Minimum Age of Sale for Tobacco Products and Electronic Cigarettes: Ethical Acceptability of US “Tobacco 21 Laws”

Abstract

Several US jurisdictions have recently passed laws that raise the minimum age of sale for tobacco products and electronic cigarettes to 21 years (Tobacco 21 laws). Although these laws have been demonstrated to be an effective means to reduce youth smoking initiation, their passage and potential expansion have provoked controversy. Critics have objected to these laws, claiming that they unduly intrude on individual freedom and that they irrationally and paternalistically restrict the freedom of those aged 18 to 20 years, who were previously able to legally purchase tobacco products.

We have examined the ethical acceptability of Tobacco 21 laws. First, we have described ethical support for such a restriction grounded in its public health benefit. We have then offered arguments that raise doubts about the soundness of critics’ objections to these regulations and described an additional ethical justification arising from concern about preventing harm to others.

On the basis of this analysis, we conclude that Tobacco 21 laws are ethically justifiable.

Recently, several US cities and states have passed laws that raise the minimum age of sale for tobacco products and electronic cigarettes to 21 years. Similar efforts are under consideration in other jurisdictions, and legislation has been introduced at the national level.

These “Tobacco 21 laws” have been demonstrated to be effective means to reduce youth smoking initiation^1–3 and are supported by a strong majority of US persons, including 68.3% of individuals aged 18 to 24 years.⁴ Nevertheless, their passage and potential expansion have provoked controversy. Critics claim that these laws unduly intrude on individual freedom^5,6 and argue that it is inconsistent to deny those aged 18 to 20 years the choice to use tobacco while simultaneously treating them as sufficiently mature, for example, to serve in the military.^5,7

We have explored the ethical acceptability of Tobacco 21 laws. We have first described ethical support for this approach grounded in its public health benefit. We have then offered arguments that raise doubts about the soundness of critics’ objections and described an additional ethical justification arising from concern about preventing harm to others. On the basis of this analysis, we conclude that Tobacco 21 laws are ethically justifiable. Although we believe this analysis also applies to e-cigarettes, we have focused on traditional cigarettes and other smoked tobacco products.

PUBLIC HEALTH RATIONALE

Ethical support for Tobacco 21 laws is primarily grounded in consequentialist considerations, specifically their anticipated public health benefit. Tobacco use is a major public health problem. Since 1964, the year Luther Terry released the first surgeon general’s report on smoking, tobacco has caused more than 20 million premature deaths in the United States.⁸ Following that report, numerous policy efforts have dramatically reduced smoking rates. These improvements notwithstanding, 40 million US persons still smoke⁹ and tobacco use remains the United States’ leading cause of morbidity and mortality; it is responsible for more than 480 000 premature deaths annually.⁸

Preventing youths from smoking is crucial for reducing overall smoking rates. Nearly all smokers start smoking before age 21 years (95.6%), with 88.0% starting before age 18 years.¹⁰ According to the 2015 Youth Risk Behavior Survey, one third of high school students had tried cigarettes, nearly half had used electronic vapor products like e-cigarettes, and one third were using tobacco products or e-cigarettes at the time of the survey.¹¹ Adolescents may be especially vulnerable to nicotine addiction, because individuals who start smoking at younger ages are more likely to become addicted, smoke daily, smoke heavily as adults, and have difficulty quitting.^2,12–14

Reducing youth access to tobacco has been effective in preventing initiation and regular use. Historically, efforts to reduce youth access have focused on disrupting retail sales to those younger than 18 years, including enhanced retailer oversight to encourage compliance with minimum age requirements and eliminating vending machines and other self-service sales. Since 2001, these strategies have led to a significant decrease (19.0%–12.6%) in the prevalence of adolescent smokers who report usually obtaining their own cigarettes from a store.¹¹ However, these regulations do not address “social sources” of tobacco, such as borrowing or buying from acquaintances who meet the minimum age of purchase. Today, smokers younger than 18 years report social sources as the most common method for obtaining cigarettes.^2,15,16 According to the National Survey on Drug Use and Health, nearly two thirds (63.3%) of those aged 12 to 17 years who had smoked in the past month had given money to others to buy cigarettes for them. Curbing access to these social sources of tobacco is therefore essential for reducing youth smoking rates.

Raising the minimum legal age to purchase tobacco products to 21 years can substantially reduce access, because minors are less likely to have peers older than 20 years who will purchase for them than they are to have peers older than 17 years. This may explain evidence from jurisdictions that were early adopters of Tobacco 21 laws indicating their efficacy in reducing youth smoking rates. An empirical study of the first US Tobacco 21 law found a 47% reduction in the smoking rate among affected high school students in the 4-year period following implementation (2006–2010).³

If implemented nationwide, Tobacco 21 laws would likely secure a substantial public health benefit. According to recent estimates from the National Academy of Medicine, raising the minimum age to 21 years nationwide would result in 223 000 fewer premature deaths from smoking among US persons born between 2010 and 2019. The largest health gains are expected to result from reducing smoking initiation among individuals aged 15 to 17 years, although rates would also decrease in other age groups, including those younger than 15 years and between 18 and 20 years.² Consequently, prohibiting tobacco sales to those aged 18 to 20 years has the potential not only to reduce smoking rates within that age group but also to substantially reduce the likelihood that those younger than 18 years will be able to access tobacco products. Supporters hold that the good consequences these restrictions should bring about are sufficient to ethically justify this liberty limitation. Opponents, of course, disagree.

ASSESSING THE LIBERTY RESTRICTION

The primary objection to Tobacco 21 laws is that they restrict the freedom of those aged 18 to 20 years, who previously could legally purchase tobacco products. Critics assert that it is inconsistent to deny those aged 18 to 20 years the freedom to purchase tobacco while treating them as sufficiently mature to marry, vote, and serve in the military. “If you’re old enough to fight and die for your country at age 18 years, you ought to be able to make the choice of whether you want to purchase a legal product or not,” argued a lobbyist during California’s legislative hearings.¹⁷ A critic of a proposal in Massachusetts deemed the laws “silly and condescending,” viewing it inconsistent to permit someone to vote but not to purchase cigarettes: “If you’re old enough to determine who should hold elected office, then you can make decisions about whether to smoke.”¹⁸

Critics also reject these laws as unduly paternalistic because they restrict the liberty of those who should be viewed as responsible for their own decisions. The editorial staff of a North Carolina newspaper argues, “When you’re an adult, you are entitled to the full range of rights and freedoms guaranteed by the United States and North Carolina constitutions. Those liberties must include the right to decide which legal substances to consume.” Raising the minimum age for tobacco, they argue, would be not only paternalistic but “infantilizing . . . depriv[ing] young people of full participation in a free society.”¹⁹ Neither of these objections is ethically compelling.

Refuting the Inconsistency Claim

The claim that Tobacco 21 laws are inconsistent with public policies that give young adults the right to vote, join the military, or get married is persuasive only to the degree that tobacco use is similar to these other activities. One might believe that they are all similar in that they involve elements of both risk (physical, emotional, or other burdens) and benefit (pleasure, satisfaction, or opportunity). Despite these similarities, however, tobacco use is importantly different from these other freedoms in at least 3 ways.

First, the risk–benefit profile of tobacco use is not analogous to the right to vote, to get married, or to join the military. When used as designed, tobacco may bring about temporary pleasure but has clearly and repeatedly been shown to cause significant harm to virtually all those who consume it. By contrast, other freedoms allowed to those aged 18 to 20 years have the potential to bring about more good than harm (although they are not guaranteed to do so).

This is a significant distinction that could make restrictions on tobacco use consistent with the protection of other types of freedom. Paternalistic policies, that is, policies that override the autonomy of individuals with the aim of benefiting them, may be ethically justifiable in some cases (when the potential benefit is significant) but not ethically permissible in others (when the benefit is minor). Because tobacco, when used as designed, is highly likely to cause harm, it is coherent to argue that restricting tobacco use is acceptable, whereas putting constraints on the other freedoms identified would not be.

It is worth noting that this argument does not entail that any harmful substance, service, or action should be prohibited or controlled by law. Rather, it highlights the fact that because tobacco use causes harm and is associated with few benefits it is not analogous to other freedoms with different risk–benefit profiles, thereby undermining the critics’ inconsistency claim.

Second, unlike voting, military service, and marriage, tobacco is highly addictive. Tobacco’s addictive nature has important implications for autonomy and so suggests that regulating it differently would be appropriate.

A third distinction between tobacco use and other freedoms granted to those aged 18 to 20 years is offered by Ronald Dworkin, who argues that different freedoms have different value. In making this argument, Dworkin critiques libertarians, such as Robert Nozick, who generally object to any government restriction on individual freedom, arguing instead that we must distinguish between total freedom, the freedom to always do just as one wants, free from any legal constraint, versus valued liberty, understood as the freedom necessary to preserve the ethical independence of individuals.²⁰ Although some choices restricted to those who have attained a minimum legal age threshold, including those involved in voting, military service, and marriage, are centrally related to establishing and maintaining the ethical independence of individuals, the choice to use alcohol or tobacco does not obviously connect to a person’s core beliefs and values. Consequently, the state should have greater latitude to regulate tobacco and alcohol, including through the use of higher minimum age thresholds, than to regulate activities pertaining to valued liberties.

Rebutting the Paternalistic Objection

A policy is paternalistic if it overrides the autonomous decision making of an individual to promote that individual’s well-being. However, both law and political theory have long recognized that immaturity can preclude the capacity for autonomous thought. If individuals younger than 21 years are not yet capable of autonomous and voluntary decision making, then their preferences are not entitled to the same respect as are the preferences of those considered fully autonomous. Thus, Tobacco 21 laws would not be truly paternalistic because they would not interfere with the decision making of a fully autonomous individual.

A growing body of evidence from psychology and neuroscience research suggests that individuals younger than 21 may not yet have this capacity, at least for some types of decisions. By age 18 years, most adolescents have attained cognitive maturity equivalent to that of adults, including the ability to understand the consequences resulting from their decisions, to process information, and to reason. Nevertheless, other aspects of psychosocial development, those that may be particularly influential to the decision to use tobacco, may still be incomplete.² For example, young adults demonstrate reduced impulse control, greater susceptibility to peer influence, and greater sensation- and reward-seeking behaviors.² This evidence related to the extended time frame for adolescent psychosocial development provides 1 form of justification for limiting the liberty of this population. As Conly has argued, respecting an individual whose cognitive development is still under way may not be best achieved by giving him or her the freedom to make poor choices.²¹

To be clear, the argument that adolescents may not be fully capable of making autonomous decisions related to tobacco should not be taken to imply that the minimum legal age for all activities should be raised to 21 years. Different brain processes evolve at different rates. Respecting adolescents may therefore require different minimum legal ages for different types of decisions.²²

Further support for the claim that Tobacco 21 laws are not truly paternalistic comes from the possibility that such laws actually serve to promote an individual’s future autonomy. If adolescents are not yet fully capable of autonomous decision making, then what is required to respect adolescents might not be giving them full latitude to make any and all decisions, but instead to delay them from making certain decisions until their capacity to do so has fully matured.²³ In the case of Tobacco 21 laws, delaying the age at which an individual can purchase tobacco can help ensure that at least the initial decision to use tobacco is, in fact, autonomous. Thus, whereas Tobacco 21 laws restrict liberty, the restriction is only temporarily compromised. Tobacco 21 laws may therefore be an example of a case in which a temporary restriction, perhaps counterintuitively, actually works to enhance long-term autonomy.²³

Delaying initiation may have additional impacts on autonomy. Adolescents are especially susceptible to tobacco addiction.² Consequently, delaying when individuals can legally initiate tobacco use may reduce the likelihood that they will become addicted. Fully addressing the relationship between addiction and autonomy is beyond the scope of this article; nevertheless, that a substantial proportion of smokers report desiring to quit but continue to smoke suggests that addiction is inconsistent with autonomous decision making.²⁴ The risk of tobacco addiction makes smoking what Taylor describes as a “constraining option,” understood as an option that, if chosen, “is likely to impair the ability of the person who chooses [it] to exercise her own autonomy post-choice.”^25(p85) Thus, selecting the option to smoke likely limits individuals’ ability to exercise their autonomy postchoice. If so, this would offer support for the argument that “it is all right to interfere with some of a person’s liberties to promote (or preserve) a greater range of liberty on balance.”^26(p95–96)

In light of these considerations, it seems reasonable to argue that, by withdrawing a short-term freedom, Tobacco 21 laws may promote long-term autonomy, helping to ensure that both the initiation and subsequent use of tobacco are autonomous decisions. As a result, the liberty burden of Tobacco 21 is less extensive than its detractors assert, and thus the criticism of unwarranted paternalism loses some of its moral force.

Justifying the Liberty Restriction

We have offered arguments to show that the critics’ claims of inconsistency and of unwarranted paternalism are not as persuasive as they first appear. Even if it were maintained that Tobacco 21 laws raise substantial concerns about liberty restriction, it would remain to demonstrate that these laws are unethical. There may be conflicting ethical considerations that support their adoption, leading to the judgment that such laws are, overall, ethically defensible.

Public health ethics, like liberalism more broadly, generally accepts that it is permissible to restrict individual freedom to prevent harm to others.^27,28 In the case of Tobacco 21, the freedom of those aged 18 to 20 years is restricted at least in part to prevent harm to younger individuals, because those aged 18 to 20 years, when able to purchase tobacco products, put younger individuals at risk. Additional ways that tobacco use may cause harm to others include exposure to secondhand smoke, distracted driving, and increased health care costs.

Assessing the permissibility of such a liberty restriction requires consideration of 3 factors.^29,30 First, how substantial is the harm that is to be prevented? Second, how likely is the proposed liberty restriction to mitigate that harm? Third, is there a less restrictive alternative that could achieve a similar benefit?

With respect to the first question, the harm to be prevented is substantial. Tobacco use is responsible for nearly half a million deaths each year. Furthermore, the harm to be prevented is harm to minors—a population that, by virtue of age, can be considered vulnerable and thus deserving of additional protections.

With respect to the second question, empirical evidence supports the claim that restricting the liberty of those aged 18 to 20 years can be an effective strategy by which to mitigate that harm. Ninety percent of persons who purchase cigarettes for distribution to minors are younger than 21 years.³¹ Preventing tobacco sales to those aged 18 to 20 years can reasonably be expected to substantially decrease both current and future tobacco use.

Finally, less restrictive means seem unlikely to secure a similarly sizable public health benefit. A common objection from opponents to public health laws is that enforcement of existing laws should be enhanced, rather than that new restrictions should be passed. In the case of Tobacco 21, this argument implies that efforts should focus on restricting social transfers of cigarettes, rather than prohibiting legal sale to individuals younger than 21 years. However, providing cigarettes to minors is already illegal in 48 states, and such transfers are nevertheless routine, suggesting substantial enforcement challenges.

An alternative objection might be that limiting restrictions to a younger age, such as 19, might reduce social transfers, with less burden on individual liberty. Because of the relative novelty of these laws, no comparative studies exist to demonstrate that the public health benefits could not be achieved with a lower age threshold. However, evidence from modeling studies suggests that a lower age would be less effective. In its report, the National Academy of Medicine estimated that raising the minimum age of sale to 19 years would have only a “modest” impact on reducing social sources for those younger than 18 years. According to the National Academy of Medicine, the impact on tobacco initiation by minors from raising the minimum age of sale to 21 years would be “substantially higher” than would raising it to 19 years.² Specifically, they estimated that the reduction in initiation by those younger than 15 and those aged 15 to 17 years would be 5% and 10%, respectively, if the age was raised to 19 years, compared with 15% and 25%, respectively, when raised to 21 years.² Among the potential reasons for this difference is that, whereas the majority of high school students are aged 18 years or younger, a substantial proportion of individuals older than 18 years are still enrolled in high school—and therefore are likely to be in social networks with those younger than 18 years.²

The 3 criteria therefore appear to be met by Tobacco 21 laws, creating an ethical justification for limiting the liberty of an individual to prevent harm to another.

An Argument by Analogy

Raising the minimum age to 21 years to prevent harm also finds strong historical precedent in the establishment of the US minimum age for alcohol consumption. Following the repeal of Prohibition, most states set their minimum legal drinking age (MLDA) at 21 years (MLDA 21), consistent with the contemporary voting age. When the Twenty-Sixth Amendment was passed in 1971, lowering the voting age to 18 years, several states also lowered their MLDA. However, numerous studies demonstrated that lower MLDA was associated with increases in motor vehicle fatalities.³² In response, many states increased their MLDA to 19, 20, or 21 years. Ultimately, the federal government passed the National Minimum Drinking Age Act in 1984, requiring states to set their MLDA at 21 years or risk losing 10% of their annual federal highway apportionment.³³

The criticisms of MLDA 21 closely resembled those of Tobacco 21, with critics arguing that individuals who were old enough to marry or die in war were similarly old enough to drink. However, these justifications were generally overridden by 3 ethical appeals in support of MLDA 21—appeals that are similarly applicable to Tobacco 21. First, a strong body of empirical evidence demonstrated that MLDA 21 was associated with fewer alcohol-related harms, particularly those associated with motor vehicles.³³ Second, as described by a federal district court in a challenge to Michigan’s MLDA 21 law, drinking alcohol is not a fundamental right, such as those guaranteed by the US Constitution.³⁴ Third, MLDA 21 was supported by the appeal of preventing harm to others, as alcohol-impaired drivers present harm not only to themselves but also to other road users. These 3 ethical arguments mirror the arguments from consequences for public health, justifications for restriction of liberty, and prevention of harm to others.

One difference between implementing a MLDA 21 law and a Tobacco 21 law relates to tobacco’s addictive nature. Those aged 18 to 20 years who use tobacco at the time the law goes into effect will lose the ability to purchase a substance they have become addicted to. One solution might be to “phase-in” the law, gradually raising the legal purchase age to ensure that no individual who was once permitted to purchase tobacco loses that ability. This approach, however, has the downside that additional youths are likely to begin legally using tobacco during the phase-in period, making the harm associated with the gradual approach significant. This is a difficult but short-term pragmatic challenge associated with implementation that will require additional analysis to resolve.

CONCLUSIONS

Convincing empirical evidence suggests that Tobacco 21 laws can bring about substantial public health benefits, preventing thousands of young people from experiencing the negative consequences of tobacco use. Although critics of these laws have argued that they are inconsistent with other public policies regarding those aged 18 to 20 years and that they are unduly paternalistic, we have argued that these objections are not persuasive. We have further described a competing ethical consideration, prevention of harm to others, that outweighs lingering concerns about liberty restriction. Ultimately, our analysis suggests that Tobacco 21 laws are ethically appropriate. Policymakers should move quickly toward adopting these laws in jurisdictions where they are not yet in effect and in so doing taking an important step to protect the health of current and future adolescents.