A day in the life of an advertising man: review of internal documents from the UK tobacco industry's principal advertising agencies (original) (raw)

BMJ. 2000 Aug 5; 321(7257): 366–371.

Centre for Tobacco Control Research, University of Strathclyde, Glasgow G4 0RQ

Contributed by

Contributors: The authors jointly discussed the conception of the paper and methodology. GH conducted the analysis and was responsible for writing the findings; he will act as guarantor for the paper. LMacF conducted the literature review. Both authors discussed and agreed to the conclusions.

In July 1999 the Health Select Committee began an investigation into the British tobacco industry to determine what action it had taken and was taking in response to the known harmful effects of smoking and the addictive nature of tobacco. One of the committee's key achievements was to force the disclosure of a large quantity of internal company documents, including—for the first time—material from the UK tobacco industry's leading advertising agencies.

These documents cover all matters to do with tobacco promotion. They shed a unique light on the social research that has been done over the past 20 years to establish if and how tobacco promotion affects smoking behaviour. This research has been extensive but by necessity limited to studying observable outputs. The documents complete the picture by disclosing the inputs. In the process the documents corroborate the key findings of research and also yield much about the motivations and mindset of the UK tobacco industry.

Summary points

Methods

The select committee identified five UK advertising agencies with clients in the tobacco industry, and using its statutory powers it requested internal documents covering tobacco related business in the previous five years. On our advice it requested particular types of document (box ​B1). Between them the agencies submitted 16 boxes containing several thousand pages of material. We have no way of knowing what proportion of the requested documents was actually sent, but given the select committee's standing and powers it is unlikely that many were withheld. It is possible that documents may have been lost or destroyed before the investigation.

: Documents sought from advertising agencies

In 1995 Silverman argued that the qualitative analysis of a large body of data is best conducted by using key themes.1 In this case the obvious source of such themes was the literature on the processes and effects of tobacco marketing. A review of this literature showed four key recurring questions: Does tobacco advertising affect consumption as well as brand share? Does the industry target young people? What part does sponsorship play? What other forms of promotion are there and what part do they play? We analysed the documents on the basis of these questions. The extent to which the themes were addressed varied between the different types of document. For example, the contact reports were simply brief minutes of business decisions made, and much of the market research material was unremarkable statistical data about market trends. In contrast, the creative briefs and qualitative market research were extremely revealing, not so much in what they said but in what they assumed about the four questions.

Key questions

Does tobacco advertising affect consumption as well as brand share?

The tobacco industry argues that its sole concern is to encourage “brand switching” among current adult smokers and not to increase consumption in itself. Studies that have focused on advertising have modelled fluctuations in advertising spend with changes in smoking prevalence,2–4 differences in smoking prevalence in countries with varying levels of advertising controls,5,6 and changes in smoking prevalence after the introduction of an advertising ban.7,8 A review of these studies undertaken by the Economics and Operational Research Division of the Department of Health concluded that advertising encourages consumption as well as brand switching among current smokers and may assist in recruiting young people to the habit.8

Does the industry target young people?

Around 80% of UK smokers begin smoking as a teenager, making children and young people vulnerable potential targets.9 Consumer research has shown that tobacco advertising is having a powerful effect on the smoking attitudes and behaviour of young people.10–13 Research has shown that the use of imagery and positive association in tobacco advertising has a notable impact on young people.14 Children are also brand conscious and, compared with adults, are much more likely to smoke the most popular and well advertised products.15–17

What part does sponsorship play?

Research has concluded that sponsorship works in a similar manner to advertising, by increasing brand awareness, promoting strong brand associations, and easing the decision to take up smoking.18–20 For instance, research has concluded that a preference for Formula One motor racing sponsored by cigarette manufacturers was a significant independent variable in progression to regular smoking.21

What other forms of promotion are there, and what part do they play?

It is not just advertising and sponsorship that have come under scrutiny. Other forms of promotion, including cigarette coupons, brand stretching (non-tobacco products with tobacco brand names), product packaging, point of sale materials, free give aways, product placement, and even the use of the internet might have effects on behaviour and attitudes to smoking.22,23 For example, Pierce et al found that tobacco promotion, including free gifts bearing tobacco brands, were causally related to the uptake of smoking among adolescents.24

Results

Does tobacco advertising affect consumption as well as brand share?

The internal documents revealed a desire to increase consumption as well as brand share. Attempts to expand the market (and slow its contraction) took three forms. Firstly, specific campaigns supporting the smoker and smoking were mooted and developed. One such effort aimed to promote the idea that “smoking can be a delight for everyone if it is done right.”w1 Other documents advocated “active support for smokers' rights”w2 and confirmed that “as smokers become more and more persecuted, they look to advertising as a friend.”w3

Secondly, particular sectors of the market can be expanded. For example, a detailed report examined ways to bring about “the reinvigoration of the cigar market.”w4 The worry was then expressed that “without this reinvigoration we will continue to see a decline in the size of the cigar market, which will hurt all brands, but particularly Hamlet.”w4 The last comment also highlights the link between brand share and market size.

A third option was to undermine tobacco control policies. Ad campaigns directly attacking particular policies were planned and researched (box ​B2), campaigns were orchestrated to minimise the impact of budget increases, and options were considered for marketing after tobacco advertising is banned. As one of the agencies put it, “why should they stop marketing their products simply because advertising is banned?”w5

: Antigovernment advertising concepts taken into consumer research(quoted from miscellaneous documents)w5

Tobacco consumption is also increased more directly by recruiting new smokers and discouraging existing ones from quitting. New recruits to smoking are seen as a crucial sector and success here is carefully monitored. For example, a decrease in their numbers is a serious problem for Silk Cut, whereas Lambert and Butler and Marlboro Lights are envied for their recruitment success (box ​B3).

: Importance of recruiting new smokers

Smoking cessation is, however, a threat, and low tar options backed by reassuring imagery can fend it off: “In conclusion, the ‘threat’ to the existing smoker base for the brand seems more likely to come from a desire to give up smoking (which is more strongly held by the SCUL [Silk Cut Ultra Light] smokers than the SCEM [Silk Cut Extra Mild] smokers), rather than from competitor brands.”w8 “A minority saw low tar cigarettes as a stage on the way to quitting smoking . . . . However, more common was a sense that low tar was a way of making quitting less urgent or necessary.”w9

Does the industry target young people?

The internal documents show that young people are a key group for tobacco marketers. The precise definition of “young” remains contentious: in many of the documents care is taken to use the phrase “young adult smokers”; however, at other times more general terms such as “young people” and the “youth market” are used. Furthermore, there is evidence of market research being done with people as young as 15. Box ​B4 shows a reanalysis by a consultancy agency of “TGI” data (a standard industry resource that is bought on a subscription basis by many advertising agencies) to provide a detailed picture of the values and aspirations of Silk Cut smokers. The sample includes 15 year olds.

: Market research with 15 year olds(quoted from Synergy Consulting)w10

Whether the industry is deliberately targeting under 16 year olds remains a matter for dispute. What the internal documents did, however, make clear was that smoking among young people was as much about image as it was about product attributes. The agencies recognise that smoking is a “rite of passage,” with young people looking for “reassurance” and “an identity” (box ​B5).

: Smoking satisfies emotional insecurities in the young

Successful brands exploit these emotional needs and insecurities. The success of Marlboro Lights, for example, “derives from its being the aspirational lifestyle brand . . . The Diet Coke of cigarettes,”w14 and “to be successful any Gallaher brand will have to tackle Marlboro's coolness of image—smokers do smoke the image as well as the taste.”w15 As one creative brief put it, “we want to engage their aspirations and fantasies—‘I’d like to be there, do that, own that.'”w16

Detailed and typically qualitative market research is therefore conducted to guide the development of “image building campaigns”w17 and provide “an infusion of style, coolness and aspiration . . . that will boost B&H's [Benson and Hedges] image with style conscious 18-24s.”w16 This takes place despite clear rules in the voluntary agreement prohibiting the association of smoking with social success or any attempts to play on the susceptibilities of those who are emotionally or physically vulnerable, especially young people.25

In addition, whether or not children are deliberately being targeted, no consideration is given to the danger that marketing aimed at adults may actually appeal to and reach those aged less than 16. This is perhaps most apparent when the documents discuss the issue of packs of 10 cigarettes. It is clearly recognised that these are bought predominantly by young adult smokers in independent outlets (corner shops) as a cheap means of acquiring premium cigarettes.

No concern is expressed that these packs, outlets, and brands may also appeal to children, although independent research has shown this to be the case.16,17,26 They do, however, recognise that “new entrants” to the market are likely to approach it through these routes: “whilst this data is not completely reliable it does reinforce the picture from old BJM data in highlighting the role of the 10s pack amongst young adult smokers and potentially new entrants to the market.”w18

What part does sponsorship play?

The internal documents confirm the similarities between advertising and sponsorship. The prime purpose of both is to create and bolster brand image. Careful consumer research is conducted to examine the image of particular sports, and the most appropriate and influential ones are then selected. Perhaps the most notable example of this (and the most worrying one, given that sponsorship of it is due to remain until 2006) is Formula One racing. One research report shows that “more active sports, with potential to create a more dynamic, exciting brand image” include “Formula One, big boat sailing, basket ball, ice hockey.”w5 The image of Formula One racing is then described in more detail as “international, glamorous, challenging, fast, furious, dangerous, living life to the full and living life on the edge.”w5 The document concludes that Formula One racing can make the Benson and Hedges' brand more “dynamic,” “macho,” and “youthful.”

A further report concludes, in similar vein, that sponsorship of Formula One racing makes the brand “very powerful” and lends “associations with young, fast, racy, adult, exciting, aspirational, but ultimately attainable environments.”w19

Other sponsorship deals are selected with equal care. Rugby league makes Silk Cut “an exciting dynamic and less pretentious brand”w20 and the Whitbread round the world yacht race makes it “masculine” and “adventurous.”w21

Sponsorship and advertising are also used in close combination: “sponsorship advertising” promoted the Silk Cut renaissance tour [of arts and raves], and publicity for the brand's Whitbread [sailing] sponsorship is maximised by placing “‘announcement’ ads in all national newspapers that run editorial on the race—to appear next to, or near, the editorial.”w21

Furthermore, the advertisers themselves find it difficult to disentangle the effects of their sponsorship from their advertising, one of them needing to conduct careful research “to identify separately the impact of Marlboro's sponsorship of the Ferrari team on the overall effectiveness of Marlboro advertising.”w22

This confusion arises because the criteria for success are identical: “as I'm sure you are aware there was excellent coverage of the new Jordan car last night on both the 9 O'Clock News and the News at Ten. The respective all men television ratings for the bulletins were 11.8 and 14.4. If we assume that the coverage equated to a 60” commercial on each station, I've estimated the equivalent advertising value to be £185 000. When the value of additional news slots on Channel 4, Channel 5 and Sky are added in, I expect the figure would exceed £250 000. Not bad to start off with!”w23

The only significant difference between sponsorship and advertising that the internal documents acknowledge is that the sales pitch in sponsorship is more hidden, enabling covert or “subliminal” messages that can get round the defences of their “wary” and media literate young targets: “at each event the level of Silk Cut branding is intended to be subliminal, with no direct reference to Silk Cut cigarettes”w24 and “they are very advertising literate, and consequently very wary of big brands latching on to aspects of their lifestyle and exploiting them. To this end Silk Cut needs to compliment the Renaissance imagery in an intriguing and stylish way.”w25

What other forms of promotion are there, and what part do they play?

Even though the internal documents come from advertising agencies—and therefore would be expected to place particular emphasis on advertising—it is clear that all aspects of marketing play a crucial role in their efforts to increase sales. The documents reveal a textbook approach to the challenge. Firstly, the population as a whole is divided into smaller, more homogeneous groups. The viability of each group is then analysed to determine if they are suitable for targeting. Finally, customised marketing strategies are devised for those groups that hold most promise. Unlike textbook marketing, however, no reference is made anywhere to the ethics of targeting particular groups or of using particular strategies. This is of concern as the poor, women, and students all emerge as front runners. The poor can be reached by a combination of price offers, gift schemes, and reassuring branding to make inferior products seem better: “almost all would rather be smoking a premium brand, and all know (because it is obvious) that a cheaper product is an inferior product. Thus, anything which implies quality is gratefully received.”w26 Women might be susceptible to their own brand: “opportunity possibly exists for overtly female targeted cigarettes (perhaps tapping into female cigar smoking trend?).”w27 Students appeal because they are image conscious, young, and readily accessed through student unions and college campuses.w5

Having identified their target market, manufacturers use all forms of communication, not just advertising, to approach them. Point of sale promotions, databases (one company claims to have over seven million namesw28), and the internet are all mentioned.w29 But perhaps the most important channel is the pack itself. Its value as both a communication tool and a “badge” is readily acknowledged and great care is taken to ensure that it continues to communicate the correct image (box ​B6).

: Importance of the cigarette pack to the marketing of cigarettes

Furthermore, efforts to meet the needs of the smoker and potential smoker do not stop at communication. Pricing, distribution, and, most fundamental of all, the product design itself all do their bit. Cheap products are offered to low income smokers, and the impact of price on brand perception is recognised.w26

Effective distribution delivers convenience, and good relationships with retailers ensure that a company's brands get maximum support: “young adult smokers find it much more important than the smoking population as a whole that things are easy to do, or buy. They want (and feel entitled to get) what they want, where they want it, when they want it”w2 and “the global objective of this research is to aid the optimisation of these communications, as a step towards maximising sell through the independent sector.”w33

And development ideas for new products ensure that the product will evolve and so continue to provide appropriate aspirational and pharmacological benefits. Ideas include an “Expresso” cigarette to fit the new “café culture,” a “macho” cigarette (complete with “a scantily clad woman on the paper”), and Scottish and Welsh cigarettes to exploit devolution (box ​B7).

: Ideas for new tobacco products(quoted from Gallagher)

That some of these suggestions may not make it to market—or indeed may be technically impossible—is beside the point. The key issue is that product design, along with all the other marketing tactics mentioned in the internal documents, is just one more opportunity to increase sales.

Conclusion

This first significant glimpse inside UK tobacco advertising highlights three things. Firstly, it confirms the key conclusions of the past 20 years of tobacco control research. This has painstakingly established that tobacco advertising can increase consumption as well as brand share, that young people are a key potential target who are particularly susceptible to psychosocial appeals, that sponsorship works in exactly the same way as advertising (only with greater subtlety), and that advertising is only a small fragment of marketing.

The internal documents show that, despite public pronouncements to the contrary, the industry has clearly been working on the assumption that all these findings are correct. Furthermore, it is deliberately exploiting the opportunities that result: market growth is a goal, brand images secure the young, the subliminal qualities of sponsorship are welcomed, and the pack (a marketing tool largely overlooked by regulators) is exploited to the full. Secondly, and perhaps more worrying than these specific practices, is the overall picture that emerges of an industry that is doing everything it can to encourage smoking. The commercial imperative is all, and the enthusiasm and competitive drive to meet its demands are palpable. Ethical doubts are never even acknowledged, and health consequences barely get a mention.

Finally, both these phenomena show the complete unworkability of voluntary agreements. In a competitive market, UK tobacco companies will always do their utmost to survive and succeed. Arguably, they have no choice—the commercial pressures are an imperative. The problem is that this imperative runs directly contrary to public health. The only solution is statutory regulation, with the overt aim of removing all tobacco marketing. This regulation has to be powerful, comprehensive, and flexible. Powerful to ensure that, in a profitable market, transgression does not pay; comprehensive to ensure that all facets of marketing are controlled, and flexible to ensure that innovations are identified and stopped. The Tobacco Regulations Authority that is recommended in the select committee report would provide a good vehicle for proposing and policing these statutory controls.

Supplementary Material

Acknowledgments

The Centre for Tobacco Control Research is core funded by the Cancer Research Campaign.

Footnotes

Competing interests: None declared.

References for the internal documents appear on the BMJ's website

References

1. Silverman D, editor. Interpreting qualitative data. London: Sage; 1995. [Google Scholar]

2. McGuiness T, Cowling K. Advertising and the aggregate demand for cigarettes. Eur Econ Rev. 1975;6:311–328. [Google Scholar]

3. Witt SP, Pass CL. The effects of health warnings and advertising on the demand for cigarettes. Scot J Pol Econ. 1981;28:86–91. [Google Scholar]

4. Radfar M. The effect of advertising on the total consumption of cigarettes in the UK. Eur Econ Rev. 1985;29:225–231. [Google Scholar]

5. Cox H, Smith R. Political approaches to smoking control: a comparative analysis. Appl Econ. 1984;16:569–582. [Google Scholar]

6. Laugensen M, Meads SC. Tobacco advertising restrictions, price, income and tobacco consumption in OECD countries, 1960-1986. Br J Addict. 1991;86:1343–1354. [PubMed] [Google Scholar]

7. Perkurinen M. The demand for tobacco products in Finland. Br J Addict. 1989;84:1183–1192. [PubMed] [Google Scholar]

8. Economics and Operational Research Division. Effect of tobacco advertising on tobacco consumption: a discussion document reviewing the evidence. London: Department of Health; 1992. [Google Scholar]

9. Thomas M, Walker A, Wilmot A, Bennett N Office for National Statistics. Living in Britain: results from the 1996 General Household Survey. London: Stationery Office; 1998. [Google Scholar]

10. Aitken PP, Leathar DS, O'Hagan FJ. Children's perceptions of advertisements for cigarettes. Soc Sci Med. 1985;21:785–797. [PubMed] [Google Scholar]

11. Aitken PP, Leathar DS, O'Hagan RJ, Squair SI. Children's awareness of cigarette advertisements and brand imagery. Br J Addict. 1987;82:615–622. [PubMed] [Google Scholar]

12. Charlton A. Children's advertisement awareness related to their views on smoking. Health Ed J. 1986;45:75–78. [Google Scholar]

13. Aitken PP, Eadie DR. Reinforcing effects of cigarette advertising on under-age smoking. Br J Addict. 1990;85:399–412. [PubMed] [Google Scholar]

14. Hastings GB, Ryan H, Teer P, MacKintosh AM. Cigarette advertising and children's smoking: why Reg was withdrawn. BMJ. 1994;309:933–937. [PMC free article] [PubMed] [Google Scholar]

15. Aitken PP, Eadie DR, Hastings GB, Haywood AJ. Predisposing effects of cigarette advertising on children's intentions to smoke when older. Br J Addict. 1991;86:383–390. [PubMed] [Google Scholar]

16. Pollay RW, Siddarth S, Siegal M, Haddix A, Merritt RK, Giovino GA, et al. The last straw? Cigarette advertising and realised market shares among youths and adults 1979-1993. J Marketing. 1996;60:1–16. [Google Scholar]

17. Arnett JJ, Terhanian G. Adolescents' responses to cigarette advertisements: links between exposure, liking and the appeal of smoking. Tobacco Control. 1998;7:129–133. [PMC free article] [PubMed] [Google Scholar]

18. Ledwith F. Does sports sponsorship act as advertising to children? Health Educ J. 1984;43:85–88. [Google Scholar]

19. Piepe AP, Charlton P, Morey J, Yerrell P. Does sponsored sport lead to smoking among children? Health Educ J. 1986;45:145–148. [Google Scholar]

20. Aitken PP, Leathar DS, Squair SI. Children's awareness of cigarette brand sponsorship of sports and games in the UK. Health Educ R. 1986;1:203–211. [Google Scholar]

21. Charlton A, While D, Kelly S. Boys' smoking and cigarette brand sponsored motor racing. Lancet. 1997;350:1474. [PubMed] [Google Scholar]

22. MacFadyen L, Hastings GB. Integrated marketing communications: a new paradigm for researching tobacco marketing and adolescent smoking. In: Tudor-Smith C, editor. Tackling tobacco. Cardiff: Health Promotion Wales; 1999. [Google Scholar]

23. Froese BD, Heone D, Lavack A, Vernon L, Madill JJ. The marketing of tobacco products. Project 502-0049. Health Canada, 15 May 1996.

24. Pierce JP, Choi WS, Gilpin EA, Farkas AJ, Berry CC. Tobacco industry promotion of cigarettes and adolescent smoking. JAMA. 1998;279:511–515. [PubMed] [Google Scholar]

25. Department of Health. The voluntary agreement on tobacco products: advertising and promotion and health warnings. London: Department of Health; 1995. [Google Scholar]

26. Barton J, Jarvis L. Smoking among secondary school children in 1996: Scotland. Office for National Statistics. London: Stationery Office; 1997. [Google Scholar]

Articles from The BMJ are provided here courtesy of BMJ Publishing Group