METHODS

In order to locate relevant industry documents, the document websites for Philip Morris, RJ Reynolds, Brown & Williamson, American Tobacco Company, Lorillard Tobacco Company, and British American Tobacco Company (Guildford Depository documents from the Minnesota trial accessed through the Centers for Disease Control and Prevention (CDC) website) as well as the Liggett & Myers documents, were accessed. A list of search terms was generated with which the term “pack*” could be linked. A full list of search terms, and the resulting number of documents returned on each website, are summarised in table 1. Multiple combinations of these search terms were used, so as to generate a practical number of documents for potential viewing on a particular subject (usually less than 50 documents from each combination of search terms). Searching was terminated when it was clear that no new documents were being returned from different combinations of search terms listed. Searching of the documents was completed in July 2001. All of the documents are available online at http://roswell.tobaccodocuments.org under the “packaging and pricing” collection.

Documents retrieved were dated from the 1950s to the mid 1990s. As table 1 shows, there were thousands of documents about packaging and packaging research. Many documents outlined plans for, and the results of, consumer research on pack design modifications. Some research was qualitative in nature, where a small number of smokers were asked to compare and appraise different packaging configurations, while others were quantitative, reporting data on smoker's responses to questionnaires about aspects of packaging. Some documents were reports and correspondence from consultant companies to the tobacco company, apparently in response to a project brief or request issued by the company. However, it was rare to find briefing correspondence from a tobacco company to one of these consultant firms—only one such document was found from Philip Morris to the Leo Burnett advertising agency in Australia.²¹ A relatively small number of documents were speeches and presentations delivered at company meetings or conferences. Multiple copies of the same documents were returned in many search combinations and a minority of illegible documents were found.

RESULTS

The process of package design and modification is neither rapid nor simple. Typically, the process takes months to several years. In their book on package design, Myers and Lubliner²² explain that the process commences with a thorough analysis of the existing product category, which involves gaining an overview of how a range of cigarettes and other products are packaged for a particular target group. Second, an analysis of existing packaging and those of competitors is generally undertaken. This involves asking consumers to rate attributes of existing packs with those of competitors. Third, clear brand positioning objectives are established to communicate the brand's “personality” and this includes looking at broader advertising and sales promotion plans for the brand. Fourth, identification of the product's attributes needs to occur to help prioritise text and visual elements of the intended communication elements on the pack. Fifth, precisely targeted package design criteria are established, including recognition, image communication, and technical requirements. All of this research and decision making is pulled together to brief the package designer, who may either work within the company or within an advertising agency (for example, Philip Morris has used the Leo Burnett advertising agency).

From this information, a broad range of brand identity and package design explorations are developed, from which a smaller range are selected for further development and consumer feedback. Feedback from consumers generally occurs through focus groups or one-on-one interviews, and this research shapes decision making about the final concepts for further development. Three dimensional mock-ups are generally made of these designs for additional consumer feedback, after which full working models and specifications would be made for the artwork, production, and printing process. New packs may be subjected to a full scale test market, where sufficient quantities of the new packs filled with cigarettes are delivered to retailers, and records are kept of sales of the new packs, with due attention being paid as to whether the new packs impact unfavourably on existing brands owned by the company. This process usually involves a lengthy period of time—several months to a year or more—to observe and monitor the movement of test packs. Documents pertaining to all of these stages of packaging research and design were found in tobacco company websites.

Role of packaging in tobacco marketing

Tobacco companies have long recognised the importance of packaging in complimenting and extending the imagery created by advertising. As early as the 1950s, some US$150 000 was spent on packaging research by Philip Morris,²³ equivalent to around $1 million in today's terms. To facilitate Marlboro's “repositioning” from a woman's cigarette (“Mild as May”) to a man's smoke in the 1950s, “more than 120 different additional superogatory designs were created, rendered and researched. They were tested for eye movement, for associated characteristics, for emotional impact, for every attribute within the power of Vienna to define or invent”.²⁴ (Here, Vienna is a reference to Freudian psychiatry.)

The influence of packaging is so great as to persist even when smokers are trained and practised in distinguishing between different brands of cigarettes when they are smoked. In 1980, British American Tobacco (BAT) investigated the influence of brand identification and imagery on subjective evaluation of cigarettes. The study exposed a panel of smokers to a control condition where the brand identification was masked and packs were absent, a condition where brand identification markings were visible on the cigarette but packs were absent, or a condition where cigarettes were contained in the pack from which they would normally be taken. The study found “ . . .even with the use of panelists who are trained to be objective in their evaluation of cigarettes, that both brand identification and pack imagery variables have a significant effect on the individual's perception of the sensory attributes of the product.”²⁵ The concept of “sensation transfer” from the pack to the product, sometimes called the “halo effect” of packaging, is an important phenomenon. As will be later illustrated, it is a critical factor in creating the impression of lower tar cigarettes.

Packaging is also particularly important in promoting the trial of a new cigarette brand and this often occurs at the point of purchase. As explained in a report commissioned by the Liggett and Myers company: “the primary job of the package is to create a desire to purchase and try. To do this, it must look new and different enough to attract the attention of the consumer. Repeat sales will depend mostly on acceptance of the product, although packaging features such as convenience in use and protection certainly play an important role.”²⁶

As advertising restrictions increase, the cigarette pack becomes ever more important as a means of communicating brand imagery. Where usual advertising channels are available, packaging works in concert with advertising imagery, complimenting the images generated by print or electronic advertising. However, in the face of advertising bans, documents show that tobacco companies realised that the pack would have to assume a much more prominent role in communicating differences between cigarettes and, particularly, brand imagery. In countries where advertising bans are comprehensive to the extent that point of purchase offers the only possibility for advertising, packaging is vital.

For example, in 1980, the senior vice president of marketing at BAT noted the dire importance of packaging where comprehensive bans on advertising were in place, as in Finland and Singapore, stating his belief that packaging would ultimately become the strongest communication device. “In a future where increasingly the product may have to sell itself through the pack, a fuller understanding of the way in which perception of such packs affects perception of their contents is desirable.”²⁵

In a speech given as part a Philip Morris international marketing meeting in Tampa, Florida in 1990, participants heard from the product innovation group in the European region that, since restrictions on conventional advertising media were forecast to increase, “our new product development program must therefore focus on those areas of opportunity which do not rely on conventional media.”²⁷ This included “new types and forms of packaging that can act as a means of communication” and also sponsorship involving “using a cigarette brand to sponsor cultural/social events so that the events together with the brand packaging become the means of communications”. The speaker went on to describe a proposed oval pack, which “projects a distinctive young masculine appearance” where concept study testing showed it to be “new, original, sensual, and striking. Test concluded: pack has tremendous appeal among young smokers.” A booklet pack and a pack shaped like a cigarette lighter were also featured. “As a conclusion, I think we have demonstrated that new cigarette products and brands have a future, even though our enemies want us to stop innovating the market. We shall overcome.”²⁷

In a highly competitive market, it is important that consumers can identify their preferred brand from among the range of brands that are available. As one BAT document explains, this is especially the case at the point of purchase. “..it is felt that given the consequences of a total ban on advertising, a pack should be designed to give the product visual impact as well as brand imagery . . .The pack itself can be designed so that it achieves more visual impact in the point of sale environment than its competitors.”²⁸ Thus, cigarette packs that are easy to identify are ideal advertisements in and of themselves, and can be designed to achieve optimum salience when arranged on retail shelves.

For some years, companies have been redesigning packs so that brand families are sufficiently similar to indicate membership of the overall parent brand, and different enough to be able to distinguish between the variants. This is well reflected in an American Tobacco Company memo which advised that “An integrated package design can provide for a greater in-store presence. One package or carton color with integrated bull-eye design can better compete as a `family' in today's cluttered in-store environment.”²⁹ In this way, the arrangement of packs at the point of purchase themselves become an advertisement for the brand family.

The design of Philip Morris' Marlboro pack represents one of the most successful and most recognisable designs in cigarette packaging. Pack recognition research used by tobacco companies involved the use of a tachistoscope containing a visual field onto which stimuli such as packs were exposed for precisely measured durations under controlled illuminations. Using this instrument, each pack was presented to the consumer in a series of short regularly spaced exposures starting at subliminal levels and proceeding in sub-millisecond increments through the stages of partial recognition of information on the pack, to the stage of full recognition. The test then measured the mean recognition time for each pack.²⁸ One BAT document describing results of pack testing using this procedure on BAT brands lamented the fact that “the performance of all the packs in the exercise is substantially outstripped by the Marlboro pack which delivers a 3.16 millisecond threshold.” This time to recognition was half the time achieved by any of the BAT brands. It goes on to say: “Consistent trends indicate that chevrons have high image prominence but with the tendency for upwards pointing chevrons to draw attention to themselves and downward pointing chevrons to draw attention to that at which they point at [sic].”³⁰

A document from the Philip Morris site in 1970 which summarised the Marlboro packaging story explained how pack recognition research was conducted by using hidden cameras on unsuspecting consumers in retail stores. “As a prospective customer entered this area, he broke the photoelectric beam immediately setting in motion the concealed camera which recorded the movement of his eyes and the final selection he made. From the thousands of films, the designers, color experts and advertising men were able to determine which package, which design and what combination of colors had the most appeal and “purchase incentive—pick me up” for the consumer.”³¹

The importance of pack graphics in recognisability is further underlined in a Leo Burnett office memo presenting the results of a pack recognition study, the prime purpose of which was to determine whether Philip Morris' Merit pack with its logo had sufficient recognition to stand alone, to the extent that the name “Merit” was no longer required. Packs which had had the brand name masked were presented one at a time to “a general cross-section of smokers”. The study found that “the older, larger share brands gravitate to the top of the list, with Marlboro and Winston leading the charge . . .Marlboro Lights was the youngest brand in the top tier, obviously buoyed by the “roof” logo, which continues to attain extraordinary recognition whether in red or gold.”³²

Because smokers carry packs around with them, taking them out many times each day to complete the ritual of lighting up, and often leaving them in an exposed and within easy reach location, the pack itself, through its association with brand image, can communicate a lot about the person who smokes. Tobacco companies undertake a considerable amount of assessment before modifying pack designs, since it is so important that smokers feel comfortable carrying the pack around and uninhibited to take it out to light up. In a memo recommending the commencement of exploratory research for the Lucky Strike cigarette line, advice to the vice-president of brand management of the American Tobacco Company was that “focus groups have indicated that the green (Lucky Strike) pack would be well received by young adult males . . .they thought it was eye-catching and a different approach to packaging . . .specifically they stated, it would be great to put down on the bar.”³³

Thus, these documents reveal that packaging plays an important role in creating and reinforcing brand imagery. In the case of advertising restrictions, pack design assumes greater importance in driving brand imagery, and plays a key role in competing for potential consumer attention at the point of purchase.

Pack design effects on perception of cigarette strength and taste

With concerns about the health risks of smoking being aired in the 1950s, tobacco companies began to introduce filter cigarettes. In the 1960s and 1970s, this concern gained momentum with the publication of the first Surgeon General's report and broadcasting of the first televised anti-smoking advertising, prompting the introduction of cigarettes that were claimed to be lower in tar, or “milder”. Research has demonstrated that these “lower tar” cigarettes did not in fact deliver lower tar when smoked, due to vent blocking and compensation through deeper inhalation.³⁴ Further, research indicates that the use of terms such as “low tar” and “light” cigarettes were purposefully designed to mislead the consumer into believing that they were smoking a lower tar or safer product.^{10, 35}

Aside from the brand name and descriptors, however, tobacco industry documents show that other aspects of pack design were purposefully used to communicate and reinforce the impression of lower tar or milder cigarettes. A wide range of colours and design configurations were market tested with consumers to determine designs that most led consumers to perceive that the cigarettes in the pack were “mild” or lower in strength. A common thread through all of this research has tobacco companies grappling with the difficulty of preserving taste satisfaction, while simultaneously reducing perceptions of cigarette strength.

Companies discovered that lighter colours on the pack appeared to promote perceptions of lower cigarette strength. For example, throughout 1979 and 1980, RJ Reynolds was testing packaging modifications to its Camel Filter brand, in an attempt to determine a design which was “able to reduce the communication of product strength among potential users of the brand while maintaining current satisfaction, taste and image perceptions”.³⁶ The tests led to packaging changes which were summarised as follows: “refinements in the package consist mainly of increasing the amount of white space on the pack and lightening the brown color tones. While other changes were made these were essential to give the revised package the appearance of reduced strength.”³⁶ Also, the colour of the word “Camel” on the pack was made a lighter brown.

The colour blue is often associated with low tar cigarettes. For example, results of a mall intercept survey of smokers' responses to new packaging for Marlboro Ultra Lights found that both Marlboro and other brand smokers “felt that both the red and blue Marlboro Ultra Lights packs were lower in tar and milder in taste than the majority of other brands, although the blue pack was reported to be somewhat lower in tar and milder in taste than the red pack”. ³⁷ In a test later that year, Philip Morris marketing research department compared smokers' responses to cigarettes packaged in a blue and a red pack. Despite the cigarettes being identical in composition, smokers appraised the cigarettes in the blue pack as “too mild”, “not easy drawing”, and “burn too fast”. Others felt the cigarettes in the red pack were “too strong” and “harsher”. This example is one of hundreds of routinely used pack tests, which employ the principle of “sensation transfer” from pack to product to influence consumer's perceptions of the smoking experience of cigarettes.³⁸ “Lower delivery products tend to be featured in blue packs. Indeed, as one moves down the delivery sector then the closer to white a pack tends to become. This is because white is generally held to convey a clean healthy association.”²⁷

Soft packs were understood by tobacco companies to create the perception of stronger cigarettes than hard packs, as evidenced in this Philip Morris memo: “The absolute ratings of both cigarettes, in terms of mildness, may be higher (that is, the cigarettes may be perceived as less intense) when the cigarettes are presented in the box.”³⁹

Menthol cigarettes were first marketed as being good for a cold or flu, as health concerns grew over cigarette smoking. Menthol originates from the peppermint plant and packaging reflected concepts of freshness and mint flavour. Green colours in menthol packaging were predominantly used to influence expectations of menthol taste.^40–43 As one of the first brands to promote menthol, RJ Reynolds' Salem brand was promoted as being “as refreshing as all outdoors”.⁴⁴

Pack designs for target groups

Basic marketing principles advocate the use of market segmentation.^{1, 45} Segmentation analysis recognises that potential consumers have diverse needs and wants, and that by carefully identifying and characterising these subgroups, product marketing can be tailored to elicit a more positive consumer response. Documents show that all of the tobacco companies use segmentation analysis as a strategy for assisting with decision making about pack design.

Packaging to appeal to young adults

Documents repeatedly show that tobacco companies are aware that brand choices are made relatively early in the life of a smoker and that packaging is an important ingredient in positioning brands to be attractive to youth. A Liggett and Myers document, for example, stated: “16–21—the formative years; smoking starts and brand preferences are developed.”²⁵

Tobacco companies constantly check with consumers to see whether cigarette packaging is perceived as old or outdated. This is especially important, when the age skew of the consumer is young, where fashions come and go so quickly. A market research presentation for Philip Morris notes the need for `new' packaging to maintain appeal for young smokers. “Once exposed to `innovative', especially young adults see their current packaging as dated and boring”.⁴⁶ This document goes on to say that the value of “new” is that it “draws attention (jealousy) from others” and that “especially young adult consumers are ready for change in packaging”.⁴⁶ Another report for Philip Morris mused that “ . . .we hear from consumers that RJR's foil wrap provides an example of something new, shiny, a little more contemporary to look at than other brands provide may be enough to tempt new smokers to try (sic).”⁴⁷

This concern with the importance of novelty for the young surfaces again in a Philip Morris report from its Research Center in Richmond, Virginia. The Center conducted a series of nine packaging preference studies up to June 1995 whereby smokers ranked a variety of new pack designs, with the final study recommending that “a plastic pack has particular appeal among the young, fitting their lifestyle of sports, outdoor activities and being `on the go'. Based on this information, we recommend that a plastic pack be pursued in future testing.”⁴⁸

In the 1970s young smokers were targeted with Lorillard's Zack cigarettes in a denim package. In a Lorillard memo proposing tracking of consumer attitudes towards Zack cigarettes in August 1974, Walter Lancaster of Lorillard explained: “This new lo-fi entry is targeted at young smokers under 30 years old . . .Marlboro smokers are the brands prime target because of the young age skew of the brand . . .Zack's strength in appealing to young adults is its unusual name, denim pack and graphic entity.”⁴⁹ In focus group discussions the following year for Lorillard, a marketing executive from the Will Graham Company reporting back from focus group discussions concluded that “Zack is a perceived as a brand for open-minded free thinking people who are young in spirit, if not in fact”. This point was reinforced by an illustrating that “one teacher said Zack seems to be replacing Marlboro as the pack her students carry”.⁵⁰

A 1973 memo which proposed a possible “Turk” blend for Camel Filter brand designed to compete with Marlboro suggested that the company “continue to explore and evaluate the need for a dramatically redesigned package that would significantly increase the brand's young adult image/appeal”.⁵¹ In market testing of alternative pack designs for Camel Lights, a market research report for the RJ Reynolds company asserted that “the alternative packs seem to `fit' better with present advertising and this advertising was, in turn, very consistent with `role model' and imagery desires of the younger smokers.”⁵²

A 1979 memo from Edward Ricci at MCA Graham Advertising to Tom Mau of Lorillard discussed the positioning of the Newport box package in the low tar market, explaining that “the Box is on target today in terms of drawing LT (low tar) prone Newport smokers and, more so long term, for attracting a greater percentage of young adult new entrants”.⁵³ Some years later, as the Newport brand was increasing its market share, Philip Morris drew the same conclusion about the appeal of box packaging for younger low tar consumers as a strategy to win back young smokers. “A brand to compete with Newport targeted toward young smokers may have stronger impact on this market if packed in a box”.⁵⁴

A qualitative investigation of Old Gold Filters by Unisearch for Lorillard in 1978, “reaction to the prototype package appeared to be the following: class, expensive, elegant” but then reflected “Biggest problem: Is this a young adult package?”.⁵⁵

Packaging for women

Packaging to appeal to women has been the subject of careful research. Cigarettes for women are often packaged in slim, long packs, often with pastel or toned down colours, to meet perceived desires to appear feminine and sophisticated. “Some women admit they buy Virginia Slims, Benson and Hedges etc. when they go out at night, to complement a desire to look more feminine and stylish”. ⁴⁷ This report went on to say: “Throughout all our packaging qualitative research, we continue to validate that women are particularly involved with the aesthetics of packaging . . .we sense that women are a primary target for our innovative packaging task, and that more fashionable feminine packaging can enhance the relevance of some of our brands.”⁴⁷

In 1992, Philip Morris was exploring different ideas for packaging of Virginia Slims Kings. In describing these plans, a Philip Morris memo indicated that “recent qualitative research indicates that female smokers, particularly young adults, show great interest in new packaging designs. In fact, this group has indicated a desire for packaging revolution, rather than evolution”.⁵⁶ In a two day workshop, Leo Burnett's Idea Lab generated more than 60 new packaging ideas for Philip Morris, including—believe it or not—headbands, belts, and shoe attachments containing cigarettes. Ultimately the company conducted further research on three of these—a “Booklet” pack (opening like a booklet with three packs that could be separated) a “ComPack” (designed to replicate a women's powder compact case), and an Oval Pack (a soft pack that could be squeezed at each end to access cigarettes).

Philip Morris also deliberated at length about revising the Alpine pack, which is a predominantly female brand. In March, 1994, Colmar Brunton conducted focus groups in Melbourne Australia for Philip Morris in order to “revitalise and strengthen the quality appeal of the Alpine brand in the 90's”. The company's brief was a to explore what kind of communication and packaging concepts would most appeal to women. It wrestled with the issue of “how to grab her attention amid the clutter?”.⁵⁷ Ultimately, the company was very pleased with its new design, as shown in fig 1.

In a 1961 report on packaging conducted by the Opinion Research Corporation for Philip Morris, the researchers concluded that “there seems to be some evidence that packaging preference is sex-linked, with the hard pack identified as best for women . . .the cleanliness of the pack is its greatest attraction for the female smoker”.⁵⁸

DISCUSSION

It is well known that the tobacco industry is highly flexible and responsive in accommodating restrictions to its ability to advertise and promote cigarettes. Previous research has shown that the industry is able to compensate for its inability to advertise in one medium by transferring advertising dollars to other marketing activities.^59–61 A recent analysis of data on 22 OECD (Organization for Economic Cooperation and Development) countries from 1970 to 1992 concluded that comprehensive bans on advertising/promotion significantly reduce smoking, while limited bans have little or no effect, because of the potential for shifting resources from banned activities to other marketing efforts.⁶² Given the progress made in many countries with tobacco advertising restrictions, cigarette packaging is becoming more important as a component of tobacco marketing. The pack, and its associated point of purchase display, is increasingly becoming a vital avenue for promoting brand imagery and advertising. Research using observational methodology in the USA, for example, has begun to empirically demonstrate that point of purchase marketing strategies are far from a randomly occurring phenomenon, but are part of a sophisticated marketing strategy designed to offset the potential beneficial effect of tax increases and tobacco control programmes and policies on tobacco consumption.^63–65

Tobacco trade and advertising magazine commentaries confirm these empirical findings. As Marc Cohen from Goldman Sachs commented: “anything that a [cigarette] company can do to stand out behind the counter is important, due to big tobacco's agreement with the states that limits advertising.”⁶⁶ The forecast for packaging, according to an April 2001 interview in a tobacco industry journal, is for even greater innovation. “Good cigarette packaging offers an additional dimension that goes beyond the mere wrapping of the product and also beyond the simple consumption of cigarettes—some kind of added value. However, I think that with regard to looming restrictions, the strategy should be completely different: there should be an image transfer for cigarette consumption as such, similar to the image change that was achieved for cigars some years ago . . .Packs with a valuable appearance may contribute to this image shift.”⁶⁷

These documents indicate that tobacco companies view packaging as an important element in cigarette marketing and have conducted an enormous amount of consumer testing and research on all elements of pack design in order to be sure about the imagery created and reinforced by each cigarette pack. Thus, tobacco companies cannot reasonably claim that they are unaware of consumer perceptions associated with particular packaging features, or that these were unintentional. If packs are effectively acting as advertisements for cigarettes, if their design characteristics make them more attractive to teenage smokers, communicate information about cigarettes that may be misleading (such as implying they are less strong or milder in some way), or minimise the salience of health warnings and contents information,^{12, 17} then pack design ought to be subject to regulation.

Comprehensive bans on tobacco advertising therefore need to include some consideration of how pack design contributes to the size and salience of point of purchase displays, as suggested by “best practice” point of purchase regulation in Tasmania, Australia.⁶⁸ The Australian experience has shown that where traditional advertising is banned at the point of sale, cigarette packs themselves can be used to make “pyramids, mechanical windmills, entire walls of display stock, designs and patterns on walls, ceilings, floors and anything else a creative advertising agency can dream up”.⁶⁸ A comprehensive ban on advertising would mean that cigarettes need to be available only from under the counter, rather than being on display. On the other hand, the unappealing images now used on cigarette packs in Canada to warn consumers of health risks²⁰ effectively represent a form of anti-smoking advertising and may lessen the need to withdraw cigarette packs from plain view in retail outlets.

Given this trend, regulation of cigarette packaging should be considered. One obvious policy option would be to require all brands to be packaged in a generic package. Without brand imagery, packs simply become functional containers for cigarettes, rather than a medium for advertising. Reports from Canada and Australia have commented upon how generic packaging, which removes brand logos from packs, increases the prominence of health warnings and may increase the believability of health warnings.^{12, 17} It is also important that consumers not be still further misled into thinking that the cigarettes in the pack are in some way “safer” than others. There may be a case to be made for requiring tobacco companies to demonstrate that cigarettes that deliver the same amounts of tar and nicotine when smoked ought to be packaged similarly. These kinds of issues are raised for regulatory consideration in a briefing paper released by the Framework Convention Alliance as part of the Framework Convention on Tobacco Control.⁶⁹ Public health research on adolescent's pack recognition, and ratings of pack attractiveness to youth should be conducted to assist policy makers to deliberate further on these policy options. Regulatory attention also needs to be directed toward banning the use of easily recognisable, pack derived brand logos without brand names, as has been used in sponsorships, including Formula 1 Grand Prix.

There have been some successful efforts by public health authorities to alert consumers to the fact that so-called “light” and “mild” cigarettes neither deliver lower tar when smoked, nor confer reduced health risks.^{70, 71} These kinds of educational initiatives might also point out how tobacco companies carefully design packs to give the appearance of cigarettes lower in strength, through using the sensation transfer phenomenon as a routine part of their product development research. Youth oriented education and advocacy approaches that have sought to publicise tobacco industry marketing approaches might usefully focus upon the way in which tobacco companies use packaging to entice young consumers to their brand.

This paper has focused upon particular aspects of pack design that contribute to brand imagery. However, there are other aspects of packaging that deserve future research. These include developing an understanding as to why cigarettes are packaged in different pack and carton sizes; exploring how price and packaging work in concert to drive brand selection, especially among low income consumers; studying the methods and results of tobacco company research designed to tailor packaging to appeal to potential consumers of different racial and ethnic groups; investigating whether any of the technical aspects of packaging, such as the substances and colourings added to the packaging material itself, might be harmful; and understanding how tobacco companies have worked to obscure required health warnings or contents labelling. Further research on all of these issues is warranted and would provide important evidence to guide regulatory initiatives.