Empirical Generalizations about the Impact of Advertising on Price Sensitivity and Price Author(s): Anil Kaul and Dick

R. Wittink Source: Marketing Science, Vol. 14, No. 3, Part 2 of 2: Special Issue on Empirical Generalizations in Marketing, (1995), pp. G151-G160 Published by: INFORMS Stable URL: http://www.jstor.org/stable/184157 Accessed: 15/08/2008 12:57
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=informs. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org.

http://www.jstor.org

MARKETING SCIENCE Vol. 14, No. 3, Part 2 of 2, 1995 Printed in U.S.A.

EMPIRICAL GENERALIZATIONS ABOUT THE IMPACT OF ADVERTISING ON PRICE SENSITIVITY AND PRICE
ANIL KAUL AND DICK R. WITTINK Cornell University
Consumers'sensitivitiesto pricechangesare an importantinput to strategicand tacticaldecisions. It has been argued that price sensitivities depend on factors such as advertising. Prior studies on the effect of advertising on consumer price sensitivity have found seemingly conflicting results. We analyze the characteristics previous studies in marketingand generatea set of three empirical of generalizations. These are ( 1 ) an increase in price advertising leads to higher price sensitivity among consumers, (2) the use of price advertising leads to lower prices, and (3) an increase in nonprice advertisingleads to lower price sensitivity among consumers. These generalizationshave important implications for managersand researchers. Managersneed to coordinatetheir advertising and pricing decisions to attain maximum profits. For researchers,our summary and discussion of empirical results provide directions for future. (Advertising; Pricing Research; Empirical Generalization)

Introduction Advertising, apart from its numerous direct effects, also has an effect on the price sensitivity of consumers and on the prices of goods in a market.1Managersand researchers are frequently interested in assessing consumers' sensitivities to price as an input to strategic and tactical decisions about market segmentation, price, marketing activities, as well as competitive marketing strategies. Many studies of the effect of advertising on the price sensitivity of consumers have been conducted. At firstglance the results of these studies appear to be conflicting. For example, some studies suggest that an increase in advertising leads to an increase in the estimated price sensitivity of consumers, while other studies have found that an increase in advertising leads to a decrease in the price sensitivity of consumers. One could blame the economists who developed two theories that predict opposite effects of advertising on prices and consumers' price sensitivity. The first view, expressed by Comanor and Wilson ( 1979), states that advertising for brands leads to (artificial) product differentiation, thus creating brand loyalty and lowering
' Price sensitivity of consumers is usually measured either by the slope of the demand curve or by the price elasticity of demand. Of these, price elasticity is the preferred measure since it is not sensitive to the units of measurement. G151 0732-2399/95/1403/G 151$01.25
Copyright ? 1995, Institute for Operations Research and the Management Sciences

G152

ANIL KAUL AND DICK R. WITTINK

sensitivity to price as a factor in brand choices. The resultant market power for the firms could result in higher prices. The second view, put forth by Stigler ( 1961 ) and Nelson ( 1970, 1974), states that advertisingby providing information about brands can increase the consideration set and lead to higher price sensitivity among consumers. Higher price sensitivity should result in lower prices charged by the firms. The controversy regardingthe possible relationship between advertising and price sensitivity has spawned many studies in marketing.Researchershave studied this relationship in different ways, employing methods varying from laboratoryand field experimentation to econometric modeling of historical data. In a few studies the relationship is examined at the consumer level. Many studies, however, are based on relationships estimated from aggregatedata (e.g., at the product market level). These studies span a large set of product situations, e.g., new products, mature products, consumer nondurables and durables, services, and even strategic business units. An overview of 18 studies in which the relationship between advertising and price sensitivity is examined in one form or another is provided in Table 1. In this table, we identify the type of study, the type of product, the number of brands, the type of advertising, the measure of advertising, the measure of price sensitivity, and the type of interaction (result) observed. The last column of Table
TABLE I
Measure of Price Sensitivity# SC SC
EL

Paper Woodside & Waddle (1975) Prasad and Ring (1976)

Lambin (1976)

Typc of Study* FE FE
EM

Type of Product** CND CND

Multiple

No. of Brands One One

Multiple

Type of Data*** B B
B

Type of adv.A P N
N

Measure of
adv.AA

Result## IN DE
DE

H/L H/L $ H/L H/L $

H/L

Staelin & Winer

(1 976)

FE FE EM
FE EM

CND CND CND

CNI)

One Multiple One

MuLltiple

B B B
B

N N N
P

EL SC EL
SC

DE IN IN
IN

Eskin & Barron (1977) Wittink (1977)

Morianty (1983)

et CGhosh al. (1983) Gatignon (1984) Krishnamurthi& Raj (1985) Shroeter et al.
(1987)

EM FE EM EM EM

CN1D Airlines CNND PS CND CND

Mlultiple Multiple One Multiple Multiple Multiple

B Consumer level B Product category

B B

N N N P N N/P

Exp. $ H/L P $ $/P

EL EL EL EL EL EL

DE IN DE IN DE DE/IN

Vanhonacker
(1989)

Popk-owskiLeszczyck & Rao

(1939)

Bolton (1989) Bemmaor & Mouchoux (I1991) Kanetkar (1992) Boulding et al. (1994) Mitra & Lynch (1993)

EM FE EM EM EX

CND CND CND Multiple CND

Multiple Multiple Multiple Multiple Multiple

Product category B Consumer level SBU Level Consumer level

P P P N N

P H/L Exp. $ H/L

EL EL EL EL EL

IN IN IN DE IN/DE

* FE = Field Experiment: EN-1 Empirical Model; EX = Consumer Level Experiment. ** CND = Consumer Nondurable; PS = Professional Service.

Brand Level Data. Nonprice/National Advertising; P = Price/Local Advertising. " H/L = High/Low; $ = Dollar Expenditure; Exp. -- No, of exposures; P = Proportion of Retailers/Firms Advertising. 'SC = Slope Coefficient; EL = Price Elasticity. Price Sensitivity Decreases with Higher Advertising. *' D= IN = Price Sensitivity Increaseswith Higher Advertising.
AN
=

*** B

THE IMPACT OF ADVERTISING ON PRICE SENSITIVITY AND PRICE

G153

1 indicates that in nine studies price sensitivity increases (IN) with higher advertising, in seven studies it decreases (DE) with higher advertising, and in two cases both effects are obtained. In this paper we discuss the effect of advertising on consumer price sensitivity and on price, and we report empirical generalizations on the basis of the results obtained in various marketing studies. By empirical generalization, we mean results that hold when replication occurs under very different conditions. We consider only those cases where at least three studies have provided the same result. Bass ( 1]994) suggests that empirical generalizations should have some measure of enhanced quantification as opposed to mere associations. Our focus is on interactions which are a higher-order phenomenon than association. The approach that we adopt in this paper is to analyze the characteristics and the results of previous studies. We provide explanations for and discuss the relevance of these generalizations. We also provide a critical perspective on the methods used, and the factors that should be kept in mind when the generalizations are interpreted. We conclude with a summary and suggestions for future research. Background One way to reconcile the apparently conflicting results is by distinguishing different types of advertisingbased upon its content (Popkowski-Leszczyckand Rao 1989, Boulding et al. 1994). The first kind of advertising, sometimes referredto as "mood" advertising, is primarily geared toward brand positioning and the communication of unique brand characteristics. We refer to this kind of advertising as nonprice advertising. Much of the manufacturers' national advertising is of this type (Resnik and Sterni1977). The second type of advertising is the kind that informs consumers about the price and availability of a brand. Since most of this advertising contains price information, we refer to this advertising as price advertising. The majority of advertising done by retailers (local advertising) is of this type. Thus, the main point of difference between nonprice and price advertising is in the content of advertising, i.e., whether price information is being communicated or not. To a large extent, local advertising is price advertising, and national advertising is nonprice advertising. In recent years local advertising has become as important as national advertising for many product categories. According to Advertising Age (Dec. 14, 1992) $76.5 billion was spent on national advertising while $55.6 billion was spent on local advertising in the United States in 1992. lience. both kinds of advertising form an important part of any brand's overall advertisinigstrategy. A change in consumer price sensitivity is usually measured either by the interaction between price and advertising in a sales response function (e.g., does the slope for price change with advertising) or by a change in the estirnated price elasticity of demand. Researchers employing experiments tend to focus on the interaction between advertising and price, while econometric researcherstend to use price elasticity as the preferredway to measure the impact of advertising on price sensitivity. The studies that we base our generalizations upon can be grouped into two distinct types-experimental studies and association studies. Experimental studies occur either in a laboratoryor in a field setting. The direction of causality is unambiguous in these studies. The strengthof the (laboratory) experimental studies is the causal direction in carefully controlled environments (internal validity), while the strength of the association studies is the generalization of results across a wider context of product categories and environments (external validity). It should be clear that field experimental studies offer the best opportunity for both internal and external validity simultaneously. Empirical Generalizations We report three empirical generalizations about advertising and its effect on consumer price sensitivity and otl price:

G154

ANIL KAUL AND DICK R. WITTINK

(1) An increase in price advertising leads to higherprice sensitivity among consumers. Marketingstudies have consistently found that when consumers are exposed to advertising which contains price information or to local advertising, their price sensitivity increases. These results have been shown to hold whether the study used experimental methods or empirical modeling of household or aggregatedata. (a) Experimental studies. Bemmaor and Mouchoux ( 1991 ) conducted a factorial experiment to test the interaction between price advertising and price. The advertising message explicitly mentioned price in it. Using price cuts of 5 percent and 15 percent across five stores, they found that price elasticities increased by 20 percent to 180 percent for various products when advertisingwas increased.The productsused in the experiment were regularground coffee, liquid cleaner, disposable diapers, cat litter, hair lacquer, and sparkling wine. Moriarity (1983), in an experimental study for a frequently bought packaged good, found a substantial and significant negative interaction between feature advertising and retail price. A similar resultis reportedby Woodside and Waddle ( 1975) for the interaction between in-store promotion and price for instant coffee. (b) Association studies. Bolton ( 1989) estimated a multivariate model relating differences in estimated price elasticity to market characteristicssuch as brand market share, advertising level, etc. The price elasticities were estimated across markets rather than within a single product class or brand. She found that higher price (retail) advertising leads to higher price elasticity. The data pertained to four product categories-frozen waffles, liquid bleach, bathroom tissue, and ketchup. Popkowski-Leszczyck and Rao (1989) have obtained similar results using aggregate data. They investigatedthe effect of price (local) advertisingon consumers' price elasticity for a consumer nondurable product. This advertising was measured by the proportion of retailers (weighted by sales volume) that cooperate in local advertising during the study period. Finally, Schroeter et al. ( 1987) studied this relationship in the market for legal services. They also found that price sensitivity goes up as advertisingincreases. Their measure of advertising is the proportion of attorneys who advertise in a specific market. Thus, six studies using different methods, from experimental analysis to econometric models, have obtained a result consistent with the idea that price advertising leads to higher price sensitivity. These studies span not only different product categories and different levels of aggregation, but also different measures of advertising intensity such as advertising spending and the proportion of firms advertising in a market. The observed higher price sensitivity among consumers could be due to two separate phenomena. One is that the price sensitivity of existing consumers might be increasing. For example, as consumers are exposed to more information, the number of brands that consumers consider increases, existing brands are compared more actively, and price as an attribute increases in importance during the decision-making process. Another is that price advertising attracts new consumers who are more price sensitive, thus leading to an overall increase in the average price sensitivity of consumers of a brand. (2) The use of price advertising leads to lower prices. Research studies primarily conducted by applied economists have examined the impact of local advertising on prices in a market. These studies consider those products (services) that are not advertised in certain areas due to legal restrictions. By comparing the prices in the areas where advertising is allowed to prices in the areas where advertising is not allowed, while controlling for other factors that affect prices, inferences are made about the effect of advertising. The finding is that local advertising leads to lower prices, relative to costs. The products in these studies are legal services,prescriptionmedicines, eye glasses,and eye examinations. Benham ( 1972) studied the impact of the presence of advertising for eye glasses and eye examinations and concluded that the presence of advertising is associated with lower

THE IMPACT OF ADVERTISING ON PRICE SENSITIVITY AND PRICE

G155

prices. Kwoka ( 1984) and Haas-Wilson ( 1986) while controlling for quality effects studied the eye glasses market and reached the same conclusion that the presence of advertising is associated with lower prices for consumers. Cady ( 1976) completed a similar study for prescription drugs and found a result consistent with the idea that local advertising leads to lower prices. One might argue that the studies which support the notion of price advertising leading to lower prices are merely association studies and thus do not imply causality. Although this is true, we want to point out that in these studies two kinds of marketsare comparedone where advertising is allowed and another where advertising is not allowed. Since the decision to allow advertising in a particulararea rests with the legislature, it appears that these results are stronger than those from other association studies. For example, if an experiment were to be conducted to test the hypothesis in question, advertising would be allowed in some states and not in other states. A comparison of prevailing prices across the two groups of states could lead to an unambiguous causal inference. In an experimental study, states would be randomly assigned to the two groups in terms of whether advertising occurs or not. If the outcomes of the legislative processes (advertising allowed or not) are not related to prevailing market characteristics(especially price sensitivity), the observed results from these quasi-experimental studies provide strong evidence. There are several reasons that can explain the result that advertising leads to a lower price. An obvious one is that local advertising leads to higher price sensitivity among existing consumers which will result in a lower optimal price level for the firms (Dorfman and Steiner 1954). For example, the advertising may have alerted consumers to outlets they would not have considered in the absence of advertising. The choice of one outlet from the larger set considered could depend especially on the prices. Also, if advertising attracts additional consumers who are more price sensitive than existing customers, the optimal price level decreases. In addition, the price may decrease because of increased competition (due to advertising)among existing firms. Retailersmay reduce pricesbecause they are cognizant of advertised prices and believe consumers will be influenced by price differences. For the applications mentioned here (legal services, eyeglasses, prescription medicines) it is unlikely that manufacturers would experience lower marginal costs of production in the presence of local advertising. However, retailers could demand manufacturer discounts with higher sales. It is interesting to note that the products examined in these studies are not ones where price is expected to be a criticalattributefor consumers. However, Urbany et al. ( 1990) document instances in which retailers'beliefs about consumers' behavior diverge from consumers' actual behavior. Thus, it may be sufficient for managers' beliefs to change with advertising. In any event, the result is clear that the presence of advertising leads to lower prices. (3) An increase in nonpriceadvertisingleads to lowerprice sensitivityamong consumers. Most studies that have included an examination of the impact of nonprice advertising on consumer price sensitivity have found that national advertising decreases price sensitivity (five studies that obtained different results are discussed later). (a) Experimental studies. Staelin and Winer ( 1976) examined the results of a heavyup nonprice advertising experiment for a frequently purchased grocery product using split cable TV. They found a significant decrease in the absolute value of the firm's price elasticity in the consumer half that was subjected to heavier amounts of advertising. Prasad and Ring ( 1976) conducted a similar field experiment where the heavy half was exposed to largeramounts of advertisingcompared to the control half. A direct comparison of the price by television advertising interaction term across the two halves of the sample showed that the interaction estimate is smaller for the heavy half, implying that the consumers exposed to heavy advertising were less price sensitive. However, Prasad and

G 156

ANIL KAUL AND DICK R. WITTINK

Ring do not show how the main price effect varies between the two groups that differ in the amount of advertising exposure. Krishnamurthi and Raj ( 1985) analyzed the results of another split cable nonprice advertising experiment for a frequently bought consumer good. They also found that increased advertising leads to a significant decrease in responsiveness to brand prices. A common factor in these studies is that the set of consumers studied in the experiments remains constant. Thus, any changes in consumer price sensitivity are for existing consumers only. (b) Association studies. Lambin ( 1976) examined the price elasticity for 22 heavily advertisedbrands in western Europe and found that the absolute values of price elasticities were (weakly) inversely related to a measure of advertising intensity. The advertising intensity measure captured manufacturer advertising that tends to exclude price information. Ghosh et al. ( 1983) examined the sales of breakfastcereals for a diary panel and found that the sales of heavily advertised brands tended to be less price elastic than the sales of less advertised brands. Advertising expenditures were collected from Broadcasting and Advertising (BAR) reports which represent national advertising. Using aggregate-leveldata, Popkowski-Leszczyc and Rao (1989) and Boulding et al. (1994) found that higher nonprice (national) advertising is associated with lower price sensitivity. Popkowski-Leszczyc and Rao ( 1989) used bimonthly data on a consumer packaged good. Boulding et al. ( 1994), using PIMS data, studied this relationship at the strategic business unit level. As no information about the contents of advertising for various firms was available, they assumed that firms having higher than average price advertise only nonprice information while firms with lower than average price would advertise price information. To allow for causal direction, they used the previous period's advertising to affect this period's estimate of price elasticity. A similar result was also obtained by Vanhonacker (1989), though he finds some nonlinearity in the estimated relationship. By estimating an econometric model he finds that as the advertising share of a brand increases, its price sensitivity decreases. Only a very high level of advertising might increase the price sensitivity. The main argument for nonprice advertising to reduce price sensitivity rests on the fact that nonprice advertising is used for positioning purposes thus making the brand more differentiated which, if successful, may result in a lower price sensitivity for the brand. This mechanism may also explain why consumer-level laboratory experiments show that price elasticities decrease when consumers receive information such as brand names and quality ratings, as has been shown by Sawyer et al. ( 1979) in an experimental study. Another interpretation is that when nonprice information is provided, consumers simply pay less attention to prices. Based on these nine studies we conclude that the results are consistent with the idea that nonprice advertising reduces the price sensitivity of consumers. This result has been obtained in studies that have used different types of data, different methods such as experimentation and econometric modeling, and different products. Though most studies support the notion that non-price advertising reduces price sensitivity, some studies (Eskin and Barron 1977, Wittink 1977, Gatignon 1983, Kanetkar et al. 1992, and Mitra and Lynch 1993) suggest that nonprice advertising can lead to an increase in price sensitivity of consumers under certain circumstances. Since there are too few studies that study specific moderating factors, no generalizations about these factors can be made. For completeness we do briefly discuss these other studies. Mitra and Lynch ( 1993 ), in an experimental study, showed that non-price advertising can increase price elasticity by increasing the number of brands considered, particularly in product markets in which consumers have to rely on memory to generate alternatives. The effect of nonprice advertising will thus be a net result of the differentiatingaspect of advertising which decreases price sensitivity and the reminder effect on consideration

THE IMPACT OF ADVERTISING ON PRICE SENSITIVITY AND PRICE

G157

sets that increases price sensitivity. Therefore, one way to interpret the results of studies showing a decrease in price sensitivity is that the reminder effect of nonprice advertising is not very strong. Alternatively, nonprice advertising in those studies may have had only a modestly positive effect on the size of the consideration set of brands. Only Wittink (1977), Eskin and Barron (1977), Gatignon (1984), and Kanetkaret al. (1992) obtained results consistent with perhaps a strong reminder effect. Since Eskin and Barron (1977) studied new products (and observed the same interaction effect in three of the four newproduct field experiments), it is quite plausible that due to an increase in the consideration sets of consumers, they found that nonprice advertisingresultedin higher price sensitivity, in their case measured by the slope coefficient for price. Wittink (1977) studied a brand in a mature consumer product category,using aggregate data. He found that the estimated price elasticity for the brand was higher in territories with higher advertising, both variablesbeing measured relative to competitors. Of course, as in many other econometric analyses, model misspecification and aggregation are potential threats to the validity of his result. In particular, Wittink et al. (1993) show that parameter estimates for a nonlinear model applied to linearly aggregated data can be severely biased. The magnitude of the bias depends on the nature of heterogeneity in marketing activities within a market. Thus, it is possible that the estimated interaction in his study is due to such differences in marketing activities across the territories.Gatignon (1984) also finds advertising to increase price sensitivity. In addition, he shows that competitive reactivity can moderate the effect of national advertising on price sensitivity. In investigating the airline market, he finds that higher competitive reactivity results in a higher price sensitivity for consumers. Advertising spending by one firm can result in a competitive reaction from its competitors. The resultant increase in messages from the firms might redirect consumers' attention to a variable on which the advertisers differ-price. High industry advertising levels can also result in increased brand comparisons, which similarly may result in increased price sensitivity of consumers. Kanetkar et al. (1992) estimated a brand choice model at the consumer level using scanner panel data and found that higher advertising leads to higher price sensitivity. Though their study does not clearly state the kind of advertising that consumers were exposed to, given the Behavior Scan data source (Eskin 1985), it seems reasonable to believe that consumers were exposed to nonprice advertising. Advertising is measured by the number of ads that a consumer is exposed to during a time period. The products used are various brands of dog food and aluminum foil. A common thread in these studies of the effect of nonprice advertising on price sensitivity is that the mix of consumers can change with a change in advertising. Thus, a change in the measured price sensitivity can be due to a change in this mix of consumers as well as to a change in individual consumers' price sensitivities. Especially for a new product (Eskin and Barron 1977) higher advertising may attract additional consumers who are more price sensitive. Although Eskin and Barron report the change in the slope coefficient for price, it is easy to show that the estimated price elasticity shows the same directional effect. Finally, we want to mention some related work. Steiner (1973) presents arguments and empirical evidence suggesting that manufacturer advertising influences both the manufacturer margin and the retailer margin. Conceptually, the argument is that manufacturer advertising increases consumer demand for the advertised brand. The selling function of retailers is facilitated by this advertising, and the advertised brand may be used to attract store traffic. The competition between retailers on the advertised brand increases, which results in lower retail margins for that brand. By the same token, the manufacturer advertising increases the power of the manufacturer relative to the retailer, allowing the manufacturer to increase its margin. Therefore, manufacturer advertising may increase manufacturer prices and simultaneously decrease retailer margins (see also

G 158

ANIL KAUL AND DICK R. WITTINK

Farris and Albion 1980). The result may be a reduction in consumers' price sensitivity (reduced competition) across brands in the presence of nonprice advertising and an increase in consumers' price sensitivity (increased competition) across stores for the same brand. The net effect on consumer prices could be positive (Farris and Reibstein 1979). Steiner (1973) presents correlational evidence that highly advertised toys have lower distribution margins. Steiner found that as advertising increased greatly between 1950 and 1970, the average retailer margin on toys declined from 49% to 35%. Steiner (1993) provides additional arguments to explain the existence of an inverse association between the margins of manufacturers and retailers. Relevance Implications These generalizations clearly state the differential effects of nonprice and price advertising on consumers' price sensitivity. Managers should consider these effects when they make decisions about advertising and pricing. The empirical results indicate that coordination between advertising and pricing decisions is important. It is easy to show that coordination will lead to maximum profits for the firm. From a research point of view, these generalizations provide support for both schools of thought regardingthe effect of advertising. Future research should concentrate on aspects such as the characteristicsof advertisingthat influence the relativerole of, for example, the differentiationand reminder effects of advertising. Also, the effects on manufacturer and retailer margins should be substantiated further. Contingencies We identify three considerations that are relevant to the examination of a relationship between advertisingand consumer price sensitivity. The firstis that we need to distinguish between the impact that an increase in advertising has on the price sensitivity of current consumers of a brand and the impact that an increase in advertising has on the average sensitivity of a possibly changing set of consumers. This distinction is important because an increase in advertising may lead to an increase in the number of consumers who purchase the brand. These new consumers may differ in price sensitivity compared with the original set of consumers. Thus, the average price sensitivity may change because of the new mix of consumers that increased advertising has attracted. Studies that use aggregate data cannot ordinarily distinguish between changes in estimated price sensitivity due to a change in the consumer mix from changes due to a change in the price sensitivity of existing consumers. On the other hand, experimental studies that measure the price sensitivity for a constant set of consumers do not suffer from this confound. Of course, both effects may be of interest to a manager or researcher.Theoretically, it is of interest to know the impact on (individual) consumers. For managerial decisions the (total) impact on marketplace behavior may be sufficient. The second aspect that has to be considered is the measure of price sensitivity. With aggregate data, the focus is often on the price effect on unit sales or market share. It is straightforwardto argue that price elasticity is the preferred measure. For disaggregate (e.g., household) data, a comparable focus may be the sensitivity of brand choice to price. The third aspect is concerned with the type of consumers that constitute the market. For example, if the product market consists of highly price-sensitive consumers, then the "ceiling effect" would suggest that price advertising can only have a limited impact on the price sensitivity of such consumers. By the same token, if the market consists of price-insensitive consumers, nonprice advertising can have little impact on these consumers' price responsiveness.

T HE IMPACT OF ADVERTISING ON PRICE SENSITIVITY AND PRICE

G 159

Conclusions and Suggestions The study of the relationship between advertising and price sensitivity has received a lot of attention in the marketing literature. We have shown that the results obtained in these studies provide useful empirical generalizations. These generalizations are based upon a large number of studies involving very differentproducts, measures,and techniques to study this relationship. Potential implications of these generalizations are relevant for managers as well as for researchers. One area that has not received much attention in the area of advertising/price interactions is the role of moderators in affecting this relationship. Balusubramanian et al. ( 1994) have identified several moderators, such as market share, similarity of brands' characteristicsor benefits, product life cycle, and the number of competitors. Potentially, the impact of these moderators is large enough to alter the marketing decisions for a brand. For example, Bemmaor and Mouchoux ( 1991) found that acrossproduct categories the changes in price sensitivity due to increased advertising varied from 20% to 180%. Thus, there is a need to find product-related and other factors that affect the amount of change in price sensitivity in such situations. Similarly,more studies are needed to identify the exact role of these and other moderators. For managers, the interesting question centers on the determination of optimal coordination levels for advertising and price, using information about advertising/price interactions. In addition, managers should consider whether changes in advertising characteristicsaffect the nature or magnitude of the interaction. References
Balusubramanian, S., A. Kaul, L. Klein, S. Sen, A. Silk, and D. R. Wittink (1994), "Price-AdvertisingInteractions," Working Paper, Yale School of Management, July. Bass, F. M. (1994), "EmpiricalGeneralizationsand MarketingScience," WorkingPaper,School of Management, The University of Texas at Dallas, February. Bemmaor, A. C. and D. Mouchoux (1991), "Measuringthe Short-term Effectof In-store Promotion and Retail Advertising on Brand Sales: A Factorial Experiment," Journal of Marketing Research, 28 (May), 202214. Benham, L. (1972), "The Effect of Advertising on the Price of Eyeglasses," Journal of Law and Economics, 4 (October), 337-352. Bolton, R. N. (1989), "The Relationship Between Market Characteristicsand Promotional Price Elasticities," AMarketing Science, 8 (2), 153-169. Boulding, W., E. Lee, and R. Staelin (1994), "Masteringthe Mix: Do Advertising, Promotion, and Sales Force Activities Lead to Differentiation," Journal of Marketing Research, 31 (May), 159-172. Cady, J. (1976), "Advertising Restrictions and Retail Prices," Journal ofAdvertising Research, 16 (October), 27-30. Comanor, W. S. and T. A. Wilson (1974), Advertisingand Market Power, Cambridge, MA: HarvardUniversity Press. and (1979), "The Effects of Advertising on Competition," Journal of Economic Literature, 17 (June), 453-476. Dorfman, R. and P. 0. Steiner (1954), "Optimal Advertising and Optimal Quality," The American Economic Review, 44, 826-836. Eskin, G. J. (1985), "Tracking Advertising and Promotion with Single-Source Data," Journal of Advertising Research, 25 (1), 31-39. and P. H. Baron (1977), "Effects of Price and Advertising in Test-Market Experiments," Journal of Marketing Research, 14 (November), 499-508. Farris, P. W. and M. S. Albion (1980), "The Impact of Advertising on the Price of Consumer Products," Journal of Marketing, 44 (Summer), 17-35. and D. J. Reibstein (1979), "How Prices, Ad Expenditures,and ProfitsAre Linked," Harvard Business Review, 57, 173-184. Gatignon, H. (1984), "Competition as a Moderatorof the Effectsof Advertisingon Sales," Journal of Marketing Research, 21 (November), 387-398. Ghosh, A., S. A. Neslin and R. W. Shoemaker (1983), "Are There Associations Between Price Elasticity and Brand Characteristics?"in Patrick E. Murphy et al. (Eds.), 1983 Educators' Conference Proceedings, Chicago, IL: AMA, 226-230.

G160

ANIL KAUL AND DICK R. WITTINK

Haas-Wilson, D. (1986), "The Effects of Commercial Practice Restrictions: The Case of Optometry," Journal of Law and Economics, 19 (April), 165-186. Kanetkar, V., C. B. Weinberg, and D. L. Weiss (1992), "Price Sensitivity and Television Advertising Exposures: Some Empirical Findings," Marketing Science, 11 (4), 359-37 1. Krishnamurthi, L. and S. P. Raj (1985), "The Effect of Advertising on Consumer Price Sensitivity," Journal of Marketing Research, 22 (May), 1.19-129. Kwoka, J. E. (1984), "Advertising and the Price and Quality of Optometric Services," American Economic Review, 74 (1), 211-216. Lambin, J. J. (1976), Advertising, Competition and Market Conduct in Oligopoly Over Time, Amsterdam: North-Holland Publishing Company. Mitra, A. and J. G. Lynch, Jr. (1993), "Toward a Reconciliation of Market Power and Information Theories of Advertising Effects on Price Elasticity and Purchase Utility," Working Paper, University of Florida. Moriarty, M. (1983), "Feature Advertising-Price Interaction Effects in the Retail Environment," Jouirnalof Retailing, 59 (Summer), 80-98. Nelson, P. (1970), "Information and Consumer Behavior," Journal of Political Economy, 78 (March/April), 311-329. (1974), "Advertising as Information," Journal of Political Economy, 82 (July/August), 729-754. Popkowski-Leszczyc, P. T. L. and R. C. Rao (1989), "An EmpiricalAnalysis of National and Local Advertising Effects on Price Elasticity," Marketing Letters, 1 (2), 149-160. Prasad, K. V. and L. W. Ring (1976), "Measuring Sales Effects of Some Marketing Mix Variables and Their Interactions," Journal of Marketing Research, 13 (November), 391-396. Resnik, A. and B. L. Stern (1977), "An Analysis of Information Content in Television Advertising," Journal of Marketing, 41 (January), 50-53. Sawyer, A. G., P. M. Worthing, and P. E. Sendak (1979), "The Role of Laboratory Experiments to Test Marketing Strategies,"Journal of Marketing, 43 (Summer), 60-67. Schroeter, J. R., S. L. Smith, and S. R. Cox (1987), "Advertising and Competition in Routine Legal Service Markets:An Empirical Investigation," The Journal of Industrial Economics, 26 (September), 49-60. Staelin, R. and R. S. Winer (1976), "An Unobservable Model for Determining the Effect of Advertising on Consumer Purchases," in Kenneth L. Bernhardt(Ed.), Marketing 1776-1976 and Beyond: 1976 Educators' Proceedings, Chicago: AMA. Steiner, R. L. (1973), "Does Advertising Lower Consumer Prices?"Journal of Marketing, 37 (October), 1926. (1993), "The inverse Association Between the Margins of Manufacturersand Retailers," Review of Industrial Organization, 8, 717-740. Urbany, J. E., P. R. Dickson, and R. Key (1990), "Actual and Perceived Consumer Vigilance in the Retail Grocery Industry,"Marketing Letters, 2 (1), 15-25. Vanhonacker,W. R. (1989), "Modelingthe Effectof Advertisingon Price Response:An EconometricFramework and Some Preliminary Findings," Journal of Business Research, 19, 127-149. Wittink, D. R. (1977), "Exploring Territorial Differences in the Relationship Between Marketing Variables," Journal of Marketing Research, 14 (May), 145-155. , J. C. Porter, and S. Gupta (1993), "Dangers in Using Market-LevelData for Determining Promotion Effects," Report Number 93-115, Marketing Science Institute, September. Woodside, A. G. and G. L. Waddle (1975), "Sales Effects of In-Store Advertising," Journal of Advertising Research, 15 (June) 29-34.