Abstract
Background
Decades of research have produced overwhelming evidence that tobacco taxes reduce tobacco use and increase government tax revenue. The magnitude and effectiveness of taxes at reducing tobacco use provide an incentive for tobacco users, manufacturers and others, most notably criminal networks, to devise ways to avoid or evade tobacco taxes. Consequently, tobacco tax avoidance and tax evasion can reduce the public health and fiscal benefit of tobacco taxes.
Objectives
First, this study aims to document, using data from the International Tobacco Control Policy Evaluation Project (ITC), levels and trends in cigarette users’ tax avoidance and tax evasion behaviour in a sample of sixteen low-, middle- and high-income countries. Second, this study explores factors associated with cigarette tax avoidance and evasion.
Methods
We use data from ITC surveys conducted in 16 countries to estimate the extent and the type of cigarette tax avoidance/evasion between countries and across time. We use self-reported information about the source of a smoker’s last purchase of cigarettes or self-reported packaging information, or similar information gathered by the interviewers during face-to-face interviews to measure tax avoidance/evasion behaviours. We use generalized estimating equations (GEE) to explore individual-level factors that may affect the likelihood of cigarette tax avoidance or evasion in Canada, United States, United Kingdom and France.
Findings
We find prevalence estimates of cigarette tax avoidance/evasion vary substantially between countries and across time. In Canada, France and the United Kingdom, more than 10% of smokers report last purchasing cigarettes from low or untaxed sources while in Malaysia, some prevalence estimates suggest substantial cigarette tax avoidance/evasion. We also find important associations between household income and education and the likelihood to engage in tax avoidance/evasion. These associations, however, vary both in direction and magnitude across countries.
Keywords: tobacco, smoking, cigarette, price, tax, avoidance, evasion, contraband, smuggling
Introduction
Decades of research have produced overwhelming evidence that tobacco taxes reduce tobacco use, save lives and increase government tax revenue.[1–3] The magnitude and effectiveness of taxes at reducing tobacco use provide an incentive for tobacco users, manufacturers and others, most notably criminal networks, to devise ways to avoid or evade tobacco taxes. Tax avoidance by tobacco users involves legal purchasing behaviour in order to pay less or no taxes. Examples include cross-border shopping, duty free shopping and internet purchases. Tobacco manufacturers can also engage in tax avoidance by changing their products and prices with the objective of reducing their tax liability. Tax evasion involves illegal methods of avoiding tobacco taxes. Such illegal activities include illicit trade or production of genuine1 or counterfeit tobacco products.[4]
Tobacco tax avoidance and evasion pose an array of challenges. Relatively inexpensive, licit or illicit, tobacco products undermine public policies that seek to render tobacco products less affordable. Moreover, tax evaded cigarettes may be sold in packs or plastic bags which do not provide the information which is required by regulations, such as health warnings and information on toxic emissions, as is the case in Canada.[5] In turn, weak tax administration can undermine the public health objectives of tobacco control measures. Tobacco tax avoidance and evasion also decrease government revenue available for health and social programs, can result in increased criminal justice expenditures, and can provide unmonitored access to cigarettes by youth.[6] Although the magnitude of the tobacco tax avoidance and evasion problem is challenging to quantify, it remains an intractable issue which merits attention due to the many implications for policy design, government revenues and public health.
The magnitude of cigarette tax avoidance and evasion can be estimated by contrasting estimates of legal cigarettes sales based on production, trade or tax revenue data with consumption estimates based on survey data while taking into account under-reporting.[7 8] The tax authorities of the United Kingdom have used this method extensively to estimate the extent of cigarette tax avoidance/evasion.[9 10] This method is also being used in Canada in an attempt to identify trends in cigarette contraband.[11 12] Additionally, Stehr (2005) used a variant of this approach to examine the tax avoidance response to tax changes in the United States.[13]
Self-reported information obtained from survey data can also be used to measure the level of cigarette tax avoidance/evasion. In many countries, information about the source of a smoker’s last or usual purchase of cigarettes can provide key tax avoidance/evasion information (e.g., First Nations reserves in Canada, out-of-state in the United States and out-of-country in the European Union). This approach has been used extensively in Canada[14–17] and also in the United States[18–20] and the United Kingdom.[21] Self-reported packaging information, or similar information gathered by the interviewers during face-to-face interviews or mail-in pack surveys can also provide key insights into tax avoidance/evasion behaviours. Examples include nonstandard or missing health warnings, tax stamps or authenticity labels. For example, in Thailand, an examination of the health warnings (i.e., the absence of warnings or warnings in a language other than Thai) during face-to-face interviews revealed if the cigarettes were likely to have been legally purchased or not.[22] A similar approach was used in Taiwan relying on self-reports from telephone interviews, using cigarette packs not bearing the tax seal as an indicator of tax evasion.[23] An inspection of discarded cigarette packs can also be used to determine whether all taxes have been paid.2 This approach has been recently used in France,[24] Canada,[25] United States[26 27] and United Kingdom.[28]
Our first objective is to document for a large number of countries, using novel data from the International Tobacco Control Policy Evaluation Project (ITC), levels and trends in cigarette users’ tax avoidance and tax evasion behaviour. Our second objective is to explore factors associated with cigarette tax avoidance/evasion.
Data and Methods
We use data from ITC surveys conducted in high-, mid- and low-income countries between 2002 and 2011; ITC surveys are longitudinal cohort surveys of tobacco use that are designed to assist policy makers in the implementation of strong evidence-based tobacco control policies. First, we use data from the first eight waves of the ITC Four-Country survey (Australia, Canada, United States and United Kingdom) conducted in 2002, 2003, 2004, 2005/06, 2006/07, 2007/08, 2008/09 and 2010/11. Second, we use data from ITC surveys conducted in four European countries: France (2006/07, 2008), Ireland (2003/04, 2004/05, 2006), Scotland (2006) and the Netherlands (2008, 2010, 2011). Third, we use data from ITC surveys conducted in seven low-and middle income countries: Bangladesh (2009, 2010), China (2006, 2007/08, 2009/10), Malaysia and Thailand (2005, 2006/07, 2008, 2009), Mexico (2006, 2007, 2008/09, 2010, 2011), Uruguay (2006, 2008/09, 2010/11) and Mauritius (2009, 2010, 2011).3 In all ITC countries (with the exception of China and Mexico4), probabilistic sampling methods were employed to obtain nationally representative samples of smokers. Stratified sampling designs were employed in high-income countries where landline telephone penetration is high. Multi-stage cluster designs were employed in low- and middle-income countries. Sampling strata were defined by geographic regions within each country. In some countries, mixed survey modes were used: in the Netherlands and in the most recent wave of data collected in the ITC 4 Country Survey, both telephone interviewing and web-based interviewing were employed; in Malaysia, face-to-face and telephone interviewing methods were used. Sampling weights were computed for each wave of each survey within all ITC countries so that results are nationally representative of smokers. Sampling weights account for survey non-response. More details on ITC’s conceptual framework and methodology are provided by Fong et al.[29] and Thompson et al.[30]
We present prevalence estimates of cigarette tax avoidance/evasion using two approaches. First, we follow Hyland et al.[31] and use responses to the question: “where did you last buy cigarettes for yourself?” to identify low or untaxed sources of cigarettes that may represent tax avoidance/evasion behaviour. We present data for individuals who responded that their last source for cigarettes was either a First Nations/Indian reserve, out of state/province/country, a duty-free outlet, a direct purchase (mail, telephone or Internet), ‘someone else’ such as an independent seller, or a military commissary. For comprehensiveness, we also present data, when relevant, for individuals who reported that their last source for cigarettes was from a friend or a relative, or who reported not knowing or refused to answer. With the exception of ITC-Netherlands, all ITC surveys include a follow up open-ended question for individuals who responded ‘other’ to the question that pertains to the source of their last cigarette purchase. All open-ended responses were manually examined and re-coded when possible. We were unable to re-code only a negligible number of responses; such responses were coded as ‘other‘. Second, we present estimates based on self-reported packaging information, or similar information gathered by the interviewers during face-to-face interviews. We use responses to the question (or a variant of the question): “do you have an empty pack handy? I need to get some information about the cigarettes you smoke? If not, can you tell me if your cigarette pack has: a standard warning label, a tax stamp, security ink, ….” Data are presented per source of information (i.e., self-reported or interviewers). All tax avoidance/evasion prevalence estimates presented were weighted using sampling weights. The two approaches described above were not practicable in all countries; hence we present estimates from a subset of the surveys. For each country or groups of countries, we present all available data that may represent cigarette tax avoidance/evasion. For high-income countries (Australia, Canada, France, Ireland, the Netherlands, Scotland, the United Kingdom, and the United States) and for Mexico and Uruguay, we present estimates that are based on the source of a smoker’s last purchase of cigarettes. For Bangladesh, China, Malaysia, Mauritius, Thailand and Uruguay, we present estimates that are based on self-reported packaging information, or similar information gathered by the interviewers during face-to-face interviews.
In addition to presenting prevalence estimates, we use generalized estimation equations (GEE) to explore factors that may affect the likelihood of cigarette tax avoidance or evasion. We use a binary indicator of potential cigarette tax avoidance/evasion defined as follows: 1 if individuals reported that their last source for cigarettes was either a First Nations/Indian reserve, out of state/province/country, a duty-free store, a military commissary, someone else or a direct purchase (mail, telephone or Internet); 0 otherwise. We exclude individuals who reported that their last source for cigarettes was from a friend or a relative or who reported not knowing or refused to answer. We do not use data obtained from self-reported packaging information, or similar information gathered by the interviewers during face-to-face interviews.5 We perform this analysis using data from the following countries: Canada, United States, United Kingdom and France. We focus on these countries because the prevalence of tax avoidance/evasion appears to be important.6 We examine associations between cigarette tax avoidance or evasion and the following individual-level factors: socioeconomic status using country-specific categorical measures of household income (low, moderate, high, refused/don’t know) and education (low, moderate, high, refused/don’t know); categorical measures of sex, daily cigarette smoking intensity (1–10, 11–20, 21–30, > 30 cigarettes/day, refused/don’t know), time to first cigarette (measured in minutes, > 60, 31–60, 6–30, < 5, refused/don’t know) and quit intention (yes, no, don’t know); and, a continuous measure of age and age squared (measured in units of ten years).
We use GEE with a logit link, a binomial family, an unstructured working correlation structure (i.e., no constraints are placed on correlations between repeated observations on individuals over time) and robust standard errors. The above specification does not converge for the United States; we instead use an exchangeable working correlation structure (i.e., all correlations within a cluster (i.e, an individual) are constrained to be identical). All models are estimated using Stata/MP 12.1 for Macintosh.
Results
Tables 1, 2 and 3 present estimates of tax avoidance or evasion based on respondents’ self-reported source of their last purchase of cigarettes. Table 1 presents data for Canada, United States, United Kingdom and Australia for eight survey waves conducted between 2002 and 2010/11. It is important to note that it is difficult to disentangle tax avoidance from tax evasion. For example, the estimates presented in Table 1 are more likely to represent tax avoidance then tax evasion, with the exception of Canada where cigarettes are likely purchased illegally from First Nations reserves. In the United States, purchases from Indian reservations and from out of state are, more often than not, legally permitted; similarly, out of country and duty free purchases are generally legal in the United Kingdom. Estimates vary substantially between countries and across time within countries. Tax avoidance/evasion appear highest in the UK: about 12.5 to 17% of respondents, between 2002 and 2010/11, reported having last purchased cigarettes from a low or untaxed source.7 These data suggest a downward trend between May-September 2003 and 2010/2011.
Table 1.
Percentage of respondents reporting the source of their last purchase of cigarettes, Canada, United States, United Kingdom and Australia
| 4 Country-Survey | Wave 1 (Nov 02 – Dec 02) |
Wave 2 (May 03 – Sep 03) |
Wave 3 (Jun 04 – Dec 04) |
Wave 4 (Oct 05 – Jan 06) |
Wave 5 (Oct 06 – Feb 07) |
Wave 6 (Sep 07 – Feb 08) |
Wave 7 (Oct 08 – Jul 09) |
Wave 8 (Jul 10 – Jun 11) |
|---|---|---|---|---|---|---|---|---|
| % | % | % | % | % | % | % | % | |
| Canada | ||||||||
| First Nations reserve | 2.0 | 2.6 | 3.5 | 6.0*** | 7.8* | 10.3* | 10.4 | 9.5 |
| Out of state/province/country | 0.3 | 0.3 | 0.3 | 0.3 | 0.1 | 0.1 | 0.0 | 0.4 |
| Duty free | 0.6 | 0.7 | 0.7 | 0.4 | 0.7 | 0.7 | 0.7 | 0.7 |
| Any direct purchase (mail, phone or Internet) | 0.0 | 0.1 | 0.0 | 0.0 | 0.0 | 0.0 | 0.1 | 0.2 |
| Someone else (e.g., independent seller) | 0.2 | 0.1 | 0.5 | 0.7 | 0.6 | 1.1 | 0.5 | 0.4 |
| Military | 0.0 | 0.1 | 0.1 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
| Friends/relatives | 0.1 | 0.3 | 1.2** | 0.5 | 0.7 | 0.7 | 1.1 | 0.3* |
| Refused/don't know/other | 0.0 | 0.3* | 0.0* | 0.1 | 0.0 | 0.1 | 0.4 | 0.2 |
| n | 2199 | 2014 | 1889 | 1774 | 1757 | 1727 | 1507 | 1245 |
| United States | ||||||||
| Indian reservation | 2.6 | 3.0 | 2.6 | 2.6 | 3.0 | 3.0 | 4.4 | 4.4 |
| Out of state/province/country | 0.7 | 0.4 | 0.8 | 0.2** | 0.4 | 0.2 | 1.3** | 0.8* |
| Duty free | 0.4 | 0.2 | 0.2 | 0.4 | 0.3 | 0.0 | 0.1 | 0.4 |
| Any direct purchase (mail, phone or Internet) | 0.8 | 1.4 | 1.9 | 1.0 | 0.6 | 0.9 | 1.0 | 0.9 |
| Someone else (e.g., independent seller) | 0.0 | 0.0 | 0.2 | 0.2 | 0.6 | 0.4 | 0.1 | 0.0 |
| Military | 0.7 | 1.0 | 1.0 | 1.0 | 0.5 | 0.6 | 0.3 | 0.5 |
| Friends/relatives | 0.1 | 0.1 | 0.4 | 0.4 | 0.2 | 0.1 | 0.2 | 0.3 |
| Refused/don't know/other | 0.1 | 0.3 | 0.1 | 0.0 | 0.1 | 0.0 | 0.1 | 0.0 |
| n | 2123 | 1909 | 1924 | 1792 | 1820 | 1766 | 1516 | 1267 |
| United Kingdom | ||||||||
| Out of state/province/country | 8.6 | 10.4 | 9.7 | 7.8* | 7.1 | 9.2 | 5.1*** | 5.4 |
| Duty free | 4.4 | 5.5 | 5.1 | 6.3 | 6.0 | 3.4 | 6.8*** | 5.8 |
| Any direct purchase (mail, phone or Internet) | 0.0 | 0.1 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
| Someone else (e.g., independent seller) | 0.6 | 1.2 | 1.6 | 1.2 | 1.0 | 1.0 | 2.5 | 1.2 |
| Friends/relatives | 2.2 | 2.0 | 1.8 | 2.5 | 2.6 | 2.2 | 3.5 | 3.8 |
| Refused/don't know/other | 0.1 | 0.2 | 0.3 | 0.0 | 0.1 | 0.0 | 0.0 | 0.0 |
| n | 2400 | 1933 | 1839 | 1738 | 1721 | 1665 | 1486 | 975 |
| Australia | ||||||||
| Out of state/province/country | 0.1 | 0.2 | 0.1 | 0.1 | 0.2 | 0.2 | 0.0 | 0.5 |
| Duty free | 0.7 | 0.5 | 0.6 | 1.2 | 0.5 | 0.9 | 1.4 | 1.5 |
| Any direct purchase (mail, phone or Internet) | 0.0 | 0.1 | 0.1 | 0.0 | 0.0 | 0.0 | 0.1 | 0.0 |
| Someone else (e.g., independent seller) | 0.2 | 0.1 | 0.4* | 0.2 | 0.6 | 0.3 | 0.2 | 0.4 |
| Friends/relatives | 0.1 | 0.4 | 0.2 | 0.3 | 0.1 | 0.2 | 0.4 | 0.3 |
| Refused/don't know/other | 0.0 | 0.0 | 0.0 | 0.1 | 0.0 | 0.0 | 0.0 | 0.0 |
| n | 2304 | 1975 | 1851 | 1714 | 1823 | 1805 | 1371 | 1116 |
Note: n represents the total number of valid responses (i.e., the denominator).
*, ** and ***: significant difference at 5%, 1% and 0.1% with previous time period. These should be interpreted with great caution as the number of comparisons is absurdly high.
Table 2.
Percentage of respondents reporting the source of their last purchase of cigarettes, France, the Netherland, Ireland and Scotland
| France | Netherlands | Ireland | Scotland | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Wave 1 (Dec 06 – Jan 07) |
Wave 2 (Sep 08 – Nov 08) |
Wave 1 (Mar 08 – Apr 08) |
Wave 4 (May 10 – Jun 10) |
Wave 5 (May 11 – Jun 11) |
Wave 1 (Dec 03 – Jan 04) |
Wave 2 (Dec 04 – Jan 05) |
Wave 3 (Feb 06 – Mar 06) |
Wave 3 (Feb 06 – Mar 06) |
|
| % | % | % | % | % | % | % | % | % | |
| Out of country (inside EU) | 12.8 | 16.7** | 1.4 | 2.7* | 2.9 | 0.8 | 1.3 | 2.8 | 3.7 |
| Out of country (outside EU) | 1.0 | 1.6 | 0.1 | 0.5 | 0.9 | ||||
| Duty free | 1.5 | 1.5 | 0.7 | 0.3 | 0.8 | 1.7 | 3.2* | 1.8 | 2.2 |
| Any direct purchase (mail, phone or Internet) | 0.0 | 0.1 | 0.0 | 0.0 | 0.1 | 0.2 | 0.0 | 0.5 | 0.2 |
| Someone else (e.g., independent seller) | 0.5 | 0.7 | 0.4 | 0.6 | 0.5 | 0.4 | 0.3 | 1.0 | 1.0 |
| Refused/don't know/other* | 0.1 | 0.0 | 1.3 | 2.4* | 3.6 | 0.3 | 0.1 | 0.2 | 0.2 |
| n | 1735 | 1540 | 2224 | 1723 | 1672 | 1071 | 912 | 578 | 507 |
Note. * = Unlike other ITC surveys, the Netherland surveys do not contain a follow up open-ended question for individuals who responded "other" as the source of their last purchase of cigarettes. See text for more details; n represents the total number of valid responses (i.e., the denominator).
*, ** and ***: significant difference at 5%, 1% and 0.1% with previous time period. These should be interpreted with great caution as the number of comparisons is absurdly high.
Table 3.
Percentage of respondents reporting the source of their last purchase of cigarettes, Mexico and Uruguay
| Mexico | Uruguay | |||||||
|---|---|---|---|---|---|---|---|---|
| Wave 1 (Sep 06 – Nov 06) |
Wave 2 (Oct 07 – Nov 07) |
Wave 3 (Oct 08 – Jan 09) |
Wave 4 (Jan 10 – Feb 10) |
Wave 5 (Apr 11 – May 11) |
Wave 1 (Oct 06 – Dec 06) |
Wave 2 (Oct 08 – Feb 09) |
Wave 3 (Oct 10 – Jan 11) |
|
| % | % | % | % | % | % | % | % | |
| Out of country | 0.6 | 0.7 | 0.2 | 1.3 | 0.4 | 0.0 | 0.2 | 0.6 |
| Duty free | 1.0 | 0.2* | 0.1 | 1.3 | 0.3 | 0.2 | 0.7 | 0.3 |
| Internet | 0.0 | 0.0 | 0.0 | 0.9 | 0.0 | 0.2 | 0.0 | 0.0 |
| Refused/don't know | 0.1 | 0.4 | 0.7 | 0.7 | 0.0 | 0.0 | 0.1 | 0.1 |
| n | 1077 | 941 | 1854 | 394 | 821 | 887 | 1294 | 1224 |
Note: n represents the total number of valid responses (i.e., the denominator).
*, ** and ***: significant difference at 5%, 1% and 0.1% with previous time period. These should be interpreted with great caution as the number of comparisons is absurdly high.
A decade ago, cigarette tax avoidance/evasion appeared to be relatively unimportant in Canada as only about 3% of respondents reported having last purchased cigarettes from a low or untaxed source. Our estimates, however, suggest a large increase – more than 4-fold – between 2002 and 2008/09 followed by a small decrease between 2008/09 and 2010/11. Cigarette tax avoidance/evasion in the US appear to have remained fairly stable during the last decade, oscillating between 5.3 and 7.6%. Unsurprisingly, because of its geographic situation and limited price differentials between states, cigarette tax avoidance/evasion appears negligible in Australia. As with levels and trends in cigarette tax avoidance/evasion, the type of cigarette tax avoidance/evasion is quite heterogeneous between countries. In Canada and the US, low/untaxed sources of cigarettes are primarily First Nations/Indian reserve purchases while in the UK, they are primarily out of country and duty free purchases.
Table 2 presents data for France, the Netherlands, Ireland and Scotland; table 3 presents data for Mexico and Uruguay. Out of country (but inside the European Union (EU)) cigarette purchases are substantial in France and appear to be growing: 12.8% of smokers in 2006/07 and 16.7% in 2008. In the Netherlands, Ireland and Scotland cigarette purchases from low or untaxed sources are low but not negligible (between about 1 and 3.8% from a source outside the country). In both Netherlands and Ireland there appears to be an upward trend in out of country purchases. Duty free purchases are not substantial but are as high as 1.5 to 3.2% in France, Ireland and Scotland. Out of country, duty free and internet purchases appear to be negligible in Mexico between 2006 and 2011, and Uruguay between late 2006 and early 2011 (Table 3).
Tables 4 and 5 present prevalence estimates of tax avoidance/evasion based on self-reported packaging information or similar information gathered by the interviewers during face-to-face interviews. Table 4 presents descriptive data for Malaysia and Thailand. The estimates suggest little evidence of tax avoidance/evasion in Thailand but substantial and increasing tax avoidance/evasion in Malaysia. Estimates for waves 2 and 3 of ITC-Malaysia also suggest self-reported information may not be reliable: estimates based on self-reports are substantially different than estimates based on pack inspection. Table 5 presents data for Mauritius, Uruguay, China and Bangladesh. The estimates suggest an important decrease in cigarette tax avoidance/evasion between early 2005 and the first half of 2008 in Mauritius and, on the whole, an increase in Uruguay. As was the case with data from Malaysia, these estimates suggest self-reported information may not be reliable. For example, the estimates based on self-reports are appreciably higher in waves 1 and 2 but lower in wave 3 in Mauritius. Estimates for China suggest substantial cigarette tax avoidance/evasion. These estimates, however, should be treated with caution. The high number of respondents and more importantly interviewers who reported or recorded not knowing suggests potentially important difficulties during the data collection process. Estimates from Bangladesh, based on a small sub-sample (about 16%) of respondents who provided a cigarette pack to the interviewers, suggest low but perhaps increasing cigarette tax avoidance or evasion.
Table 4.
Percentage of smokers’ cigarette packs showing evidence of possible tax avoidance or evasion in Malaysia and Thailand
| Source of information | Malaysia | Thailand | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Wave 1 (Jan 05 – Mar 05) |
Wave 2 (Aug 06 – Jun 07) |
Wave 3 (Jan 08 – Aug 08) |
Wave 1 (Jan 05 – Mar 05) |
Wave 2 (Aug 06 – Jun 07) |
Wave 3 (Jan 08 – Aug 08) |
Wave 4 (Apr 09 – Nov 09) |
||||||||
| pack | self- report |
pack | self- report |
pack | self- report |
pack | self- report |
pack | self- report |
pack | self- report |
pack | self- report |
|
| % | % | % | % | % | % | % | % | % | % | % | % | % | % | |
| Health warnings | ||||||||||||||
| Non standard warning labels | 3.5 | 3.4 | 7.6 | 3.7 | 8.1 | 1.0* | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.2 | 0.2 | 0.0 |
| No warning labels | 3.8 | 3.1 | 8.4* | 7.1 | 2.3* | 3.1 | 0.0 | 0.5 | 0.0 | 0.0 | 0.9 | 0.2 | 2.7 | 0.0 |
| Refused/don't know | n/a | n/a | 7.8 | 2.9 | 0.4*** | 2.5 | 0.0 | 0.0 | 0.0 | 0.3 | 0.0 | 0.4 | 0.0 | 0.0 |
| n | 633 | 1048 | 379 | 871 | 957 | 498 | 260 | 754 | 31 | 829 | 211 | 784 | 299 | 618 |
| Tax stamps | ||||||||||||||
| No tax stamps or security ink | 9.2 | 9.2 | 13.7* | 10.0 | 39.7** | 14.3* | 0.0 | 0.0 | 0.0 | 0.0 | n/a | n/a | n/a | n/a |
| Refused/don't know | n/a | n/a | 10.8 | 3.6 | 2.0 | 22.9*** | n/a | n/a | 0.0 | 0.7 | n/a | n/a | n/a | n/a |
| n | 617 | 1035 | 376 | 880 | 374 | 248 | 260 | 754 | 31 | 828 | n/a | n/a | n/a | n/a |
Note: n represent the total number of valid responses per item (i.e., the denominator); n/a = not applicable.
*, ** and ***: significant difference at 5%, 1% and 0.1% with previous time period. These should be interpreted with great caution as the number of comparisons is absurdly high.
Table 5.
Percentage of smokers’ cigarette packs showing evidence of possible tax avoidance or evasion, Mauritius, Uruguay, China, Bangladesh
| Source of information | Mauritius | |||||
|---|---|---|---|---|---|---|
| Wave 1 (Jan 05 – Mar 05) |
Wave 2 (Aug 06 – Jun 07) |
Wave 3 (Jan 08 – Aug 08) |
||||
| pack | self-report | pack | self-report | pack | self-report | |
| % | % | % | % | % | % | |
| No tax stamps or authenticity labels | 11.0 | 16.6 | 7.6 | 10.2 | 3.1 | 1.7 |
| Don't know | 0.8 | 10.6 | 0.2 | 17.2 | 0.1 | 17.8 |
| n | 210 | 388 | 335 | 218 | 305 | 230 |
| Uruguay | ||||||
| Wave 1 (Oct 06 – Dec 06) |
Wave 2 (Oct 08 – Feb 09) |
Wave 3 (Oct 10 – Jan 11) |
||||
| Non standard warning labels | 6.4 | 4.2 | 3.1 | 5.1 | 12.0*** | 5.1 |
| No warning labels | 4.6 | 0.6 | 3.6 | 1.8 | 8.5*** | 5.6 |
| Don't know | 0.2 | 0.0 | 0.7 | 0.8 | 3.4*** | 3.7*** |
| n | 305 | 426 | 361 | 137 | 765 | 459 |
| China | ||||||
| Wave 1 (Mar 06 – Dec 06) |
Wave 2 (Oct 07 – Feb 08) |
Wave 3 (May 09 – Mar 10) |
||||
| No authenticity labels | 10.7 | 8.5 | 13.4* | 7.2 | 6.3*** | 7.2 |
| Don't know | 10.3 | 0.2 | 5.8*** | 0.2 | 2.5*** | 8.9*** |
| n | 2308 | 1089 | 2455 | 817 | 2553 | 1191 |
| Bangladesh | ||||||
| Wave 1 (Feb 09 – May 09) |
Wave 2 (Mar 10 – Jun 10) |
|||||
| No tax stamps | 0.4 | n/a | 3.7** | n/a | ||
| n | 368 | n/a | 401 | n/a | ||
Note: n represent the total number of valid responses per item (i.e., the denominator).
*, ** and ***: significant difference at 5%, 1% and 0.1% with previous time period. These should be interpreted with great caution as the number of comparisons is absurdly high.
Table 6 presents the results of the GEE models. In Canada, we find a statistically significant negative association between household income and the odds of engaging in cigarette tax avoidance/evasion (middle-, high-income are jointly significant; Wald test: chi-square=7.6 and p=0.020) while in the UK we find a statistically significant positive association (Wald test: chi-square=41.4 and p<0.001). In the United States, we find that individuals with higher education have higher odds of engaging in cigarette tax avoidance/evasion (middle-, high-education are jointly significant; Wald test: chi-square=16.2 and p<0.001). More specifically, in the United States the odds of engaging in cigarette tax avoidance/evasion for individuals with moderate and high levels of education are 1.6 and 1.5 times higher than those of individuals with low levels of education. We do not find statistically significant associations between socioeconomic status (measured by income or education) and the odds of engaging in cigarette tax avoidance/evasion in France. On the whole, the results suggest, that if anything, in France there are positive associations between household income and education and cigarette tax avoidance/evasion.
Table 6.
Generalized estimating equations models: factors associated with the odds of engaging in cigarette tax avoidance or evasion
| Canada | United States | United Kingdom | France | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Odds Ratio | 95% CI | p-value | Odds Ratio | 95% CI | p-value | Odds Ratio | 95% CI | p-value | Odds Ratio | 95% CI | p-value | |
| Income (ref. low) | ||||||||||||
| mid | 0.92 | (0.73, 1.16) | 0.49 | 0.88 | (0.70, 1.10) | 0.26 | 1.27 | (1.05, 1.54) | 0.01 | 1.05 | (0.80, 1.38) | 0.75 |
| high | 0.72 | (0.56, 0.92) | 0.01 | 1.10 | (0.85, 1.42) | 0.47 | 1.86 | (1.53, 2.25) | 0.00 | 1.21 | (0.87, 1.68) | 0.25 |
| dk/refused | 0.79 | (0.56, 1.11) | 0.17 | 0.65 | (0.40, 1.05) | 0.08 | 1.31 | (1.02, 1.68) | 0.04 | 0.66 | (0.32, 1.38) | 0.27 |
| Education (ref. low) | ||||||||||||
| mid | 1.00 | (0.80, 1.24) | 0.98 | 1.58 | (1.26, 1.99) | 0.00 | 1.09 | (0.92, 1.30) | 0.33 | 1.13 | (0.87, 1.47) | 0.37 |
| high | 1.06 | (0.79, 1.44) | 0.69 | 1.48 | (1.10, 1.99) | 0.01 | 1.15 | (0.91, 1.44) | 0.24 | 1.03 | (0.75, 1.42) | 0.84 |
| dk/refused | 0.35 | (0.02, 5.03) | 0.44 | 3.10 | (0.35, 27.74) | 0.31 | 0.54 | (0.25, 1.19) | 0.13 | 1.34 | (0.28, 6.41) | 0.72 |
| Sex (ref. men) | 1.11 | (0.91, 1.35) | 0.32 | 0.73 | (0.59, 0.90) | 0.00 | 1.08 | (0.93, 1.25) | 0.30 | 0.83 | (0.66, 1.05) | 0.12 |
| Age | 1.95 | (1.29, 2.93) | 0.00 | 2.18 | (1.44, 3.28) | 0.00 | 1.69 | (1.25, 2.29) | 0.00 | 0.66 | (0.36, 1.19) | 0.17 |
| Age squared | 0.95 | (0.91, 0.99) | 0.02 | 0.95 | (0.91, 0.99) | 0.02 | 0.96 | (0.93, 0.99) | 0.02 | 1.04 | (0.96, 1.12) | 0.33 |
| No. cigarettes per day (ref. 1–10 cig.) | ||||||||||||
| 11–20 cig. | 1.62 | (1.35, 1.95) | 0.00 | 1.40 | (1.09, 1.80) | 0.01 | 1.04 | (0.90, 1.19) | 0.61 | 1.17 | (0.89, 1.53) | 0.26 |
| 21–30 cig. | 1.44 | (1.13, 1.85) | 0.00 | 1.61 | (1.19, 2.16) | 0.00 | 1.23 | (1.02, 1.50) | 0.03 | 1.24 | (0.75, 2.08) | 0.40 |
| > 30 cig. | 1.91 | (1.30, 2.82) | 0.00 | 1.69 | (1.18, 2.41) | 0.00 | 1.18 | (0.91, 1.53) | 0.21 | 1.34 | (0.54, 3.32) | 0.53 |
| dk/refused | 1.70 | (0.44, 6.64) | 0.44 | 1.05 | (0.21, 5.17) | 0.96 | 1.22 | (0.56, 2.64) | 0.62 | - | - | - |
| Time to first cigarette (ref. > 60 min.) | ||||||||||||
| 31–60 min. | 0.66 | (0.50, 0.88) | 0.00 | 1.33 | (0.98, 1.79) | 0.07 | 0.91 | (0.75, 1.11) | 0.34 | 0.99 | (0.74, 1.33) | 0.96 |
| 6–30 min. | 0.85 | (0.66, 1.09) | 0.20 | 1.25 | (0.93, 1.69) | 0.15 | 0.93 | (0.77, 1.11) | 0.41 | 1.36 | (1.02, 1.81) | 0.04 |
| ≤ 5 min. | 1.07 | (0.80, 1.42) | 0.67 | 1.37 | (0.99, 1.90) | 0.06 | 0.98 | (0.80, 1.21) | 0.87 | 1.13 | (0.73, 1.75) | 0.58 |
| dk/refused | 0.80 | (0.52, 1.21) | 0.29 | 0.87 | (0.36, 2.09) | 0.76 | 1.01 | (0.64, 1.59) | 0.97 | 3.58 | (2.71, 4.73) | 0.00 |
| Intention to quit (ref. no) | ||||||||||||
| yes | 0.91 | (0.76, 1.09) | 0.30 | 0.71 | (0.59, 0.85) | 0.00 | 0.80 | (0.71, 0.89) | 0.00 | 0.79 | (0.64, 0.97) | 0.02 |
| dk/refused | 0.97 | (0.63, 1.48) | 0.89 | 0.66 | (0.40, 1.09) | 0.11 | 0.57 | (0.37, 0.89) | 0.01 | 0.68 | (0.15, 3.12) | 0.62 |
| Wave (ref. wave 1) | ||||||||||||
| wave 2 | 0.38 | (0.26, 0.55) | 0.00 | 0.81 | (0.55, 1.21) | 0.31 | 1.86 | (1.36, 2.56) | 0.00 | 0.78 | (0.56, 1.08) | 0.13 |
| wave 3 | 0.44 | (0.31, 0.62) | 0.00 | 0.81 | (0.56, 1.18) | 0.27 | 1.79 | (1.31, 2.45) | 0.00 | - | - | - |
| wave 4 | 0.50 | (0.36, 0.68) | 0.00 | 0.85 | (0.59, 1.21) | 0.37 | 1.64 | (1.23, 2.19) | 0.00 | - | - | - |
| wave 5 | 0.76 | (0.57, 1.01) | 0.06 | 0.75 | (0.53, 1.07) | 0.11 | 1.47 | (1.11, 1.95) | 0.01 | - | - | - |
| wave 6 | 0.84 | (0.65, 1.09) | 0.20 | 0.80 | (0.57, 1.12) | 0.20 | 1.45 | (1.11, 1.88) | 0.01 | - | - | - |
| wave 7 | 1.23 | (0.97, 1.56) | 0.09 | 0.62 | (0.45, 0.86) | 0.00 | 1.37 | (1.05, 1.78) | 0.02 | - | - | - |
| wave 8 | 1.12 | (0.89, 1.39) | 0.33 | 0.89 | (0.66, 1.21) | 0.47 | 1.19 | (0.92, 1.54) | 0.19 | |||
| Time-in-sample (ref. 1)* | ||||||||||||
| 2 | 1.18 | (0.99, 1.41) | 0.06 | 1.16 | (0.96, 1.40) | 0.12 | 1.35 | (1.15, 1.59) | 0.00 | 1.08 | (0.77, 1.50) | 0.67 |
| 3 | 1.45 | (1.18, 1.79) | 0.00 | 1.02 | (0.80, 1.30) | 0.89 | 1.50 | (1.26, 1.78) | 0.00 | - | - | - |
| 4 | 1.33 | (1.05, 1.68) | 0.02 | 0.94 | (0.71, 1.25) | 0.67 | 1.57 | (1.27, 1.93) | 0.00 | - | - | - |
| 5 | 1.65 | (1.27, 2.15) | 0.00 | 1.04 | (0.73, 1.49) | 0.81 | 1.66 | (1.29, 2.14) | 0.00 | - | - | - |
| 6 | 1.48 | (1.07, 2.05) | 0.02 | 1.15 | (0.73, 1.81) | 0.56 | 1.50 | (1.11, 2.04) | 0.01 | - | - | - |
| 7 | 1.69 | (1.15, 2.48) | 0.01 | 0.81 | (0.43, 1.55) | 0.53 | 1.88 | (1.30, 2.72) | 0.00 | - | - | - |
| 8 | 2.35 | (1.49, 3.69) | 0.00 | 0.82 | (0.37, 1.81) | 0.63 | 2.21 | (1.37, 3.56) | 0.00 | - | - | - |
| Intercept | 0.01 | (0.01, 0.04) | 0.00 | 0.00 | (0.00, 0.01) | 0.00 | 0.02 | (0.01, 0.03) | 0.00 | 0.63 | (0.19, 2.11) | 0.46 |
Note: age and age squared measured in units of ten years; For Canada, UK and France: GEE (link=logit, family=binomial, working correlation structure=unstructured), for US GEE (link=logit, family=binomial, working correlation structure=exchangeable);
average time-in-sample: Canada = 2.5, US = 2, UK = 2.5, France = 1.5.
Age is found to increase (at a decreasing rate) the odds of engaging in cigarette tax avoidance/evasion in Canada, the US and the UK but not in France. In France, age appears to decrease (at a decreasing rate) the odds of engaging in cigarette tax avoidance/evasion; this association between age and cigarette tax avoidance/evasion is, however, not statistically significant. More specifically, in Canada, the odds of a 40 years old are 1.33 times higher than those of a 30 years old, while the odds of a 50 years old are 1.20 times higher than those of a 40 years old. We do not find that men have statistically significant higher odds to engage in cigarette tax avoidance/evasion than women in Canada, the UK and France. We do find, however, that women in the US have statistically significant higher odds than men. On the whole, we find that heavier smokers tend to have higher odds of engaging in cigarette tax avoidance/evasion. These associations, however, are only statistically significant in Canada and the United States. We find that time to first cigarettes is likely associated with cigarette tax avoidance/evasion (the shorter the time to first cigarettes, the higher the odds). Finally, we find statistically significant associations between quit intentions and cigarette tax avoidance/evasion in the US, UK and France; smokers who intend to quit have lower odds of engaging in cigarette tax avoidance/evasion.
Discussion
Principal findings
We find the prevalence of cigarette tax avoidance/evasion varies substantially between countries, and across time within countries. In Canada, France and the United Kingdom, more than 10% of smokers report last purchasing cigarettes from low or untaxed sources while in Malaysia, some estimates suggest substantial cigarette tax avoidance/evasion between 2005 and 2008. In Canada, the percentage of smokers reporting buying cigarettes from low or untaxed sources increased more than four-fold between late 2002 and late 2007, early 2008 but seem to have leveled off or decreased since. We find that the sources of low or untaxed cigarettes is very much country-specific. In Canada, and to a lesser extent in the United States, the main source of low or untaxed cigarettes are First Nations /Indian reserves; in the United Kingdom, it is out of country and duty free purchases; while in France, it is out of country, but within the EU, purchases. It is important to note that our estimates of the prevalence of cigarette tax avoidance/evasion likely under estimates the proportion of total cigarettes that are purchased from low or untaxed sources as smokers who buy cigarettes from low or untaxed sources tend to be heavier smokers.8
As expected we find that, on the whole, heavier and more addicted smokers are more likely to engage in tax avoidance/evasion while smokers who intend to quit are less likely. We find important associations between household income and education and the likelihood of engaging in tax avoidance/evasion. These associations, however, vary both in direction and magnitude across countries. These results highlight the importance of taking into consideration country specific contextual factors. For example, the differentials in cigarette taxes or prices between neighboring jurisdictions are often assumed to be the key driver of cigarette tax avoidance/evasion. Although theoretically sound, such assertions ignore an array of other factors that may be equally or more important. For example in Canada, a province-level examination of the data reveals that cigarette tax avoidance/evasion is substantial in only two provinces, Ontario and Québec, the two provinces with the lowest cigarette taxes in Canada.[32] Such findings suggest that the proximity to opportunities for tax avoidance/evasion may be one of the key factors. Other important country-level factors may include the tax effectiveness of governments’ law enforcement, geographic situation (e.g., long and porous borders) and the extent to which the tobacco industry tolerates or actively encourages tax evasion.
Limitations
A number of limitations merit discussion. First, most prevalence estimates presented are based on self-reported responses and consequently, social desirability bias (when respondents provide socially desirable answers) cannot be ruled out. Responses may represent true behaviours or perceptions about what respondents thought investigators wanted to hear, or a combination of both. Cigarette tax evasion, an illegal activity, may be particularly affected.[33] Imperfect recall, especially with respect to pack information, may also introduce measurement error. It is not possible to disentangle the effects of social desirability bias and imperfect recall. Both may, at least in part, explain the differences we observed between estimates based on self-reports and estimates based on pack inspections. Second, unequal spacing between waves, and differences in the duration and timing of fieldwork suggest caution should be exercised when examining temporal differences. For example, as pointed out by Hyland et al., wave 2 of the Four Country surveys took place in the UK’s summer holiday season when respondents would have been more likely to visit neighboring countries where low or untaxed cigarettes may be more readily available compared with Wave 1, which took place late in the European autumn.[31] Third, it is difficult to disentangle tax avoidance from tax evasion. For example, carrying cigarettes across state or international borders may or may not be illegal, depending on the allowance permitted. Similarly, purchasing cigarettes on First Nations/Indian reserves is not necessarily illegal. Additional limitations include the inability to capture industry evasion, sales of untaxed cigarettes that are made in legitimate stores, and counterfeit cigarettes or cigarette with counterfeit stamps.
Implications for policy and research
Article 6.2 of the Framework Convention on Tobacco Control (FCTC), an international treaty negotiated by the member states of the World Health Organization with objective to reduce the prevalence of tobacco use and exposure to tobacco smoke, calls upon each Party to adopt or maintain measures that prohibit or restrict, sales to or importations by international travelers of tax- and duty-free tobacco products.[34] The proportion of smokers who report obtaining duty-free cigarettes at last purchase is substantial only in the UK (between about 3.5 and 7% during the past decade). Duty-free cigarette purchases are appreciably lower in other high-income countries but by no means negligible. Moreover, even if the true prevalence of duty-free cigarette purchases is on average only 1% (or even lower), such prevalence estimate represent considerable tax revenue losses. Additionally, there is some evidence that the availability of duty-free cigarettes has enabled cigarette tax evasion.[4] Duty-free cigarette allowances are limited in several countries; some countries such as Barbados, Singapore and Sri Lanka have gone further and no longer permit any duty-free allowances for cigarettes.[4] In November 2012, the Protocol to Eliminate Illicit Trade in Tobacco Products was adopted by the Parties to the FCTC. The new protocol’s aim is to combat illegal trade in tobacco products through control of the supply chain and international cooperation. A key measure of the protocol is the commitment of each Party to establish a global tracking and tracing system.
Most of the prevalence estimates of cigarette tax avoidance/evasion presented are national averages (with the exception of China and Mexico, ITC-surveys are nationally representative). National estimates may hide important and policy-relevant regional differences. For example, the four-fold increase between 2002 and 2008/09 in Canada was due in great part to an increase in purchasing on First Nations reserves within the province of Ontario. Similarly, an inspection of the pattern of tax avoidance in France shows that it was almost entirely due to smokers in border regions purchasing their cigarettes from the bordering countries.[35] Additional promising avenues for research include the examination of the effects of country-level factors on tax avoidance/evasion and the effects of tax avoidance/evasion on purchasing behavior (e.g., prices paid and quantities purchased) and on the pass-through rate for cigarettes.[36]
What this paper adds.
The magnitude and effectiveness of taxes at reducing tobacco use provide an incentive for tobacco users, manufacturers and others, most notably criminal networks, to devise ways to avoid or evade tobacco taxes. The magnitude of cigarette tax avoidance/evasion is challenging to quantify.
We find that the extent and the type of cigarette avoidance/evasion varies substantially between countries and across time. We also find important associations between household income and education and the likelihood of engaging in cigarette tax avoidance/evasion.
Acknowledgments
We thank Cuiping Long for her research assistance, Emily McGirr for research and editing assistance, and Grace Li and Ruth Loewen for guidance in using ITC data.
Funding: The data collection for the ITC Project is supported by grants R01 CA 100362 and P50 CA111236 (Roswell Park Transdisciplinary Tobacco Use Research Center, and P01 CA138389, R01 CA090955) from the National Cancer Institute of the United States, Robert Wood Johnson Foundation (045734), Canadian Institutes of Health Research (57897, 79551, and 115016), Commonwealth Department of Health and Aging, Canadian Tobacco Control Research Initiative (014578), National Health and Medical Research Council of Australia (265903), the International Development Research Centre (104831-002), the International Development Research Centre (African Tobacco Situational Analysis), New Zealand Health Research Council (06/453), New Zealand Ministry of Health, Mexican Consejo Nacional de Ciencia y Tecnologia (Salud-2007-C01-70032), Bloomberg Global Initiative—International Union Against Tuberculosis and Lung Disease, the Chinese Center for Disease Control and Prevention, the French Institute for Health Promotion and Health Education (INPES), the French National Cancer Institute (INCa), Observatoire français des drogues et toxicomanies (OFDT), the Netherlands Organisation for Health Research and Development (ZonMw) (the Netherlands), German Federal Ministry of Health, Dieter Mennekes-Umweltstiftung and Germany Cancer Research Center (DKFZ), Cancer Research UK (C312/A6465 and C312/A3726), NHS Health Scotland (RE065), Flight Attendants’ Medical Research Institute (FAMRI), GlaxoSmithKline (3516601), Pfizer (Ireland), the Korean Ministry of Health and Welfare, the Malaysian Ministry of Health, and ThaiHealth Promotion Foundation. GT Fong is supported by a Senior Investigator Award from the Ontario Institute for Cancer Research and a Prevention Scientist Award from the Canadian Cancer Society Research Institute. The SILNE Project is funded by the European Commission through FP7 HEALTH-F3- 2011-278273.
Footnotes
The production of genuine brands may be illegal if legal manufacturers declare only a fraction of their production to the tax authorities. [4]
It is worth noting that the littered pack approach will likely overstate the extent of tax avoidance/evasion given that it picks up commuting, tourism, etc. that have nothing to do with tax avoidance. Moreover, smokers that litter may be more likely to engage in tax avoidance/evasion.
ITC surveys were also conducted in Germany (2007, 2009, 2011),New Zealand (2007/08, 2008/09) and in the Republic of Korea (2005, 2008, 2010): ITC-Germany omitted the most useful response categories to identify low or untaxed sources of cigarettes in all its waves; ITC-New Zealand omitted the most useful question in wave 2 while nearly 90% of respondents failed to answer in wave 1; and, estimates of tax avoidance/evasion in the Republic of Korea were not distinguishable from zero. Consequently, we do not present or use any of ITC-Germany, ITC-New Zealand and ITC-Korea data in our analyses.
ITC-China was conducted in six cities (Beijing, Changsha, Guangzhou, Shanghai, Shenyang and Yinchuan); ITC-Mexico was conducted in four cities (Mexico City, Guadalajara, Tijuana, and Ciudad Juárez).
Although the prevalence of tax avoidance/evasion appears to be important in China, Malaysia, Mauritius, Thailand, and Uruguay; we do not use data from these countries because of the differences we observed between estimates based on self-reports and estimates based on pack inspections.
Data on last purchased cigarettes from a low or untaxed source are also available from Australia, Ireland, the Netherlands,, Mexico and Uruguay; in these countries, the prevalence of cigarette tax avoidance/evasion is small to negligible.
Estimates for the UK more likely represent tax avoidance rather than tax evasion as most of the sources of last cigarette purchases are Out of state/province/country and duty free.
This will be the case inasmuch as heavier smokers purchase a large proportion of their cigarettes from low or untaxed sources.
Contributors: GEG led the writing of the manuscript, the data analysis and interpretation of data and contributed substantially to the design of the study. PD contributed substantially to the data analysis, interpretation of data and the design of the study and revised the article critically for important intellectual content. FJC conceived the study and contributed substantially to the interpretation of data and revised the article critically for important intellectual content. GTF is Founder and Chief Principal Investigator of the International Tobacco Control Policy Evaluation Project and contributed to the design of the study. All of the authors gave final approval of the version to be published.
Competing interests: None.
References
- 1.Chaloupka FJ, Warner KE. The Economics of Smoking. In: Culyer AJ, Newhouse JP, editors. Handbook of Health Economics. Amsterdam: Elsevier Science, North-Holland; 2000. pp. 1539–1627. [Google Scholar]
- 2.Gallet CA, List JA. Cigarette demand: a meta-analysis of elasticities. Health Econ. 2003;12(10):821–835. doi: 10.1002/hec.765. [DOI] [PubMed] [Google Scholar]
- 3.International Agency for Research on Cancer. IARC Handbooks of Cancer Prevention: Tobacco Control. Volume 14. Effectiveness of Price and Tax Policies for Control of Tobacco. Lyon: International Agency for Research on Cancer; 2011. [Google Scholar]
- 4.World Health Organization. WHO Technical Manual on Tobacco Tax Administration. Geneva: WHO; 2010. [Google Scholar]
- 5.Health Canada. Contraband Cigarettes: Tobacco Smoke Analysis. Ottawa: Health Canada; 2010. [Google Scholar]
- 6.RCMP. Contraband Tobacco Enforcement Strategy. Ottawa: Royal Canadian Mounted Police; 2008. [Google Scholar]
- 7.IARC Working Group on Methods for Evaluating Tobacco Control Policies. IARC Handbooks of Cancer Prevention, Tobacco Control, Vol. 12: Methods for Evaluating Tobacco Control Policies. Lyon International Agency for Research on Cancer; 2008. [Google Scholar]
- 8.Merriman D. Tool 7. Smuggling. Understand, Measure and Combat Tobacco Smuggling. World Bank Economics of Tobacco Toolkit. In: Yurekli A, de Beyer J, editors. World Bank Economics of Tobacco Toolkit. World Bank; 2002. [Google Scholar]
- 9.HM Customs and Excise. Measuring Indirect Tax Fraud. London: HMCE; 2001. [Google Scholar]
- 10.Customs HRa. Tackling Tobacco Smuggling – building on our success. A renewed strategy for HM Revenue & Customs and the UK Border Agency. London: HMRC, UKBA; 2011. [Google Scholar]
- 11.Physicians for a Smoke-Free Canada. The Canadian Tobacco Market Place. Estimating the volume of Contraband Sales of Tobacco in Canada. Updated – April 2010. Ottawa: Physicians for a Smoke-Free Canada; 2010. [Google Scholar]
- 12.Physicians for a Smoke-Free Canada. The Canadian Tobacco Market Place. Estimating the volume of Contraband Sales of Tobacco in Canada in 2009. Updated – January 2011. Ottawa: Physicians for a Smoke- Free Canada; 2011. [Google Scholar]
- 13.Stehr M. Cigarette tax avoidance and evasion. J. Health Econ. 2005;24(2):277–297. doi: 10.1016/j.jhealeco.2004.08.005. [published Online First: Epub Date]|. [DOI] [PubMed] [Google Scholar]
- 14.Callaghan RC, Veldhuizen S, Ip D. Contraband cigarette consumption among adolescent daily smokers in Ontario, Canada. Tob. Control. 2011;20(2):173–174. doi: 10.1136/tc.2010.037507. [published Online First: Epub Date]|. [DOI] [PubMed] [Google Scholar]
- 15.Callaghan RC, Veldhuizen S, Leatherdale S, Murnaghan D, Manske S. Use of contraband cigarettes among adolescent daily smokers in Canada. CMAJ. 2009;181(6–7):384–386. doi: 10.1503/cmaj.090665. [published Online First: Epub Date]|. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Leatherdale ST, Ahmed R, Vu M. Factors associated with different cigarette access behaviours among underage smoking youth who usually smoke contraband (native) cigarettes. Can. J. Public Health. 2011;102(2):103–107. doi: 10.1007/BF03404156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Luk R, Cohen JE, Ferrence R, McDonald PW, Schwartz R, Bondy SJ. Prevalence and correlates of purchasing contraband cigarettes on First Nations reserves in Ontario, Canada. Addiction. 2009;104(3):488–495. doi: 10.1111/j.1360-0443.2008.02453.x. [published Online First: Epub Date]|. [DOI] [PubMed] [Google Scholar]
- 18.DeCicca P, Kenkel DS, Liu F. NBER Working Paper No. 15941. NBER Working Paper Series. Cambridge: National Bureau of Economic Research; 2010. Excise Tax Avoidance: The Case of State Cigarette Taxes. [Google Scholar]
- 19.Emery S, White MM, Gilpin EA, Pierce JP. Was there significant tax evasion after the 1999 50 cent per pack cigarette tax increase in California? Tob. Control. 2002;11(2):130–134. doi: 10.1136/tc.11.2.130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Chiou L, Muehlegger E. Crossing the Line: Direct Estimation of Cross-Border Cigarette Sales and the Effect on Tax Revenue. B.E. Journal of Economic Analysis and Policy: Contributions to Economic Analysis and Policy. 2008;8(1) [Google Scholar]
- 21.Taylor AJ, Langdon M, Campion P. Smuggled tobacco, deprivation and addiction. European journal of public health. 2005;15(4):399–403. doi: 10.1093/eurpub/cki006. [published Online First: Epub Date]|. [DOI] [PubMed] [Google Scholar]
- 22.Sarntisart I. HNP Discussion Paper, Economics of Tobacco Control Paper No 15. Washington: The World Bank; 2003. An Economic Analysis of Tobacco Control in Thailand. [Google Scholar]
- 23.Chen H-F, Chen S-H, Lee J-M, Jeng H-Y. Who Are the Potential Smokers of Smuggled Cigarettes? Asian Economic Journal. 2010;24(3):221–234. [published Online First: Epub Date]|. [Google Scholar]
- 24.Lakhdar CB. Quantitative and qualitative estimates of cross-border tobacco shopping and tobacco smuggling in France. Tob. Control. 2008;17(1):12–16. doi: 10.1136/tc.2007.020891. [published Online First: Epub Date]|. [DOI] [PubMed] [Google Scholar]
- 25.Barkans M. Department of Community Health Sciences, Faculty of Applied Health Sciences. St. Catharines: Brock University; 2010. Contraband tobacco on post-secondary campuses in Ontario. Master of Arts, Applied Health Sciences. [Google Scholar]
- 26.Kurti MK, von Lampe K, Thompkins DE. The illegal cigarette market in a socioeconomically deprived inner-city area: the case of the South Bronx. Tob. Control. 2012 doi: 10.1136/tobaccocontrol-2011-050412. [published Online First: Epub Date]|. [DOI] [PubMed] [Google Scholar]
- 27.Merriman D. The Micro-Geography of Tax Avoidance: Evidence from Littered Cigarette Packs in Chicago. Am Econ J-Econ Polic. 2010;2(2):61–84. [published Online First: Epub Date]|. [Google Scholar]
- 28.Wilson N, Thomson G, Edwards R, Peace J. Estimating missed government tax revenue from foreign tobacco: survey of discarded cigarette packs. Tob. Control. 2009;18(5):416–418. doi: 10.1136/tc.2009.031278. [published Online First: Epub Date]|. [DOI] [PubMed] [Google Scholar]
- 29.Fong GT, Cummings KM, Borland R, et al. The conceptual framework of the International Tobacco Control (ITC) Policy Evaluation Project. Tob. Control. 2006;15(Suppl 3):iii3–iii11. doi: 10.1136/tc.2005.015438. [published Online First: Epub Date]|. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Thompson ME, Fong GT, Hammond D, et al. Methods of the International Tobacco Control (ITC) Four Country Survey. Tob. Control. 2006;15(Suppl 3):iii12–iii18. doi: 10.1136/tc.2005.013870. [published Online First: Epub Date]|. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Hyland A, Laux FL, Higbee C, et al. Cigarette purchase patterns in four countries and the relationship with cessation: findings from the International Tobacco Control (ITC) Four Country Survey. Tob. Control. 2006;15(Suppl 3):iii59–iii64. doi: 10.1136/tc.2005.012203. [published Online First: Epub Date]|. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Ontario Tobacco Research Unit. The Tobacco Control Environment: Ontario and Beyond. Monitoring and Evaluation Series. Tobacco Taxes: Monitoring Update (April 24, 2011) Toronto: Ontario Tobacco Research Unit; 2011. [Google Scholar]
- 33.Krumpal I. Determinants of social desirability bias in sensitive surveys: a literature review. Quality & Quantity. 2011:1–23. [published Online First: Epub Date]|. [Google Scholar]
- 34.World Health Organization. Framework Convention on Tobacco Control. Geneva: World Health Organization; 2003. [Google Scholar]
- 35.Nagelhout GE, van den Putte B, Allwright S, et al. Socioeconomic and country variations in cross-border cigarette purchasing as tobacco tax avoidance strategy. Findings from the ITC Europe Surveys. Tob. Control. 2013 doi: 10.1136/tobaccocontrol-2012-050838. [published Online First: Epub Date]|. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Harding M, Leibtag E, Lovenheim MF. The Heterogeneous Geographic and Socioeconomic Incidence of Cigarette Taxes: Evidence from Nielsen Homescan Data. American Economic Journal: Economic Policy. 2012;4(4):169–198. [Google Scholar]