Possible Changes in Tobacco Products: Considering Consumer and Supply Responses
So far this report has focused exclusively on illicit markets that are generated by tax evasion and avoidance. This focus reflects the fact that tax differentials among states (and on Indian reservations) increase incentive for tax evasion and tax avoidance and contribute to existing illicit tobacco markets in the United States. However, other kinds of restrictions can also generate illicit markets: in particular, regulations that restrict or ban features of some cigarette design, formulation, or packaging can generate illicit markets in what are presently legal products. As part of the charge to this committee, this chapter reviews the available research that could inform whether and how the demand and supply for illicit tobacco might be affected by nonprice regulations on legal tobacco products.
The chapter first briefly reviews the current policy and regulatory environment for tobacco in the United States, specifically, the authority and responsibilities of the Food and Drug Administration (FDA). The next two sections examine what is known about consumer behavior in response to changes to tobacco products, particularly in regard to possible regulatory requirements for cigarette design, formulation, or packaging. The chapter then considers the impact of possible regulatory changes on the supply of illicit tobacco and the role of electronic cigarettes (e-cigarettes) as an alternative to tobacco products. The final section of the chapter presents the committee’s recommendations for research to better understand the possible effects of product regulation on consumer behavior and illicit markets.
As discussed in Chapter 1, under the 2009 Family Smoking Prevention and Tobacco Control Act (FSPTCA), the FDA has regulatory authority for tobacco products through provisions designed to protect public health. A key feature of the act requires that new tobacco products do not have greater potential for initiating or maintaining dependence than existing products (see Henningfield et al., 2011). Such provisions may be used, for example, to implement product standards for nicotine levels or to place limits on mentholation. The FDA also has authority to set standards for tobacco pack messaging.
The new tobacco regulatory environment ushered in by the FSPTCA in the United States is likely to result in changes to the way in which some tobacco products are engineered, as well as the manner in which they are packaged and the way messages about them are communicated to consumers. Thus, a key question for the future is whether tobacco consumers are likely to engage in the illicit market if their preferred product characteristics are altered or eliminated by regulation.
That key question has not been directly addressed in the research to date. Although surveys and qualitative studies have examined the characteristics of users of illicit tobacco in comparison with users of licit products (see Chapter 3), this research has addressed only price-induced illicit markets and rarely considers the question of why only some consumers purchase illicit tobacco products. However, there is a body of research on the effects on consumer preferences and smoking behavior of variations in product composition, design, and packaging. This research provides useful knowledge on how tobacco consumers may respond when faced with changes in tobacco product characteristics. However, its nature makes it indicative rather than conclusive for the question of the emergence of illicit markets.
PRODUCT APPEAL AND CONSUMER RESPONSE
In this section, we first discuss product appeal—the complex array of factors that affect what consumers experience when they smoke. We then examine three avenues of regulatory changes to cigarette design and formulation: (1) regulation of product design features and constituents, (2) regulation of nicotine level, and (3) regulation of menthol. It is important to note that these explorations do not imply any forthcoming policy of the FDA: rather, they provide analysis of the potential impact of a limited number of different tobacco product regulatory approaches on consumer behavior on which some evidence is available. International experiences with restrictions on cigarette pack messaging (e.g., requirements for plain packaging or
large graphic warning labels) and research on consumer response to such messaging are also discussed.
The pervasive and enduring success of the tobacco industry over the past century is in large part a function of its capacity to create tobacco products, especially cigarettes, that appeal to consumers and promote addiction (Brandt, 2009). More than other tobacco products, such as pipes, smokeless tobacco, and dry snuff, cigarettes have the capacity to rapidly deliver a dose of nicotine that is within the optimal human physiological range for initiating and maintaining dependence (Ferris Wayne and Carpenter, 2009). In recent decades, tobacco manufacturers have used increasingly sophisticated technology to engineer cigarettes to deliver optimal doses of nicotine to consumers. Tobacco manufacturers have sought to enhance what the Institute of Medicine (2012) refers to as addictive potential (also known as abuse liability): the potential for tobacco products to initiate and maintain tobacco dependence.
By manipulating product features tobacco manufacturers optimize the speed and amount of nicotine dosing while providing additional appealing chemosensory characteristics (Ferris Wayne and Connolly, 2002; Carpenter et al., 2005, 2007; Ferris Wayne and Carpenter, 2009; Kreslake and Yerger, 2010). Those features include filter, ventilation, circumference, length, tobacco blend, processing, paper, and additives. Variations among cigarette brands also may involve the use of technological innovations to alter smoke chemistry, mechanisms of delivery and bioavailability of nicotine, to shape consumers’ sensory perceptions of smoking and smoking behavior (Connolly et al., 2000; Ferris Wayne and Connolly, 2002, 2004; Cook et al., 2003; Keithly et al., 2005).
Product appeal can be influenced by the use of additives, such as nicotine analogs and synergists, to modify nicotine effects (Bates et al., 1999; Ferris Wayne et al., 2004). Another important example is the modification of nicotine bioavailability by increasing the pH of the smoke in tobacco processing: this increase and other changes increased the proportion of “free” (unprotonated) nicotine (Hurt and Robertson, 1998; Henningfield et al., 2004). The general purpose of such product design modifications is to influence the chemosensory characteristics of tobacco products, to make them easier and more pleasant to consume. At the same time, the idiosyncratic chemosensory cues of a given product provide feedback for smokers about nicotine and tar strength, which gives smokers a way to tailor the amount of nicotine delivered to a preferred range by modifying the intensity of their puffing behavior (Rose, 2006; Centers for Disease Control and Prevention, 2008; Rees et al., 2012; Vansickel et al., 2012; Hoffman and Evans, 2013).
Manufacturers have optimized product appeal by targeting products with specific design features to subgroups of tobacco users (Ferris Wayne and Connolly, 2002; Carpenter et al., 2005; Kreslake et al., 2008b). Variations in the smoothness/harshness balance, nicotine effect, mouth feel, and draw resistance have been used to accommodate the preferences of subgroups of smokers, such as women and youth (Ferris Wayne and Connolly, 2002; Kreslake et al., 2008a). The rapid uptake of Camel cigarettes among youth was due not solely to marketing, but also to modifications in product design that affect sensory perceptions (Ferris Wayne and Connolly, 2002). Another example is mentholated cigarettes, which have cooling properties that decrease smoke irritation. Manufacturers determined that lower levels of menthol are better tolerated by youth, allowing for easier use by novice smokers and enhancing the potential for initiation (Bates et al., 1999; Connolly et al., 2000; Kreslake and Yerger, 2010).
Opportunities for tobacco product marketing are limited or banned in many countries, particularly those that ratified the Framework Convention on Tobacco Control (FCTC) of the World Health Organization. In the United States, tobacco product promotion on electronic broadcast media has been banned since the early 1970s, and public billboard advertising was banned under the Master Settlement Agreement of 1998 (see Chapter 1). With restrictions on many forms of mass marketing, tobacco manufacturers have placed increasing importance on the cigarette pack as a way to differentiate their products from competitors’ brands. Cigarette smokers are widely considered to be brand loyal (Alsop, 1989; Pollay, 2000; Carter, 2003), particularly in the United States, committed to a specific brand for years. While studies of brand loyalty link consumer selection more to satisfaction, taste, and price tier than to pack appearance (Cowie et al., 2013; Saeed et al., 2013; Dawes, 2014), the pack design is believed to convey and remind consumers of the product features they desire.
When a new regulation reduces product appeal, consumers can make several choices: quit using the product altogether, switch to another legally available product, switch to an illicit product, or continue using the modified product (see Figure 1-1 in Chapter 1). Any reduction in product appeal in response to regulated changes in product characteristics may in principle increase the illicit tobacco market to accommodate consumer demand for the original (unregulated) product. The rest of this section considers possible regulatory changes to features of tobacco products and reviews research relevant to consumer response and behavior. The available research on product changes is reviewed in the context of Figure 1-1 (in Chapter 1), to inform the implications for demand for illicit (unregulated) products. We note again that these scenarios are not meant to suggest or endorse any forthcoming regulations by FDA: they are offered as explorations in understanding factors that may affect the illicit tobacco market.
Regulation of Product Design Features and Constituents
In the current regulatory environment, product features that pose great public harm or promote initiation and dependence on tobacco are considered potential targets for regulation. Cigarette features that have recently been considered include ignition propensity, filter ventilation, characterizing flavors, and new designs that have the potential to reduce emission of toxicants.
A variety of technological innovations have been used to potentially reduce the emission of toxicants while maintaining the capacity of the product to deliver nicotine, including low- or non-combusted cigarettes and enhanced filtration systems. These “potential reduced exposure products”1 have been developed and marketed to appeal to health-conscious smokers. In general, the modification of cigarettes to reduce harmful smoke constituents has tended to reduce the chemosensory qualities that contribute to the appeal of the product. Studies of smokers’ reactions to potential reduced-exposure cigarettes have suggested that they have low appeal for consumers, who quickly discontinue use or use them infrequently, often in conjunction with their usual cigarette brand (Breland et al., 2006; Caraballo et al., 2006; Rees et al., 2008).
These findings suggest that when taste and other chemosensory characteristics of cigarettes are negatively affected, the appeal of the product is substantially diminished for some smokers. Other changes in product design, including modifications that lower cigarette ignition propensity, change filter ventilation, and remove characterizing flavors, have produced no more than modest reductions in product appeal in the aggregate. Although the direct impact of regulations on consumer behavior has not been widely assessed, the available evidence, as discussed below, suggests that regulations that have been implemented to reduce ignition propensity and ban cigarette flavors have not led to demand for unregulated products from illicit sources.
Reduced Ignition Propensity
One example of product regulation is the legally mandated requirement for ignition performance standards for cigarettes that has been introduced
1 Potential reduced-exposure products is a term of art used to describe modified tobacco products, cigarette-like products (whether or not they contain tobacco), and pharmaceutical products and medical devices (whether or not they contain nicotine) developed for their potential to reduce harms from tobacco. The term “potential” is used to avoid misinterpretation because whether exposure to tobacco toxicants is reduced depends on the users’ behavior, such as frequency and intensity of use: reduced exposure does not necessarily ensure reduced risk to the user or reduced harm to the population (Institute of Medicine, 2001).
in a number of U.S. and international jurisdictions in the past decade. This requirement represents one of the first regulations pertaining to the physical design of cigarettes. In the United States, the “New York Standard,” which requires that no more than 25 percent of 40 cigarettes burned under standard conditions demonstrate a full-length burn, has been widely adopted, and all 50 U.S. states now have measures to limit the ignition propensity of cigarettes (Hall, 2013). To meet this standard, most manufacturers introduced “banded” wrapping paper, which slows the rate of combustion by restricting the supply of oxygen to the burning ember (Connolly et al., 2005).
Initial opposition to the introduction of reduced ignition propensity cigarettes included complaints that the modified product would not meet consumer expectations and that taste and other characteristics would be compromised. Other concerns included the potential for increased exposure to tobacco toxicant because of the possibility that more intensive puffing would be required to keep the cigarette lit. Concerns were also raised that the change in product design would prompt disaffected consumers to seek supplies of cigarettes from neighboring states where an ignition regulation had not been adopted (Connolly et al., 2005).
Research has found these concerns were not warranted. Consumers in Massachusetts reported no change in taste or satisfaction for their usual brand, nor any change in obtaining cigarettes from a different source, after the introduction of the ignition law (Seidenberg et al., 2012). In a laboratory-based study, smokers showed no change in puffing intensity or exposure to carbon monoxide after switching to a reduced ignition propensity cigarette of the same brand. The overall number of cigarettes smoked even decreased among smokers who switched to the new product (O’Connor et al., 2010): this outcome suggests that a difference in performance was detectable to smokers and that some smokers decided to quit or reduce tobacco use. Other research has demonstrated that the consumer market has been largely unaffected by the introduction of cigarettes with reduced propensity to ignite (Connolly et al., 2005).
Cigarettes with higher filter ventilation (produced by placing concentric rows of tiny perforations in the filter tipping paper) allow fresh air to be drawn into the cigarette and mixed with the smoke to create a smoother, cooler sensory effect (Kozlowski and O’Connor, 2002; King and Borland, 2004). The reduction in harshness is generally associated with lower risk or harm from tobacco (National Cancer Institute, 2001; O’Connor et al., 2013; Elton-Marshall et al., 2014).
When the European Commission established maximal values for tar (10 mg), nicotine (1 mg), and carbon monoxide (10 mg) yields, manufactur-
ers met this standard by primarily increasing filter ventilation (O’Connor et al., 2006). There has been no research directly examining the impact of this standard on the illicit market. Increased filter ventilation does reduce machine yields and changes the smoking sensation; however, research has shown that smokers engage in compensatory smoking behaviors.2 Highly ventilated filters encourage consumers to puff more intensively, which produces similar tar and nicotine yields as a low-ventilation cigarette (National Cancer Institute, 2001; Kozlowski and O’Connor, 2002; Hammond et al., 2005). There have been suggestions to ban filter vents in combination with low maximum standards for tar, nicotine, and carbon monoxide yields in order to make cigarettes less appealing and encourage cessation (Kozlowski et al., 2006). No such bans have been initiated, but several countries have banned misleading descriptors that indicate low tar yields, such as “light” and “mild,” from packaging (see discussion on packaging, below).
The FSPTCA currently bans certain types of tobacco products, including a requirement under Section 907 of the act that cigarettes may not be sold with characterizing flavors, including candy flavors. Evidence shows that more than two-fifths of adolescent smokers prefer flavored tobacco (King et al., 2014) and that use of flavored products among youth may shape long-term tobacco use preferences (Villanti et al., 2013). These findings suggest that the appeal of flavored cigarettes may be sufficiently high among youth to promote switching to alternative products. However, there is no good source of data to determine whether an illicit market in flavored cigarettes has proliferated since the ban. There is some evidence that vendors circumvent the flavor ban through the sale of flavored little cigars, which are not subject to a flavor ban (see, e.g., Jo et al., 2015). Studies on other types of bans have also shown that alternative legal products may be adopted when a preferred product is banned, such as the adoption of smokeless tobacco products where the use of cigarettes is restricted (Klesges et al., 2010; Boyle et al., 2012).
2 Yields of tar, nicotine, and carbon monoxide can be measured on standardized smoking machines. In the past, the tobacco industry categorized cigarettes with low tar yields on standardized smoking machines (<15mg) as low yield and used descriptors such as light, low, or mild. Design changes have affected tar and nicotine measurements: different size and density filters, ventilation holes, chemical additives, and different types of tobacco. Many smokers modify their behaviors when smoking low-yield cigarettes to compensate and take in more nicotine than estimated by a smoking machine (National Cancer Institute, 2001).
Regulation of Nicotine
The FSPTCA allows FDA to implement product standards around reduction of nicotine, as long as the nicotine level is not reduced to zero. This provision necessitates determining, for public health benefits, the threshold for nicotine delivery below which initiation will not occur (Benowitz and Henningfield, 1994, 2013; Hatsukami et al., 2013). Regulations that require nicotine to be reduced to a dose threshold below which tobacco dependence cannot be initiated could foreseeably reduce the addictive potential of cigarettes, but it would also reduce the appeal of products and so it might cause consumers to seek alternatives (Benowitz and Henningfield, 1994, 2013; Hatsukami et al., 2013).
Several ongoing research initiatives are investigating the questions of how consumers respond to products with lowered nicotine: Would cigarette use be reduced by lowering nicotine? Will smokers change their smoking frequency and intensity? Will smokers quit? Will smokers adopt alternative, unregulated products or modify existing products to obtain preferred nicotine dosing? (Hatsukami et al., 2013). More definitive findings on consumer response to reduced levels of nicotine are anticipated from this research.
To date, laboratory research has shown that cigarettes without nicotine3 may continue to function as a reinforcer—that is, to maintain smoking behavior—because of the presence of other smoking-related sensory cues (Shahan et al., 1999). Indeed, both nicotine and non-nicotine cigarettes have demonstrated similar reinforcing capacity, although when directly compared, a nicotine cigarette was preferred by participants (Shahan et al., 2001). These findings were replicated when a monetary reward was introduced. That is, participants’ consumption of both nicotine and non-nicotine cigarettes was reduced by similar amounts when a financial incentive was provided to forgo smoking (Shahan et al., 2001). An earlier comparison of nicotine and non-nicotine cigarettes also showed that while smokers preferred the nicotine cigarette, there was no difference in its capacity to reduce craving or withdrawal in comparison with a non-nicotine cigarette (Pickworth et al., 1999).
Switching studies have been conducted using both acute designs (with one session) and sustained designs (for 2 weeks or more). Findings from studies of switches to low-nicotine cigarettes suggest that while very low-nicotine cigarettes support smoking behavior and have the capacity to ameliorate craving, reduction in nicotine levels in cigarettes to very low levels affects product appeal. Smokers who switched to very low-nicotine cigarettes tend to puff more intensively to extract adequate nicotine (Strasser
3 Technically, it is very low nicotine as it is not possible to eliminate nicotine entirely. It is appropriate to say non-nicotine because the actual level is negligible, but we use “very low” as technically accurate.
et al., 2007; Hatsukami et al., 2010). In a laboratory study with an acute design, Strasser et al. (2007) observed increases in puffing volume for low-machine-yield nicotine (0.3 mg/cigarette) and very low-nicotine (0.05 mg/cigarette) products in comparison with a regular one (0.6 mg/cigarette). This finding was partially replicated in another study, which found compensatory puffing with a 0.3-mg machine-yield-nicotine cigarette, but no compensation with a 0.05-mg product (Hatsukami et al., 2010). The direct effect of switching to a reduced-nicotine cigarette on product appeal was not reported in these studies. However, compensatory increases in puffing intensity might be taken to imply that smokers changed their smoking behavior to overcome a reduction in product appeal caused by lower-than-expected nicotine.
Another study examined the effects of progressively reducing cigarette nicotine content, from 12 mg to 1 mg per cigarette (equal to approximately 0.9- to 0.1-mg machine-yield-nicotine per cigarette), over a 4-week period (Benowitz et al., 2012). The study found that smokers continued to smoke a similar number of cigarettes per day, even after 6 months of smoking a 1-mg cigarette in comparison with control participants who continued to smoke their usual cigarettes. Nonetheless, nicotine intake (as measured by plasma cotinine concentration) declined as the cigarette nicotine content was reduced. The findings suggest that progressive nicotine reduction and use of very low-nicotine cigarettes can be sustained over an extended time period. In another sustained, 11-day switching study using very low-nicotine cigarettes, Donny and colleagues (2007) observed both a decrease in daily consumption and a decrease in motivation to smoke. Participants smoking very low nicotine cigarettes reported amelioration of craving following smoking, but maintained generally negative perceptions of the product throughout the study period. These findings were extended in a study that examined the role of nicotine reinforcement with a transdermal patch in the sustained use of very low-nicotine cigarettes (Donny and Jones, 2009). This study’s findings suggested that non-nicotine cigarettes were smoked less when transdermal nicotine was provided, and the cigarettes were perceived as having low positive effects and high negative effects (Donny and Jones, 2009). Other research has shown that sustained use of very low-nicotine cigarettes may support cessation. In a 6-week switching study, Hatsukami and colleagues (2010) found that a 0.05-mg cigarette produced a higher cessation rate than a 0.3-mg nicotine product, and the cessation rate was similar to that of a 4-mg nicotine lozenge.
Overall, the available evidence suggests that changes in cigarette nicotine content produce only modest changes in smoking behavior and that smokers are able to tolerate substantial reductions in nicotine. In the short term, very low-nicotine or non-nicotine cigarettes have similar reinforcing potential as conventional cigarettes. Over a sustained period, very
low-nicotine cigarettes have been shown to support smoking behavior for 6 months. However, the evidence also suggests that non-nicotine cigarettes have low appeal for smokers.
The next step for understanding the potential for illicit markets is to examine how consumers view the permanent loss of specific product features that they have previously found desirable and if that loss is enough to cause them to enter the illicit market. There is a need to go beyond the small array of studies showing that, notwithstanding smokers’ preference for cigarettes with nicotine, they also report that very low-nicotine cigarettes moderate their cravings and withdrawal symptoms and may support quitting. The constituents of cigarette smoke may provide sufficient secondary reinforcement to support smokers as they adjust to a low-nicotine product. These studies have shown short-term responses. The investigation of longer-term use of very low-nicotine cigarettes is needed to understand whether such use is sustained and what factors support prolonged use of low-nicotine products or cessation.
To better understand the role of nicotine in smoking behavior, research would have to assess the effects of nicotine reduction on measures of appeal and product choice. Research would also need to study how very low-nicotine products affect specific sensory features preferred by smokers and to what extent very low-nicotine cigarettes prompt smokers to either quit smoking or seek alternative products. The limited evidence to date suggests that cessation rather than sustained use of a very low-nicotine cigarette is more likely among smokers who plan to quit.
Regulation of Menthol
Menthol is an additive whose primary function is to alter the chemosensory qualities of smoking. It is responsible for producing a cooling sensation through activation of the trigeminal and other sensory nerves. Menthol modulates responses to physical stimuli, including temperature and irritation. Menthol also produces a characteristic “minty” taste, although the characteristics of menthol are different than mint flavor itself.
Mentholation of cigarettes has received attention as a product formulation feature that has the capacity to enhance the additive potential of tobacco products (Tobacco Products Scientific Advisory Committee, 2011). Mentholation has raised concerns for its appeal among subgroups of smokers targeted by cigarette manufacturers, including youth, African Americans, and women (Kreslake and Yerger, 2010). Among youth, menthol enhances initiation to smoking because it reduces the harshness of tobacco smoke, making inhalation easier for novice users (Kreslake et al., 2008b). Indeed, manufacturers have marketed menthol subbrands with lower levels of menthol to youth (Bates et al., 1999; Kreslake et al., 2008a;
Kreslake and Yerger, 2010). Brands with higher levels of menthol have been designed and marketed to older smokers, including African Americans, to capitalize on their chemosensory preferences for a smoother, cooler product (Kreslake et al., 2008b; Kreslake and Yerger, 2010). In international markets, such as Japan, menthol has been used to target women smokers and has been attributed with rapid growth in developing new consumer markets (Connolly et al., 2011).
There have been proposals to ban mentholation because of its capacity to enhance addictive potential of cigarettes, especially the initiation of dependence among youth (Tobacco Products Scientific Advisory Committee, 2011). In response to these proposals, a number of reviews of the function and effects of menthol in cigarettes have been conducted, supporting the conclusion of the positive role of menthol in initiation and maintenance of smoking behavior. However, relatively little research is available to guide understanding of how consumers who prefer mentholated products might respond if menthol in cigarettes was removed or reduced by regulation.
In an active smoking study using a switching design, menthol smokers smoked a mentholated Camel Crush product (which delivers mentholation in response to “crushing” a flavor pellet embedded in the filter) for 15 days followed by 15 days of smoking a nonmentholated Camel Crush (Strasser et al., 2013). Puffing intensity increased following the switch from menthol to nonmenthol, but there was no corresponding increase in daily smoking or biomarkers of exposure. Participants found the nonmenthol product less appealing than their usual brand. These findings suggested that change in mentholation had relatively minor influence on product use.
A survey of smokers about a potential menthol ban found that support for a ban was high among Latino and African American smokers, who are often associated with menthol use (Pearson et al., 2012). Most telling, 39 percent of menthol smokers stated that they would quit if menthol cigarettes were unavailable, and an additional 25 percent said they would switch to a nonmentholated product and try to quit. Only 13 percent stated that they would switch to a nonmentholated brand with no thought of quitting. The remainder reported that they did not know what they would do.
One survey has addressed the question of smoker behavior in the context of a menthol cigarette ban (O’Connor et al., 2012). Using an online survey format, 471 current smokers were asked about their likely response if menthol cigarettes were banned, and they also completed a simulated purchase task to estimate demand for menthol and nonmenthol cigarettes across a range of price points. Over 40 percent of menthol smokers indicated that they would miss their brand, and a similar percentage reported that they would be angry. One-quarter of menthol smokers indicated that they “would find a way to buy a menthol brand” (suggesting possible engagement with the illicit market), but more than one-third indicated
that they would try to quit. Menthol and nonmenthol smokers reported similar demand elasticity for their preferred product type. Of particular interest was the observation that menthol smokers who indicated a willingness to obtain contraband cigarettes reported a higher median consumption (cigarettes per day) and lower demand elasticity in comparison with smokers who did not express interest in contraband products.
The available evidence, while limited, shows that many smokers would pursue licit alternative outcomes, including switching to a nonmentholated cigarette or quitting, if their preferred mentholated product was not legally available. However, heavier smokers were more likely to report that they would seek mentholated cigarettes through the illicit market.
To directly assess the effects of switching from a mentholated to an unmentholated product, controlled studies would have to be done. Another important research question is whether the adoption of nonmentholated products or quitting are achievable and sustainable outcomes for smokers who prefer menthol products. Other research issues of importance are to what extent consumers will use legal strategies to acquire a mentholated product, in contrast to seeking menthol cigarettes in the illicit market. Legal options could include the use of mentholated e-cigarettes or the use of legal menthol preparations that can be added to a cigarette to produce a “self-mentholated” cigarette.
Regulation of Product Packaging
Many countries have policies that require health warnings on packaging and that restrict packaging and labeling that are misleading or deceptive. Warning labels on tobacco products are intended to alert consumers to toxic substances and other risks to their health and dissuade them from smoking. Marketing descriptors on cigarette packs such as “light” and “ultra-light” have been shown to promote false perceptions of reduced health risks (Ashley et al., 2001; National Cancer Institute, 2001; Etter et al., 2003; Hammond and Parkinson, 2009). “Light” descriptors are now banned by more than 80 countries, including Australia, Canada and EU nations. In June 2010, the FDA, acting in response to its mandate under the FSPTCA, prohibited the labeling or advertising of tobacco products with descriptors such as “light,” “mild,” or “low.”
There has been limited research to evaluate the influence of the ban on light descriptors in any market. Evidence has shown that manufacturers have responded to the ban by adopting text descriptors and colors that continue to distinguish the target brand from its competitors, using nonbanned language. For example, in the United States and Mexico, “light” has been replaced with “gold,” and consumers may have no problem identifying their usual brand (Thrasher et al., 2010; Connolly and Alpert, 2014).
Graphic health warnings, which use color pictorial images to communicate health information, were first introduced in Canada in 2001, and more than 60 countries to date have either implemented pictorial warnings or have passed legislation to do so (Cunningham, 2014). Appearance of warnings, including size, placement, and content, varies from country to country. Some packaging also includes contact information to direct consumers to resources to quit smoking.
In the United States, Section 201 of the FSPTCA directed FDA to adopt more accurate and effective health warnings on cigarette packs. In its final ruling of June 2011, FDA issued its requirement for manufacturers to include color graphics to accompany the nine new health warnings and released nine graphic warning messages. Two legal challenges by tobacco manufacturers and retailers quickly followed: one ruling upheld the graphic warnings requirements of Section 201 of the FSPTCA (see Reinberg, 2013); the other found that the labels violated tobacco manufacturers’ First Amendment free speech rights (see Pelofsky, 2012). FDA has announced that new labels will be developed to comply with the First Amendment.
In December 2012, Australia became the first country to require plain packaging for tobacco products. At present, a number of other countries, including Ireland, New Zealand, the United Kingdom, and France, are considering plain packaging. The tobacco industry is aggressively challenging Australia’s plain packaging law in multiple jurisdictions.4
Research has seldom been undertaken to investigate directly the likelihood that a consumer will engage in the illicit tobacco market when the preferred pack design is unavailable. Therefore, one can consider existing evidence on broader issues of consumer responses. The available research, as discussed below, shows that graphic health warnings reduce the appeal of tobacco products and increase public awareness of the health risks of smoking. Limited evidence suggests that a high proportion of smokers would prefer a brand without a graphic warning and that smokers take steps to conceal or avoid their exposure to graphic health warnings. There are numerous legal ways in which smokers can overcome the negative effects of graphic warnings (many of which are promoted by the tobacco industry), including the use of stickers to cover the packs or branded tins or other containers to which cigarettes sold in packs with graphic warnings can be transferred (Zacher et al., 2014). These alternatives would presum-
4 The industry challenge on Australia’s plain packaging law was overwhelmingly defeated in Australia’s High Court. But as of July 2014, there are two pending actions involving trade and investment agreements. Philip Morris Asia, based in Hong Kong, has mounted a challenge to Australia’s plain packaging law through the bilateral investment treaty between Hong Kong and Australia. In parallel, five countries—Cuba, Dominican Republic, Honduras, Indonesia, and Ukraine—are challenging the law at the World Trade Organization.
ably serve as mitigating factors against participation in the illicit market to obtain packs without graphic warnings.
The rest of this section considers in more detail the research on consumer behavior when graphic warnings and plain packaging are introduced.
Tobacco Health Warnings
Warning labels on product packaging communicate the risks and dangers of using the product to potential consumers. Graphic warnings usually feature highly evocative images of diseased organs or persons in advanced stages of tobacco-related illness. The vivid pictorial warnings are intended to correct widespread misconceptions about the risks of tobacco use and to enhance recall of those messages. The use of pictures and larger warnings (whether text or pictorial) has increased public awareness, information recall, and perception of risk (Strahan et al., 2002; Moodie et al., 2010; Bansal-Travers et al., 2011; Miller et al., 2011; McCool et al., 2012; Strasser et al., 2012; Hammond et al., 2013b).
Prominent pictorial warnings have been found to be more effective than text-based warnings in increasing perceptions of the risk of tobacco use and in preventing use and promoting cessation (Hammond et al., 2006; Borland et al., 2009; Centers for Disease Control and Prevention, 2011; Hammond, 2011; Azagba and Sharaf, 2013). Other studies have found similar responses among youth (Goodall and Appiah, 2008; White et al., 2008; Germain et al., 2010; Miller et al., 2011; Hammond et al., 2014). Although the tobacco industry has often mounted major campaigns against and legal challenges to graphic health warnings, the labels have been widely supported by the public—both smokers and nonsmokers (Miller et al., 2011).
For consumers that continue to smoke, cigarettes in packs with large graphic health warnings have been perceived as less attractive, less smooth, higher in tar, and having greater health risk than cigarettes in packs with smaller warnings (Hammond et al., 2014). A survey of Italian smokers found that more than half of them would change brand if faced with packs with graphic warnings, and two-thirds would feel uncomfortable showing the package (Mannocci et al., 2013). The discomfort experienced by viewing graphic warnings may lead smokers to actively avoid looking at the graphic warning component of cigarette packs (Maynard et al., 2014), and consumers may be more likely to conceal cigarette packs that contain prominent graphic warnings (Moodie and Mackintosh, 2013; Zacher et al., 2014). Other research has shown that avoidance of graphic health warnings by adolescents may be greater among smokers than those who have never smoked or smoke occasionally. A study among high school students in the United Kingdom found that daily smokers were more likely to avoid paying attention to graphic health warnings than students who were experiment-
ing with smoking and weekly smokers (Maynard et al., 2013). Consistent with this finding, a separate survey of UK youth ages 11-16 found that daily smokers were more likely than experimental smokers or nonsmokers to conceal packs to avoid warnings after graphic warnings were placed on the back of the pack (Moodie et al., 2013).
Plain (Standardized) Packaging
Plain packaging is a tobacco control policy measure to address concerns about the influence of cigarette pack descriptors and brand imagery on positive product appeal. Plain packaging requires the removal of corporate logos, trademarks, colors, and imagery; it permits manufacturers to print only the brand name in a mandated size, font, and place on the pack. Prominent health warnings and any other legally mandated information, such as toxic constituents and tax-paid stamps, may also be included. A generic color and layout are specified for the packs. Plain packaging is a more aggressive approach to reduce the appeal of tobacco products and increase the prominence of mandated health warnings.
The transnational tobacco industry has claimed that standardized packaging makes it more difficult to distinguish legitimate products from counterfeit ones, thereby encouraging illicit trade. At a recent earnings briefing for British American Tobacco, financial analysts were told by the company that plain packaging in Australia had no impact on business, but that there was an increase in the sales of counterfeit cigarettes (Greenblat, 2014). However, the Australian Customs and Border Protection Service had already seen a steady rise in the detection of illicit cigarettes, which predated and continued after the introduction of plain packaging in December 2012. Moreover, the Customs and Border Protection Service in Australia has detected only one incident involving plain packaging among the 17 million cigarettes it seizes monthly of illicitly traded tobacco (Corderoy, 2014).
Early research following the introduction of plain packaging in Australia in December 2012 has found no evidence for an increase in the sale or use of illicit tobacco products. The sale of illicit cigarettes (those that were irregularly packaged or suspiciously priced) was assessed in a retail purchase surveillance study conducted across the country before, during, and for 8 months after the introduction of plain packaging (Scollo et al., 2014a). In the months after plain packaging was introduced, less than 1 percent of packs purchased (5 of 878 packs) were deemed likely to be illicit, compared with the observed rate of 2 percent (13 of 598 packs) prior to the introduction of plain packaging.
A telephone survey of Australian smokers conducted in 2011 and 2013 found that a similar proportion reported recent use of unbranded illicit tobacco in both years, 2.3 and 1.9 percent, respectively (Scollo et al., 2014b).
In 2013, a similarly small proportion of cigarette smokers, 2.6 percent, reported purchase of one or more packs in noncompliant packaging in the past 3 months, and 1.7 percent reported purchasing cigarettes from an informal seller in the past year (Scollo et al., 2014a). While these data were collected in the early period following introduction of a highly publicized policy, a period in which implementation and enforcement efforts are expected to be unusually rigorous, the results nonetheless suggest that there has not been an increase in illicit market purchases following the introduction of plain packaging in Australia.
Considering the recency of the first implementation of policies on plain packaging, most research on the potential effects of plain packaging on consumer behavior has focused on perceptions of risk, product appeal, and future intentions to quit. Studies by Wakefield and colleagues demonstrated that the progressive reduction of pack descriptors decreased product appeal and increased negative perceptions of product taste (Wakefield et al., 2008; Germain et al., 2010). Other research suggested that plain packaging has a potentially greater negative effect on product appeal and purchase intent than increasing the size of a graphic health warning on the same pack (Wakefield et al., 2012). Plain packaging can reduce positive brand imagery among smokers of low socioeconomic status (Guillaumier et al., 2014), perhaps by undermining smokers’ personal identification with brands that define social attributes and standing (Hoek et al., 2012). The effect of plain packaging on consumer perceptions of product appeal appears to show some universality across gender and cultural setting: similar negative impacts of plain packaging have been observed among young female smokers in the United Kingdom (Hammond et al., 2013a) and Brazil (White et al., 2012) and youth in New Zealand (Hoek et al., 2013).
One study has addressed the potential effects of plain packaging on consumer intentions to engage in the illicit tobacco market. Using focus groups, Moodie and colleagues (2012) asked 54 young adult smokers in the United Kingdom about their perceptions of illicit tobacco products and the effect of plain packaging on future intentions to purchase illicit tobacco. There was no evidence that plain packaging would make illicit cigarettes harder for smokers to identify, nor would it substantially change the price of illicit cigarettes given the low manufacturing cost. Smokers regarded illicit tobacco as inferior in quality, and their decision to purchase was influenced by availability and price. The findings, which require replication in a larger, controlled survey or behavioral laboratory setting, provide preliminary evidence that plain packaging is unlikely to make illicit products comparatively more appealing to consumers.
As is the case with graphic warnings, there are numerous legal ways in which smokers can overcome the negative effects of plain packaging, including the use of stickers or branded containers. Australia has witnessed
an increase in such behavior, often encouraged and supported by the tobacco industry (Zacher et al., 2014). While these strategies subvert the law and limit the public health impact of the policy, they could also serve to discourage engagement in illicit trade for branded packages.
This section considers supply-side responses to possible product regulation. The relevant policy question is whether, if FDA did regulate product features and there is a substantial demand for the original product, what factors would determine whether there is a supply to meet that demand? Although this discussion separates issues of supply from those of demand, the two sides of the market cannot be segmented so neatly, nor seen in terms of a time sequence. The emergence of demand does not necessarily precede the creation of a supply system. For example, supply can precede demand in the sense that people who supply some other illegal commodities, such as marijuana, could add illicit high-nicotine cigarettes to their offerings, speculating that the market can emerge. In addition, samples may be given away free or at reduced price to stimulate demand.
Established distribution networks are important for allowing changes in supply (see Joossens and Raw, 1998, p. 67; Shleynov et al., 2008). As discussed in Chapter 2, the existing illicit market in the United States consists largely of bootlegging from low-tax jurisdictions to high-tax jurisdictions, which results in cheaper cigarettes to consumers in high-tax jurisdictions. If features and constituents of cigarettes are prohibited at the federal level, bootlegging across state borders would not be the means to supply illegal products. The source of such illegal products would have to come from either countries in which such products are legal, counterfeits, or domestic illegal production.
In cases of large-scale smuggling, contraband cigarettes, similar to other contraband, such as illegal drugs, would enter the United States under the guise of international commerce or international noncommercial traffic, or they would be brought across the border clandestinely outside of regular border crossings. In small amounts, contraband cigarettes might also be transported by mail or parcel service. As noted in Chapter 2, there is evidence that smuggling of illegal cigarettes from abroad and the presence of counterfeit cigarettes of foreign origin has occurred in the United States (U.S. General Accounting Office, 2004; T. Chen, 2008). There is also evidence that the global tobacco industry has been complicit in the large-scale smuggling.
Domestic illegal production could arise in several ways. Some native tribes in the United States already produce their own cigarettes and could potentially produce cigarettes with prohibited features. Native American
brands produced on tribal land have the potential to play the same role as cheap whites: they could be sold in bulk for distribution outside of tribal lands. The existence of a distribution system for illegal cigarettes makes Indian reservations a potential source of illegal manufacture within the United States. As discussed in Chapter 7, Indian reserves in Canada have played this role. Fourth-tier manufacturers are also a potential source of domestic illegal production (see discussion in Chapter 2).
An important policy question is whether smuggling from outside the United States or illegal production within the country are likely to emerge as major sources of illicit tobacco if the formulation or design of specific products is restricted. It is difficult to consider what might occur because there are few instances in which a legal product in wide use is suddenly banned for reasons of public health. One instance is the prohibition of chlorofluorocarbons (CFCs), a major chemical (used as a refrigerant), following the adoption of the Montreal Protocol in 1987, aimed at preventing expansion of the ozone hole. The ban led to the emergence of a market in the prohibited substances, mostly from developing to developed countries (De Sombre, 2000). It is claimed that at one point in the 1990s smuggled CFCs were second in value only to drugs as illegal imports (Saab, 1998). There are a few ways in which banned CFCs and cigarettes are similar. For example, both illicit products are similar to their legal imports, and it takes considerable effort for customs officers to distinguish legal from illegal shipments. There are of course differences as well: CFCs are used in manufacturing within the United States while cigarettes go through a distribution system that includes no additional processing.
Currently, the legal supply of tobacco products includes a range of controls on manufacturers and distributors that would inhibit illegal production (see Chapter 5). Similarly, with regard to international smuggling, general customs enforcement appears to be robust. Furthermore, cigarettes are bulky relative to even the least compact of illegal drugs, marijuana. For comparison: marijuana sells at the import level for about $2.50 per cubic centimeter, and the same volume of cigarettes would generate only about 2 cents.5 To provide just 10 percent of the current estimated U.S. illicit market of 124 million cartons (see Chapter 4) would require smuggling a volume equivalent to 142.6 million cubic meters of conventionally packaged cigarette cartons. From a logistical point of view, this kind of smuggling into the United States would be an order of magnitude more challenging than
5 Volume is the principal determinant of risk for smugglers. Prices are of course usually quoted per unit weight. We have converted $10,000 per kilogram for marijuana into a measure per unit volume by assuming that a kilogram of commercial-grade marijuana occupies 3,939 cubic centimeters (Center for Investigative Reporting, 2013). A carton of cigarettes has a volume of 1,150 cubic centimeters. We use a smuggled import price of $20 per carton, based on current base prices, minus 80 percent for federal taxes, as an approximation.
current major smuggling activities. However, cigarettes are smuggled into many countries, even those with strict border controls, so it is not impossible to do so on a large scale.
The Internet may also play a role in connecting domestic buyers with foreign suppliers of illegal tobacco products. In recent years, government agreements with credit card and major shipping companies to ban transactions and shipments of all Internet cigarette sales have been effective in limiting the Internet as a means for illegal purchases (Ribisl et al., 2011). Currently, however, there are ways to avoid the agreements: Internet vendors can accept other forms of payment (than credit cards), and they can use other delivery options.
The discussion so far has focused on the products of an illegal market. However, there is also a question as to whether domestic wholesale and finance distribution systems would develop to support such a market. Some apparently profitable illegal market niches are never filled: for example, as discussed in Chapter 4, there are smuggling corridors (pairs of high-tax–low-tax states) in the United States for which there is surprisingly little activity. The reasons that such apparently profitable markets do not develop are not clear.
There has been no systematic research that would help assess what factors are likely to determine the supply response to product regulation. The likelihood that a large-scale illicit supply will develop in response to product regulations will be influenced by the potential profitability of supplying smokers with illicit products. The profit potential may be limited by the development of close substitutes that are likely to remain legal. For example, with the ready availability of liquid nicotine capsules to supply the e-cigarette market, it is easy to imagine devices that allow smokers to add nicotine to the flow of smoke from non-nicotine cigarettes.
E-CIGARETTES: A TOBACCO ALTERNATIVE
One possible outcome from regulation of conventional cigarettes may be the adoption of e-cigarettes as an alternative legal product. E-cigarettes and other electronic nicotine delivery systems have attracted considerable attention among the general public as well as the public health community since the original prototype was exported from China in 2004: see Box 8-1. The emergence of e-cigarettes has greatly increased the complexity of the nicotine market from the relative simplicity of the cigarette market. Added to the complexity is the fact that prominent tobacco companies are entering the e-cigarette market (Esterl, 2013).
There is wide variation among countries in their approach to electronic cigarette regulation. Some countries, including Argentina and Singapore, have banned electronic cigarettes completely (Agence France-Presse, 2011;
E-Cigarettes: Benefits and Risks
Ecigarettes have been promoted as a safer alternative (than conventional cigarettes) for delivering nicotine to smokers. Because ecigarettes deliver nicotine to a user through vaporization of a nicotine solution, the harmful byproducts of combustion are eliminated. This potential for lowered exposure risk has promoted substantial public interest. Recent evidence suggests that the prevalence of ecigarette use, especially among adolescents who currently smoke, has grown rapidly (Centers for Disease Control and Prevention, 2013).
Although ecigarettes have been linked to helping smokers quit traditional cigarettes (Bullen et al., 2013; Etter and Bullen, 2014), as well as lowering the number of cigarettes smoked per day in those not seeking to quit (Caponnetto et al., 2013), the data are by no means conclusive or without contradiction (Bialous and Sarma, 2014; Carr, 2014; Doyle et al., 2014; Grana et al., 2014). Public health concerns include the new use of ecigarettes by nonsmokers or former smokers (which may increase the likelihood of conventional cigarette use); dual use of ecigarettes and conventional cigarettes, which undermines the full potential for exposure reduction and undermines cessation; and a negative effect on what has been progress in reducing and stigmatizing tobacco use, particularly in public locations (Fairchild et al., 2014).
There is also concern that liquid nicotine devices might pose new health risks. Vegetable glycerin, added to make smoke visible when the nicotine solution is vaporized from ecigarettes, has resulted in a documented case of lipoid pneumonia, a consequence of lipid substances being inhaled and causing inflammation in the lungs (McCauley et al., 2012). Eliquids—contained in the cartridges to fill ecigarettes—are not regulated by FDA, yet purportedly contain neurotoxins that could induce seizures or death (Richtel, 2013). Eliquid poisonings are increasingly being reported, particularly among young children.
The ability of ecigarettes to be used as a drug delivery device is another major concern. Ecigarette cartridges are commonly sold containing liquid nicotine, but they can easily be refilled with a liquefied illegal drug (Cobb and Abrams, 2011). Such reuse has occurred in the United States, especially among youth, with users replacing the nicotine with drugs, such as marijuana or heroin, to create a smokeless and odorless drug delivery system (Erwin, 2013; Walser, 2013).
Given the recent emergence of ecigarettes and changing technology, research on them and their use is just getting under way, and the broad implications of widespread adoption of ecigarettes are not yet known.
Euromonitor International, 2013c); others have not regulated the products at all (Euromonitor International 2013b). Some countries have adopted a two-tier approach: Australia, for example, permits non-nicotine cartridges but outlaws those containing nicotine (Grace, 2008). Still others have a more complex approach, regulating e-cigarettes as medicinal devices (Euromonitor International, 2013a) or imposing restrictions on advertising and flavors (Keating, 2014).
In the United States, a proposed deeming regulation was issued April 2014,6 and it is expected that e-cigarettes will eventually fall under the regulatory authority of the FDA. In its proposed regulation for e-cigarettes, FDA included a requirement for FDA premarket review, restrictions on sale to minors, prohibition of scientifically unsupported health claims, and mandatory package warning labels. However, the proposed deeming rule did not include current plans for regulation of characterizing flavors or restrictions on marketing strategies for e-cigarettes.
A regulatory ban on menthol or on high levels of nicotine may be difficult to enforce given current and emerging technologies. Several cigarette brands produced by major U.S. manufacturers feature an embedded menthol capsule in the filter that provides mentholation when crushed. It is possible that a cigarette holder could be developed that would hold capsules of menthol, nicotine, or other additives. In the current regulatory environment, preparations containing nicotine and menthol are readily and legally available from e-cigarette retailers and other sources, and devices are available that facilitate the administration of these substances through vaporization. That is, there may be little technical impediment to the development of devices that bridge a gap to allow modification of conventional cigarettes using e-cigarette technology. As a consequence, regulation of high nicotine levels or menthol flavoring may lead not to illicit cigarette markets, but, instead, may increase demand for electronic cartridges that contain nicotine or menthol concentrates. 7
The relatively recent introduction of e-cigarettes makes it difficult to forecast what impact this new product will have on both legal and illicit tobacco consumption. Until recently, the e-cigarette market had been made up of over 200 companies, most of which were small and independent from large tobacco companies. With the entrance of large tobacco companies, increased competition, advertising, and international commerce are expected to significantly alter the e-cigarette market (Esterl, 2013; Sebastian and McDermott, 2013). Product databases, such as UN’s Comtrade system for coding such products in international commerce, currently do not have unique codes for e-cigarette devices, their cartridges, or their refillable contents, making it difficult to measure changes in the trade of these products.
In addition, there are no detailed data on e-cigarette use patterns, such as frequency of use, duration, dual use of conventional cigarettes, and motivations for use (Grana et al., 2014). If some feature or use of cigarettes
6 A deeming regulation is a proposed regulation that would include products meeting the definition of a “tobacco product” under the Tobacco Control Act to be subject to the FDA’s jurisdiction.
7 The proposed rule giving FDA the authority to regulate e-cigarettes would also include tobacco product components or parts used in consumption, such as e-cigarette cartridges.
or e-cigarettes are regulated, one research approach might be to compare prevalence and use data with sales data to measure change in activity, but existing data sources are not detailed enough for accurate measurement. Some of these data needs can piggyback onto country-level surveys that do cover tobacco use. The National Youth Tobacco Survey, National Health and Nutrition Examination Survey, and the HealthStyles Survey ask several questions about e-cigarette awareness and use (King et al., 2013). However, surveys would need to be more detailed with respect to the measurement of e-cigarette use, not just whether an e-cigarette has ever been used, for these data to serve as useful measurement instruments. Furthermore, unlike cigarettes, e-cigarettes do not have an excise tax levied on them, so an independent measurement of sales volume or tax revenues does not exist.8
There is a complex set of issues with regard to e-cigarettes. Strong research designs that could provide information have not, so far, been employed. The uncertainty suggests a need for research on the impact of such regulatory action in other countries: one possibility is Brazil’s pending ban on tobacco additives (including menthol). A shift to electronic delivery systems containing additives that are banned from conventional tobacco products may not be a perfect substitute for the current products, but could limit both the effectiveness of potential regulations in terms of reducing tobacco use in general and the profitability and, consequently, the size of an illicit tobacco market.
Under the FSPTCA, FDA has authority to regulate tobacco products in order to protect public health, including a requirement that new tobacco products do not have greater potential for initiating or maintaining dependence than existing products. To fulfill its responsibilities, FDA is considering a wide range of policy options, which includes trying to estimate the possible effects of those options on the illicit tobacco market.
Many regulatory options designed to reduce demand for tobacco products will necessarily involve strategies to reduce the addictive potential of those products, which is likely to reduce their appeal to consumers. According to the model presented in Figure 1-1 (in Chapter 1), if product appeal is reduced, consumers might try to quit tobacco use, seek a legal version of the same or an alternative product, or engage in the illicit mar-
8 There are two sources of data that could be helpful—TechNavio and Research and Markets: data made available to researchers might provide useful market and sales forecasts to help identify areas that may be most inclined to see a rise in e-cigarette use. See TechNavio at http://www.technavio.com/report/global-e-cigarette-market-2014-2018 and Research and Markets at http://www.researchandmarkets.com/research/gx8h33/us_electronic [January 2015].
ket. Policy makers need information on the factors that promote choice of one option over another and the proportions of smokers that are likely to choose each option. Current research efforts that have focused on changes to consumer satisfaction and smoking behavior in response to variations in product composition, design, or packaging provide some information, but the findings are suggestive, not conclusive.
Product modifications, such as increased ignition propensity and reduced filter ventilation, have been shown to have limited impact on product appeal when considered in isolation from other product features. Reductions in nicotine level and mentholation may have a more significant effect on reducing product appeal. However, findings on consumer behavior in response to very low levels of nicotine are mixed: some studies found smokers are more likely to quit than seek alternative products and some studies found little or no change in cigarette consumption. The only research to date on menthol has been short-term switching studies, which suggest that changes in smoking behavior and habits are minimal for those switched to unmentholated cigarettes for short periods. Research on graphic warnings and plain packaging has shown that consumers report lower product appeal for both packaging possibilities, but the effects on product preference and consumption appear to be modest.
Overall, the limited evidence to date suggests that demand for illicit versions of the current conventional cigarette, if they are modified through regulations, may be small. The availability of alternative options, including use of the now-regulated products and other licit products and cessation, may diminish demand for illicit products. Although some smokers may seek more appealing illicit products if available and accessible, established distribution networks and new sources of product (which would either have to be smuggled from other countries or produced illegally) would be necessary to create a supply of cigarettes with prohibited features. The profit potential of a new type of illicit tobacco market would likewise be limited by the availability and development of legal products that are close substitutes, as well as the robustness of enforcement.
Research on many aspects of product demand and supply in response to changes in tobacco products would be very valuable to policy makers. It would be useful to test whether certain product characteristics are more appealing to consumer subgroups, including youth, women, racial/ethnic minorities, and smokers with specific chemosensory preferences. In addition, studies could explicitly consider whether or not individual desire for a particular characteristic outweighs the costs of acquiring a product on the illicit market. Such research would benefit from the use of both traditional and innovative assessment of consumer perceptions, including measures of appeal, chemosensory preferences, risk perceptions, nicotine effect and liking, willingness to pay, and future use intentions, as well as willingness to
obtain a product from an illicit source. It would be useful to have studies that simultaneously assess the influence of price, product quality, ease of access, and risk of social or legal penalty on consumers’ product preferences and intent to use. The new challenges and alternatives posed by the availability of e-cigarettes deserve special attention.
RECOMMENDATION 8-1 Research is needed to examine how smokers respond to the permanent loss of specific product features that they have previously found desirable, as a result of bans and restrictions on key constituents and additives as well as changes to packaging. Research should assess consumers’ intentions to seek products with banned features through the illicit market in comparison with other options, such as quitting and using alternative products. Factors that promote individual variation in response should also be examined.
RECOMMENDATION 8-2 Research is needed on the relationship between the use of e-cigarettes and the use of conventional tobacco products and on the role of e-cigarettes as an alternative to participation in the illicit tobacco market. Longitudinal studies are needed to understand the dynamics of the relationship and to determine the extent of full substitution of e-cigarettes compared with dual use or reversion to conventional products. Such work will require improvements to sources of data, including unique coding for e-cigarettes in international commerce. Furthermore, although some current surveys include questions on e-cigarette use and awareness, more detailed questions are needed on factors that affect use and their relationship to the use of conventional cigarettes.
RECOMMENDATION 8-3 The paucity of studies on the supply side of the illicit tobacco market presents challenges for research, and creative methodologies will be needed. One potential source of needed information may come from reviews of analogous markets, perhaps in other countries, where existing products have been removed from the market, but similar or related products continue to be available in legal commerce, to determine what factors influenced the emergence of illegal supply.