Article Text
PDF
Abstract
Background We investigated the impact of cigarette price differences across the European Union (EU) on cross-border tobacco purchasing because of cheaper price among current cigarette smokers.
Methods Individual-level tobacco-related data (including cross-border tobacco purchasing behavior) were from the Special Eurobarometer 385 (V.77.1), a cross-sectional survey of persons aged ≥15 years from 27 EU Member States during 2012. Country-specific weighted average prices (WAP) per 1000 cigarettes (as of 1 July 2012) were obtained from the European Commission, and divided by 50 to yield WAP per cigarette pack. The dispersion in EU cigarette prices was measured with the coefficient of variation. Multivariate logistic regression was applied to measure the relationship between EU-wide cigarette price differential and cross-border tobacco purchasing because of cheaper price among current cigarette smokers (n=6896).
Results The coefficient of variation for cigarette WAP within the EU was 0.39 (mean price=€3.99/pack). Of all current cigarette smokers in the EU, 26.2% (27.5 million persons) engaged in a cross-border tobacco purchase within the past 12 months, of which 56.3% did so because of cheaper price in another country. EU-wide cigarette price differential was significantly associated with making a cross-border tobacco purchase because of cheaper price (adjusted OR=1.34; 95% CI 1.22 to 1.47).
Conclusions Reducing differences in cigarette tax and price within the EU, coupled with a stricter limitation on the quantity of cigarettes that it is possible to carry from one Member State to another, may help reduce cross-border tax avoidance strategies.
- Economics
- Global health
- Prevention
- Price
- Taxation
Statistics from Altmetric.com
Introduction
There is ample evidence that increasing tobacco taxes reduces tobacco use, particularly among youths and individuals of low socioeconomic status.1–6 Tobacco users have, however, been known to adopt various tax avoidance strategies in response to increased tobacco taxes, including switching to relatively cheaper tobacco products, or engaging in distance tobacco purchases (eg, internet-based or cross-border purchases) because of cheaper prices.7–10 Tobacco price differences within a country or an economic region may reflect differences between its provinces or Member States (MS) regarding tobacco taxation, income distribution or other economic factors such as tobacco production. Such tobacco price heterogeneity could potentially undermine local efforts to reduce tobacco use,10 and could exist even in the presence of robust tax harmonisation systems. For example, despite having one of the most successful regional tax harmonisation structures, the MS of the European Union (EU) show greater variability in cigarette prices than the USA, which has no harmonised tax structure among its 50 states and Washington DC.11–13 Specifically, during 2010, the coefficient of variation for cigarette prices within the USA was 0.20 compared with 0.44 for the EU.11 ,12
This marked cigarette price variability across the EU coupled with its single economic market system creates a favourable environment for cross-border tobacco purchases. Within the EU, no limits exist on what private persons can buy and take with them when travelling between EU MS, as long as the products purchased are for personal use and not for resale.14 Taxes (value added tax and excise) are included in the price of the products in the EU MS of purchase and no further payment of taxes can be due in any other EU MS provided the amount of tobacco purchased is below certain specified thresholds, namely: 800 cigarettes, 400 cigarillos, 200 cigars, or 1 kg of smoking tobacco—each amount representing 100% of the total allowance for tobacco products.14 For purchases originating from non-EU countries and transported into an EU MS, the maximum quantities for which private individuals are exempted from value added tax and excise duty imports are 200 cigarettes, 100 cigarillos, 50 cigars, or 250 g of smoking tobacco.15
In recent years, extensive tobacco control reforms have been instituted in the EU which have the potential to narrow tobacco price differences, and ensure stronger regulation of cross-border tobacco transactions. These regulatory reforms are embodied within two recent EU Directives, namely: (1) The Tobacco Taxation Directive (of 21 June 2011) which became operational by 1 January 2014, providing a mandate to MS regarding the structure and rates of excise duty applied to cigarettes and certain other manufactured tobacco products.16 Under the new rules, excise duties must account for ≥60% of the weighted average price (WAP), and be ≥€90/1000 cigarettes. Furthermore, the specific component of the excise duty on cigarettes must be at least 7.5%, and up to 76.5% of the total tax share. It is expected that the relatively large increase in the excise tax floor under the new tax Directive may help to narrow current differences in tobacco taxes and prices,11 ,13 which in turn may reduce any economic incentives for consumers to engage in cross-border purchases. (2) The Tobacco Products Directive (TPD; of 21 March 2014), proposed to become operational by May 2016, provides a mandate for stronger regulation of the manufacture, marketing and sales of tobacco products, including cross-border sales.17 The Directive notes “Cross-border distance sales of tobacco products could facilitate access to tobacco products that do not comply with this Directive. There is also an increased risk that young people would get access to tobacco products. Consequently, there is a risk that tobacco control legislation would be undermined. MS should, therefore, be allowed to prohibit cross-border distance sales. Where cross-border distance sales are not prohibited, common rules on the registration of retail outlets engaging in such sales are appropriate to ensure the effectiveness of this Directive.”17
In view of these unfolding regulatory reforms across the EU tobacco control landscape, nuanced information on the prevalence, patterns and determinants of cross-border tobacco purchasing across the EU is warranted. This study therefore assessed cross-border tobacco purchasing behaviour among current cigarette smokers aged ≥15 years within the EU, as well as the impact of existing EU-wide cigarette price differences on cross-border purchasing of tobacco products.
Methods
Data source
Individual-level tobacco-related data were from the Special Eurobarometer 385 (V.77.1),18 a face-to-face cross-sectional survey of 26 566 persons aged ≥15 years from 27 EU countries, conducted between 25 February and 12 March 2012 using the appropriate national language in the respective countries.18 Although the EU currently has 28 MS, the 28th country (Croatia) only became a member on 30 June 2013,19 and so was not included in this survey. All 27 MS surveyed were categorised into four subregions using the United Nations’ grouping: Western Europe (France, Belgium, Austria, Germany, the Netherlands and Luxembourg), Southern Europe (Greece, Italy, Malta, Portugal, Slovenia, Spain and the Republic of Cyprus), Northern Europe (Denmark, Ireland, UK, Latvia, Lithuania, Finland, Sweden and Estonia), and Eastern Europe (Hungary, Poland, Bulgaria, Romania, Slovakia and Czech Republic). Participants were sampled to yield representative estimates for each EU MS as well as for the EU overall.
Measures
Sociodemographic characteristics
Sociodemographic variables included age (15–24; 25–39; 40–54 or ≥55 years); sex (male or female); residence type (rural or small/large town); educational attainment, measured as age at stopping full-time education (≤15 years/no education, 16–19 years, ≥20 years, or currently studying); and, financial difficulty, assessed with the question “During the last twelve months, would you say you had difficulties to pay your bills at the end of the month…?” Response options were: ‘Most of the time’, ‘From time to time’ or ‘Almost never/never’.
Since the Eurobarometer survey did not collect information on individual-level income, we included per capita Gross Domestic Product (GDP) in Euro during 2012 as an indicator of country-level income using World Bank data.20 (Country fixed effects were not included in any of the models.) Inclusion of per capita GDP was deemed important because of the approximation of demand as a function (primarily) of price and income.
Current cigarette smoking and cross-border tobacco purchasing
Current cigarette smoking was defined as use of ‘boxed cigarettes’ (ie, manufactured cigarettes) ‘everyday’, ‘weekly’, ‘monthly’ or ‘less than monthly’. Respondents who reported that they had “never” used cigarettes, or had used it only “once or twice”, as well as those who indicated that they “used to smoke but had stopped” were categorized as current non-smokers. A cross-border tobacco purchase was assessed with the question “In the past 12 months, have you bought tobacco products in another country than (OUR COUNTRY)?” Any of the following multiple response options was categorised as a positive indication of a cross-border tobacco purchase “Yes, in another EU country”; “Yes, in a country located outside the EU”; or “Yes, but you were not sure where it came from.”
The rationale for engaging in a cross-border tobacco purchase was determined with the question “For what reasons have you bought tobacco products in another country?” Multiple response options were “The product is forbidden in (OUR COUNTRY)”; “The price of this product was cheaper than in (OUR COUNTRY)”; or “You were in another country and needed tobacco products.” The response option of “The product is forbidden in (OUR COUNTRY)” was taken to indicate smokeless tobacco, e.g., Snus, since this is the sole tobacco product type permitted for sale within an EU MS (Sweden), while being prohibited elsewhere within the EU.17 The mode of making a cross-border tobacco purchase was determined with the question “How did you buy tobacco products in another country?” Multiple response options were “Via the Internet”; “Via telephone”; “You were there” or “Someone else was there.” The amount of tobacco consumed within a year that was from cross-border tobacco purchases was determined with the question “On the following scale, what proportion of your annual tobacco consumption do you buy in another country?” Categorical responses were recoded as <5%, 5–50% or >50%.
Cigarette prices across the EU
We obtained cigarette prices (in Euro) for the 27 EU MS from two different sources. First, we used the European Commission's benchmark WAPs (per 1000 cigarettes) as of 1 July 2012. WAPs were calculated by dividing the total value of all cigarettes consumed in the preceding year (based on the retail selling price including all taxes), by the total quantity of cigarettes consumed.21 For our analyses, we divided the value of the WAP per 1000 cigarettes by 50 to yield WAP per cigarette pack. Second, we obtained prices per cigarette pack for the most popular price category (MPPC) as of July 2012 from the WHO surveillance database.22
Data analyses
The dispersion in EU cigarette prices was measured with the coefficient of variation, calculated as the ratio of the SD to the mean. Since the WAP and MPPC prices were specific for cigarettes, the denominator for all analyses within our study was restricted to current smokers of manufactured cigarettes (n=6896). Statistical comparisons of prevalence estimates were performed using the χ2 test (p<0.05). As an indicator of the reliability of prevalence estimates, the relative SE (RSE) was assessed by dividing the SE of the estimates by the estimates themselves. Estimates with RSEs ≥50% were considered statistically unreliable and not reported.
Within multivariate analyses, the primary outcome of interest was a report by a current cigarette smoker that they had engaged in a cross-border tobacco purchase because of cheaper price. We did not use a measure of ‘any cross-border tobacco purchase’ (ie, cross-border purchases made for any reason) as the outcome in order to reduce confounding by socioeconomic status since wealthy persons or working professionals are more likely to travel abroad23 and hence may be more likely to have bought tobacco products abroad while on holidays or on business. To measure the impact of EU cigarette price differences on cross-border tobacco purchase because of cheaper price, we fitted multiple logistic regression models, each adjusting for sex, age, financial difficulty, per capita GDP, residence type and educational attainment. In model 1, we assessed the general impact of EU-wide differential in cigarette WAP on cross-border tobacco purchasing because of cheaper price. In model 2, we explored the impact of country-specific differentials in cigarette prices by computing the difference in MPPC price between each country and its lowest price land or maritime boundary neighbour (whether that was an EU MS or a non-EU country). MPPC price rather than WAP was used within this second model because WAP data were available for only EU MS. For comprehensiveness, we also modelled the relationship between the number of bordering countries with cheaper cigarette MPPC price surrounding each country and the likelihood of engaging in cross-border tobacco purchasing because of cheaper price. Data were weighted using the sampling weights provided in the Eurobarometer survey, and all analyses were performed with STATA V.12.
Results
Baseline results provided by the Eurobarometer 385 survey report noted 50.6% of respondents as never smokers, 27.9% as current smokers and 21.3% as ex-smokers. Additional country-specific estimates of smoking and other descriptive results are available through the official report.24
Of all current cigarette smokers in the EU, 26.2% (27.5 million persons) engaged in any cross-border tobacco purchase within the past 12 months while 14.8% (15.5 million persons) engaged in a cross-border tobacco purchase because of cheaper price within the past 12 months (table 1).
Variation in cigarette price and prevalence of cross-border tobacco purchasing among current cigarette smokers in 27 EU MS, 2012 (n=6896)
Of current cigarette smokers who engaged in any cross-border tobacco purchase within the past 12 months, 80% purchased the tobacco products from another country within the EU while 24.9% made the purchase from a country located outside the EU (table 2). The reasons for engaging in a cross-border tobacco purchase were as follows: because of cheaper price, 56.3%; respondents were in another country and needed tobacco products while there, 44.6%; or the product purchased was prohibited or not available in their home country, 2.9%. Approximately 60.9% of current cigarette smokers who had engaged in any cross-border tobacco purchase within the past 12 months reported that they typically purchased <5% of their annual tobacco consumption from another country; 20.7% typically purchased between 5% and 50% of their annual tobacco consumption from another country; while 14.9% typically purchased >50% of their annual tobacco consumption from another country. By mode of making a cross-border purchase, 85.7% of cigarette smokers who had engaged in any cross-border tobacco purchase within the past 12 months visited the country themselves for the purchase while 15.3% bought the products through another person who was visiting the country. Notably, virtually no current cigarette smokers reported making a cross-border tobacco purchase over the internet or phone.
Characteristics of cross-border transactions made by current cigarette smokers in 27 EU MS who engaged in any cross-border tobacco purchase within the past 12 months, 2012 (n=2582)
Effect of EU-wide cigarette price differential on cross-border tobacco purchase
The coefficients of variation for WAP and MPPC cigarette prices during 2012 within the EU were 0.39 (mean price=€3.99/pack) and 0.42 (mean price=€4.70/pack), respectively. As shown in table 3, the adjusted results from model 1 indicated that the EU-wide differential in cigarette WAP was significantly associated with increased likelihood of making a cross-border tobacco purchases because of cheaper price elsewhere (adjusted OR (aOR)=1.34; 95% CI 1.22 to 1.47). Sociodemographic factors that were predictive of making a cross-border tobacco purchase because of cheaper price included being older in age and having higher levels of education. Compared with younger smokers aged 15–24 years, the odds of making a cross-border purchase because of cheaper price were significantly higher among current cigarette smokers aged 25–39 years (aOR=1.83; 95% CI 1.14 to 2.92); 40–54 years (aOR=1.97; 95% CI 1.21 to 3.20), or ≥55 years (aOR=2.30; 95% CI 1.41 to 3.76). Similarly, the odds of making a cross-border tobacco purchase because of cheaper price were significantly higher among current cigarette smokers who stopped full time education at ages 16–19 years (aOR=1.64; 95% CI 1.12 to 2.40), or ≥20 years (aOR=1.77; 95% CI 1.16 to 2.70) compared to those that had stopped full-time education at age ≤15 years or had no education.
Model 1: Impact of overall EU-wide differential in cigarette WAP on cross-border tobacco purchasing because of cheaper price among current cigarette smokers in 27 EU MS (n=6896)
Geographic contiguity with countries with lower cigarette prices and cross-border tobacco purchasing because of cheaper price
The four countries with the highest prevalence of cross-border tobacco purchasing because of cheaper price in which >25% of all cigarette smokers reported making such a purchase (France, Finland, Germany and Austria), were each bordered by at least one country with >25% lower MPPC pack price (table 1, figure 1). Across the EU, the odds of engaging in a cross-border tobacco purchase because of cheaper price were 7.97, 12.24, 23.74 and 33.25; higher among current cigarette smokers in countries bordered by 1, 2, 3 or ≥4 countries with lower cigarette MPPC prices, compared with those not bordered by any country with lower cigarette price (all p<0.05; data not shown). For each euro difference between the MPPC price of a country relative to its lowest price neighbouring country, the odds of making a cross-border purchase increased by 15% (aOR=1.15; 95% CI 1.09 to 1.22; table 4).
Model 2: Impact of country-specific differences in cigarette MPPC price relative to bordering countries* on cross-border tobacco purchasing because of cheaper price among current cigarette smokers in 27 EU MS (n=6896)
Variation in the overall proportiona of current cigarette smokers in 27 EU MS who engaged in a cross-border purchase within the past 12 months because of cheaper price, 2012. Note: Cigarette prices (pack) shown are for the most popular price category as of July 2012, obtained from the WHO database. EU, European Union; MS, Member States. aNumerator in each country was the number of cigarette smokers that had engaged in a cross-border tobacco purchase within the past 12 months because of cheaper price, the denominator was the total number of current cigarette smokers.
Discussion
Our results showed that wide cigarette price differentials existed between EU MS in 2012, and this increased the incentives for cigarette smokers to engage in cross-border tobacco purchasing because of cheaper prices elsewhere. These results are consistent with a previous study conducted within six European countries during 2006–2008, which also found higher cross-border cigarette purchasing in European regions bordering countries with cheaper prices.25 Taken together, these findings underscore the need for stronger efforts to homogenise cigarette prices between EU MS in order to strengthen national and EU-wide tobacco control efforts. A stricter limitation on the number of cigarette packs (or grams of tobacco) that it is possible to carry from one MS to another may also be a practical measure to help reduce cross-border tax avoidance strategies.
The most recent EU tobacco tax reform which raised the tax floor on cigarettes has been projected to reduce cigarette price differences as well as tobacco consumption within the EU.13 ,16 Increasing tobacco taxes reduces tobacco use because of decreases in smoking initiation, increases in smoking cessation, and reductions in daily cigarette consumption among those who still continue to smoke.4 ,6 The revised tax structure also has the potential to reduce the volume of the illicit tobacco market (eg, smuggled, bootlegged or counterfeit cigarettes) in the EU, which currently ranks as one of the largest markets for illicit tobacco trade. During 2007, over 58 billion cigarettes were illegally traded in the EU.26 A recent study also showed that more than 10% of smokers in the UK and France reported last purchasing cigarettes from low or untaxed sources.27 The tobacco industry often tries to discourage increased tobacco taxation by propagating a narrative that increased taxes are likely to increase illicit trade because of demand among consumers with the desire to save money by purchasing cheaper illegal products, as well as supply by criminal elements with the desire to make money by evading high levels of taxation.28 However, Joossens et al26 showed that the countries with the highest levels of illicit trade do not necessarily have the highest levels of tax. Lowering overall demand is an evidence-based strategy of reducing illicit trade, and increasing tobacco taxation remains one of the most effective ways of reducing demand.5 ,29
The revised EU TPD introduced new measures to combat illicit tobacco sales within the EU. Specifically, the Directive proposes a future EU tracking and tracing system at packet level for tobacco products so that compliance with the Directive can be monitored and better enforced.17 Under this directive, all unit packets of tobacco products would be required to be irremovably marked with a unique identifier containing information on the date, time and place of manufacture; the product name as well as the intended market of retail sale and shipment route, among other details. In addition to this unique identifier, all unit packets of tobacco products on the EU market would be required to carry a visible, tamper proof security feature to facilitate the identification of authentic products.15 The incorporation of such security features may help to ensure stronger tobacco control regulation among EU MS, while allowing consumers legitimate access to tobacco products not available in their domestic market.
Our results further revealed that younger smokers aged 15–24 years were less likely to engage in a cross-border tobacco purchase because of cheaper price compared with older smokers. Younger smokers in the EU are more likely to be occasional or non-daily smokers, and also smoke fewer cigarettes per day compared with older smokers.24 This lower heaviness of smoking among adolescents and young adults, coupled with the fact that cross-border tobacco purchases may be associated with additional expenses (eg, transportation costs) might make such cross-border transactions unaffordable or unattractive for youths who are generally price-sensitive. Indeed, previous research has shown that the predominant source of obtaining cigarettes among youths is through social contacts such as their peers.30 Additionally, the legal age limits for purchasing and using tobacco products which vary across the EU31 might limit the ability of adolescents to buy cigarettes from certain MS. At the same time though, these variations in legal age limits might allow individuals deemed as being underaged and ineligible to purchase tobacco products in MS with higher age limits to be able to purchase tobacco from MS with lower age limits through cross-border purchases. Implementing and enforcing harmonised age limits for purchasing tobacco products across EU MS may benefit public health by reducing youth access to tobacco products.
The strengths of the study include the use of nationally representative data within 27 EU MS. We recommend that future iterations of the Eurobarometer retain the question assessing cross-border tobacco purchase, to allow for assessments of trends in cross-border purchasing patterns and help in evaluating the effectiveness of ongoing tobacco control measures. Some limitations exist to this study. First, the cross-sectional data only permits inferences about associations rather than causal effects. Second, despite high agreement between self-reported cigarette smoking and biochemical tests.32 the self-reporting of tobacco use and cross-border purchasing behaviour could have resulted in misreporting. Third, small sample sizes may have resulted in imprecise estimates, while the absence of some individual-level data (eg, income) may have resulted in inability to completely control for some confounders. Finally, it is reasonable to assume that cross-border purchase of tobacco may be more prevalent among smokers residing close to national borders, as travel costs may be considerably lower. However, we only analysed data at a country level and did not assess for within-country variations because of lack of data.
Conclusion
This study demonstrated that cross-border tobacco purchases because of cheaper price were higher in MS with relatively higher tobacco prices in comparison to their neighbouring countries. These findings underscore the need to further reduce differences in tobacco price and tax among EU MS in order to strengthen national and EU-wide tobacco control efforts. Furthermore, additional research is needed to assess if upcoming tracking and antismuggling security features on tobacco packages may help to reduce illicit trade across the EU.
What this paper adds
This study demonstrated that existing cigarette price differential in the European Union (EU) was significantly associated with making a cross-border tobacco purchase because of cheaper price within the past 12 months in 2012.
Of current cigarette smokers who engaged in a cross-border tobacco purchase within the past 12 months, 80% purchased the tobacco product from another country within the EU while 24.9% made the purchase from a country located outside the EU.
Subpopulation differences in cross-border tobacco purchasing patterns were noted among population subgroups defined by geographic region, age and educational attainment.
Findings call for enhanced measures to reduce cigarette price differential across the EU to strengthen the effectiveness of individual Member States’ tobacco taxation strategies.
References
Footnotes
Contributors ITA designed the study, performed the analyses and produced the initial draft. UTO, FTF, EB, AV, and CIV critically reviewed the manuscript and performed data interpretation and manuscript preparation. All authors read, edited and approved the final manuscript and are responsible for the overall content as guarantors.
Competing interests ITA initiated the reported research while affiliated with the Center for Global Tobacco Control at Harvard University. He is currently affiliated with the Centers for Disease Control and Prevention's Office on Smoking and Health. This report was completed and submitted outside of the official duties of his current position and does not reflect the official policies or positions of the Centers for Disease Control and Prevention.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement All data used in this study are publicly available secondary data.