A Preliminary Investigation of the Information of Cigarette Advertising: a Logitudinal Analysis

Debra Jones Ringold, The American University
ABSTRACT - This paper examines whether cigarette advertising has exclusively provided information regarding the pleasurable aspects of smoking or whether it also raised the possibility that cigarette smoking is unhealthy and/or dangerous. This is relevant given the evolution of product liability theory and the commercial free speech controversy. The findings provide empirical support for Calfee's (1985) assertion that prior to the FTC's 1955 cigarette advertising guides, health claims constituted a considerable portion of the information disseminated through cigarette advertising.
[ to cite ]:
Debra Jones Ringold (1987) ,"A Preliminary Investigation of the Information of Cigarette Advertising: a Logitudinal Analysis", in NA - Advances in Consumer Research Volume 14, eds. Melanie Wallendorf and Paul Anderson, Provo, UT : Association for Consumer Research, Pages: 269-273.

Advances in Consumer Research Volume 14, 1987      Pages 269-273


Debra Jones Ringold, The American University

[The author wishes to acknowledge the efforts of Stefan Schechter and Sarah Kvale expended in the collection of the advertisements used in this study. Ms. Kvale and Raheel Masood served as the raters and are due my thanks. The helpful comments of Gary T. Ford and John E. Calfee are gratefully acknowledged.]

[The American University, Kogod College of Business Administration, 4400 Massachusetts Avenue. N.W., Washington, D.C. 20016.]


This paper examines whether cigarette advertising has exclusively provided information regarding the pleasurable aspects of smoking or whether it also raised the possibility that cigarette smoking is unhealthy and/or dangerous. This is relevant given the evolution of product liability theory and the commercial free speech controversy. The findings provide empirical support for Calfee's (1985) assertion that prior to the FTC's 1955 cigarette advertising guides, health claims constituted a considerable portion of the information disseminated through cigarette advertising.


Conventional wisdom has long recognized that cigarettes may be unhealthy and/or addictive. Cigarettes have historically been referred to as "coffin nails" and were the target of antismoking crusaders in the early part of this century (Calfee 1985). As early as 1604, Ring James is said to have been disgusted at the fact that tobacco users become "obstinately addicted" to the substance (Garner 1980). It was even suggested in a 1953 Business Week article that the cigarette industry itself, had discouraged smoking by advertising "at the top of their lungs about nicotine, cigarette hangovers, smoker's cough, mildness and kindred subjects." While Calfee (1985) has examined the effect of regulation on health information in cigarette advertising, no one has ever systematically examined the extent to which cigarette advertising has provided information leading to the conclusion that smoking may be unhealthy.

Such an examination of past and current cigarette advertising is interesting given the evolution of product liability theory and the commercial free speech controversy. Moreover, Calfee (1985) has suggested that prior to the Federal Trade Commission's (FTC) 1955 imposition of cigarette advertising guidelines, health related claims constituted a significant portion of information conveyed in cigarette advertising. The purpose of this paper then, is to contribute to our understanding of what types of information, or lack thereof, was/is provided by cigarette companies, and was/is available to smokers, through print advertisements. More specifically, this paper examines whether cigarette print advertising has provided information about the health consequences of smoking. Since the adverse consequences of smoking often take several decades to reveal themselves, cigarette print advertising throughout the last sixty years is examined.

These findings are particularly relevant in the context of current product liability litigation wherein plaintiffs have alleged that they relied on advertisements for information and were induced by them to continue smoking notwithstanding the warnings (Harvard Law Review 1986). In the context of the commercial free speech controversy, which frequently focuses on the right of the consumer to receive commercial information (Welkowitz 1986), this study will help identify the types, and relative prominence, of information conveyed. And lastly, these findings will assist in quantifying the degree to which health claims were made prior to the FTC 1955 cigarette advertising guidelines.

The paper is organized as follows. Research objectives and methodology are nest discussed and described. This is followed by a presentation of the findings. The final section discusses the implications of this study and presents suggestions for additional research.


With the foregoing in mind, the objective of this research was to determine: (1) the extent to which cigarette advertising included information about the health consequences of smoking; (2) the type of health information contained in such claims; and (3) how the extent and type of health information varied over time and by brand.

Given these objectives, content analysis was thought to be the most appropriate methodological approach. Information provided in advertisements from 1926 to 1985 was analyzed. This time period was selected because magazine advertisements for these products were uncommon prior to 1926 and were not yet available for 1986. Data collection proceeded as follows. Five cigarette brands offered for sale during this time period were selected. The original product type, manufacturer, and introduction date for each of the brands chosen was: Camel (unfiltered, R.J. Reynolds, 1913), Chesterfield (unfiltered, Liggett-Meyers, 1912), Old Gold (unfiltered, Lorillard, 1926), Viceroy (filtered, Brown and Williamson, 1936) and Lucky Strike (unfiltered, American Tobacco, 1916). In addition the advertisements associated with a later filter brand (Rent, Lorillard, 1952) were examined.

These brands were chosen on the basis of market share and size of firm. Camel, Lucky Strike, and Chesterfield together commanded over 80% of the market in both 1925 and 1935 (Calfee 1985). By 1950, those three plus Old Gold and Phillip Morris had a combined share of over 80% of the market. Viceroy, therefore, was a less popular brand and was included to contrast the other four. By 1963, the filter brands reduced the combined share of the original big three brands to 21% (Calfee 1985), and thus, Rent was selected to represent these entries. Further, brands offered by both large and small firms were desirable. Camel and Lucky Strike were produced by the two consistently largest firms (R.J. Reynolds and American Tobacco), while the remaining four brands were produced by smaller actors (Calfee 1985).

One print advertisement for each of these brands was sought for each of the sixty years. It was felt that one advertisement per year was sufficient to illustrate how advertising content changed over time. Advertisements were obtained from Time magazine (first published 1923) dated as close as possible to July 1. July 1 was selected in order to avoid seasonal variation and year-end budgetary effects. If an ad for any of the above brands could not be located in Time, an ad was pulled from the corresponding volume of The New Yorker (first published 1925, ceased cigarette advertising in 1964), The Saturday Evening Post (published 1821 to 1969 and 1971 to present), or Life (published 1936 to 1971 and 1969 to present), in that order. Of these publications, only Time is considered to be a "news and opinion magazine", the remaining three are categorized as "general editorial" (Ulrich's International Periodical Directory 1985).

Advertisements could not be located, in the publications chosen, for each brand in every year. Further research indicated that the primary reason for a missing advertisement was the total absence of magazine advertising for a given brand in a given year (Publishers' Information Bureau/Leading National Advertisers). The next most common reason was, according to LNA, the absence of advertising in the publications selected for a given brand in a given year.

Advertisements for Camel were most successfully located with 0.77 of all years represented (n=46, 60 years). The other brands are represented as follows: Kent 0.76 (n-25, 33 years), Lucky Strike 0.73 (n-44, 60 years), Viceroy 0.68 (n-34, 50 years), Chesterfield 0.63 (n-34, 60 years) and Old Gold 0.40 (n-24, 60 years). [All four of the unfiltered brands later added a filter. When advertisements for the unfiltered variety could not be located, the filtered variety of the same brand was substituted and noted. The years in which these substitutions began were: Camel 1971, Lucky Strike 1965, Chesterfield 1967, and Old Gold 1955.] Of the 211 advertisements collected, 70.6 percent case from Time, 8.5 percent from The New Yorker, 8.5 percent from The Saturday Evening Post, and 12.3 percent from Life,

A coding scheme for the information conveyed in the headline, subhead and copy was then developed. A review of the literature, [In particular see Doron (1979); Miles (1982); Taylor (1984); Calfee (1985); Pollay (1985); Tye (1985); Warner (1985): and The American Medical Association (1986).] and an examination of a variety of ads, suggested the following general categories of information related to: health, cigarette construction, taste, exhortation to purchase/slogans, hedonic satisfaction, tar and nicotine content, price/ availability, coupons/contests, and celebrity/athlete endorsements. Each claim or statement (hereafter referred to as claims) [An exhortation to buy a brand is not strictly an advertising claim, it is a directive. Likewise, encouraging contest participation is not an advertising claim, it is an invitation. However, the majority of the information found in these advertisements could reasonably be described as claims.] was entered only once regardless of the number of times that particular claim was repeated within a single advertisement. For example, if a claim appeared in the headline, the subhead, and the copy, the claim was coded for the headline only. The focus of the study was the number and type of distinct claims or statements, not the amount of repetition within the ads. This is consistent with the approach adopted by Pollay (1985).

Two raters (judges), who were unaware of the purpose of the study, were trained in the use of the coding scheme. Each judge examined and coded the entire set of ads and the inter-rater reliability vas found to be 0.81. [As suggested by Kassarjian (1977), the measure of reliability used was the ratio of coding agreements to the total number of coding decisions. Kassarjian asserts that researchers can be satisfied with coefficients of reliability above 85 percent.] Differences in code assignments were then resolved and the resulting consensus served as the final data base.


The results will be presented as follows. First, the most common categories of claims will be described and compared across brands and time periods. This will be followed by an examination of the specific health claims made and an evaluation by brand and by decade.

Most Common Claims. Taste claims (e.g. mildness, flavor, aftertaste) were the most frequently made in the total set (all brands, all years) of advertisements: they constituted 22.3 percent of all claims made. Next in frequency (19.7%) were statements associated with cigarette construction (e.g. filter, quality of tobacco, package). Health claims were the third most frequently made, with 18.2 percent of all claims being health related (e.g. reduced throat irritation, reduced coughs, protection against adverse health effects). Thus, nearly one-fifth of all claims involved health (health claims exclude the Surgeon-General's Warning and tar and/or nicotine figures). Statements relating to pleasure accounted for 10.7 percent. All remaining information fell into categories constituting 33.1 percent of claims made. In the 211 total advertisements, an average of 5.5877 claims were made (s.d. - 2.2857) and ranged from 1 to 12 claims per advertisement.

Since an advertisement was not available for all brands in all years, comparisons of claims by brand and over time were made on the basis of the percent of total claims constituted by a particular category of claims. The time periods used were as follows: 1926-29, 1930-39, 1940-49, 1950-54, 1955-59, 1960-69, 1970-79, and-1980-85. Analysis of claims by decade is convenient and is consistent with the work of advertising historians such as Fox (1984) and Pollay (1985). The decade of the fifties was split in order to more closely examine the effect of the 1955 Federal Trade Commission ban on health claims in cigarette advertising. This allows for a pre- and post-1955 analysis consistent with Calfee (1985). Advertising of tar and nicotine figures by certain brands (Camel, Lucky Strike and Old Gold) was banned by the FTC in 1950-51 (Calfee 1985;1986); tar and nicotine figures associated with brands not effected by earlier FTC actions were banned by the FTC in 1960; and the inclusion of ear and nicotine figures was subsequently mandated in 1970. Thus, changes in policy regarding the inclusion of tar and nicotine figures coincide well with the decade breakdown used.

As is shown in Table 1, from 1926 to 1954 the three most frequently made claims in cigarette advertising were those



related to health, taste, and cigarette construction. Health claims were the most frequently made type of claim in three out of the four periods prior to 1954. A typical health claim made by A. J. Reynolds was "Not one single case of throat irritation due to smoking CAMELS." This was stated in conjunction with the findings "of noted throat specialists after a total of 2,740 weekly examinations of throats of hundreds of men and women who smoked Camels--and only Camels--for 30 days" (Time 1950). Viceroy, a Brown and Williamson product, suggested that..."Filtered cigarette smoke is better for your health" (Time 1951) and Lorillard's Old Gold asked..."Why Risk Sore Throats?" when Old Gold promised "Not A Cough In A Carload" (New Yorker 1929).

From 1955 to 1969, taste claims were most frequently made, followed by cigarette construction, exhortation to buy/slogan, and pleasure. From 1970 to 1985, tar and nicotine figures, taste claims, and the Surgeon-General's warning were most frequently conveyed. During the first five years of the eighties, car and nicotine figures, the Surgeon General's warning, and claims associated with reduced tar and/or nicotine were most frequently presented. Thus, health related information constituted one of the four most frequently made types of claim/statement in every period studied with the exception of 1955 to 1969. See Table 1 for more detail.



As discussed above, prior to 1955, health claims were an important component of cigarette advertising for five out of the six brands studied. See Table 2. Two of these six brands (Old Gold, Lorillard and Lucky Strike, American Tobacco) incorporated health claims in their advertising in the 1920's. Camel, Chesterfield and Viceroy began to utilize such claims in the 1930's. Calfee (1985) has suggested that advertising which included "less unhealthy" claims (suggestions that the advertised brand is less unhealthy than a competitor's) would be a more advantageous strategy for smaller firms and that smaller firms would be likely to engage in such advertising earlier than larger competitors. This study neither confirms nor refutes this suggestion since health claims were present in the early advertisements sponsored by both American Tobacco and Lorillard, a large and small firm respectively. Because of the small number of ads available in the twenties and thus included in this study, it is impossible to establish which of these companies incorporated these claims first. For Rent, introduced in 1952, such claims represented the central focus of its introductory advertising campaign. Only one brand, Chesterfield, failed to utilize health claims frequently. For all of the remaining brands, health claims comprised either the most or the second most frequently used category.

After 1955, taste and construction claims, together with information associated with tar and nicotine, were most common for five out of the six brands examined. Chesterfield was again exceptional in that tar and nicotine information was not presented by this brand until the inclusion of such information was mandated by the FTC. See Table 3.



Health Claims. Of the health claims utilized in the various brand advertisements prior to 1955, those associated with throat irritation, coughs, the irritating properties of cigarettes, and protection from adverse side effects were most co hon. In the forties and early fifties, it became a common practice to cite "scientific" or "medical" research "substantiating" these types of claims. In the early fifties the notion of health risk reduction was not an uncommon one. It was the fourth most frequently used health claim in the first years of that decade. See Table 4.

Lorillard, makers of Kent, incorporated many of these assertions in a 1953 advertisement: "Here's how you can get the protection you need against nicotine and tars--and the smoking pleasure you want...if you're a sensitive smoker-- and published medical reports show that 1 out of every 3 smokers is unusually sensitive to the tars and nicotine in tobacco--you can now enjoy real health protection with every cigarette you smoke...If smoking dulls your sense of taste, gives you a 'raw' throat or 'bunched-up' nerves, chances are you're sensitive to nicotine and tars. So, for your own health, as well as pleasure, you should try KENT" (Time 1953).



An examination of Table 5 suggests that, with few exceptions, there is little to distinguish the various brands in terms of the health claims frequently used.



Five out of six brands referred to throat irritation; risk reduction or protection was a frequent theme with five of the six; four out of six referred to "scientific" findings. Camel was unique, however, in that it presented brand endorsements by medical personnel.

The actual placement and/or prominence of health claims within the advertisements was also of interest. Of the total health claims made, 235 were presented in the headline, 31% were presented in subheads, and 46% were presented in the copy. Prior to 1955, brands did exhibit differences in the placement and/or prominence of health claims within these ads. Chesterfield and Kent placed the majority of health related claims in the copy, while the four remaining brands (Camel, Old Gold, Viceroy and Lucky Strike) accorded the majority of such claims the prominence associated with headlines and subheads. See Table 6.



After 1955, health related claims were very seldom made (6.5% of all health claims made) and when they were, over 70% were presented in the copy. The Surgeon-General's warning and tar/nicotine figures were presented, almost without exception, in copy. Claims associated with a reduction in tar and/or nicotine however, were most often presented in the headline or subhead.


While this study is not one of consumer perception or information processing, it does identify the dominant advertising themes associated with several cigarette brands and examines them over time. It has been demonstrated that until 1954, health claims were a consistent and important consideration in the advertisement of the brands analyzed here. Moreover, in the years following 1970, tar and nicotine figures, the Surgeon-General's warning, and claims regarding reduced ear and/or nicotine were prevalent. Only in the period from 1955 to 1969 was there a paucity of commercial information relating to health.

These findings provide considerable empirical support for Calfee's (1985) assertion that prior to the imposition of the FTC's 1955 cigarette advertising guidelines, cigarette manufacturers voluntarily addressed the issue of cigarettes and health. m e prominence accorded to health related claims and the constancy with which these claims appeared leaves little doubt that such claims were a major component of the cigarette advertising strategy adopted by the majority of brands examined here.

Consideration of consumer interpretation of the health related information found in these ads is beyond the scope of this study. However, Calfee (1985) has suggested that brands claiming to be less unhealthy draw attention to the harmful aspects of smoking. Consistent with this view, the present study found that until 1955, ads which incorporated health claims typically emphasized the harmful effects of the product class in an effort to make other brands appear more harmful by comparison. It is Calfee's (1985) conclusion that in the presence of such advertising, it was probably impossible for any smoker, or potential smoker, to ignore or forget that cigarettes lead to throat irritation, smoker's cough, and other symptoms frequently enumerated in the cigarette advertising of the period.

It is the period 1955 to 1969 wherein health related information in cigarette advertising was lacking. In part, this paucity of health related information may be a function of the brands examined herein. Calfee (1985) found that, in general, tar and nicotine claims were frequently made in the period 1957 to 1960. This was not the case in this study. However, it must be pointed out that three out of the six brands examined herein were subject to earlier FTC bans on tar and nicotine figures, and thus it is not surprising that such assertions comprise a very small percent of total claims made from 1955 to 1959 (less than 1%). Tar and nicotine figures were completely absent after 1960, when the FTC uniformly prohibited their use in cigarette advertising. Moreover, as a result of the 1955 FTC ban, health claims were very rarely made (less than 2% of total claims) from 1955-69.

In 1970, the cigarette industry agreed to use Federal Trade Commission approved figures for tar and nicotine in all advertisements. Claims associated with reduced tar ant/or nicotine began to appear in the seventies as well. The first health warning to appear in cigarette advertisements, "Warning--the Surgeon General has Determined that Cigarette Smoking is Dangerous to Your Health", did so in 1972. With the uniform inclusion of health warnings and tar and nicotine figures it is not surprising that these statements are among the four most frequently made in advertisements after 1970.

Implications for Future Research. This study suggests that further research is warranted in order to appreciate the types, and relative amounts, of information contained in cigarette advertising over time. In addition to expanding (including more brands) and improving (including additional print media) the existing longitudinal data base, it has been suggested that a "cross-sectional" examination of cigarette advertising be undertaken (Calfee 1986). Using a methodology similar to that developed by Pollay (1985), it is possible to survey cigarette advertising associated with the product class in selected periods. Such an examination would permit the recognition of behavior idiosyncratic to the brands examined longitudinally and permit an appreciation for the types, and relative amounts, of information generally available to smokers regardless of brand.

If, as has been alleged, plaintiffs in product liability litigation relied on cigarette advertising when making decisions to begin or continue smoking, then an appreciation for when this decision was made seems prudent. If these findings are generalizable, or if plaintiffs smoked the brands examined herein, a case could be made that cigarette advertising did, or did not, provide information relevant to health depending upon when the smoking decision was made. This of course is only the most rudimentary information issue, since how consumers evaluate such information weights heavily on its usefulness. Calfee's (1985) work suggests that an empirical investigation of how consumers process "less harmful" claims is warranted Do such claims, as he suggests, inform users that consumption of the product class is harmful and therefore provide information pertinent to the assessment of risk? It is also important to examine longitudinally the number, and location, of articles appearing in the popular press which discuss the effects of cigarette smoking on health. This seems particularly relevant for the period 1955-69 in which commercial presentation of health related information was virtually prohibited.

This study has interesting implications for the commercial free speech arguments concerned with consumers' right to commercial information. This study demonstrates that for these Six brands in the years 1980-85, tar and nicotine figures, claims regarding reduced tar ant/or nicotine, and the Surgeon-General's warning constitute three out of the four most frequently made claims or 58.492 of all claims mate. One might argue therefore, that the information content in recent cigarette ads is considerable and that the public would suffer a loss of health related information as a result of a ban on remaining cigarette advertising. Pertinent to such an argument, as well as to those individuals seeking recovery from the cigarette industry, is the issue of consumer evaluation of the health warnings and the tar and nicotine figures incorporated in every advertisement in the recent past. Are such warnings in fact vitiated by surrounding at copy and visuals? Can consumers make use of tar and nicotine figures in such a way as to evaluate their consumption of both? Clearly, additional research is warranted.


The American Medical Association Board of Trustees (1986), "Media Advertising for Tobacco Products," Journal of the American Medical Association, 255 (8), 1033.

Calfee, John E. (1985), "Cigarette Advertising, Health Information and Regulation Before 1970," working paper, Bureau of Economics, Federal Trade Commission, 1-79.

Calfee, John E. (1986), Bureau of Economics, Federal Trade Commission. personal communication.

Doron, Gideon (1970), The Smoking Paradox, Cambridge, Massachusetts: Abt Books.

Fox, Stephen (1984), The Mirror Makers: A History of American Advertising and Its Creators, New York: William Morrow and Company, Inc.

Garner, Donald W. (1980), "Cigarette Dependency and Civil Liability: A Modest Proposal," Southern California Law Review, 53 (1371), 1423-1465.

Harvard Law Review (1986), "Plaintiffs' Conduct as a Defense to Claims Against Cigarette Manufacturers," Harvard Law Review. 89 (February), 809-827.

Kassarjian, Harold E. (1977), "Content Analysis in Consumer Research," Journal of Consumer Research, 4 (June), 8-18.

Miles, Robert R. (1982), Coffin Nails and Corporate Strategies, Englewood Cliffs: Prentice-Hall, Inc.

Pollay, Richard W. (1985), "The Subsiding Sizzle: A Descriptive History of Print Advertising, 1900-1980," Journal of Marketing, 49 (Summer), 24-37.

Publishers' Information Bureau/Leading National Advertisers (all years extant), Magazine Advertising Expenditures: New York: Publishers' Information Bureau.

Taylor, Peter (1984), Smoke Ring: The Politics of Tobacco, London: The Bodley Read.

Tye, Joe B. (1985), "Cigarette Marketing: Ethical Conservatism or Corporate Violence?" New York State Journal of Medicine, July, 324-327.

Ulrich's International Periodical Director, 24th Edition (1985), New York: R.R. Bowker Company.

Warner, Kenneth E. (1985), "Tobacco Industry Response to Public Health," Health Education Quarterly, 12 (2), 115-127.

Walkowitz, David (1986), "The Ban on Television Advertising of Cigarettes in Light of Efforts to Do the Same in Print," New York Law Journal, 195 (April 4), 1, 27.