Introduction Understanding the impact of prices for tobacco and nicotine products is critical for creating policies to prevent use among young people. This study examines the impact of electronic cigarette (e-cigarette) and cigarette prices on current e-cigarette and cigarette use among youth and young adults.
Methods Data were from a national probability-based sample aged 15–21 collected in 2014 and followed every 6 months for 2.5 years through 2016. We conducted separate conditional likelihood logistic regression models with past 30-day e-cigarette use and past 30-day cigarette use outcomes on the sample of individuals who participated in at least two survey waves (n=11 578) with linked Nielsen market-level price data for rechargeable e-cigarettes and cigarettes. Models controlled for time-varying variables at the individual and state policy levels, and fixed effects at the individual, wave and market levels.
Results Higher cigarette prices were associated with increased past 30-day e-cigarette use, indicating e-cigarettes may serve as a substitute for cigarettes. We did not find a statistically significant relationship between rechargeable e-cigarette prices and past 30-day e-cigarette use; neither did we find a significant relationship between rechargeable e-cigarette prices and past 30-day cigarette smoking.
Conclusion This is the first study to examine e-cigarette and cigarette prices on e-cigarette and cigarette behaviour longitudinally among young people. Findings suggest the need for better measuring the costs associated with e-cigarette use among this population, as well as a careful assessment of price and tax policies that takes into account cross-product impact to sufficiently discourage e-cigarette and cigarette use among young people.
- electronic nicotine delivery devices
- non-cigarette tobacco products
- public policy
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
A key public health goal is to prevent electronic cigarette (e-cigarette) use among young people while also preventing and reducing cigarette smoking. Among the most effective strategies for reducing tobacco use are tobacco tax and price policies. It is well demonstrated that higher cigarette prices are associated with lower cigarette use. Substantial increases in cigarette prices deter potential users from initiating and stimulate current smokers to reduce consumption or quit.1 Further, young people are generally more responsive to changes in cigarette prices than adults, particularly for those who are transitioning from experimentation to regular use.1 2
While the literature on cigarette price, cigarette demand and use is substantial, there is less evidence on the impact of prices on use of alternative tobacco and nicotine products such as e-cigarettes. Further, there is little research regarding the demand for e-cigarettes and cigarettes in the context of varying prices for each. If the two products are substitutes, a higher price for one relative to the other can decrease use of the former and increase use of the latter. If the products are complements, or commonly used together, a price increase in one will lead to reduced use of both. Recent literature examining price and sales data at the market level in the USA and European Union has found that the price elasticities of demand for disposable and rechargeable e-cigarettes are generally more elastic than that for cigarettes,3–7 indicating that changes in e-cigarette prices will induce relatively larger changes in e-cigarette demand. Data on product substitution and complementarity have been mixed, with some studies finding that e-cigarettes are substitutes for cigarettes3 4 8–10 and others finding that the products are neither substitutes nor complements.5–7
To the best of our knowledge, only one published peer-reviewed study has looked at the impact of e-cigarette and cigarette prices on e-cigarette use behaviour among individuals in a non-experimental setting. Using the national Monitoring the Future study, Pesko et al 11 found that market-level disposable e-cigarette prices in 2014–2015 were not associated with past 30-day use among middle and high school youth, although higher disposable prices were significantly associated with a reduction in days vaping among youth overall and among current e-cigarette users. Pesko et al used e-cigarette refills (ie, bottles of liquid used with rechargeable cigarettes) as a proxy for rechargeable e-cigarette prices and found that refills were not associated with past 30-day e-cigarette use or days vaping. The impact of cigarette prices on e-cigarette use had mixed effects, with the primary models showing no relationship with e-cigarette use or days vaping, and secondary models finding a statistically significant positive relationship for days vaping only.11 However, this study was cross-sectional and could not evaluate the causal impact of prices on youth e-cigarette use over time. The study did not examine the relationship between e-cigarette prices and cigarette use.
As of January 2019, nine states and Washington DC have implemented taxes on e-cigarette products and/or related and component parts.12 13 Given the rapidly changing policy environment and a dynamic tobacco and nicotine marketplace, examining the impact of prices on e-cigarette and cigarette use longitudinally is important for understanding evolving youth and young adult tobacco use patterns and informing policy. This study examines the impact of e-cigarette and cigarette prices on current e-cigarette use and cigarette use among youth and young adults using a nationally representative longitudinal cohort.
This study uses data from the Truth Longitudinal Cohort (TLC), a nationally representative, probability-based sample. Detailed TLC methods and response rates have been described and are available elsewhere.14 Briefly, the cohort consists of youth and young adults aged 15–21 at baseline recruited via address-based sampling with online data collection. Respondents completed the baseline survey from April to July of 2014, and were followed for four additional waves at 6-month intervals, with wave 5 data collected from July to October 2016. Refreshment samples of 1000 additional participants were added to the total sample at waves 2 through 5, and followed for all remaining waves. The sample for this study reflects data collected at baseline through wave 5, but includes only participants who completed two or more surveys during this period and who had Nielsen price data (n=11 578). Analytical samples included only those individuals with change over time on the outcome variable for each analysis.
Two outcomes were examined: current e-cigarette use and current cigarette use. Current e-cigarette use was defined as self-reported past 30-day e-cigarette use among the full sample: ‘During the past 30 days, on how many days did you use e-cigarettes (like Blu, NJOY, etc)?’ Those who reported using e-cigarettes on one or more days were categorised as current e-cigarette users (1=e-cigarette user, 0=non user). Current cigarette use was defined as self-reported past-30 day cigarette use among the full sample: ‘During the past 30 days, on how many days did you smoke a cigarette?’ Those who reported using cigarettes on one or more days were categorised as current cigarette smokers.
This study used quarterly price data of rechargeable e-cigarettes and cigarettes from Nielsen Company’s retail store scanner database from 2014 to 2016. Nielsen data provide representative dollar sales and volume estimates of products using in-store scanner equipment and field audits. The sample includes Nielsen’s convenience store channels (ie, independent and chain), food/grocery stores (eg, Publix, Safeway), drugstores (eg, Walgreens, Rite Aid) and mass merchandisers (eg, Kmart, Target) across 54 markets. We used universal product codes to identify rechargeable devices and cigarettes and obtained the total retail dollar sales and volume for each product for each market. The average rechargeable e-cigarette price per unit and cigarette price per pack was calculated by dividing the total dollar sales by the volume sold per year. Prices for rechargeable e-cigarettes were standardised based on the number of devices in each sales unit, which incorporated volume-based discounts and multi-item pack prices. Prices were adjusted to 2016 dollars using the Consumer Price Index Inflation Calculator obtained from the Bureau of Labor Statistics. We examined disposable e-cigarette prices but found extremely limited variation across markets, so we did not include disposables in this analysis. Quarterly rechargeable e-cigarette and cigarette prices were linked to respondents by location (state and county geocodes) and survey completion dates.
Models incorporated only time-varying variables. These covariates included state-level cigarette and e-cigarette policy variables for 2014–2016, including cigarette clean air laws, tobacco control expenditures per capita, e-cigarette minimum legal sale age laws and retail licensure requirements to sell e-cigarettes. Cigarette clean air laws were from the Americans for Nonsmokers’ Rights Foundation,15 and all other policy data came from the Centers for Disease Control and Prevention State Tobacco Activities Tracking and Evaluation system.12 We also looked at e-cigarette clean air laws,12 but this variable was highly correlated with cigarette clean air laws, so we conducted analyses using only the cigarette clean air policies. We did not include state-level cigarette taxes as this variable was highly correlated with the cigarette price variable. We also did not include state-level smoking prevalence and measures of individual-level cigarette use (in the e-cigarette models) or e-cigarette use (in the cigarette models) as these variables are endogenous to price variables and inclusion would bias price coefficients. We also included individual age. Detailed descriptions of variables are included in the footnotes of tables 1–3.
We calculated weighted proportions or means for sociodemographic, policy and price variables for the full sample. We ran separate conditional likelihood logistic regression models with past 30-day e-cigarette use and past 30-day cigarette use outcomes, controlling for all time-varying variables and fixed effects at the individual, wave and market levels. In these models, only individuals with within-person variation in the outcomes over time remain in the model and all others are dropped. These models control for all constant predictors, reducing potential omitted variable bias, and produce appropriate estimates of standard errors that are corrected for dependence.16 Missing data on any variable were less than 1% and these cases were deleted. We also calculated own-price and cross-price elasticity for all models.
Table 1 shows weighted individual and observation-level sociodemographics. A total of 55.7% were non-Hispanic white, 14.3% were African–American and 20.0% Hispanic. Baseline past 30-day cigarette smoking was 10.1% and baseline past 30-day e-cigarette use was 9.7%. The average rechargeable price at baseline was $18.05 (in 2016 dollars), but over the study period declined to an overall average of $13.66. Prices for cigarette packs, which contain 20 cigarettes in the USA, remained relatively stable over time at $6–$7 (see figure 1 for prices per quarter). Only 5.3% lived in a state that required licences for over-the-counter e-cigarette sales at their baseline, and this increased to an over time average of 9.5% during the course of the study. Three-quarters lived in a state with a minimum age for legal e-cigarette sales at their baseline, and this increased to an overall average of 82.4%.
Table 2 shows the odds of past 30-day e-cigarette use as a function of rechargeable and cigarette prices among the analytical sample. Models 1–3 show past 30-day e-cigarette use as a function of prices and age, with individual, wave and market fixed effects. Models 4–6 show past 30-day e-cigarette use as a function of prices, age and state tobacco and e-cigarette policy variables, with individual, wave and market fixed effects. Model 1 shows no significant effect of rechargeable prices on past 30-day e-cigarette use, model 2 shows a positive and significant effect of cigarette prices on past 30-day e-cigarette use, and the associations remain the same in model 3 with both rechargeable prices and cigarette prices. In models 4 and 6, rechargeable prices remain non-significant. In models 5 and 6, cigarette prices are attenuated with the inclusion of state policy variables, with ORs still being positive but marginally significant. In partially adjusted model 3 and fully adjusted model 6, the cross-price elasticity of cigarette prices with respect to past 30-day e-cigarette use is 0.58 and 0.94, respectively. These results indicate that a 10% increase in cigarette prices leads to a 5.8%–9.4% increase in past 30-day e-cigarette use, respectively. Age was found to be a significant predictor of past 30-day e-cigarette use; the older a youth, the more likely he/she will use e-cigarettes in the past 30 days. State-level e-cigarette minimum legal purchase age laws were found to reduce the likelihood of e-cigarette use among young people in models 4–6.
Table 3 shows past 30-day cigarette use as a function of rechargeable prices, cigarette prices and covariates. Again, models 1–3 show prices and age covariates, and models 4–6 show prices, age and state tobacco and e-cigarette policy variables. All models include individual, year and market fixed effects. Rechargeable prices and cigarette prices are not significantly associated with past 30-day cigarette use in any of the models. Similar to the results in table 2, age is a significant predictor of cigarette smoking. In addition, state-level e-cigarette over-the-counter licensing provisions were found to be negatively and significantly associated with cigarette smoking in the past 30 days.
This is the first study to examine e-cigarette and cigarette prices on youth and young adult e-cigarette use and cigarette use longitudinally. Study findings reveal that higher rechargeable prices are not associated with either current e-cigarette use or current cigarette use. In partially adjusted models, cigarette prices were positively and significantly associated with e-cigarette use, and in fully adjusted models this relationship was attenuated but remained marginally significant. Depending on model specifications, the results indicated that a 10% increase in cigarette prices could increase e-cigarette use by 5.8%–9.4%. These findings indicate a trend of e-cigarettes as a possible substitute for cigarettes. However, substitution effects may be limited. The results indicated no significant relationship or trend for e-cigarette prices and cigarette use. These findings suggest that cigarettes may not be a substitute for e-cigarettes. In other words, individuals may be more inclined to use e-cigarettes when cigarette prices increase, while an increase in e-cigarette prices alone may not provide enough incentive to use cigarettes.
Our findings are similar to Pesko et al.11 Both studies found no significant relationship between rechargeable prices and past 30-day e-cigarette use. To be clear, lack of a significant relationship between e-cigarette prices and past 30-day e-cigarette use does not necessarily mean there is no influence of e-cigarette price on e-cigarette use among youth and young adults given that our models primarily captured aspects of e-cigarette participation, not levels of consumption. Studies that have examined the impact of e-cigarette prices on overall consumption of e-cigarettes using retail sales data have clearly demonstrated that higher e-cigarette prices reduce the sales and consumption of e-cigarettes.3–7 We believe the lack of a relationship between e-cigarette prices and e-cigarette participation reflects the difficulties in properly identifying and estimating the impact of prices on youth e-cigarette use behaviours. Several factors make these effects difficult to detect. Some research suggests that past 30-day e-cigarette use may largely reflect experimentation on the part of young people.17 If this is the case, individuals may be less likely to purchase their vaping devices for use versus using devices from friends and/or others. In addition, young individuals are less likely in general to purchase e-cigarettes at retail stores. Data from the 2016 National Youth Tobacco Survey show that among older teens, who range in age from 15 to 19 years old in the survey, only 26% of those who tried e-cigarettes in the past 30 days purchased them while the rest obtained e-cigarettes directly from another person (ie, family member, friend or other person). Of those who purchased, 27% purchased at a retail store, whereas 44% purchased at vape shops.18 Patterns are similar among young adults. A national survey found that only 24% of vapers aged 18–24 bought their e-cigarettes at retail, the lowest among all adult age groups, while 48% purchased at a vape shop.19 The e-cigarette price measure used in our study was constructed using retail store sales data, which may not accurately capture the costs of e-cigarettes used by young people and the wide range of products sold in vape shops. This e-cigarette price measure also did not capture the differences in the fixed costs of devices and subsequent recurring costs of e-liquid refills.
Youth and young adult e-cigarette purchasing patterns underscore the limitations of using Nielsen price data as a proxy for all e-cigarette prices. These data do not include price data from vape shops, online e-cigarette websites and places such as mall kiosks. Price data from these latter sources are not available, and this exclusion could limit the ability to detect relationships between prices and use. New and creative research is needed to capture data on e-cigarette prices across vape shop and online e-cigarette retail sites in order to improve our ability to evaluate the impact of e-cigarette prices on use, especially among younger individuals.20 E-cigarette price data from new sources should be compared with Nielsen e-cigarette prices in similar locations. Such data could be used in studies such as this one to explore via sensitivity analyses and other strategies21 a feasible range or set of price values that would incorporate average prices for e-cigarettes across vape shops, online sites and Nielsen retail stores.
The findings also showed a trend that e-cigarettes may serve as a substitute for cigarettes in the e-cigarette use model, but there was no such reverse relationship of cigarettes serving as a substitute for e-cigarettes in the cigarette use model. These findings align with studies of aggregate market demand that have examined cross-price elasticity of e-cigarette and cigarette prices. Zheng et al 4 found that e-cigarettes and cigarettes are substitutes, but the impact of cigarette prices on e-cigarette sales was much stronger than the impact of e-cigarette prices on cigarette sales. This may be partially due to differences in market share for cigarettes and e-cigarettes, but it may also be due to differences in actual substitution behaviours. Cigarette smokers may be more likely to substitute e-cigarettes for cigarettes with higher cigarette prices either in an attempt to quit cigarettes,22–24 or due to consideration of others,22 24 the flavour appeal of e-cigarettes 22 23 or perceptions that e-cigarettes are less harmful than cigarettes.22–25 Those who use e-cigarettes may be less likely to use cigarettes as a substitute given widespread knowledge of the harms of cigarettes25 as well as cigarette-related stigma.26 Grace et al’s experimental study of simulated demand among smokers found that smokers rate e-cigarettes 81% as favourably as own-brand cigarettes.8 No studies have looked at favourability of cigarettes among e-cigarette users or potential users, but findings here suggest it may be lower than Grace et al’s findings for e-cigarette favourability among smokers.
Reviews of price elasticity studies show that youth and young adults are generally highly price-sensitive with respect to cigarette consumption.2 27 28 In this study, however, we did not find a statistically significant relationship between cigarette price and cigarette smoking in the past 30 days among the youth cohort from 2014 to 2016. Several time-varying factors not included in this analysis may either mask or change the relationship between cigarette prices and cigarette use among young people. For example, large national antismoking campaigns targeting youth, young adults and adults ran during this time period. Evidence indicates these campaigns successfully reduced smoking among all age groups,29–36 and may have also indirectly impacted smoking-related social norms through parental and peer effects and smoking denormalisation.30 37–41 These factors may have masked the relationship between price and rate of smoking. Other factors include more general changes in the economy that may have made cigarettes more affordable. During the time period of this study, there was limited change in average cigarette prices (see figure 1) and cigarette taxes.42 At the same time, inflation43 and gas prices44 declined and remained low, thus increasing individual purchasing power for products such as cigarettes. Finally, the tobacco market-place was changing rapidly during this time, with the introduction of new products45 and evolving tobacco product use patterns among young people.1 46–49 Experimental research finds that the availability of alternatives in the market-place can alter demand for cigarettes in unexpected ways.10 Further research including experimental, observational and qualitative research on youth tobacco use patterns and purchasing behaviours would be valuable in better understanding the changing effects of price across multiple tobacco products in the current market.
Strengths of this study include a rigorous analysis and the use of a probability-based national longitudinal sample of youth and young adults followed for more than 2 years and the use of exogenous market-level e-cigarette and cigarette price data on all tobacco retail outlets tracked by Nielsen nationally. Limitations include the use of Nielsen price data as a proxy for all e-cigarette prices, as discussed earlier, and the fact that market-level price data were missing for some respondents. Retail pricing may not be as relevant for youth who are currently accessing e-cigarettes through non-retail sources such as friends and family. In addition, rechargeable devices are a diverse product class and generally require other products such as cartridges and/or e-liquid refills for use. Thus including only device cost does not capture all the costs and payments related to use. As described above, Pesko et al 11 used e-cigarette refills as a proxy for rechargeable device and also did not find relationships with past 30-day e-cigarette use. We did not ask youth about the type of e-cigarette devices used, so we were not able to link rechargeable prices with the specific product type used. Some youth who reported e-cigarette use may have been using disposable products. At the time this study was conducted, however, rechargeable e-cigarette use was much more common than disposable use among youth and young adults.50–52 With the rapid rise of pod mod systems that currently dominate the market,45 future research should ask specifically about device type used when assessing e-cigarette use.
Future longitudinal research is needed to assess many critical questions regarding youth and young adult tobacco and nicotine use in the context of varying prices for a wide range of e-cigarette products, as well as cigarettes and other combustible and non-combustible tobacco products. Examples of such research include analyses on important questions of e-cigarette initiation and progression among youth and young adults overall and among specific populations. For example, research examining the impact of e-cigarette and cigarette prices on cigarette consumption among young low-risk and high-risk non-smokers, as well as among cigarette smoking youth and young adults, is needed. The controversial 53 yet growing body of evidence that links e-cigarette use among youth to later cigarette use54 ,53 underscores the need to continue to monitor the impact of price on both e-cigarette and cigarette use among youth and young adults.
As products in the tobacco and nicotine market-place multiply and consumers develop complex new patterns of use, the issue of relative prices for tobacco and nicotine products becomes increasingly important.55 This study’s findings demonstrate that e-cigarettes may serve as substitutes for cigarettes but cigarettes did not appear to serve as substitutes for e-cigarettes during the time period from 2014 to 2016. Findings suggest the need for better measuring the costs associated with e-cigarette use among youth and young adults, as well as a careful assessment of price and tax policies that takes into account cross-product impact to sufficiently discourage e-cigarette and cigarette use among young people. Evidence on price policies and e-cigarette use is still in its infancy. Research is critically needed to understand the impact of e-cigarettes, cigarettes and other tobacco product prices on elasticity and substitution among specific groups of users, particularly users of combustible and non-combustible products with varying patterns of tobacco use.
What this paper adds
Understanding the impact of prices for tobacco and nicotine products is critical for creating policies to prevent use of these products among young people.
This is the first study to examine the impact of electronic cigarette (e-cigarette) and cigarette prices on e-cigarette and cigarette behaviour longitudinally among young people.
The findings demonstrate that e-cigarettes may serve as a substitute for cigarettes.
Rechargeable prices were not associated with e-cigarette use.
Study findings suggest the need for better measuring of the costs associated with e-cigarette use among youth and young adults, as well as the need for a more thoughtful approach of price and tax policies that takes into account cross-product impacts to discourage initiation and use of cigarettes and e-cigarettes among young people.
Given the rapidly changing technology of the e-cigarette market, the relationship between e-cigarette prices and tobacco product use among young people needs further research.
Contributors JC conceptualised the study and analyses and wrote the paper. JH contributed to the analyses and revisions. HX conducted the analyses. MSG, ECH and DV contributed to revisions.
Funding This study was funded by Truth Initiative.
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement Data are available upon reasonable request. All data relevant to the study are included in the article or uploaded as supplementary information.