs, it obviously is biased because the seller is using the message both to provide information and to persuade. Therefore, it is to be expected that consumers will be skeptical of advertising. Despite such skepticism, advertising must be worthwhile because sellers continue to spend billions of dollars on it. Thus, the first major extensions of Stigler's work examined issues associated with "verification" of advertising claims.

The verification issue formed the starting point for Akerlof's important paper on markets in which all claims are suspect. Akerlof demonstrates that when consumers cannot verify any advertising claims, one result can be a "lemons" market in which all products are of the lowest possible quality despite consumer willingness to pay for higher quality. In a lemons market there is no conflict between price and quality because both are at the lowest possible level. (Akerlof, 1970.) Akerlof's largely theoretical paper inspired a few attempts to assess empirically the tendency for lemons markets to occur in practice. (Bond 1982; Lynch, et al., 1985.)

The verification problem is also the motivation behind the development of attempts to categorize advertising claims as having either search, experience or credence qualities. Nelson (1974) wondered how advertising provides information to consumers when, in his words, 'The producer is not interested in providing information for consumers. He is interested in selling more of his product. Subject to a few constraints, the advertising says anything the seller of a brand wishes.' (1974, p. 729.) Nelson hypothesized that consumers will tend to be skeptical of all advertising claims, but will be least skeptical of claims about attributes that can be verified through prepurchase search activity (called "search attribute claims"), and most skeptical of claims that can be verified only after the product has been purchased and used for some period of time ("experience attribute claims"). (Nelson 1970.) He reasoned that sellers take consumer skepticism into account and therefore have an incentive to be truthful when making claims that can be verified before purchase. Consumers should therefore expect search attribute advertising to be true. Parenthetically it might be mentioned that this conclusion holds whenever the cost of evaluating the truthfulness of a claim is low, as when evaluating experience attribute claims for low-priced, frequently purchased goods.

Darby and Karni (1973) suggested that for certain technically complex products and services (gall bladder surgery, for example), most consumers can never know whether a claim is true even after purchase and consumption. They suggested a third category of advertising claims which they termed "credence attribute claims." Because such claims can never be verified by average consumers, it follows that consumers should be even more skeptical of credence-attribute claims than experience attribute claims.

The second extension from the Stigler formulation stems from the fact that information is transmitted by sellers and gathered by buyers in ways other than those normally included in models of information search. Researchers have been puzzled for many years by the consistent empirical result that consumers do not search as much as might be expected. Consumer researchers have contributed much to this literature with their efforts concentrating primarily on consumer search, price dispersion and the relation between price and quality. (See, for example, the brief review in Ratchford and Gupta 1987; and Jacoby 1984.) Generally, this research concludes that price dispersion exists for goods of similar quality; price and quality are only weakly positively correlated and in the words of Ratchford and Gupta, "consumers appear to behave as if search costs are high." (at p. 312)

We are left with two conflicting results. On the one hand, theory with a high degree of face validity argues that consumers will search for information until marginal benefit equals marginal cost, and argues that consumers will be skeptical of advertising for experience and credence attributes (especially for high-priced products). This suggests a high level of search. But as noted above, consumers do not search much.

A potential explanation for these apparently conflicting results is that consumers perceive their current level of information is sufficient, i.e., the marginal gain from search is low no matter what the cost of search. This could happen for three reasons: consumers trust advertising to be unbiased, consumers believe there are ways of extracting useful information from admittedly biased sources, or consumers passively "search" for information while talking with friends, shopping for other products and the like. In another paper in this session, Calfee and Ringold (1988) demonstrate that consumers are indeed quite skeptical of advertising and consistently have been so over many years. Thus, the other two explanations are of great interest.

The topic of how consumers may obtain useful information from biased sources has been discussed in a largely theoretical literature on market "signals," where a signal is defined as a bit of information that can improve the predictability of a second bit of information (Spence 1973). The signaling literature first analyzed labor markets, but has since expanded to include consumer markets. The central point is that markets may allow information to be transmitted credibly because consumers understand how to identify high quality sellers and understand when sellers have an incentive to keep their promises (Spence, 1973, 1974 and 1976; Ippolito 1986a, 1986b, and references therein).

Nelson also discussed signals when stating that consumers should use the mere fact that a firm advertises, the volume of advertising and the market share of the advertiser as indicants of quality (Nelson 1974). Signalling per se has not been the focus of much consumer research (Kirwani 1987). But a fairly large literature indicates that consumers routinely use inference formation strategies and cues when confronted with incomplete information. (Olson and Jacoby 1972: Ford and Smith 1987.) It would be neither surprising nor inconsistent to find that consumers also routinely use market signals.

The passive search explanation is also interesting. The argument is that while engaged in everyday activities such as talking to friends, shopping for other products, reading magazines and so OD, consumers are learning incidentally and almost unconsciously about products for which they are not yet in the market. When the time comes to buy one of these products, consumers may search memory and decide they have enough information to make a decision, or at least to narrow the choices to a reasonable number (Lynch and Srull 1982).

A third economic problem with information is the difficulty of enforcing property rights. A producer of information has great difficulty controlling dissemination of his product, (witness the continuing attempts by Consumer Reports to restrict use of their ratings by advertisers), particularly when the main practical use of the information involves only the "bottom" line (which brand is most or least reliable overall. for example.) Closely related to this problem is the fact that information often involves externalities, i.e., spillover costs and benefits for persons who do not produce or pay for information.

Particularly notable in consumer markets are positive externalities. Both buyers and sellers are affected buyers because, for example, search by some individuals tends to improve the market for all consumers (Salop 1976, Butters 1977), and sellers because, for example, advertised information that applies to more than a single brand may help sellers of competing brands or even competing products. Hence an important role is played by "free riders," that is, parties who do not share in the costs of production yet who enjoy benefits from such production. The effect is that the seller who gathers and disseminates information usually cannot recoup anywhere near the marginal value of his output. Externalities or spillovers can also be negative. An example is the fact that false claims by one seller can reduce the credibility of all advertisers.

A fourth remarkable economic problem with information not fully considered by Stigler is the extraordinary economies of scale associated with its production and dissemination. There are often huge fixed costs associated with the creation of information and very small marginal costs of distribution. Marginal cost is therefore often far below average cost. This fact, coupled with the property rights problem discussed above, prevents fully efficient pricing and carries the implication that sellers have an incentive to produce less than the optimal amount of information. A by-product is that buyers often must use signals because they are all that is available, even when buyers would prefer specific information.

SKETCHING THE BEHAVIORAL ASPECTS OF THE ECONOMICS OF INFORMATION

It was stated at the outset that the economic theory of information is a theory of consumer and seller behavior. In the paragraphs above we have attempted to point out some of the more unique aspects of information that affect the way information markets work. We have also attempted to describe why these peculiarities of information influence consumer and seller behavior. The sum of these effects can be described as follows.

Consumers think rationally about other actors in the market, especially sellers. They look beyond advertising claims to assess the incentives of advertisers and expect advertisers to understand that consumers do this! Consumers will, for example, take into account the fact that advertising is expensive and will pay off only for sellers that ultimately satisfy buyers, and hence, the volume of advertising for a brand is a reliable clue to product quality. (Nelson, 1974.) Consumers do more than simply distrust advertisers. They differentially distrust advertisers, being least distrustful when the advertiser has an incentive to speak truthfully because the market will inflict penalties on untruthful behavior. And this rational behavior goes even further. Consumers realize that sellers know how consumers think, and buyers and sellers both take this into account. It is this iterative process -- sellers expect buyers to expect that sellers will behave in a certain way -- that allows market signals to work. This seemingly delicate process can theoretically lead to plausible market equilibriums that balance skepticism and sophistication. Some empirical evidence seems to fit, an example being research on selective consumer discounting of price advertising. (Blair and Landon, 1981; Liefeld and Heslop, 1985.)

Consumers also understand how information economics works. They realize, or so the theory goes, that sellers, in an effort to avoid advertising information beneficial to competitors, will seek to attach information to their own brand even when the information would apply equally to others. Thus consumers will not necessarily conclude that competing products lack the advertised attribute. Consumers also expect that sellers will not provide all information about their products, especially deleterious information, and that as a result much product information will have to come from other sources. On the other hand, the lack of certain favorable information in advertising may be a signal that the brand does not have the positive attribute, and the brand may be downrated accordingly. (Finn, 1981.)

It is not clear that the consumer expectations described above actually exist for most consumers, or even that the net result of the behavior of all consumers is consistent with this theory. What is clear is that a number of testable hypotheses about individual consumer behavior can be derived from the economics literature. That is the subject of the next section.

SOME IMPLICATIONS FOR CONSUMER BEHAVIOR

Several of the hypotheses and research questions about consumer behavior presented in this section have been alluded to in the literature, while others have escaped attention. We attempt to demonstrate, however, that all follow quite naturally from the theory that has been presented. The first set of research questions and hypotheses are concerned with consumers' skepticism of sellers' claims and the verification issue. The assumption that consumers are skeptical of advertising is by itself uninteresting. More interesting is the extent to which whether certain factors modulate the level of skepticism. For example, has consumer skepticism changed over time in response to, or as a prelude to, increases or decreases in advertising regulation?

Any of three hypotheses about the role of regulation appear plausible. The first is that regulatory bodies react to public concern with deceptive advertising and step up their actions accordingly. Under this explanation, changes in public opinion about the truthfulness of advertising precede and cause changes in regulatory activity. The second hypothesis is that the public perception about the need for regulation is a reaction to increased awareness that regulatory agencies are taking action. With this explanation, the public would notice that the FTC, for example, is increasing its regulatory activities and would conclude there must be an increasing amount of deceptive advertising. Under this scenario, changes in the public's skepticism with advertising would lag changes in regulatory activity. A third hypothesis is that public opinion about the truthfulness of advertising is independent of regulatory activity. Under this explanation, the primary concern is whether skepticism is increasing, decreasing or remaining the same over time. Obviously, advertisers are vitally concerned with this issue because if the perceived truthfulness of advertising is decreasing, it will take more effort and be more costly for reputable sellers to market their goods. The paper by Calfee and Ringold in this session strongly suggests that advertising regulation has had little if any effect on consumer skepticism.

A second important set of hypotheses concern whether the degree of consumers' skepticism with advertising claims varies inversely with their ability to verify claim veracity. The economics of information literature assumes that consumers routinely examine claims and automatically array them on some type of internal claim verifiability scale. The work of Nelson (1974) and Darby and Karni (1973) leads to the conclusion that search claims should be discounted the least and credence claims the most, but no empirical evidence supporting this perspective has yet been published. As noted briefly in the prior section, the economics literature also assumes that consumers consider the price of goods when judging the veracity of a claim. The argument is again based on the market's ability to discipline untruthful sellers. For low-priced goods consumers can try the product and (in the case of search and experience attribute claims) determine whether it performs as advertised. In this situation, trial substitutes for search. Of course, this strategy cannot be used in a cost effective manner for high-priced goods or for credence attribute claims for low-priced products. Nonetheless, it is predicted that consumers will perceive that search and experience claims for low-cost goods are truthful.

No explicit hypotheses have been developed in economics about how price and type of attribute (search, experience and credence) interact. Based on the literature, however, some predictions can be developed. For example, given the six possible combinations of low/high price and SEC attributes, it might be expected that consumers are least skeptical of all about search attribute claims for low-priced goods, that high-priced search claims are perceived as second most truthful and so on ending with the prediction that consumers are most skeptical about credence attribute claims for high-priced goods. This 2 X 3 framework can be extended to include additional dimensions such as whether the claim is objective (i.e., factual) or subjective and to include covariates such as involvement or consumer expertise.

Signalling theory also leads to interesting predictions. Most obvious is the hypothesis drawn directly from the economics of information that consumers use signals appropriately. In some ways it may be premature to discuss this hypothesis since very little conceptual work has taken place identifying the types of signals that consumers prefer to use. That is, although economists have offered some lists of signals, e.g., volume of advertising, manufacturer market share, and warranty length, no well tested theory of signals exists. Nonetheless, the idea that signals are used is so intuitively appealing that, without in any way diminishing the importance of that issue, other points can be discussed.

Most intriguing to us is the issue of how signals interact with claims. One of the clearest, and we expect, most reliable signals attached to many products is the brand name. The name Sony or IBM provides such important information to consumers that the perceived truthfulness of credence and experience claims is, we expect, enhanced tremendously. Does such enhanced credibility transfer to lesser known- sellers? That is, when IBM makes a claim about what is essentially an experience or credence attribute such as productivity gains from using personal computers, are similar claims by the sellers of "clones" enhanced? Are there really, as the literature suggests, free rider effects in credence claims?

Finally, an extremely important research question involves whether consumers feel comfortable making decisions based on their current levels of product information. Suppose consumers believe that the combination of explicit information from advertising, reliable signals, and information stored in memory from other sources is sufficient to choose among products. Then policy efforts aimed at reducing search costs may show marginal effects at best. As noted above in another context, consumers draw inferences from available information. In addition, use of signals may be a simplifying strategy that reduces the cost of thinking (Shugan 1980), or partially accounts for the finding that consumers "tend to examine only small proportions of the brand- and attribute information that is available." (Jacoby 1984.) Adult consumers may believe they are so experienced at extracting useful information from the market that the value of search is bound to small. If so, this suggests the hypothesis that, ceteris paribus, more mature consumers will exhibit greater confidence in their abilities to interpret market signals than will less mature consumers. As an aside, it might also suggest that this type of research not be done with student subjects; they may lack sufficient experience as consumers.

SOME IMPLICATIONS FOR SELLER BEHAVIOR

Examining seller behavior from the standpoint of information economics raises at least two sets of implications. The first focusses on how sellers can make information credible so as to avoid "lemons" markets. Sellers have a great incentive to solve the lemons problem because sellers of average or superior products lose as much from informationally barren markets as do consumers.

A number of hypotheses arise directly from the classification of claims into search, experience and credence attribute claims. We would expect sellers of high price products to emphasize search attributes and sellers of low price goods to emphasize search and experience attributes. No seller would be expected to emphasize pure credence attributes except insofar as such claims can be made credible by appealing to unimpeachable sources (an example being the recent Kellogg All-Bran advertising campaign based on National Cancer Institute conclusions about the connections between cancer and dietary fiber.) Sellers will also make use of third party evaluations for experience attributes, especially for high price products. Finally, reputation will be emphasized by sellers of products rich in credence and experience attributes.

A second group of research questions is concerned with signals. We noted earlier that sellers can use market signals to communicate credibly with buyers. Signals may be classified as "pure signals" and "bonds." A pure signal is simply a piece of information that implies another piece of information about a market participant. The classic example comes from Spence's analysis of labor markets, which showed that under plausible assumptions more capable individuals will invest in education as a signal to potential employers even when the education does nothing to improve job ability. (Spence 1973.) The archetype example of a pure signal in consumer markets is Nelson's model in which the sheer volume of advertising for a brand serves as a signal for consumer satisfaction with the brand (because of the role of repeat purchasing.) (Nelson 1970.)

Bonding refers to situations in which a seller has undertaken a visible investment that will be lost if he does not perform as-promised in advertising claims. Thus bonding, unlike pure signals, involves incentives. Ippolito, 1986a and 1986b.) Sellers could even find it advantageous to advertise their own incentives (as Frank Perdue does when he bemoans his fate if one of his chickens turns out not to be fresh). These points raise fascinating empirical questions, of which we mention three. One is the extent to which markets actually produce effective bonding mechanisms. Another question has to do with costs. Because bonding and signalling both impose costs -- costs that are deadweight losses in comparison to a market in which all information is credible -- a natural question is the magnitude of such costs. The third question, of course, is the magnitude of consumer benefits that derive from the use of bonds and other signals. Examination of these questions would do much to move consumer information economics beyond such first-order considerations as price dispersion.

A third major issue, quite different from the lemons problem, is how sellers will deal with attenuated property rights, positive externalities, and related difficulties in the economics of information. We can expect that sellers will seek to internalize the benefits of providing information by advertising, say, "Ajax chicken has less fat than hamburger" rather than simply "chicken has less fat than hamburger." Sellers will wish to "educate" consumers on such matters as the availability of unbiased third party information. We would expect such efforts to be strongly affected by the market share of the advertiser, because the education effort will tend to benefit other sellers of products that are evaluated more easily because of the new consumer consciousness. A striking recent example is a series of ads by General Motors that essentially introduces consumers to the Insurance Institute for Highway Safety and the safety statistics produced by that organization.

We might also expect advertising to "free-ride" on information from other sources. This can done by using "inferred" claims, i.e., claims that are created in consumers' minds by a combination of information in ads and "ambient information" that consumers acquire from other sources. This has certainly happened in the recent surge of health claims for foods rich in fiber and calcium or containing little cholesterol and saturated fat. Thus "high in fiber" may convey to some consumers the claim, "helps prevent colon cancer." This process poses many problems for advertising regulation and in fact for years the FDA banned the word "cholesterol" from ads and labels in order to prevent inferred health claims. (Hutt, 1986.) But the inferred claim process can increase the efficiency with which both product and health information are disseminated to consumers. Little is known, however, of the practical importance of these developments.

CONCLUSIONS AND NEXT STEPS

There is little doubt that the economic theory of information contains many sophisticated assumptions about how consumers search for, evaluate and use information. For the most part the crucial assumptions of this theory have not come to the attention of consumer researchers (Wright's (1986) "schemer schema" being a notable exception), while the empirical findings in consumer research, marketing and psychology have been ignored by information economists. In this paper we have endeavored to identify some of the more intriguing characteristics of information that influence information markets and to discuss the testable assumptions about consumer behavior contained in the theory.

As we see it the logical next steps to advance the theory of the economics of information are (1) to compare what is known about consumer behavior with what is assumed in the theory about how consumers obtain and use information and (2) to begin a program of research aimed at examining some of the as yet as untested assumptions and implications. Clearly, the methodological sophistication that consumer researchers bring to problems of this type can make a substantive contribution to assessing the economic theory of information.

REFERENCES

Akerlof, G., (1970), "The Market for 'Lemons': Quality Uncertainty and the Market Mechanism," 84 Quarterly Journal of Economics, 84, 488-500.

Beales, Howard, Richard Craswell, and Stephen Salop (1981) "The Efficient Regulation of Consumer Information," Journal of Law and Economics, Vol. 24, pp. 491-539 (December.)

Blair, Edward A., and E. Laird Landon, Jr. (1981) "The Effects of Reference Prices in Retail Advertisements." Journal of Marketing, 45, 61-69.

Bond, E., (1982) "A Direct Test of the 'Lemons' Model: The Market for Used Pickup Trucks," American Economic Review, 72, 836.

Butters, Gerard R. (1977) "Equilibrium Distribution of Sales and Advertising Prices," Review of Economic Studies, 44, 465-491 (October.)

Darby, M., and Karni, E. (1973) "Free Competition and the Optimal Amount of Fraud," 16 J. of Law and Economics 67-88 (April.)

Federal Trade Commission (1979) Consumer Information Remedies, staff report.

Finn, Adam (1981) "Inferential Belief Formation Through the Use of Non-Information: An Example," Advances in Consumer Research, 8, 344-48.

Ford, Gary T. and Ruth Ann Smith (1987), "Inferential Beliefs in Consumer Evaluations: An assessment of Alternative Processing Strategies," Journal of Consumer Research, (December), in press.

Hutt, Peter (1986) "Government Regulation of Health Claims in Food Labeling auld Advertising," Food Drug Cosmetic Law Journal, 41, 3-75.

Ippolito, P. (1986a) "Consumer Protection Economics: A Selective Survey," in P. Ippolito and D. Scheffman, ed's, Empirical Approaches to Consumer Protection Economics, FTC Staff Report

Ippolito, P. (1986b) "Advertising and Product Quality: The Role of the Bonding Characteristics of Advertising," Working Paper No. 148, Federal Trade -Commission, Dec. 1986.

Jacoby, Jacob (1984), "Perspectives on Information Overload," Journal of Consumer Research, 10, (March), 432-35.

Kirwani, Amna (1987), "Advertising as a Signal of Product Quality," dissertation proposal, Stanford University.

Liefeld, J., and Heslop, L. (1985) "Reference Prices and Deception in Newspaper Advertising," Journal of Consumer Research, 11, 868-876.

Lynch, John and Thomas Srull (1982), "Memory and Attention Factors in Consumer Research: Concepts - and Research Methods," Journal of Consumer Research, 9 (June), 18-37.

Lynch, M., Miller, R., Plott, C., and Porter, R. (1985) "'Lemons' in Experimental Markets," in P. Ippolito and D. Scheffman, ed's, Empirical Approaches to Consumer Protection Economics, FTC Staff Report.

Nelson, P. (1970) "Information and Consumer Behavior," 78 Journal of Political Economy, 311 (March/ April.)

Nelson, P. (1974) "Advertising as Information," 82 Journal of Political Economy 729 (July/August.)

Olson, Jerry and Jacob Jacoby (1972) "Cue Utilization in the Quality Perception Process," Proceedings of the Third Annual Conference of the Association For Consumer Research, 167-79.

Ratchford, Brian T., and Pola B. Gupta (1987) "On Measuring the Informational Efficiency of Consumer Markets," in P. Anderson and M. Wallendorf, ed's, Advances in Consumer Research 14, 309-313.

Salop, S. (1976) "Information and Monopolistic Competition," 66 American Economic Review 240 (papers and proceedings.)

Salop, S. (1979) "Parables of Information Transmission in Markets," in A. W. Mitchell, ed., The Effects of Information on Consumer and Market Behavior, American Marketing Association, Chicago.

Schwartz, Alan, and Louis Wilde (1979) "Intervening in Markets on the Basis of Imperfect Information: A Legal and Economic Analysis," 127 University of Pennsylvania Law Review, 630-682.

Shugan, Steven, M. (1980), "The Cost of Thinking," Journal of Consumer Research, 7 (No. 2) 99-111.

Spence, M. (1973) "Job Market Signalling," 87 Quarterly Journal of Economics 355-374 (Aug.)

Spence, M. (1974) "Competitive and Optimal Responses to Signals: An Analysis of Efficiency and Distribution," 7 Journal of Economic Theory 296-332.

Spence, A. Michael (1976) "Informational Aspects of Market Structures: AD Introduction," 90 Quarterly Journal of Economics 591-597.

Stigler, George (1961) "The Economics of Information," 69 Journal of Political Economy 213-225.

Wright, Peter (1986), "Schemer Schema: Consumers' Intuitive Theories About Marketers' Influence Tactics," Advances in Consumer Research, 13, 1-3.

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