Article Text

Download PDFPDF

The prevalence of illicit cigarette consumption and related factors in Turkey
  1. Bekir Kaplan1,
  2. Ana Navas-Acien2,
  3. Joanna E Cohen1
  1. 1 Institute for Global Tobacco Control, Johns Hopkins University Bloomberg School of Public Health, Baltimore, Maryland, USA
  2. 2 Department of Environmental Health Science, Columbia University Mailman School of Public Health, New York City, New York, USA
  1. Correspondence to Dr Bekir Kaplan, Institute for Global Tobacco Control, Johns Hopkins University Bloomberg School of Public Health, 2213 McElderry St 4th Fl, Baltimore, Maryland 21205-2103, USA; bkaplan9{at}jhu.edu, drbekir_ankara{at}hotmail.com

Abstract

Background The tobacco industry claims that high cigarette taxes drive illicit trade and that governments should therefore not increase tobacco tax because it will increase the level of illicit trade. This study examines illicit cigarette consumption in Turkey after a tobacco tax increase and its related factors.

Method This national cross-sectional survey was conducted in March-June 2013 and 9717 people aged ≥18 years participated in the interviewer-administered survey. Smokers were asked to show their last used cigarette pack to the interviewers and price paid for their cigarettes. Factors associated with smoking cigarettes with a tobacco tax stamp and paying ≥5 TL (Turkish lira) for a pack of cigarettes were analysed with logistic regression.

Results Among the observed cigarette packs, 12.1% did not have the Turkish tax stamp. More illicit cigarettes were observed in the East region than in other regions (p<0.001). The reported average amount paid for a pack of cigarettes was €2.12 (€2.15 for men and €1.97 for women, p<0.001). The amount paid for cigarettes with a tax stamp (€2.15) was higher than the amount paid for cigarettes without tax stamp (€1.08) (per cigarette pack) (p<0.001).

Conclusion Just over 1 in 10 smokers (12%) had an illicit cigarette pack about 5 months after the final tax increase; this was most common in the East region of Turkey. Estimates are comparable to those in previous studies and do not indicate that an increase occurred in the prevalence of illicit cigarette use compared with before the tobacco tax increase.

  • Illicit Cigarette
  • Prevalence
  • Turkey

Statistics from Altmetric.com

Introduction

Increasing tobacco product taxes is one of the most effective ways of reducing tobacco consumption,1 and thus the tobacco industry tries to prevent tax increases. The tobacco industry argues that an increase in taxes will negatively affect the country’s economy and that it will cause losses in tax revenue due to increased illicit trade of tobacco.2–4 While the tobacco industry argues that increasing taxes will lead to an increase in illicit trade of cigarettes, it is known that tobacco companies use illicit trade as a method of sourcing markets in low and middle-income countries with tobacco products; using this method, for instance, one tobacco company succeeded in introducing cigarettes in some African countries where cigarette import was prohibited.5 6

Joossens and Raw emphasise that illicit cigarette trade has a number of negative consequences for tobacco control— including loss of tax revenue, lower priced cigarettes and thus higher levels of consumption.7 There is also evidence that transnational tobacco companies have used illicit trade as a tool to create new markets in low-income populations and in countries where their products have traditionally had limited reach.8–12

Illicit trade of tobacco products is a major public health problem as lower prices of illicit cigarettes lead to increased cigarette consumption.6 Illicit tobacco consumption was estimated to be 11.6% worldwide in 20076 and almost 10% in 2015.13 Illicit tobacco trade is not directly related to tobacco price8 or tax increases.12 For example, many countries with high tobacco excise taxes have a much smaller illicit trade problem than countries with much lower excise taxes.12 Other factors, including the presence of informal distribution networks, organised crime, industry participation and corruption, probably contribute more to cigarette smuggling than price levels.6

Turkey is one of the first countries to implement all MPOWER strategies (Monitor tobacco use, Protect people from second-hand smoke, Offer help to quit, Warn about the dangers of tobacco, Enforce bans on tobacco advertising, promotion and sponsorship, and Raise taxes), and smoking prevalence began to decline since 2008.14 In Turkey, Special Consumption Tax on tobacco was 28% in 2004 and 58% during 2005 and 2009.15 In January 2010, Turkey increased the Special Consumption Tax on tobacco by 20% and this increase resulted in the total tax on tobacco being 79% of the retail price.16 After the increase in the tobacco tax, the average price paid for cigarettes in Turkey increased by 42% during 2008–2012, cigarettes became less affordable, and the average smoking prevalence declined by 15%.17 Furthermore, Turkey’s tax on tobacco products increased to 82% of the retail price in January 2013, becoming one of the highest tobacco taxes in the world.14

It was estimated that illicit cigarette consumption was 14% in Turkey in 2004 before tax increases.18 The Ninth Development Plan published by the Turkish Government in 2006 estimated that illicit cigarettes represent approximately 10%–15% of total consumption.19 Following the cigarette price increase, rumours about the increased use of illicit cigarettes began to surface.20 There were reports in the Turkish media that cigarette smuggling increased 20% because of the raised taxes.21 22 The National Committee on Cigarette and Health, however, explained that it was the tobacco industry that claimed a 20% increase in cigarette smuggling in the countries where cigarette taxes increased.23 The aim of this study was to determine the extent of illicit cigarette consumption in Turkey and identify the characteristics of smokers who use these products.

Methods

Study population

The Turkish National Health Promotion Survey is a national cross-sectional survey conducted between March and June 2013. A multistage, stratified, weighted cluster sampling method was used to obtain a representative sample of the Turkish population. The selection of clusters was conducted by the TURKSTAT (Turkish Statistical Institute) according to the ‘Address-based population registration system, 2012’ and one person from each household was assigned by the Kish method.24

The sample size was calculated as 11 986 households countrywide to predict in Statistical Region Unit Classification 1 (Nomenclature of Units for Territorial Statistics 1) levels, considering type 1 error of 5% and 20% non-response rate.

A total of 9717 (81%) people aged ≥18 years participated in the survey, 1280 (11%) refused to participate, 731 (6%) addresses were not a dwelling and 274 (2%) households could not be found at the addresses during the study period. After the data were obtained and cleaned, all analyses were conducted with sampling weights that were adjusted for the complex design of the survey, considering designation probabilities and non-response rates. We present unweighted numbers, but weights were applied to obtain percentages.

Data collection

The face-to-face interviewer-administered survey consisted of 73 questions including sociodemographic, lifestyle and medical items. Participants who smoked were also asked to show their last used cigarette pack. Cigarette packs in Turkey are required to have a Turkish tax stamp (TTS) to show that the cigarettes were produced or imported legally in Turkey; packs without the TTS are considered illicit. The interviewer asked to see the last used cigarette pack and recorded whether the respondent’s last used cigarette pack had a TTS. Respondents were also asked how much they paid for their last cigarette pack. The price was recorded in Turkish lira (TL) and converted to euro using the lira–euro exchange average during the project data collection period.

Respondents were placed in one of the ‘five regions demographic classification’ description of the Turkish Demographic and Health Survey (2008) based on the address of their residence.25

Using the complex samples module of SPSS V.20.0, weighted percentage distribution, mean, SE calculations and χ2 and logistic regression analyses were performed; p values <0.05 were considered statistically significant. Binary logistic regression analysis was performed to evaluate the predictive values of the following potential variables in relation to presence of having a TTS on last used cigarette package: sex, age, education status, employment status, marital status, region of residence and settlement area. The effects of confounders, interactions and multicollinearity were evaluated in the model. Ethical approval for the survey was obtained from the Zekai Tahir Burak Women’s Health Education and Research Hospital Ethics Committee, with the approval number 82/2012.

Results

A total of 9717 people aged ≥18 years participated in the survey. Forty-four per cent of respondents were aged 25–44 years, 50.5% were female, 55.5% had graduated from primary/secondary school, 28.6% were housewives, 43.2% lived in the West region and 65.3% lived in urban areas. Just over a quarter (27.9%) of respondents used any kind of tobacco product every day, 1.6% smoked occasionally and 94.3% of tobacco users smoked cigarettes. Twelve per cent (12.1% (95% CI 9.8 to 14.8)) of cigarette smokers showed a cigarette pack that was illicit. Showing an illicit cigarette pack was more common among less educated smokers, East region residents, rural residents and the unemployed (table 1). While only 18.8% of cigarettes costing <5 TL (€1.71) had the TTS, 97.9% of cigarettes costing ≥5 TL had the tax stamp (p<0.001) (table 1).

Table 1

Characteristics of smokers by presence of the Turkish tax stamp on their cigarette pack, Turkey, 2013 (n=1900)

The odds of having a TTS on their cigarette pack was 3.56 (95% CI 1.47 to 8.63) times higher (OR) among primary/secondary school graduates, 2.83 (95% CI 1.18 to 6.82) times higher among high school graduates and 5.65 (95% CI 1.00 to 32.05) times higher among university graduates compared with those who did not complete primary school. The odds of having the TTS on their cigarette pack was 3.70 (95% CI 1.22 to 11.18) times higher among people from the Central-North region compared with people from the East region. The odds of having the TTS on their cigarette pack was 114.32 (95% CI 54.43 to 240.10) times higher among the packs with high cigarette price (≥5.0 TL) compared with packs with low prices (<5 TL) (table 1).

The mean price paid for a pack of cigarettes with a tax stamp was 6.65±0.04 TL (€2.26), while the price was 3.18±0.15 TL (€1.08) for cigarettes without the tax stamp (p<0.001).

Buying cigarette packs costing >5 TL (€1.71) was higher among men than women, the more educated, smokers living in regions other than the South and East regions, workers and students compared with housewives, and for cigarettes with the tax stamp compared with cigarettes without it (p<0.001) (table 2).

Table 2

Characteristics of smokers by price paid for cigarettes, Turkey, 2013 (n=2222)

The odds of paying ≥5 TL (€1.71) for a pack of cigarettes was 3.11 (95 CI% 1.34 to 7.22) times higher for men than women; 8.32 (95% CI 3.31 to 20.87) times higher among people living in the West region and 8.80 (95% CI 3.78 to 20.68) times higher among people from the North-Central region compared with those living in the East region; and 101.99 (95% CI 49.20 to 211.41) times higher among people who bought cigarettes with a TTS compared with those who bought cigarettes without the stamp (table 2).

Discussion

This study evaluated the prevalence of illicit cigarette consumption, the price paid for a pack of cigarettes and the factors associated with these circumstances in Turkey. We found that 12.1% of smokers had cigarettes that were illicit. According to a report by Euromonitor, the prevalence of illicit cigarette consumption in Turkey in 2004 was 14%.18 A study evaluating the frequency of illicit cigarette consumption in 2006 reported that 11.0% of cigarettes consumed in Turkey were illicit.26 According to GATS (Global Adult Tobacco Survey) 2012, the prevalence of consuming illicit cigarettes, also based on the absence of the TTS, was 9.1% (95% CI 6.8 to 12.0).14 Although statistical comparison between studies cannot be made, the values are similar enough to suggest that illicit use has been relatively stable over time and does not show a strong dependence to changes in tax policy.

According to a review paper published in 1998,27 one-third of the worldwide cigarette market in 1996 comprised illicit products. Another study found that 11.6% of the worldwide cigarette market in 2007 comprised illicit products.7 The prevalence of illicit cigarette consumption in Turkey in this study was comparable to their estimate of illicit use in middle-income countries (11.8%).

In the present study, the use of illicit cigarettes was higher among people with low levels of education and East region residents. The East region borders countries with lower cigarette prices, such as Syria and Iraq. Moreover, the geographic features, relatively high unemployment rate and long border28 in the East region may be enabling factors for illicit trade. Any form of illicit consumption is also more common in the East than in other regions.29 In a multicountry study in Europe, illicit cigarette consumption was more frequent among those living in a country that shared a land or sea border with Ukraine, Russia, Moldova or Belarus, which are major suppliers of cheap and illicit cigarettes.8 Higher education and income level have also been associated with cross border cigarette purchasing in Europe.30 Comparable with the present study, GATS 2012 found that the rate of illicit cigarette consumption was highest among poorly educated people.14 The fact that the rate of illicit cigarette consumption is higher among less educated people indicates that smokers with lower incomes prefer illicit cigarettes owing to their lower cost. In a survey in Thailand, younger age and higher average monthly expenditure on cigarettes were associated with illicit cigarette consumption.31 While not statistically significant, the use of illicit cigarettes was higher among women (14.5%) than men (11.2%). In GATS 2012,14 the rate of illicit cigarette consumption was also higher among women (9.3%) than men (9.0%). The reason for sex differences in the consumption of illicit cigarettes is unclear. Although it could be related to differences in income and education between men and women, in our study the differences persisted after adjusting for those variables.

According to a cross-sectional survey conducted in 18 European countries in 2012, illicit tobacco trade is not directly related to tobacco prices.8 There is no relationship between illicit trade and price when prices are adjusted by gross domestic product per capita in purchasing power standards. Illicit trade might be influenced by factors other than price such as cost of operating in a country, industry participation, how well crime networks are organised, the likelihood of being caught, the punishment if caught and corruption levels.8 32

Analyses by the World Bank have shown that high levels of illicit tobacco products are linked more closely to corruption and tolerance of contraband sales than to tobacco tax increases.33 Research in Central and Eastern Africa suggests that illicit cigarette consumption in this region is not caused by difference in tax levels but more by weak state capacity, high levels of corruption and the activities of rebel groups.34

Limitations

Twenty-four per cent of cigarette users did not show their cigarette pack to the study interviewer. The distribution of people who did not show their cigarette pack was not equal across regions; 42.6% lived in West region. Because illicit cigarette use was relatively low (4%) in the West region, we believe that if all smokers showed their packs, the prevalence of illicit cigarette consumption would be lower than what we reported. It is also possible, however, that people with illegal cigarettes would try to hide them, being less likely to show them during the interview. In this scenario, the prevalence of illicit cigarette use would be higher. Second, we did not ask respondents where they purchased their most recent pack of cigarettes. For the cross-validation of illicit tobacco consumption prevalence, the provenance of the last purchased cigarette pack could be a predictor of illicit cigarette consumption. We do not have the data required to determine price demand flexibility of illicit cigarette consumption. Third, we categorised the packs without TTS as illicit. Because there would be a small proportion of cigarettes brought in legally from other countries for personal use, we might have overestimated the prevalence of illicit consumption. Fourth, because we could not find illicit cigarette prevalence estimated with a similar methodology before tax increases, we could not compare the illicit consumption level with other estimates that used a similar methodology. We used Euromonitor International’s estimation for 2004 and 2006 and the Turkish Government Plan for 2006 for the comparison; neither report based their estimate on survey data. Finally, our data did not provide information on brand name, which could have allowed us to better understand the different types of illicit practices in the Turkish market.

Conclusion

The prevalence of illicit cigarette consumption in Turkey in 2013 was 12.1% and our study confirmed that the prevalence has not increased following the increase in tobacco taxes in Turkey. The prevalence of illicit cigarette consumption was higher among people with lower education levels and varied according to region of residence, indicating that this problem is due in part to the social, cultural and geographical characteristics of the place of residence. Similar to illicit trade of other products, the most effective way to reduce the amount of illicit product is to enhance the fight against illicit trade rather than lowering tobacco product taxes.1 When developing future policies, the fact that women use cheaper cigarettes and thus acquire cigarettes easily should be taken into consideration. To eliminate all forms of illicit trade in tobacco products, there is need to develop scientific, technical and institutional capacity to plan and implement appropriate national, regional and international measures. Effective action to prevent and combat illicit trade in tobacco products requires a comprehensive international approach, and close cooperation on all aspects of illicit trade, including illicit trade in tobacco, tobacco products and manufacturing equipment.

What this paper adds

  • Tobacco tax increase is one of the most effective ways of reducing tobacco consumption.

  • The tobacco industry argues that an increase in taxes will cause losses in tax revenue due to increased illicit trade of tobacco.

  • The paper’s illicit use estimate is comparable to earlier estimates for Turkey, and thus does not indicate that a significant change in illicit use occurred after the tax increase.

  • Illicit consumption is related to region of residence, with illicit cigarette consumption more common in the East region that borders countries with lower cigarette prices.

Acknowledgments

The authors would like to thank the Turkish Ministry of Health for supporting this survey.

References

Footnotes

  • Contributors BK contributed to study design, data collection, data analysis, result interpretation and drafting of the manuscript. ANA and JEC contributed to the data analysis, interpretation and drafting of the manuscript.

  • Funding This study was supported by the Ministry of Health of Turkey, General Directorate of Health Research.

  • Competing interests None declared.

  • Ethics approval Zekai Tahir Burak Women’s Health Education and Research Hospital Ethics Committee, with the approval number 82/2012.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Correction notice This paper has been amended since it was published Online First. Owing to a scripting error, some of the publisher names in the references were replaced with ’BMJ Publishing Group'. This only affected the full text version, not the PDF. We have since corrected these errors and the correct publishers have been inserted into the references.

Request Permissions

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.