Tobacco Documents Online

EFFECTS OF CIGARETTE ADVERTISING ON CONSUMER BEHAVIOR JOEL B. COHEN, Ph.D.* University of Florida Y ~ ~i. (!//,l Imperial Tobacco Limited & RJR-Macdonald, Inc. C. Le Procurer General du Canada Requete pour Judgement Declaratoire *Distinguished Service Professor of Marketing, Adjunct Professor of Anthropology and Director, Center for Consumer Research, University of Florida N CA 0 a a co ~ ~ 0 ~

• . EFFECTS OF CIGARETTE ADVERTISING ON CONSUMER BEHAVIOR 1.00 ECONOMIC ASPECTS OF ADVERTISING . . . . . . . . . . . . . . . . . . . 1 1.10 Market Expansion and Brand Switching Effects 3 1.20 Econometric Studies of Advertising Effects 4 1.30 Experiences in Other Countries 5 1.40 Advertising Effects Can't Be Limited To Brand Switching 7 1.50 Nonsmokers Cannot Be Shielded From Advertising 8 1.60 New Smokers Are Being Added 9 2.00 EFFECTS OF ADVERTISING ON CONSUMER BEHAVIOR . . . . . . . . . . . . . 11 2.10 Advertising and Communications Goals 11 2.20 Asking People to Assess Advertising Effects on Themselves 14 3.00 ADVERTISING AND ITS EFFECTS ON CONSUMER ATTITUDES .......... 16 3.10 What is Emphasized in Advertising 17 3.11 Product attribute claims 18 3.12 Brand image strategies 19 3.20 Executional Strategies 22 3.30 Motivational Processes Affecting Cigarette Ad Effectiveness 24 3.31 Instrumental needs 24 3.32 Value-expressive needs 25 3.33 Ego defensive needs 26 3.34 Cognitive consistency needs 28 3.40 High vs. Low Involvement Cognitive Processes 29 3.50 Persuasion by Association _ ~ 31 4.00 A STAGE OF SMOKING ANALYSIS OF CIGARETTE ADVERTISING .........32 4.10 The Initiation Stage 33 4.11 Risk assessment issues 34 4.12 Advertising distorts heal`'- consequences 35 4.20 The Regular Use Stage 36 4.21 Deception by implication 36 4.22 Advertising provides reminders to smoke 38 4.30 The Withdrawal Stage 39 4.31 Freedom of choice is illusory 39 4.32 Quitting vs. brand switching 41 5.00 CONCLUSION . . . . . . . . . • • • • • • • • • • • • • • • . . . . .43 5.10 A Summary of Key Points 44 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 7

. Cigarette advertising, whether by one company or another, for one brand or another, or one advertising campaign or another, needs to be first understood as, simply, advertising. It is not necessary to examine the differences in approach and style of execution to understand the essence of what advertising does. As Rossiter and Percy (1987) note, "advertising" comes from the latin word "advertere," meaning to turn toward: "Advertising aims at turning the buyer or consumer's mind toward purchase" (page 4). A further analysis of differing advertising philosophies, various advertising agency recipes for successful advertising, how communications goals are turned into concrete executions, how such executions are understood by consumers and why consumers respond more favorably to certain executions can be quite useful. Such analyses tell us how particular types of ads affect perceptions, beliefs, attitudes and ultimately purchase behavior. ' This report will focus initially on the broader and mor.e,basic question of what effects advertising has so that its role in consumers' purchase and use of cigarettes can be better understood. It will then examine how cigarette advertising influences consumer behavior. 1.00 ECONOMIC ASPECTS OF ADVERTISING Firms think about advertising as both an expense item and an investment. As an expense item, advertising represents an allocation of resources to a particular use -- a use that is justified against alternative uses of company resources. Instead of spending a certain amount of money on advertising, the firm could lower its price, provide added profit margins to retailers, and, in short, do any number of things to produce sales and profits. Thus advertising is viewed as one component of the marketing mix. N r.n If expenditures for o ~ a co N O ~A

. advertising could not be justified economically, firms would either spend the money on more productive activities or they would simply reduce expenses. Put simply, advertising is expected to produce sales revenues, or the money would not be spent (Rossiter and Percy, 1987, page 20; Rothschild, 1988, pages 625- 626). When advertising is thought of as an investment there is a more long- term view of its effects. A number of studies have indicated that some firms spend more on advertising than can be justified in terms of immediate sales since they know it has a carryover effect in subsequent years (Rothschild, 1988, pages 622-623; Kotler, 1988, page 621). Advertising is thought to build a consumer franchise in the sense of creating a favorable disposition toward both the company and its products. This favorable disposition represents a potential for purchase which is activated over time whenever needs for the • product become salient or product/company cues are strong enoQgh. As with any other investment, companies have certain return-on-investment criteria. If resources are used for advertising this creates an opportunity cost, since the funds are not invested elsewhere. Hence, the investment in advertising is expected to "pay off" over time. This comes about primarily through increased sales, but so-called "institutional advertising" can also support personnel (e.g., executive hiring) and financial (e.g., public investment) activities. Regardless of whether the firm views advertising as an expense or an investment, advertising's economic role is the same. There is no way to produce sales revenues without convincing people to purchase the product. So, unlike other marketing expenditures which strictly push the product through the sales channel by making it more attractive for middlemen to sell the ~ c5 0 0 N 2 ~

product, advertising is used to pull the product through the sales channel by directly affecting consumers' demand for the product. 1.10 Market Expansion and Brand Switching Effects When the analysis of economic effects is expanded to an industry level, distinctions are sometimes made between advertising that "expands the market" and advertising that "promotes brand switching/defends against brand switching." Such an analysis obscures a very important point. Even though the market in total may not be expanding (or may even be shrinking) customers who leave the market still are being replaced. Describing the competitive conditions of the cigarette industry as a'°mature market" is more useful for marketing managers -- who must decide which elements of their marketing mix to emphasize -- than it is for understanding advertising effects in this market. Even assuming a lack of aggregate industry growth, cigarette.Industry advertising and promotional programs must be strong enough to replace the approximately 5% of cigarette users who leave the market each year merely to maintain sales at existing levels. Advertising plays an important role in attracting new replacement customers. Without replacing people who no longer use the product, stable markets would be in decline and declining markets would be in ruin. The need to replace smokers who either quit smoking or are casualties of smoking has led many to question cigarette industry claims that their advertising is intended primarily to encourage brand switching. Some defend the economic value of brand switching by underscoring the profitability of cigarettes. However, the U.S. evidence indicates that the tobacco industry spends about $9 per person per day for advertising and promotion, and only 102 3

1% of smokers switch brands in an average year (Warner, 1986; 1989 Surgeon General's Report page 503). Most consumers merely switch among brands of the same company (e.g., Imperial has approximately 547% of the market) or move back and forth among the three companies that control 99% of the market. Accordingly, such expenditures -- purely for brand switching -- would seem to be difficult to justify economically. By drastically reducing their advertising and promotion expenditures the industry would appear to be better off financially. If the relative share of such expenditures were maintained, company shares of brand switchers would be about the same, hence profits should be higher. Perhaps cigarette companies have estimated the likelihood of people discontinuing smoking at such reduced levels and have decided that the additional expenditures are justified economically on that basis. We will have much more to say about advertising's important additional role in the initiation phase of smoking behavior. - -- 1.10 Econometric Studies of Advertising Effects There have been frequent attempts to estimate the effects of industry- wide advertising on cigarette consumption. However, no statistical analysis can be safely generalized beyond the scope and sensitivity of the data on which it is based. Econometric studies essentially examine the effect of incremental year-to-year changes in advertising on sales. Such data are lacking in scope: there is no way to extrapolate from them to a ban on cigarette advertising and promotion and the cumulative effects such a ban is likely to have over time. It took many years to develop the cigarette market and to cultivate particular segments of the market (e.g., women). If analysts are really talking about the effects of advertising -- or the effects of a ban 4

1% on advertising -- on cigarette sales, the scope of the analysis can not be limited to short duration effects. Bagozzi (1986) makes a similar point in his analysis of the facilitating role of advertising on the smoking behavior of women (pages 405-406). Also, new smokers are a small proportion of total smokers. We know that incremental changes in advertising expenditures are unlikely to alter the amount smoked for existing smokers (e.g., someone smoking a pack a day is not likely to move to one and a half packs a day). So, even substantial effects on smoking initiation would be swamped by the absence of effects on the larger group of existing smokers. This could lead such statistical analyses to conclude that advertising effects in expanding the market for cigarettes were weak. Unless tobacco companies were willing to provide data which allowed statisticians to isolate the effects of advertising and promotion on first- time smokers and particular population subgroups (e.g., adoleicents), mo§t econometric studies will be insensitive to the otherwise masked effects on smoking initiation. Despite the difficulty of isolating advertising effects on sales, a recent assessment of 14 econometric studies (which took advertising expenditures, tobacco price and personal income into account) concluded that 11 showed that, "advertising significantly affected national cigarette sales" (New Zealand Toxic Substances Board, 1989, page 33). 1.30 Experiences in Other Countries While industry-level effects have been estimated in other countries, based on their experiences with different types of advertising bans, the "experiments" on which such analyses are based are confounded by circumstances unique to each country (e.g., demographic and economic changes, related 5

i* government anti-smoking policies) and the absence of appropriate comparison groups. For example many such "bans" are only partial (i.e., leaving large loopholes for effective tobacco promotion through sponsorship of sporting events and other activities, providing few restrictions on package design and the use of nontraditional forms of advertising) or poorly enforced. Italy is generally considered to be a classic example in which fines for illegal advertising are simply viewed as a routine cost of doing business, even assuming the violation is "noticed." More basically, each country is to some extent unique. Comparing changes in smoking rates across countries requires careful consideration of cultural and attitudinal factors affecting the use of cigarettes. As a general rule, comparisons between individual countries are not likely to be meaningful unless such factors have been taken into account.' The previously cited report of the New Zealand Toxic Substances Board joins with the World Health Organization in criticizing a 16-country study carried out under the auspices of the Children's Research Unit and presented through the International Advertising Association, using the following strong language, "In short, the data in the IAA brochure were deficient and the conclusions simplistic, unjustified, erroneous and misleading" (New Zealand Toxic Substances Board, 1989, p. 61). Owing to such weaknesses, New Zealand commissioned its own 33-country study and reported that, "When countries were grouped according to the degree of governmental restriction of tobacco promotion, the greater the degree of restriction the greater the average annual fall in tobacco consumption. This was also true for the rate of decrease in the percentage of adults and young people who smoke" (New Zealand Toxic Substances Board, 1989 p. 64). 6

1.40 Advertising Effects Can't Be Limited To Brand Switching In truth, there is no such thing as "only brand switching advertising." Any advertising effective enough to entice consumers to switch brands has, by definition, made a particular cigarette more appealing. While it is possible to debate how much more appealing a particular cigarette brand has become as a result of advertising (and more will be said about how firms examine this question later on), if the advertising did not add significantly to the appeal of the brand it would be changed. If no advertising could be found that would add significantly to the appeal of the brand then the money would not be spent. It may be relevant here to respond to the question, "Couldn't the advertising only make the particular cigarette attractive for existing smokers?" Though the reasons for this will be clearer a little later in our' discussion, the answer is essentially, "No." There are many_.products ranging all the way from home electronics to garden tools for which the answer would be somewhat different. In those cases, the product itself meets the relevant set of needs and wants. The secondary choice of brand, model and store is made on the basis of a tradeoff among performance features and terms of sale. The brand, then, does not have the drawing power to bring people into the market for these products. Such brand advertising and promotion primarily affect choice once a consumer has decided to purchase the product. This is not the case for cigarettes and a number of products for which the different brand images draw consumers uniquely to them. As the legendary advertising executive, David Ogilvy, put it: "The greater the similarity between brands, the less part reason N cn plays in brand selection. There isn't any significant difference ~ 0 m 0

between the various brands of whiskey, or cigarettes, or beer. They are all about the same. And so are the cake mixes and the detergents, and the margarines. The manufacturer who dedicates his advertising to building the most sharply defined personality for his brand will get the largest share of the market at the highest profit." (as cited in Bagozzi, 1986, page 386.) Though cigarettes may also differ among themselves in appearance and performance aspects, an adolescent becomes a Marlboro smoker, a Virginia Slims smoker or an Export "A" smoker, not just a cigarette smoker. Thus, not only is cigarette smoking itself given a boost each time a brand of cigarettes is presented in an attractive manner, but the brand acquires additional value as a means of supporting or he:ping to establish a desirable personal identity.. 1.50 Nonsmokers Cannot Be Shielded From Advertising Nonsmokers, and particularly adolescents, cannot be made immune to advertising effects -- even if the primary goal of such advertising were to attract smokers of other brands. We will shortly take up the question of the cigarette industry's stated lack of intention to attract new smokers. For now it is only necessary to consider the argument that cigarette advertising functions primarily to motivate brand switching. In order to accept this idea we would need to believe that the attractive and heavily researched images and symbols connected to cigarette smoking and the various brands only "work" on people who already smoke. It is as if a magic curtain could be put in place to shield children, teenagers and others from the impact of these appeals. No convincing theoretical argument or empirical evidence has yet been introduced 8 N ttt 0 0 0 w tj NJ ~ ~