Lesbian, gay, bisexual, and questioning (LGBQ) teenagers face unique psychological challenges.1 Religious critiques of homosexuality (Blosnich et al. 2020; Hatzenbuehler et al. 2012), societal discrimination (Badgett 1995; Badgett et al. 2021), and family conflict surrounding sexual identity (Donahue et al. 2017; Ryan et al. 2009) generate substantial unique stresses that contribute to sharp disparities in mental health between LGBQ and heterosexual youth (Eckstrand et al. 2022; Hatzenbuehler et al. 2024). LGBQ teens are more likely to experience depressive symptoms (Irish et al. 2019) and engage in suicidal behaviors (Liang et al. 2023; Rees et al. 2022; Sun et al. 2023) than their heterosexual counterparts. These disparities likely contribute to higher rates of risky health behaviors for sexual minorities, in part as a coping mechanism for psychological trauma (Earnshaw et al. 2017; Friedman 2020; Rosario et al. 2014).

Nicotine use among LGBQ-identifying teenagers is also substantially higher than among heterosexual youth. In 2023, prior-month use of electronic nicotine delivery systems (ENDS) among LGBQ teenagers surveyed in the State Youth Risk Behavior Surveys was 45 percent higher (20.6 percent versus 14.2 percent) and combustible cigarette smoking was 81 percent higher (4.7 percent versus 2.6 percent) than among heterosexual teens (Centers for Disease Control and Prevention [CDC] 2025a). These differences may lead to disparities in mortality and morbidity later in life, as combustible tobacco use is the leading cause of preventable death in the United States (U.S. Surgeon General 2014; 2020). While ENDS are generally considered less harmful than combustible tobacco products, vaping may still carry health risks relative to abstinence from all products containing nicotine (McNeill et al. 2018; Mohammadi et al. 2022; U.S. Surgeon General 2016).

LGBQ and heterosexual individuals may differ in both motivations for tobacco use and responsiveness to tobacco control policies. The “minority stress hypothesis” (MSH) suggests that tobacco use among LGBQ individuals may serve as a coping response to stressors associated with stigma, discrimination, bullying, and social pressure (Meyer 1995; Badgett et al. 2024; Blosnich et al. 2020; Donahue et al. 2017; Kosciw et al. 2016; Liang et al. 2023; Ryan et al. 2009; Schuster et al. 2015). LGBQ youths may be less responsive to policies intended to curb vaping and smoking if they turn to an addictive good to cope with or escape from stress-related psychological trauma (Saffer and Dave 2005). Differences may also arise from social context. LGBQ young adults are more likely to smoke in social settings, reflecting the historical importance of bars and clubs as safe spaces within LGBQ communities, which may increase exposure to tobacco use and peer influences relative to heterosexual individuals (Jannat-Khah et al. 2018; Remafedi 2007). Whether analogous mechanisms operate for nicotine vaping is unclear. In addition, LGBQ youths’ peer-based social networks may facilitate access to illicit markets, potentially attenuating the effects of ENDS and cigarette policies (McPherson et al. 2001). Finally, these patterns may be reinforced by targeted marketing, as ENDS producers, similarly to combustible tobacco firms, actively sponsor LGBQ-oriented events such as pride festivals (Truth Initiative, 2024).

Each of these factors suggests a stronger attachment to nicotine that can be framed using economic models of addiction. Models of both myopic and rational addiction assume “tolerance,” in which a stock of addiction accumulated from past consumption yields a negative marginal utility in the present period, and “reinforcement,” which means that the positive marginal utility from current consumption rises with the stock of past consumption (Cawley and Ruhm, 2011). These features create a feedback loop in which greater past consumption leads to a stronger addiction, which creates a greater disutility that can be alleviated by consumption in the present period, but this new consumption reinforces the addiction. Therefore, consumption at any given time depends not only on current prices but also on past prices, as those impact the current addictive stock. This in turn implies that the long-run price elasticity exceeds the short-run elasticity. If minority stress induces LGBQ teens to, on average, start using nicotine earlier and consume it more frequently than heterosexual teens, this would translate to a larger accumulated addiction stock and therefore stronger reinforcement to continue using it. Intuitively, then, we would expect LGBQ teens to have more difficulty quitting when prices rise.2

Importantly, these models of addiction do not incorporate the availability of a close substitute with the same addictive properties as the good subject to the price intervention. This availability could make the own-price responsiveness of more attached users similar to that of less attached users, with more attached users instead being more likely to switch to the substitute rather than quit entirely. Thus, the availability of close substitutes could mute differences in own-price responsiveness between LGBQ and heterosexual teens.

In this paper, we study the effects of ENDS and cigarette policies on ENDS and cigarette use among LGBQ and heterosexual individuals, paying particular attention to whether the results support or refute the MSH. We focus on four policies that have been widely studied and shown in previous work to affect youth or young adult vaping or smoking: ENDS taxes (Abouk et al. 2023a; Allcott and Rafkin 2022; Cotti et al. 2022; Friedman and Pesko 2022; Pesko et al. 2020), cigarette taxes (Carpenter and Cook 2008; Cotti, Nesson, and Tefft 2017, 2018; Nesson 2017), ENDS flavor bans (Cotti et al. 2025; Friedman et al. 2024a; Friedman et al. 2024b; Friedman et al. 2025; Saffer et al. 2024), and ENDS minimum legal sale age (MLSA) laws (Abouk and Adams 2017; Dave et al. 2019; Friedman 2015; Pesko 2023). While economic models of addiction focus on prices, the logic that greater attachment makes quitting harder when incentives change also applies to non-price-based interventions such as MLSA laws and flavor bans. These regulations impact the “full” price of ENDS use, which includes time and legal risk. Moreover, under full enforcement and compliance, these regulations would effectively lead to infinite prices for the focal age group and products.

Using data from the State Youth Risk Behavior Survey (YRBS), we connect variation in ENDS use and cigarette smoking among heterosexual and LGBQ teens to changes in the tobacco control policy environment in a generalized differences-in-differences framework. We find that ENDS taxes and MLSA laws generate economically meaningful and statistically robust reductions in vaping among heterosexual teens, while the corresponding estimates for LGBQ teens, who exhibit substantially higher baseline ENDS use, are uniformly small and statistically indistinguishable from zero. In contrast, cigarette taxes have no effect on ENDS use or cigarette smoking among either demographic group, consistent with Hansen et al. (2017), who find that cigarette taxes “lost their bite” on youth tobacco use in the 2010s. With respect to ENDS flavor bans, we find imprecise negative effects on ENDS use, consistent with Cotti et al. (2025a), but no conclusive evidence of differential impacts on LGBQ and heterosexual youth.

We further explore the MSH by investigating heterogeneity tied to mental health-related stressors among LGBQ and heterosexual teens. Among LGBQ teens without mental health stressors (measured using reports of school bullying victimization, persistent sadness/depression, or suicidal ideation), ENDS taxes and flavor bans produce large and statistically meaningful reductions in vaping that are, in some cases, even larger than those for heterosexual teens. In contrast, LGBQ teens who do report these stressors exhibit small and non-statistically significant responses to the same policies and, for flavor bans, show substitution toward cigarette smoking that nearly offsets reductions in ENDS use. This pattern of strong policy responsiveness among psychologically healthier LGBQ teens, but muted or offsetting behavioral responses among those facing bullying or mental health challenges, provides further evidence consistent with the MSH. Auxiliary analyses of adults using the Behavioral Risk Factor Surveillance System (BRFSS) provide weaker statistical evidence for differential ENDS policy responses by sexual identity. We conclude that evidence for minority stress as the primary mechanism to explain heterogeneous treatment effects among LGBQ and heterosexuals is present among youths but mixed across the set of policies we explore and largely absent for adults.

Our work provides two main contributions to the literature. First, we build on the work of Carpenter and Sansone (2021) by being the first, to our knowledge, to estimate the impact of ENDS and cigarette policies on LGBQ youth and adult use of nicotine products. Given the potentially important differences in motivations for nicotine vaping among LGBQ youth as compared to heterosexual youth, and potentially substantial long-run health costs of addiction to tobacco products, examining how these policies differentially impact sexual minority teens is important. Second, we explore possible explanations for heterogeneity in the policy effects across sexual minorities and heterosexual teens, particularly the role of the MSH and mental health.