Introduction Package quantity refers to the number of cigarettes or amount of other tobacco product in a package. Many countries restrict minimum cigarette package quantities to avoid low-cost packs that may lower barriers to youth smoking.
Methods We reviewed Truth Tobacco Industry Documents to understand tobacco companies’ rationales for introducing new package quantities, including companies’ expectations and research regarding how package quantity may influence consumer behaviour. A snowball sampling method (phase 1), a static search string (phase 2) and a follow-up snowball search (phase 3) identified 216 documents, mostly from the 1980s and 1990s, concerning cigarettes (200), roll-your-own tobacco (9), smokeless tobacco (6) and ‘smokeless cigarettes’ (1).
Results Companies introduced small and large packages to motivate brand-switching and continued use among current users when faced with low market share or threats such as tax-induced price increases or competitors’ use of price promotions. Companies developed and evaluated package quantities for specific brands and consumer segments. Large packages offered value-for-money and matched long-term, heavy users’ consumption rates. Small packages were cheaper, matched consumption rates of newer and lighter users, and increased products’ novelty, ease of carrying and perceived freshness. Some users also preferred small packages as a way to try to limit consumption or quit.
Conclusion Industry documents speculated about many potential effects of package quantity on appeal and use, depending on brand and consumer segment. The search was non-exhaustive, and we could not assess the quality of much of the research or other information on which the documents relied.
- packaging and labelling
- socioeconomic status
- tobacco industry documents
- advertising and promotion
Statistics from Altmetric.com
Tobacco package quantity refers to the number of cigarettes or cigars, the amount of smokeless tobacco (SLT) or the quantity of other product in a package. Local, state and national governments have passed laws limiting tobacco package quantities because of concerns that package quantity can influence tobacco use (21 CFR 1140.16, 1–5 p. 168, 6). In the USA, it is illegal to sell cigarette packages with fewer than 20 cigarettes, a restriction enacted in most states by the Master Settlement Agreement (1998–2001) and later by federal law (21 CFR 1140.16). A small but growing number of laws restrict minimum package quantities for non-cigarette tobacco products such as cigars (eg, 1–3), and few laws restrict maximum package quantities.
The rationale for restricting minimum tobacco package quantities is that small packages may lower barriers to use by children.2 5–7 Critics have used the term ‘kiddie pack’ to refer to cigarette packs with fewer than 20 sticks7 8 and roll-your-own (RYO) tobacco packages with less than 16.8 g of tobacco,9 arguing that such packages are designed for youth who are on a tight budget or must conceal the product from adults.10 Indeed, a 2006 study in Ireland found that youth smokers (aged 12–17) tended to buy cigarettes in packs of 10 (76%), whereas young adults (age 18–24) and older adults (age 25+) tended to buy cigarettes in packs of 20 (60% and 77%, respectively) (11, p. 24). A 1987 study also found that smokers aged 14 and 15 years in one Australian community were more likely than adult smokers to report purchasing cigarettes in a 15-pack (56% vs 9%, respectively).10
Aside from these suggestive results, little research has evaluated how package quantity influences tobacco product use or appeal. A recent systematic review of published research on the effects of tobacco package quantity, size and shape on consumption found only three eligible studies, all of which concerned cigarette length.12
Given the dearth of published information, we examined tobacco companies’ previously secret internal documents concerning tobacco package quantity. Millions of industry documents are publicly available in the Truth Tobacco Industry Documents (TTIDs).13 14 Few studies have examined documents on package quantity, and only in passing.15–17 We conducted a review of industry documents to understand tobacco companies’ rationales for changing or introducing new package quantities, including their expectations and research regarding how package quantity may influence consumer behaviour.
We searched the TTID in three phases (see table 1). In phase 1, we used a standard snowball sampling technique18 typical for industry document reviews (eg, 19–21). In phase 2, we searched a single string of terms. In phase 3, we conducted a follow-up snowball search of additional terms and strings based on documents found in phase 2. In each phase, we identified documents for full-text review if the title or first few pages contained information about research on package quantity or stated a reason for choosing or changing a package quantity; this is opposed to, for example, invoices simply listing products. We reviewed the full text of potentially relevant documents. The search was not limited to consumer research documents, nor was it limited to any particular tobacco product type or geographic location. We restricted the search to documents from 1980 or later to focus on more recent documents. For all relevant documents, we wrote a short summary and extracted information concerning reasons why a company may choose a particular package quantity for one of its products, and research on whether (or why) consumers may use particular package quantities. All authors participated in summarising documents; the first author read all document summaries and viewed all documents to identify emergent themes.
We identified 216 unique documents (see online supplementary table for descriptions) including research reports, memos and presentations (n=88); memos describing product ideas, marketing plans and marketing decision rationales (n=112); and miscellaneous documents including speech scripts, retailer packets, and trade magazine and news articles (n=16). Much of the consumer research was qualitative; documents describing quantitative research generally did not provide significance tests or effect sizes. Documents included information from the USA as well as Australia, Belgium, Canada, France, Germany, Japan, Korea, Malaysia, Paraguay and others. Most documents concerned cigarettes (n=200), with the remainder on RYO tobacco (n=9), SLT (n=6) and ‘smokeless cigarettes’ (n=1).1 Dates ranged from 1980 to 2009, with most from the 1980s (n=117) and 1990s (n=88), and the remainder from the 2000s (n=11).
Companies did not regard either small or large packages as universally superior. Rather, documents discussed the suitability of package quantity changes for specific brands, markets and consumer segments. For example, research evaluated consumers’ reasons for liking particular cigarette package quantities separately by brand22–25 and among smokers differing in attitudes and psychographic characteristics.26–34 For non-value brands, large packages could potentially detract from perceived quality.26 28 33 Cigarette brands with low proportions of ‘committed’ users were viewed as more price-sensitive and vulnerable to competitors’ use of package quantity changes to stimulate brand-switching.34 Smokers were classified into segments such as commodity smokers (‘price is everything’) and contemporaries (smokers concerned about health effects), with commodity smokers preferring large packages and contemporaries averse to them.29 Documents acknowledged that consumer characteristics and preferences change over time.29 35–38
Package quantity constraints
Companies were constrained in their ability and willingness to change package quantities. For cigarettes, some jurisdictions banned ‘sample-pack’ sizes,39 40 and tax laws created logistical challenges and price penalties for non-standard quantities.39 41–44 Package quantity changes sometimes required converting or buying new packing machinery.45–50 Before launching a product in a new package quantity, companies considered the risk that the product would cannibalise the market share of their other products.32 49 51 52 Package quantity changes could also provoke negative reactions from competitors such as price wars.35 38 53–56 Companies seemed to initiate quantity changes in markets where a brand had low or declining market share,35 37 52 57–62 or where competitors threatened to gain market share using novel package quantities or price cuts.51 53 55 62–77
Price, affordability and value
A key reason why companies changed package quantities was to influence consumer behaviour through effects on price.35 Smaller packages could be sold for lower per-package prices,23 32 36 41 42 49 51 53 54 57 58 60 61 63 66–71 74–76 78–132 and large packages provided per-unit value.25 29 36 38 42 45 50 52 53 55–59 61–65 73 83 90 91 117 119 122 128 133–182
Small, low-priced packages were aimed at stimulating purchase by users of competitor brands and maintaining the brand’s current user base.23 32 36 41 42 49 51 53 54 57 58 60 61 63 66–71 74–76 78–132 For example, one document recommended introducing a 10-pack of Raffles cigarettes to ‘facilitate trial, particularly by younger smokers’, and to avoid losing Raffles’ current smokers to competitors’ 10-pack offerings.49 In qualitative research, some smokers viewed a 10-pack as ‘an ‘introductory’ or ‘trial pack’ offer’ that ‘provides a ‘low price’ way to ‘experiment’ with a new brand’.81 Similarly, cigarette research in Belgium found that “St. Michel 20’s should be able to attract more new users than St. Michel 25’s because of the overall lower per pack price”.23 The 20-pack was also estimated to have a lower price elasticity than the 25-pack, meaning that 20-pack smokers would be less likely to quit smoking or switch brands if the price increased.23 For non-value brands, small packages were used to reduce prices without a discount that could ‘denigrate the brand image’.78 116 183
Large packages offering a value were similarly used as competitive tools to maintain and expand companies’ market share through effects on brand-switching. For instance, a mall intercept study and group interviews found high interest in a 25-pack of cigarettes as an ‘economical option’ that ‘should be cheaper per unit’.133 In some cases, documents described the potential effects of large-package value offerings on a country’s overall smoking rate and sales volume.38 50 62 Most notably, one cigarette company executive was asked about the effects of a new 25-pack on US cigarette consumption.38 He responded that cigarette prices had been rising, which ‘can increase the quit rate and also decrease the consumption of cigarettes per day’.38 By providing five extra cigarettes per package, he predicted that the 25-pack could ‘address the quitting issue to a degree and also address the rate per day by offering this value’.38 Another document similarly stated that the 25-pack could ‘stabilise the industry volume decline’ caused by cigarette price increases.50 This document compared the idea of a 25-pack to prior innovations such as filtered, ‘low tar’ and 100 mm cigarettes, which were said to have reversed industry volume declines caused by consumer awareness of smoking’s health effects and by cigarette tax increases.50 However, 25-pack value offerings ultimately captured little US market share, in part because they ‘ran counter to smokers’ stated objectives of wishing to cut down on smoking’.42 In contrast, the introduction of a 25-pack in Australia met high consumer demand and sparked a price war as competitors reacted with 30-pack, 35-pack and 40-pack offerings.62
Response to tax increases
Companies introduced new package quantities in response to excise tax increases—both to offset effects on their sales volume and to capitalise on consumers’ increased price consciousness and likelihood of switching brands.30 32 35 38 50 51 53 58 62 78 90 95 102 123 125 132 134 184–186 In one case, a 15-cigarette pack was introduced because ‘taxation had driven up the price… The 15’s pack gave new smokers an opportunity to try Export for under two dollars’.78 In anticipation of increased cigarette excise taxes, one company evaluated a RYO product in two tin sizes as alternatives to manufactured cigarettes for price-sensitive smokers.184 187 The company introduced a 30 g product into a marketplace dominated by 50 g pouches expecting that ‘growth will be generated by its attractive price point in a market characterised by increasing government taxes’.32 Other documents similarly described how package quantity changes had been used to increase cigarette brands’ market share following excise tax increases (eg, 134) or to retain consumers at risk of switching to value brands (eg, 185).
Matching current and desired consumption rates
Companies considered large and small packages as appealing to consumers by matching high25 31 42 45 83 138–144 188–195 or low68 72 82 84–89 108 110 111 128 137 188 194 196–201 daily use rates. In some cases, consumers’ daily use rates were said to differ from standard package quantities.24 31 33 190 198 202–204 In other cases, consumers were said to prefer non-standard package quantities that matched their desired use rate. Some smokers believed that larger packs would encourage them to smoke more, which they saw as a downside.27 28 31 45 117 144 162 180 For instance, in one study, a top reason for disliking a carton of 20 packs (19 cigarettes each) was that it ‘tempts you to smoke more’.27 Conversely, smokers believed that smaller packs would help them ration their consumption and smoke less, potentially benefiting their health.22 79 83 84 89 128 188 189 205–209
Several documents stated that, because of effects on consumption rates, marketing small packages may decrease overall industry sales volumes.71 94 122 208 A focus group report concluded that introducing a 12-pack of cigarettes may hurt volume because any new sales from brand-switching ‘would be more than offset by smokers who use the twelve pack to cut back’.208 Other documents similarly noted that ‘widespread availability of 10’s may result in slightly lower industry volume’ because ‘light smokers may consume less, given a convenient rationing unit’.71 122 One company hired analysts to forecast the effects of banning sales of cigarette packs with fewer than 20 sticks in Puerto Rico: The document predicted that ‘total market consumption will increase’.94 This was based on the company’s observation that ‘total market consumption increases with average pack size’, which they theorised was because “small pack sizes are purchased in part, by ‘Quitters’ and ‘Occasional’ users, who are using the small pack sizes as a self-discipline to reduce consumption or quit smoking altogether”.94 Another document also stated that market modelling had found a positive association between package quantity and sales volume.210 Relatedly, in Australia, large packs were credited with helping maintain industry sales volumes even as smoking prevalence fell.62
Inaccurate premarket predictions
In some cases, companies struggled to accurately predict the effects of package quantity changes on purchase behaviour.211–214 Research examined the effect of replacing 20-packs of cigarettes in vending machines with 12-packs, priced at per-cigarette parity.211 The 12-packs led to a 47% decline in the brand’s total share of cigarette sales in test-markets.211 Similarly, Skoal Long Cut suffered market losses after increasing its SLT can size from 1.0 to 1.2 ounces.212–214 The company’s research suggested that the change would not affect consumer behaviour.212 A document explained the mistake by noting that the research had examined consumption but neglected purchase patterns: ‘the study provided the cans to the consumers rather than requiring them to actually purchase the product’,212 leading to ‘a significant underestimation of what increasing the can size of Skoal by 20% would do’.214
Companies used consumer demographics such as age to understand the market for small packages.24 27 110 118 199 For example, documents noted that young adult smokers tended to smoke less heavily than older ones.24 27 Two documents contained data on smokers under age 18: German research collected data on heaviness of smoking and cigarette purchase behaviours among smokers in various age groups, including ages 16–24,27 and US research examined intentions to buy a 10-pack of KOOL cigarettes among Black smokers in various age groups, including ages 16–20.118 A Malaysia document noted that packs with fewer than 20 cigarettes ‘addressed an important consumer segment’ of ‘younger consumers or starters with lower consumption needs’.199 Package quantity surveillance in Puerto Rico found that 14-packs were especially likely to be used by young adults 18–24 years old.215 Research on ‘Project BUCK’—an effort to launch a cigarette 5-pack for $1.00 in Canada—found that the concept had especially high appeal among smokers who were young (aged 18–34), ‘starters’ or ‘switchers’.110 These groups’ higher interest was thought to be driven by low sensitivity to per-stick prices and low daily consumption rates.110 The launch of a cigarette 10-pack in Korea sought to generate trial among young adult smokers and ‘starters’.72
Race and ethnicity were also examined.24 216–219 One company considered introducing a 10-pack of ‘Inner-City Black Targeted Brand’ cigarettes because ‘Blacks smoke fewer cigarettes per day and have less money’216 and because ‘this smoker group is cost conscious, but rarely uses coupons or buys generic products’.217 A Black Smoker Study and a Brand Switcher study assessed whether Black smokers would buy KOOL 10-packs.218 Test-marketing of a Newport 10-pack found that sales were strongest in ‘economically depressed areas’ regardless of racial make-up.219
Smokers viewed small packages as unique.24 81 83 109 198 208 220 221 In focus groups, a cigarette 12-pack was seen as ‘cute/a novelty’ and ‘something which sparked enough curiosity to stimulate a trial’.208 A subsequent market test found that the 12-pack could be used as a promotional device for several months per year: ‘By doing this on a sporadic basis the novelty (the second most mentioned reason for buying) would not wear off’.222 The imagery associated with small packages was not always appealing223 224: Focus group research on cigarette 10-packs and 14-packs suggested that ‘modifications would need to be made to the package to mitigate both ‘feminine-looking’ and/or ‘sample size’ feelings’.224
Ease of carrying or concealing
Smaller packages were sometimes appealing because they made products easier to carry.30 83–85 88 89 128 137 189 193 224–226 A RYO pouch that was 30% smaller than others on the market was seen as ‘a convenient size for social occasions’.30 Qualitative research in Paraguay found that 10-packs of cigarettes were ‘easier to carry around’225; focus group participants stated that “‘youngsters’ (ages 10–13) like the ‘10’s’ pack very much because, since they ‘aren’t supposed to be smoking’, they are ‘easier to hide’”.225 Other qualitative research suggested this was also a reason for buying single cigarettes: ‘students claimed they preferred cigarettes in loose form, as they smoked on the sly’.207
Smaller packages may have a consumer benefit in terms of perceived or actual freshness.22 89 110 136 184 187 188 193 197 206 227–229 This applied to consumers who expected to finish the product in a shorter period of time.
Free sample packages
Small cigarette packs were used as free samples for three reasons.230–236 State tax laws created logistical and cost advantages to giving away free packs that were small rather than large.231–234 Marketers expected giveaways of small rather than standard packs to be less threatening to retailers231 and harder for retailers to misappropriate (ie, sell rather than give away).231 235 Finally, small packages were cheaper to give away. Companies studied the profitability of their sampling programmes in terms of generating competitive trial and conversion (ie, use by smokers of competitor brands), estimating the conversion rate required to offset sampling costs within a given period of time.230 235 237 A promotion test comparing cigarette 10-packs and 20-packs as free samples concluded that the 20-pack led more smokers to purchase the brand in the next week but did not lead to a significantly higher conversion rate 6 weeks later.230
Previously secret industry documents described the consumer appeal of small and large packages. Large packages were more economical (per-unit), matched the consumption rates of long-term, heavy users, and were targeted at ‘smart shoppers’ and value brand users. Small packages were more affordable in terms of immediate cash outlay (overall price); matched the consumption rates of newer and lighter users and the desired consumption rates of others; attracted interest through novelty, convenience and perceived freshness; and were targeted at people with low income, young people, minorities and premium brand users. Targeting package quantities using brand, psychographic, lifestyle and demographic characteristics led companies to develop highly differentiated product portfolios (cf, 238 239). Even so, companies’ premarket predictions about product uptake were sometimes highly inaccurate, with package quantity changes sometimes leading to large sales reductions.
As suggested previously,15 16 companies used novel package quantities to lower the price, either per-package or per-unit. Often, companies did this in response to excise tax increases. The stated goal of introducing new package quantities was to increase and defend brands’ market share through effects on brand-switching—and, at times, quitting—among increasingly price-conscious consumers. Package quantity changes aimed at enhancing affordability were sometimes targeted at people with low income, young people and minorities, which may harm public health if it discourages quitting (cf, 240–242). Prior research shows the importance of price in tobacco use, including among youth,240 241 243–248 and highlights industry efforts to promote value-for-money.249
We did not find documents in which companies explicitly discussed the introduction of small packages as a way to encourage trial by non-users, or by youth. Documents suggested that companies strove to motivate brand-switching among users of competitor products and to react to competitive threats and changing consumer preferences that could erode market share or sales volume. No documents specifically described targeting non-users of products, although some referred to ‘starters’ and ‘new users’. A prior review of packaging-related industry documents also found none describing the use of ‘kiddie packs’ to target minors.17 However, in another prior review, some documents appeared to use ‘young adult’ as a catch-all term for youth and young adults (15, pp. i7–i10), and so we cannot rule out the possibility that ‘young adults’ in some of our documents included youth. Also, we cannot rule out that companies’ marketing decisions would have the effect of encouraging trial among these populations. While documents did not explicitly discuss youth, they suggest several explanations for prior findings that youth prefer small packages.10 11 These include the traditional explanations of low price and ease of concealment2 5–7 ,10 as well as the notion that younger and newer users may simply prefer small packages because of the fit with their lower consumption rate.
Industry documents suggested that some current smokers preferred small cigarette packs as a tool to try to quit, cut down or maintain a desired consumption level. This accords with a prior study finding that many consumers—particularly those who see themselves as impulsive—prefer to purchase ‘vices’ in small quantities in order to try to constrain consumption.250 Research on US smokers’ theoretical willingness-to-pay for 10-packs also suggests a desire to ration consumption.251 However, it is also possible that the availability of small packs would tempt people to make purchases that they would not have made if only larger packs were available, as small packages may ‘fly under the radar’ of people’s self-control efforts.252 The role that these processes (consumption rationing, ‘flying under the radar’) may play in the purchase of small packages of tobacco products remains a topic for future research.
This was a non-exhaustive search of the TTID. We reviewed and described all relevant documents we found, but we limited our search to 1980 or later and reviewed a small number of documents (216). The documents may not be representative of other TTID documents describing package quantity. Different companies may use different terminology, which could influence the documents we retrieved. Moreover, document authors may have self-censored their language to obscure particular marketing objectives as a matter of company policy or to avoid legal liabilities (15, pp. i7–i10). All documents in this review concerned cigarettes, RYO, SLT and ‘smokeless cigarettes’. Most of the documents were old, with almost all from the 1980s and 1990s. Finally, we could not assess the veracity of research findings and many of the market predictions made in the documents.
Tobacco companies tailored package quantities to specific brands, consumer segments and markets, seeking to meet a complicated mix of preferences among current users. Documents also discussed starters, new users and younger users as potential target markets for small package quantities, primarily based on these individuals’ lower consumption rates and lower sensitivity to per-unit (vs per-package) prices. As an industry document review, our analysis should be viewed tentatively and not as the conclusive word on any aspect of tobacco package quantity. Given this, we encourage continued scientific investigation of the role that package quantity may play in purchase and consumption behaviour among current users and non-users of tobacco products.
What this paper adds
Little published research has evaluated the effects of marketing small or large packages on tobacco appeal, purchase or use.
We reviewed previously secret industry documents to understand companies’ rationales for changing or introducing new package quantities, as explained in companies’ internal records, such as marketing plans and research documents.
Package quantity changes were highly targeted by brand and consumer characteristics. Small and large packages were primarily marketed to reduce the per-package and per-unit prices of products, respectively, and to match consumers’ actual or desired consumption rates.
Documents described package quantity changes as a tool for increasing and defending a brand’s market share through effects on brand-switching, but such changes also had potential implications for quitting behaviour.
Contributors All three authors contributed by conceptualising the project; searching for, reviewing and summarising documents; and writing and revising the report.
Funding This work was not grant-supported but was funded via internal FDA resources.
Disclaimer This publication represents the views of the authors and does not represent FDA/CTP position or policy.
Competing interests None declared.
Patient consent Not required.
Provenance and peer review Not commissioned; externally peer reviewed.