Illicit Cigarette Trade in Five South American Countries: A Gap Analysis for Argentina, Brazil, Chile, Colombia, and Peru

Illicit Cigarette Trade in Five South American Countries: A Gap Analysis for Argentina, Brazil,... Abstract Introduction Because of its nature, it is very hard to measure illicit tobacco trade in any product. In the case of Latin American countries, there is scant information on the magnitude and characteristics of this cigarette trade. The goal of this article is to provide estimates on the evolution of the illicit cigarette trade in five South American countries: Argentina, Brazil, Chile, Colombia, and Peru. Methods Gap analysis estimates for cigarette tax evasion/avoidance (a comparison on the evolution of the difference between registered cigarette sales and measured population consumption) are developed for Argentina, Brazil, Chile, Colombia, and Peru. Nationally representative surveys, conducted regularly, are used to measure population consumption. Confidence intervals constructed by bootstrapping sample estimates are generated to statistically evaluate the evolution of the gap. Results Illicit cigarette trade has increased as a percentage of total sales in Brazil in recent years. In the case of Argentina, after a relative decrease between 2005 and 2009, it seems to have stabilized. There is no statistical evidence to argue that there has been an increase of illicit cigarette trade in Chile, Colombia, and Peru, despite substantial price increases in Chile and tax increase in both Colombia and Peru. Conclusions Using simple statistical methods, it is possible to assess the trend in illicit tobacco trade over time to better inform policy makers. Getting reliable and regular population consumption surveys can also help to track illicit tobacco trade. Claims by tobacco industry of a positive association between price/tax changes and illicit trade are unsubstantiated. Implications Evolution of illicit cigarette trade in five Latin American countries shows different trajectories, not in line with tobacco industry estimates, which highlight the importance of producing solid, independent estimates. There are inexpensive methodologies that can provide estimates of the evolution of the relative importance of illicit trade and can be used to inform policy makers. Introduction The World Health Organization Framework Convention on Tobacco Control (FCTC) defines illicit tobacco trade as “any practice or conduct prohibited by law which relates to production, shipment, receipt, possession, distribution, sale or purchase including any practice or conduct intended to facilitate such activity.”1 This trade usually involves criminal organizations smuggling in and distributing, large quantities of tobacco products in domestic markets, though it can also involve individuals bringing in smaller quantities for resale. This should not be confused with certain forms of tax avoidance (ie, the purchase of tobacco products in accordance with countries’ customs and tax regulations), which include cross-border shopping for personal consumption (ie, individuals purchasing tobacco products in other countries, in compliance home countries’ regulations regarding quantities, and not for resale), duty-free shopping (ie, similar to cross-border shopping but acquired in duty-free shops or areas), Internet and E-mail/phone purchases (ie, remote purchases that are allowed in the buyer’s home country), etc.2 While tax avoidance is regulated and accepted by the authorities (because of the inefficiency of controlling the circulation of small quantities for personal use), illicit trade usually involves tax evasion, counterfeiting, etc.2 and is explicitly combated by authorities. The reasons for this are that it erodes two of the main goals of tobacco taxes: raising tobacco taxes to reduce consumption (as illicit products are usually cheaper than legitimate ones) and generating fiscal revenues. The tobacco industry usually associates the existence of illicit trade with tobacco taxes, claiming that increasing the price of tobacco products via taxes incentivizes criminals to increase illicit trade, which, in turn, defeats the objectives of reducing consumption and increasing fiscal revenues,3,4 though these claims have been debunked.5 In fact, a number of studies have shown that smuggling in tobacco products has, on occasions, been closely linked to the tobacco industry’s interests and even its participation.6,7 Because of its nature, it is very hard to measure illicit trade in tobacco (or any) product, though a few methods have been proposed, all with advantages and disadvantages. They range from surveys of tobacco users, examination of discarded cigarette packages or those obtained from smokers, monitoring of tobacco trade flows, econometric modeling, etc.2 In the case of Latin American countries, there is scant information on the magnitude and characteristics of illicit cigarette trade, by far, the most-sold tobacco product in the region.8 Some studies have attempted to quantify it for some countries using diverse sources of information and different methodologies. In some cases (eg, Argentina, Bolivia, and Brazil), aggregate annual data on cigarette exports and imports are used to estimate the volume of illegal trade9–11; in others (eg, Chile, Brazil, and Uruguay), legal cigarette sales are compared with the results reported in consumption surveys12–15; others (eg, Panama and Colombia) undertake ad hoc surveys of smokers to measure contraband.16–18 In the case of Argentina, estimates are made assuming different prevalence rates for cigarette smoking (because authors were unaware of the real ones), so the estimates of illicit trade are too broad and not very informative.9 For example, they range from 29% to 40% of the total market for 1996, while it is from 36% to 46% for 2002. A recent study in Brazil uses 2008 Global Adult Tobacco Survey (GATS) and the 2013 National Health Survey (PNS) and defines a minimum sales price for legal cigarettes (considering costs and profits for distributors and retailers).14 When these minimum prices are compared with those actually paid by consumers (prices below the minimum would imply illegal cigarettes), the authors estimate that the illicit cigarette trade grew from 17% in 2008 to 31% in 2013. These estimates, though obtained using a sound methodology, do not appear to be consistent with those of previous studies for Brazil, which put illicit trade at 28% in 2002.11 A study carried out in Chile compares consumption reported in the 2002 Survey of Drug Use in the General Population (EDPG) with domestic sales (provided by the tobacco industry), concluding that illicit trade that year was reportedly 4.2% of the local market.12 This does not consider the phenomenon of underreporting of cigarette smoking, which is usually present in consumption surveys.5 Another survey for Chile estimates tobacco tax evasion, though without distinguishing between domestic evasion and smuggling, using a private survey on the consumption of cigarettes and alcoholic beverages.13 This is used to estimate the amount of specific tax that should have been collected in 2011, which is compared with the actual amount of the tax collected, thus, estimating that tax evasion (not necessarily because of smuggling) accounted for 17.3% of total sales. A survey of smokers was recently carried out in Colombia to measure penetration of illicit trade in the country’s five largest cities.18 The results of this survey show that average penetration of illegal trade is 3.5% of the total market. Lastly, no independent studies were found on the size or the evolution of the illicit cigarette trade in Peru. As can be seen, there is scant evidence and estimates for countries with more than one study are not always consistent with each other. This means that public debate is often dominated by dubious or imprecise numbers supplied by the tobacco industry, as diverse authors have reported,19,20 especially in the form of press reports that are widely reproduced by local media outlets.21–24 Given the importance of illicit trade to public policy, it is inconvenient to base the debate on data whose origin is not very transparent and evolution is unreliable. Thus, the objective of this article is to provide estimates on the evolution of the illicit cigarette trade in five South American countries: Argentina, Brazil, Chile, Colombia, and Peru. These estimates use secondary data and publicly available data sources and allow a gap analysis of said evolution to be carried out. Despite the limitations of this methodology, it allows transparent and easily replicable estimates to be obtained that can serve as benchmarks for public debate. The use of these estimates can serve to compensate the estimates provided by the tobacco industry. Methods A gap analysis estimates tax evasion/avoidance based on a mathematical identity that holds that total consumption—of cigarettes, for example—in a country is equal to the consumption registered (which paid the corresponding taxes), plus what was not registered (contraband, tax evasion, tax avoidance, and product counterfeiting).2 The official statistics on units that pay taxes ought to supply registered consumption, while total consumption can be estimated from population consumption surveys. They must provide information on the total number of smokers and the average number of cigarettes they consume in a given period of time (eg, 1 month). Multiplying both numbers allows one to estimate total consumption in said period and unregistered consumption can be obtained by subtracting registered consumption from this number. This method is simple, easily replicable and does not require expensive data collection processes as it uses available data sources. Given the uncertainty regarding the extent to which consumption is underreported, an issue that often affects consumption surveys, this method only allows one to estimate deviations from the central trend, assuming that the proportion of underreporting remains constant over time, a reasonable assumption when the surveys used are relatively close to each other in time.2,25 For example, previous studies in South America using this method do not consider the possibility of underreporting.12,13 Working with just one survey and failing to consider another point in time makes the estimates obtained in this way unreliable, especially because underreporting can be high.5 Having obtained the gap between the estimated consumption and registered sales, it is necessary to produce confidence intervals on such a gap, to assess if changes over time are statistically significant. The simplest and more intuitive way of estimating these intervals is by bootstrapping the point estimate of the product of the reported smoking prevalence and the intensity of smoking. Each round of bootstrapping creates a new random sample (with replacement) from the original, considering weights and sampling strategies. By repeating 1000 times this procedure, it is possible to obtain a distribution of such a parameter that is close to the actual distribution and to estimate confidence intervals. This is conducted for each survey/country, such that 99% confidence intervals are calculated. Data The information used to estimate consumption at the population level comes from official surveys that are carried out regularly. In the case of Argentina, the National Risk Factors Survey (ENFR) for the years 2005, 2009, and 2013 contains information on average monthly consumption and intensity of cigarette smoking, among other variables, in the population more than the age of 18 years (with nationally representative samples of 41 392, 34 732, and 32 365 observations, respectively). There is also the 2008 and 2011 National Survey on the Prevalence of Psychoactive Substance Consumption (EnPreCosp), with nationally representative samples for the population between 16 and 65 years of age (34 203 and 34 343 observations, respectively). Both editions of the survey ask about the number of days that cigarettes were smoked during the last month, and the average number of cigarettes smoked on those days. The ENFR and the EnPreCosp are not directly comparable, given that they cover different population groups, but they can be used individually for the gap analysis. The surveys used in Brazil are the 2008 GATS and the 2013 PNS. The former is nationally representative for the population more than the age of 15 years (39 425 observations), while the latter is nationally representative for the entire population (205 545 observations). To make them comparable, people under the age of 15 years are eliminated from the 2013 PNS and it is implicitly assumed that the young population not covered by the surveys does not substantially alter their relative consumption compared with the rest of the population over time. In both the surveys, people are asked about the average number of cigarettes smoked per day by daily smokers and per week for those who do not smoke every day. For Chile, the EDPG from 2008, 2010, 2012, and 2014 was used (with samples of 17 133, 15 576, 17 154, and 19 512, respectively). It is nationally representative of the population from 12 to 65 years of age. In all the cases, people were asked about monthly tobacco consumption and the intensity of use on an average day. For Colombia, the Psychoactive Substance Consumption Survey (ECSP) for 2008 and 2013 was used, with national coverage of the population aged 12–65 years and with samples of 29 164 and 32 605, respectively. The 2008 ECSP does not count the number of cigarettes individuals reported smoking on an average day (something that is reported in the 2013 edition). In contrast, this information is reported by intervals: (1) less than 1 cigarette/day, (2) from 1 to 5, (3) from 6 to 10, (4) from 11 to 20, and (5) more than 20. Since this variable is essential to calculating the amounts consumed, a number needs to be assigned to each interval. For the first interval, it is assumed that those who reported smoking less than one cigarette a day smoke one a day. For intervals (2), (3), and (4), a random number is taken (with uniform distribution) between the limits of the ranges and that amount is assigned. Lastly, for the open interval (5), a number is taken (with uniform distribution) between the lower limit of the range (20 cigarettes/day) and a number of cigarettes such that the average of this interval coincides with the average number of cigarettes smoked by those who reported consuming more than 20 cigarettes/day in the 2013 ECSP. Thus, the average number smoked by people who reported consuming more than 20 cigarettes/day in the 2013 ECSP is equal to 47 cigarettes. These data are used to estimate the upper limit, considering the formula of averages in a uniform distribution. Assigning a consumption of one cigarette to those who consume less than one cigarette/day is likely to be a conservative assumption in terms of the evolution of illicit trade, as it raises the number of cigarettes consumed in the base year. Lastly, in the case of Peru, the drug use surveys carried out by the National Commission for Development and Life without Drugs (DEVIDA) in 2006 and 2010 were used, which is nationally representative of the population 12–65 years of age (with samples of 11 825 and 20 275, respectively). The survey contains complete information on the smoking population and average daily intensity of tobacco use. Table 1 shows the main variables taken from each survey to calculate annual consumption estimates. Table 1. Descriptive Variables of National Surveys Country  Survey  Year  Total population  Month prevalence (%)  Number of smokers (past month)  Average intensity  Annual estimated consumption  Lower bound  Upper bound  Argentina  ENFR (18 y old and more)  2005  22 935 297  30.0  6 890 222  11.0  27 695 319 985  25 851 269 324  29 539 370 646  2009  24 434 595  28.5  6 975 173  11.2  28 005 396 000  26 342 428 288  29 668 363 712  2013  25 777 587  26.8  6 920 767  11.1  27 645 902 000  25 310 222 192  29 981 581 808  EnPreCosP (16–65 y old)  2008  21 669 915  31.0  6 714 423  11.2  25 506 871 000  23 371 745 672  27 641 996 328  2011  22 520 925  28.9  6 510 124  10.8  25 263 861 000  23 679 438 104  26 848 283 896  Brazil  GATS/PNS (15 y old and more)  2008  142 998 657  17.2  24 552 398  10.0  88 146 859 982  83 292 130 382  93 001 589 582  2013  146 278 273  14.7  21 515 018  9.9  77 032 649 061  71 402 391 477  82 662 906 645  Chile  EDPG (12–65 y old)  2008  9 047 804  40.2  3 638 122  7.6  7 376 469 872  6 465 687 004  8 287 252 740  2010  9 426 076  35.3  3 324 577  8.5  7 630 861 705  6 773 205 398  8 488 518 012  2012  9 871 872  33.4  3 296 218  8.1  6 926 764 927  6 113 687 745  7 739 842 109  2014  9 683 210  33.8  3 268 955  8.6  7 286 892 666  6 201 028 754  8 372 756 579  Colombia  ECSP (12–65 y old)  2008  19 764 799  17.1  3 371 875  9.2  9 662 052 000  8 355 023 088  10 969 080 912  2013  23 301 463  13.0  3 017 539  8.0  7 225 791 000  6 490 067 368  7 961 514 632  Peru  DEVIDA (12–65 y old)  2006  10 748 960  17.5  1 875 854  4.2  2 856 799 389  2 349 876 077  3 363 722 701  2010  12 107 628  13.3  1 607 070  4.6  2 696 604 670  2 117 156 654  3 276 052 686  Country  Survey  Year  Total population  Month prevalence (%)  Number of smokers (past month)  Average intensity  Annual estimated consumption  Lower bound  Upper bound  Argentina  ENFR (18 y old and more)  2005  22 935 297  30.0  6 890 222  11.0  27 695 319 985  25 851 269 324  29 539 370 646  2009  24 434 595  28.5  6 975 173  11.2  28 005 396 000  26 342 428 288  29 668 363 712  2013  25 777 587  26.8  6 920 767  11.1  27 645 902 000  25 310 222 192  29 981 581 808  EnPreCosP (16–65 y old)  2008  21 669 915  31.0  6 714 423  11.2  25 506 871 000  23 371 745 672  27 641 996 328  2011  22 520 925  28.9  6 510 124  10.8  25 263 861 000  23 679 438 104  26 848 283 896  Brazil  GATS/PNS (15 y old and more)  2008  142 998 657  17.2  24 552 398  10.0  88 146 859 982  83 292 130 382  93 001 589 582  2013  146 278 273  14.7  21 515 018  9.9  77 032 649 061  71 402 391 477  82 662 906 645  Chile  EDPG (12–65 y old)  2008  9 047 804  40.2  3 638 122  7.6  7 376 469 872  6 465 687 004  8 287 252 740  2010  9 426 076  35.3  3 324 577  8.5  7 630 861 705  6 773 205 398  8 488 518 012  2012  9 871 872  33.4  3 296 218  8.1  6 926 764 927  6 113 687 745  7 739 842 109  2014  9 683 210  33.8  3 268 955  8.6  7 286 892 666  6 201 028 754  8 372 756 579  Colombia  ECSP (12–65 y old)  2008  19 764 799  17.1  3 371 875  9.2  9 662 052 000  8 355 023 088  10 969 080 912  2013  23 301 463  13.0  3 017 539  8.0  7 225 791 000  6 490 067 368  7 961 514 632  Peru  DEVIDA (12–65 y old)  2006  10 748 960  17.5  1 875 854  4.2  2 856 799 389  2 349 876 077  3 363 722 701  2010  12 107 628  13.3  1 607 070  4.6  2 696 604 670  2 117 156 654  3 276 052 686  ENFR = National Risk Factors Survey; GATS/PNS = Global Adult Tobacco Survey/ National Health Survey; EDPG = Survey of Drug Use in the General Population; ECSP = Psychoactive Substance Consumption Survey; DEVIDA = National Commission for Development and Life without Drugs. View Large Table 1. Descriptive Variables of National Surveys Country  Survey  Year  Total population  Month prevalence (%)  Number of smokers (past month)  Average intensity  Annual estimated consumption  Lower bound  Upper bound  Argentina  ENFR (18 y old and more)  2005  22 935 297  30.0  6 890 222  11.0  27 695 319 985  25 851 269 324  29 539 370 646  2009  24 434 595  28.5  6 975 173  11.2  28 005 396 000  26 342 428 288  29 668 363 712  2013  25 777 587  26.8  6 920 767  11.1  27 645 902 000  25 310 222 192  29 981 581 808  EnPreCosP (16–65 y old)  2008  21 669 915  31.0  6 714 423  11.2  25 506 871 000  23 371 745 672  27 641 996 328  2011  22 520 925  28.9  6 510 124  10.8  25 263 861 000  23 679 438 104  26 848 283 896  Brazil  GATS/PNS (15 y old and more)  2008  142 998 657  17.2  24 552 398  10.0  88 146 859 982  83 292 130 382  93 001 589 582  2013  146 278 273  14.7  21 515 018  9.9  77 032 649 061  71 402 391 477  82 662 906 645  Chile  EDPG (12–65 y old)  2008  9 047 804  40.2  3 638 122  7.6  7 376 469 872  6 465 687 004  8 287 252 740  2010  9 426 076  35.3  3 324 577  8.5  7 630 861 705  6 773 205 398  8 488 518 012  2012  9 871 872  33.4  3 296 218  8.1  6 926 764 927  6 113 687 745  7 739 842 109  2014  9 683 210  33.8  3 268 955  8.6  7 286 892 666  6 201 028 754  8 372 756 579  Colombia  ECSP (12–65 y old)  2008  19 764 799  17.1  3 371 875  9.2  9 662 052 000  8 355 023 088  10 969 080 912  2013  23 301 463  13.0  3 017 539  8.0  7 225 791 000  6 490 067 368  7 961 514 632  Peru  DEVIDA (12–65 y old)  2006  10 748 960  17.5  1 875 854  4.2  2 856 799 389  2 349 876 077  3 363 722 701  2010  12 107 628  13.3  1 607 070  4.6  2 696 604 670  2 117 156 654  3 276 052 686  Country  Survey  Year  Total population  Month prevalence (%)  Number of smokers (past month)  Average intensity  Annual estimated consumption  Lower bound  Upper bound  Argentina  ENFR (18 y old and more)  2005  22 935 297  30.0  6 890 222  11.0  27 695 319 985  25 851 269 324  29 539 370 646  2009  24 434 595  28.5  6 975 173  11.2  28 005 396 000  26 342 428 288  29 668 363 712  2013  25 777 587  26.8  6 920 767  11.1  27 645 902 000  25 310 222 192  29 981 581 808  EnPreCosP (16–65 y old)  2008  21 669 915  31.0  6 714 423  11.2  25 506 871 000  23 371 745 672  27 641 996 328  2011  22 520 925  28.9  6 510 124  10.8  25 263 861 000  23 679 438 104  26 848 283 896  Brazil  GATS/PNS (15 y old and more)  2008  142 998 657  17.2  24 552 398  10.0  88 146 859 982  83 292 130 382  93 001 589 582  2013  146 278 273  14.7  21 515 018  9.9  77 032 649 061  71 402 391 477  82 662 906 645  Chile  EDPG (12–65 y old)  2008  9 047 804  40.2  3 638 122  7.6  7 376 469 872  6 465 687 004  8 287 252 740  2010  9 426 076  35.3  3 324 577  8.5  7 630 861 705  6 773 205 398  8 488 518 012  2012  9 871 872  33.4  3 296 218  8.1  6 926 764 927  6 113 687 745  7 739 842 109  2014  9 683 210  33.8  3 268 955  8.6  7 286 892 666  6 201 028 754  8 372 756 579  Colombia  ECSP (12–65 y old)  2008  19 764 799  17.1  3 371 875  9.2  9 662 052 000  8 355 023 088  10 969 080 912  2013  23 301 463  13.0  3 017 539  8.0  7 225 791 000  6 490 067 368  7 961 514 632  Peru  DEVIDA (12–65 y old)  2006  10 748 960  17.5  1 875 854  4.2  2 856 799 389  2 349 876 077  3 363 722 701  2010  12 107 628  13.3  1 607 070  4.6  2 696 604 670  2 117 156 654  3 276 052 686  ENFR = National Risk Factors Survey; GATS/PNS = Global Adult Tobacco Survey/ National Health Survey; EDPG = Survey of Drug Use in the General Population; ECSP = Psychoactive Substance Consumption Survey; DEVIDA = National Commission for Development and Life without Drugs. View Large Regarding information on registered consumption, official information from cigarette sales is used whenever it is available. In the case of Argentina, the information on annual registered cigarette consumption comes from the Ministry of Agro-industry. According to this source, 36.9 billion, 43.3 billion, 42.4 billion, 43.1 billion, and 41.8 billion cigarettes were sold in 2005, 2008, 2009, 2011, and 2013, respectively.26 For Brazil, the information on registered cigarette sales comes from the Federal Revenues Service, which reports that 105.9 billion and 75.9 billion cigarettes were sold in 2008 and 2013, respectively.27,28 There is no official information on registered sales for Chile, Colombia, and Peru. Because of this, sales information from Euromonitor International (EI) is used. While there is evidence that EI’s illicit trade estimates are inconsistent over time and tend to overestimate real illicit trade,29 the same does not seem to be the case with the companies’ numbers on production, sales, or market share. These numbers are frequently used30–32 and there is no suspicion that they might be manipulated, basically because there are not significant incentives to do so. In the case of Chile, EI reports registered sales equal to 14.3 billion, 14.5 billion, 14.2 billion, and 12.6 billion cigarettes in 2008, 2010, 2012, and 2014, respectively.33 In the case of Colombia, EI reports registered sales equal to 18.7 billion and 13.1 billion cigarettes in 2008 and 2013, respectively.34 Lastly, in the case of Peru EI reports registered sales equal to 3.1 billion and 2.7 billion cigarettes in 2006 and 2010, respectively.35 Results Figures 1 and 2 show the results of the gap analyses carried out for the different countries. Given that what matters in this type of analysis are the deviations from the original trend, the initial data on registered sales and estimated for the initial year of each survey are equaled to 100. The consumption estimated in the surveys and their confidence intervals are estimated based on these initial values, meaning that the compared evolution of both trends provides information on the evolution of illicit trade. Figure 1. View largeDownload slide Gap analysis for Argentina. ENFR = National Risk Factors Survey. Figure 1. View largeDownload slide Gap analysis for Argentina. ENFR = National Risk Factors Survey. Figure 2. View largeDownload slide Gap analysis for Argentina. EnPreCosP = National Survey on the Prevalence of Psychoactive Substance Consumption. Figure 2. View largeDownload slide Gap analysis for Argentina. EnPreCosP = National Survey on the Prevalence of Psychoactive Substance Consumption. Figures 1 and 2 show the results for the two Argentine surveys considered. In the case of the ENFR, Figure 1 shows that registered sales increased more quickly than the ENFR’s estimated consumption between 2005 and 2009, meaning that there was a decline in the illicit cigarette trade. This gap is statistically significant and can be seen graphically, with the evolution in registered consumption outside the confidence interval built around estimated consumption. Both consumption trends are practically parallel between 2009 and 2013, leading to the assertion, under the assumptions made, that there was no change in the illicit cigarette trade between those two years. This is to a certain extent confirmed by the analysis of the EnPreCosp (Figure 2), which shows that there are no statistically significant differences between the evolution of registered and reported consumption between 2009 and 2011. This, in the context of a practically constant tobacco month prevalence, leads to the conclusion that illicit trade is likely to have remained constant in both absolute and relative terms (ie, total market share). These results are robust if EI figures for registered sales are used instead of official figures.8 In the case of Brazil, Figure 3 shows that the gap between registered cigarette sales and estimated consumption increases significantly over time (in the chart, registered sales are below the lower limit of the confidence interval of estimated consumption), suggesting that the proportion of illicit cigarettes consumed increased over time. This behavior is verified in the context of a statistically significant decline in cigarette consumption (in total quantities consumed). These results are robust if EI figures for registered sales are used, though in this case the increase in contraband is smaller.8 Figure 3. View largeDownload slide Gap analysis for Brazil. Figure 3. View largeDownload slide Gap analysis for Brazil. Figure 4 shows the results for Chile. The surveys’ consumption estimates in Figure 4 show an oscillating and slightly downward trend in the number of cigarettes consumed (assuming constant underreporting), while the consumption registered by EI shows a downward trend. However, the evolution of registered consumption is within the confidence interval for estimated consumption, meaning that one cannot reject the null hypothesis that they are equal and that illicit trade in this period had remained constant in relative terms during this period. Figure 4. View largeDownload slide Gap analysis for Chile. Figure 4. View largeDownload slide Gap analysis for Chile. Figure 5 shows the evolution of registered and estimated consumption for Colombia. Contrary to the previous case, here there is a clear downward trend in both consumptions. Similar to the previous case, the EI consumption estimate is within the confidence interval of the consumption observed, so the null hypothesis that they are equal cannot be rejected. In this case, one can also assert that there are no statistically significant relative variations in illicit trade during this period, despite claims of sharp increases made by the tobacco industry.36–38 Figure 5. View largeDownload slide Gap analysis for Colombia. Figure 5. View largeDownload slide Gap analysis for Colombia. Lastly, Figure 6 shows the evolution of registered consumption according to EI and the estimated consumption for Peru. In the case of registered consumption, according to EI, there were reportedly no significant differences between the consumption estimated in the surveys, meaning that the illicit cigarette trade in 2010 was statistically similar in relative terms to that of 2006. Figure 6. View largeDownload slide Gap analysis for Peru. Figure 6. View largeDownload slide Gap analysis for Peru. Discussion The lack of information on the illicit cigarette trade in South American countries has allowed the tobacco industry to directly present figures of its own on the problem (with its own studies) or indirectly through international consultancy firms with clear conflicts of interest. The presentation of these figures seems to have more to do with current debates on tax increases in countries than with real movements or concerns about the scale of illegal trade. The results confirm what was known in the case of Brazil, that the relative size of the illicit cigarette trade has grown recently, and they reveal that said increase is statistically significant, even if using EI figures on reported sales.14 This analysis does not tell us the reasons for this increase, which does not necessarily have to do with recent tax increases, but could also be because of a deficient institutional context and widespread corruption.5,39 In fact, there is evidence that tax increases have managed to lower total consumption and increase government revenues.14 Thus, it is clear that this policy must continue, but while strengthening oversight institutions and improving public practice related to transparency. It could otherwise be hard to obtain political support for subsequent tax increases. The results are novel in the case of the other countries analyzed. The result for Argentina should come as no surprise: this country has made little progress with tools to control tobacco, such as tax increases; it is one of the few in the region that has not ratified the FCTC and, thanks to the high levels of inflation, the real price of legal cigarettes shows a significant decline in the period analyzed. Thus, the real average price fell by around 23% between 2005 and 2009, while the decline was approximately 30% between 2005 and 2013, though there is controversy surrounding official inflation numbers.40 There were no tax increases in Chile between 1998 and 2010, though the real price of cigarettes increased by more than 50%. The increase in real price was 17% between 2008 and 2010 alone. These price increases were entirely decided by British American Tobacco Chile (BAT Chile), which controls over 95% of the market.33 During this time, the tobacco company showed no concern at all about the effects that these increases could have on illicit trade and it only appeared in the public debate when taxes began increasing moderately after 2010.41–43 Between 2010 and 2014, real prices of cigarette increased by about 68%. At any rate, the analysis presented shows that illicit trade did not change significantly between 2008 and 2014. EI, which takes BAT Chile as a source for its estimates, reported an illicit market of just 1.8% in 2008.44 There is no certainty about the way EI obtained this estimate (or any estimate of cigarette illicit trade), as the methodology is obscure, at times inconsistent, and has been subjected to criticism.29 According to the analysis carried out, there are no statistical reasons to assume that there were any significant changes in this between 2008 and 2014. In the case of Colombia, there was a tax change in 2010, as the specific excise tax was changed (from its 2006 level) and linked to inflation. In addition, a 10% surcharge was created to finance universal health coverage.45 These changes do not seem to have had any significant impact on the illicit trade trend. Moreover, the result presented here takes on greater significance if one relates it to the recently collected evidence that the relative size of the illegal cigarette market is reportedly just 3.5% on average in Colombia’s main cities.18 There is no evidence to suggest that there were structural changes in the cigarette market that could have altered this share between 2013, the last year analyzed here, and mid-2016. In fact, the real price of cigarettes increased by just 1.3% between January 2013 and January 2016. In the case of Peru, there was a tax change in early 2010, as the excise tax for tobacco went from an ad valorem rate of 30% to a specific amount of S/0.07 per cigarette, which implied an immediate change in nominal prices of about 8%. More importantly, the real price of cigarettes increased by slightly less than 6% between 2006 and 2010. Considering that real salaries grew by around 3% in this period,46 one finds that affordability remained constant or declined slightly. While the results presented here reveal previously unknown data for five South American countries, or which had not been systematized until now, there are limitations. The main limitation to consumption gap analyses is that they do not provide an estimate of the size of the market for illicit products, but rather a trend in its evolution. Thus, one cannot know the number of illicit cigarettes or the lost tax revenue. In addition, this analysis does not allow one to distinguish between the different components of non-registered consumption, such as smuggling, tax evasion, tax avoidance (bootlegging, duty-free shop purchases, etc.), or product counterfeiting.2,25 As with underreporting, one must assume that the proportional share of each of these components remains constant over time, meaning that deviations in the trend imply deviations in illicit trade. Another assumption that has to be made is that the proportion of illicit cigarette consumption for the population groups not covered by the usage surveys (typically, youths under 16 years and adults over the age of 65 years) is similar to what is registered for the rest of the population. It would be ideal to have several waves of surveys that allow this to be done, but only five countries in South America have two or more waves of population consumption surveys. It is to be hoped that the analysis conducted here can be extended as new surveys are collected. In summary, this study conducted with data of five South American countries found that illicit trade increased in only one country, decreased in another, and was largely unchanged in three others—despite the latter having substantial increases in tobacco excise levels or tobacco product prices during the study period. This adds to the evidence that the level of illicit trade is influenced by a myriad of factors and that there is no simple relationship, as claimed by the tobacco industry, that increasing tobacco excise levels results, inevitably, in increased illicit tobacco trade. Funding The author is grateful for the funding of the Pan American Health Organization (PAHO). The views and results presented here do not necessarily coincide with those of the PAHO. Declaration of Interests None declared. Acknowledgments The author thanks Rosa Sandoval, Itziar Belausteguigoitia, Roberto Iglesias, Mathieu Poirier, and Daniel Araya for their comments, along with those of the participants of seminars in Panama, Costa Rica, and a Congress in Cape Town, where preliminary results were presented. The usual disclaimer applies. References 1. World Health Organization, Framework Convention on Tobacco Control, 2005. Available from: http://www.who.int/fctc/en/ 2. Ross H. Understanding and measuring cigarette tax avoidance and evasion: A methodological guide . 2015. Available from: https:// tobacconomics.org/wp-content/uploads/2015/03/Ross_Methods_to_Measure_Illicit-Trade_03-17-15.pdf 3. British American Tobacco. Comercio Ilícito Argentina. Available from: http://www.batargentina.com/group/sites/BAT_9YXKEP.nsf/vwPagesWebLive/DO9T5K4G Accessed December 27, 2017. 4. La Tercera. Comercio Ilegal de Cigarrillos en Chile Creció un 386% en Cinco Años . La Tercera. 2017. 5. U.S. National Cancer Institute and World Health Organization. The Economics of Tobacco and Tobacco Control. National Cancer Institute Tobacco Control Monograph 21. NIH Publication No. 16-CA-8029A. Bethesda, MD: U.S. Department of Health and Human Services, National Institutes of Health, National Cancer Institute; and Geneva, CH: World Health Organization; 2016. 6. Lee K, Collin J. “Key to the Future”: British American tobacco and cigarette smuggling in China. PLoS Med . 2006; 3( 7): e228. Google Scholar CrossRef Search ADS PubMed  7. Collin J, LeGresley E, MacKenzie R, Lawrence S, Lee K. Complicity in contraband: British American tobacco and cigarette smuggling in Asia. Tob Control . 2004; 13( suppl 2): ii104. Google Scholar CrossRef Search ADS PubMed  8. Euromonitor International. Passport Data. 2017. Available from: https://www.portal.euromonitor.com/portal/magazine/homemain Accessed September 15, 2017. 9. González-Rozada M. Economía del Control del Tabaco en los Países del Mercosur y Estados Asociados: Argentina: 1996–2004 . Washington, DC: Organización Panamericana de la Salud; 2006. 10. Alcaraz V. Economía del Control del Tabaco en los Países del Mercosur y Estados Asociados: Bolivia . Washington, DC: Organización Panamericana de la Salud; 2006. 11. Iglesias RM, Szklo AS, Souza MC, et al.  . Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control . 2017; 26: 53– 9. Google Scholar CrossRef Search ADS PubMed  12. Debrott Sanchez D. Economía del Control del Tabaco en los Países del Mercosur y Estados Asociados: Chile . Washington, DC: Organización Panamericana de la Salud; 2006. 13. Jorratt M. Estimación de la evasión tributaria en los impuestos selectivos al consumo: El caso de Chile. Rev Adm Tribut . 2012( 34): 33– 45. 14. Iglesias RM, Szklo AS, Souza MCD, de Almeida LM. Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control . 2016. 15. Ramos A, Curti D. Economía del Control del Tabaco en los Países del Mercosur y Estados Asociados: Uruguay . Washington, DC: Organizacion Panamericana de la Salud; 2006. 16. Herrera Ballesteros VH. Análisis de la Demanda de Tabaco en Panamá y el Control del Efecto Asequibilidad con Medidas Fiscales y Control del Contrabando: Implicaciones Para Política Fiscal 2000–2011 . Panama: IDRC Digital Library; 2013. 17. Herrera Ballesteros VH, Zúñiga J, Gómez B, Roa R. Factores socioeconómicos asociados a la compra ilegal de productos del tabaco en Panamá. Salud Pública de México. 2016; 59( suppl 1):S88–S96. 18. Maldonado N, Llorente BA, Iglesias RM, et al.  . Measuring illicit cigarette trade in Colombia. Tob Control . 2018. doi:10.1136/tobaccocontrol-2017-053980 19. Chen J, McGhee SM, Townsend J, Lam TH, Hedley AJ. Did the tobacco industry inflate estimates of illicit cigarette consumption in Asia? An empirical analysis. Tob Control . 2015; 24( e2): e161– 167. Google Scholar CrossRef Search ADS PubMed  20. Stoklosa M. Is the illicit cigarette market really growing? The tobacco industry’s misleading math trick. Tob Control . 2016; 25( 3): 360. Google Scholar CrossRef Search ADS PubMed  21. Ibarra V.http://www.economiaynegocios.cl/noticias/noticias.asp?id=325219 El Mercurio. 2017. 22. Redacción Economía y Negocios. Coltabaco advierte que más impuestos al cigarrillo dispararán el contrabando . El Espectador. 2016. 23. Redacción Gestión. Contrabando de cigarrillos en el Perú concentra el 18% de venta total forma. Gestion . 2015. 24. Diario Financiero. Comercio Ilícito de Cigarrillos en Chile Creció Cuatro Veces en los Últimos Cinco Años . Accessed May 5, 2017. 25. Merriman D. Comprender, Medir y Combatir el Contrabando de Tabaco: Herramienta 7: Contrabando . The World Bank; 2005. 26. Ministerio de Agroindutstria de la República Argentina. 2017. Available from: http://www.agroindustria.gob.ar/sitio/areas/tabaco/estadisticas/_ archivos//000001-Volumen%20de%20Paquetes%20de%20Cigarrillos%20Vendidos%20por%20Rango%20de%20Precio%20(2004-2017).pdf 27. Receita Federal do Brasil. 2015. Available from: http://idg.receita.fazenda.gov.br/orientacao/tributaria/regimes-e-controles-especiais/producao-de-cigarros-no-brasil-anos-anteriores 28. Receita Federal do Brasil. 2016. Available from: http://idg.receita.fazenda.gov.br/orientacao/tributaria/regimes-e-controles-especiais/producao-de-cigarros-no-brasil-2013 29. Blecher E, Liber A, Ross H, Birckmayer J. Euromonitor data on the illicit trade in cigarettes. Tob Control . 2014; 24( 1): 100. Google Scholar CrossRef Search ADS   30. Linegar DJ, van Walbeek C. The effect of excise tax increases on cigarette prices in South Africa. Tob Control. 2018;27:65–71. 31. Seidenberg AB, Behm I, Rees VW, Connolly GN. Cigarette sales in pharmacies in the USA (2005–2009). Tob Control . 2012; 21( 5): 509. Google Scholar CrossRef Search ADS PubMed  32. Greenland SJ. Cigarette brand variant portfolio strategy and the use of colour in a darkening market. Tob Control . 2015; 24( e1): e65. Google Scholar CrossRef Search ADS PubMed  33. Euromonitor International. Tobacco in Chile. 2016. Available from: http://www.portal.euromonitor.com 34. Euromonitor International. Tobacco in Colombia. 2015. Updated August 2015. Available from: http://www.portal.euromonitor.com 35. Euromonitor International. Tobacco in Peru. 2015. Updated August 2015. Available from: http://www.portal.euromonitor.com 36. Caracol Radio. Precios de los Cigarrillos Subirían Ante Aumento de Impuestos. 2011. Available from: http://caracol.com.co/radio/2011/01/03/economia/1294047480_405801.html 37. El Espectador. Colombia Sigue Fumando Ilegal. 2011. Available from: http://www.elespectador.com/noticias/temadeldia/colombia-sigue- fumando-ilegal-articulo-292829 38. Portafolio. Contrabando de Cigarrillos Sigue en Aumento en el país. 2013 Available from: http://www.portafolio.co/economia/finanzas/contrabando-cigarrillos-sigue-aumento-pais-85638 39. Transparency International. Corruption Perception Index 2016: Transparency International. 2017. Available from: http://www.transparency.org/whatwedo/publication/corruption_perceptions_index_2016 40. Statement by the IMF Executive Board on Argentina. Press release. 2013. Available from: https://www.imf.org/en/News/Articles/2015/09/14/01/49/pr1333 41. Serrano G. Reforma tributaria: Mañana sube el precio de cigarrillos, alcohol y bebidas azucaradas . La Tercera. 2014. Accessed September 30, 2014. 42. 24horas.cl. Aumenta Contrabando de Cigarros en Nuestro País. 2013. Available from: http://www.24horas.cl/nacional/aumenta-contrabando- de-cigarros-en-nuestro-pais-927199] 43. La Segunda. El Gigantesco Negocio de los Cigarrillos “Ilegales”: Se Venden en Kioscos y Ferias. 2012. Available from: http://www.lasegunda.com/Noticias/Nacional/2012/07/768557/el-gigantesco-negocio-de-los-cigarrillos-ilegales-se-venden-en-kioscos-y-ferias] 44. Euromonitor International. Tobacco in Chile. 2010. Available from: http://www.portal.euromonitor.com 45. Ministerio de Salud de Colombia. Impuestos al Tabaco. In: Ministerio de Salud de Colombia, editor. Papeles en Salud . Bogotá, Colombia: Ministerio de Salud de Colombia; 2016. 46. Instituto Nacional de Estadísitca e Informática de Perú. Sistema de información económica: INEI, Peru; 2017. Available from: http://iinei.inei.gob.pe/iinei/siemweb/publico/ © The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nicotine and Tobacco Research Oxford University Press

Illicit Cigarette Trade in Five South American Countries: A Gap Analysis for Argentina, Brazil, Chile, Colombia, and Peru

Loading next page...
 
/lp/ou_press/illicit-cigarette-trade-in-five-south-american-countries-a-gap-en8IBPxFVZ
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
1462-2203
eISSN
1469-994X
D.O.I.
10.1093/ntr/nty098
Publisher site
See Article on Publisher Site

Abstract

Abstract Introduction Because of its nature, it is very hard to measure illicit tobacco trade in any product. In the case of Latin American countries, there is scant information on the magnitude and characteristics of this cigarette trade. The goal of this article is to provide estimates on the evolution of the illicit cigarette trade in five South American countries: Argentina, Brazil, Chile, Colombia, and Peru. Methods Gap analysis estimates for cigarette tax evasion/avoidance (a comparison on the evolution of the difference between registered cigarette sales and measured population consumption) are developed for Argentina, Brazil, Chile, Colombia, and Peru. Nationally representative surveys, conducted regularly, are used to measure population consumption. Confidence intervals constructed by bootstrapping sample estimates are generated to statistically evaluate the evolution of the gap. Results Illicit cigarette trade has increased as a percentage of total sales in Brazil in recent years. In the case of Argentina, after a relative decrease between 2005 and 2009, it seems to have stabilized. There is no statistical evidence to argue that there has been an increase of illicit cigarette trade in Chile, Colombia, and Peru, despite substantial price increases in Chile and tax increase in both Colombia and Peru. Conclusions Using simple statistical methods, it is possible to assess the trend in illicit tobacco trade over time to better inform policy makers. Getting reliable and regular population consumption surveys can also help to track illicit tobacco trade. Claims by tobacco industry of a positive association between price/tax changes and illicit trade are unsubstantiated. Implications Evolution of illicit cigarette trade in five Latin American countries shows different trajectories, not in line with tobacco industry estimates, which highlight the importance of producing solid, independent estimates. There are inexpensive methodologies that can provide estimates of the evolution of the relative importance of illicit trade and can be used to inform policy makers. Introduction The World Health Organization Framework Convention on Tobacco Control (FCTC) defines illicit tobacco trade as “any practice or conduct prohibited by law which relates to production, shipment, receipt, possession, distribution, sale or purchase including any practice or conduct intended to facilitate such activity.”1 This trade usually involves criminal organizations smuggling in and distributing, large quantities of tobacco products in domestic markets, though it can also involve individuals bringing in smaller quantities for resale. This should not be confused with certain forms of tax avoidance (ie, the purchase of tobacco products in accordance with countries’ customs and tax regulations), which include cross-border shopping for personal consumption (ie, individuals purchasing tobacco products in other countries, in compliance home countries’ regulations regarding quantities, and not for resale), duty-free shopping (ie, similar to cross-border shopping but acquired in duty-free shops or areas), Internet and E-mail/phone purchases (ie, remote purchases that are allowed in the buyer’s home country), etc.2 While tax avoidance is regulated and accepted by the authorities (because of the inefficiency of controlling the circulation of small quantities for personal use), illicit trade usually involves tax evasion, counterfeiting, etc.2 and is explicitly combated by authorities. The reasons for this are that it erodes two of the main goals of tobacco taxes: raising tobacco taxes to reduce consumption (as illicit products are usually cheaper than legitimate ones) and generating fiscal revenues. The tobacco industry usually associates the existence of illicit trade with tobacco taxes, claiming that increasing the price of tobacco products via taxes incentivizes criminals to increase illicit trade, which, in turn, defeats the objectives of reducing consumption and increasing fiscal revenues,3,4 though these claims have been debunked.5 In fact, a number of studies have shown that smuggling in tobacco products has, on occasions, been closely linked to the tobacco industry’s interests and even its participation.6,7 Because of its nature, it is very hard to measure illicit trade in tobacco (or any) product, though a few methods have been proposed, all with advantages and disadvantages. They range from surveys of tobacco users, examination of discarded cigarette packages or those obtained from smokers, monitoring of tobacco trade flows, econometric modeling, etc.2 In the case of Latin American countries, there is scant information on the magnitude and characteristics of illicit cigarette trade, by far, the most-sold tobacco product in the region.8 Some studies have attempted to quantify it for some countries using diverse sources of information and different methodologies. In some cases (eg, Argentina, Bolivia, and Brazil), aggregate annual data on cigarette exports and imports are used to estimate the volume of illegal trade9–11; in others (eg, Chile, Brazil, and Uruguay), legal cigarette sales are compared with the results reported in consumption surveys12–15; others (eg, Panama and Colombia) undertake ad hoc surveys of smokers to measure contraband.16–18 In the case of Argentina, estimates are made assuming different prevalence rates for cigarette smoking (because authors were unaware of the real ones), so the estimates of illicit trade are too broad and not very informative.9 For example, they range from 29% to 40% of the total market for 1996, while it is from 36% to 46% for 2002. A recent study in Brazil uses 2008 Global Adult Tobacco Survey (GATS) and the 2013 National Health Survey (PNS) and defines a minimum sales price for legal cigarettes (considering costs and profits for distributors and retailers).14 When these minimum prices are compared with those actually paid by consumers (prices below the minimum would imply illegal cigarettes), the authors estimate that the illicit cigarette trade grew from 17% in 2008 to 31% in 2013. These estimates, though obtained using a sound methodology, do not appear to be consistent with those of previous studies for Brazil, which put illicit trade at 28% in 2002.11 A study carried out in Chile compares consumption reported in the 2002 Survey of Drug Use in the General Population (EDPG) with domestic sales (provided by the tobacco industry), concluding that illicit trade that year was reportedly 4.2% of the local market.12 This does not consider the phenomenon of underreporting of cigarette smoking, which is usually present in consumption surveys.5 Another survey for Chile estimates tobacco tax evasion, though without distinguishing between domestic evasion and smuggling, using a private survey on the consumption of cigarettes and alcoholic beverages.13 This is used to estimate the amount of specific tax that should have been collected in 2011, which is compared with the actual amount of the tax collected, thus, estimating that tax evasion (not necessarily because of smuggling) accounted for 17.3% of total sales. A survey of smokers was recently carried out in Colombia to measure penetration of illicit trade in the country’s five largest cities.18 The results of this survey show that average penetration of illegal trade is 3.5% of the total market. Lastly, no independent studies were found on the size or the evolution of the illicit cigarette trade in Peru. As can be seen, there is scant evidence and estimates for countries with more than one study are not always consistent with each other. This means that public debate is often dominated by dubious or imprecise numbers supplied by the tobacco industry, as diverse authors have reported,19,20 especially in the form of press reports that are widely reproduced by local media outlets.21–24 Given the importance of illicit trade to public policy, it is inconvenient to base the debate on data whose origin is not very transparent and evolution is unreliable. Thus, the objective of this article is to provide estimates on the evolution of the illicit cigarette trade in five South American countries: Argentina, Brazil, Chile, Colombia, and Peru. These estimates use secondary data and publicly available data sources and allow a gap analysis of said evolution to be carried out. Despite the limitations of this methodology, it allows transparent and easily replicable estimates to be obtained that can serve as benchmarks for public debate. The use of these estimates can serve to compensate the estimates provided by the tobacco industry. Methods A gap analysis estimates tax evasion/avoidance based on a mathematical identity that holds that total consumption—of cigarettes, for example—in a country is equal to the consumption registered (which paid the corresponding taxes), plus what was not registered (contraband, tax evasion, tax avoidance, and product counterfeiting).2 The official statistics on units that pay taxes ought to supply registered consumption, while total consumption can be estimated from population consumption surveys. They must provide information on the total number of smokers and the average number of cigarettes they consume in a given period of time (eg, 1 month). Multiplying both numbers allows one to estimate total consumption in said period and unregistered consumption can be obtained by subtracting registered consumption from this number. This method is simple, easily replicable and does not require expensive data collection processes as it uses available data sources. Given the uncertainty regarding the extent to which consumption is underreported, an issue that often affects consumption surveys, this method only allows one to estimate deviations from the central trend, assuming that the proportion of underreporting remains constant over time, a reasonable assumption when the surveys used are relatively close to each other in time.2,25 For example, previous studies in South America using this method do not consider the possibility of underreporting.12,13 Working with just one survey and failing to consider another point in time makes the estimates obtained in this way unreliable, especially because underreporting can be high.5 Having obtained the gap between the estimated consumption and registered sales, it is necessary to produce confidence intervals on such a gap, to assess if changes over time are statistically significant. The simplest and more intuitive way of estimating these intervals is by bootstrapping the point estimate of the product of the reported smoking prevalence and the intensity of smoking. Each round of bootstrapping creates a new random sample (with replacement) from the original, considering weights and sampling strategies. By repeating 1000 times this procedure, it is possible to obtain a distribution of such a parameter that is close to the actual distribution and to estimate confidence intervals. This is conducted for each survey/country, such that 99% confidence intervals are calculated. Data The information used to estimate consumption at the population level comes from official surveys that are carried out regularly. In the case of Argentina, the National Risk Factors Survey (ENFR) for the years 2005, 2009, and 2013 contains information on average monthly consumption and intensity of cigarette smoking, among other variables, in the population more than the age of 18 years (with nationally representative samples of 41 392, 34 732, and 32 365 observations, respectively). There is also the 2008 and 2011 National Survey on the Prevalence of Psychoactive Substance Consumption (EnPreCosp), with nationally representative samples for the population between 16 and 65 years of age (34 203 and 34 343 observations, respectively). Both editions of the survey ask about the number of days that cigarettes were smoked during the last month, and the average number of cigarettes smoked on those days. The ENFR and the EnPreCosp are not directly comparable, given that they cover different population groups, but they can be used individually for the gap analysis. The surveys used in Brazil are the 2008 GATS and the 2013 PNS. The former is nationally representative for the population more than the age of 15 years (39 425 observations), while the latter is nationally representative for the entire population (205 545 observations). To make them comparable, people under the age of 15 years are eliminated from the 2013 PNS and it is implicitly assumed that the young population not covered by the surveys does not substantially alter their relative consumption compared with the rest of the population over time. In both the surveys, people are asked about the average number of cigarettes smoked per day by daily smokers and per week for those who do not smoke every day. For Chile, the EDPG from 2008, 2010, 2012, and 2014 was used (with samples of 17 133, 15 576, 17 154, and 19 512, respectively). It is nationally representative of the population from 12 to 65 years of age. In all the cases, people were asked about monthly tobacco consumption and the intensity of use on an average day. For Colombia, the Psychoactive Substance Consumption Survey (ECSP) for 2008 and 2013 was used, with national coverage of the population aged 12–65 years and with samples of 29 164 and 32 605, respectively. The 2008 ECSP does not count the number of cigarettes individuals reported smoking on an average day (something that is reported in the 2013 edition). In contrast, this information is reported by intervals: (1) less than 1 cigarette/day, (2) from 1 to 5, (3) from 6 to 10, (4) from 11 to 20, and (5) more than 20. Since this variable is essential to calculating the amounts consumed, a number needs to be assigned to each interval. For the first interval, it is assumed that those who reported smoking less than one cigarette a day smoke one a day. For intervals (2), (3), and (4), a random number is taken (with uniform distribution) between the limits of the ranges and that amount is assigned. Lastly, for the open interval (5), a number is taken (with uniform distribution) between the lower limit of the range (20 cigarettes/day) and a number of cigarettes such that the average of this interval coincides with the average number of cigarettes smoked by those who reported consuming more than 20 cigarettes/day in the 2013 ECSP. Thus, the average number smoked by people who reported consuming more than 20 cigarettes/day in the 2013 ECSP is equal to 47 cigarettes. These data are used to estimate the upper limit, considering the formula of averages in a uniform distribution. Assigning a consumption of one cigarette to those who consume less than one cigarette/day is likely to be a conservative assumption in terms of the evolution of illicit trade, as it raises the number of cigarettes consumed in the base year. Lastly, in the case of Peru, the drug use surveys carried out by the National Commission for Development and Life without Drugs (DEVIDA) in 2006 and 2010 were used, which is nationally representative of the population 12–65 years of age (with samples of 11 825 and 20 275, respectively). The survey contains complete information on the smoking population and average daily intensity of tobacco use. Table 1 shows the main variables taken from each survey to calculate annual consumption estimates. Table 1. Descriptive Variables of National Surveys Country  Survey  Year  Total population  Month prevalence (%)  Number of smokers (past month)  Average intensity  Annual estimated consumption  Lower bound  Upper bound  Argentina  ENFR (18 y old and more)  2005  22 935 297  30.0  6 890 222  11.0  27 695 319 985  25 851 269 324  29 539 370 646  2009  24 434 595  28.5  6 975 173  11.2  28 005 396 000  26 342 428 288  29 668 363 712  2013  25 777 587  26.8  6 920 767  11.1  27 645 902 000  25 310 222 192  29 981 581 808  EnPreCosP (16–65 y old)  2008  21 669 915  31.0  6 714 423  11.2  25 506 871 000  23 371 745 672  27 641 996 328  2011  22 520 925  28.9  6 510 124  10.8  25 263 861 000  23 679 438 104  26 848 283 896  Brazil  GATS/PNS (15 y old and more)  2008  142 998 657  17.2  24 552 398  10.0  88 146 859 982  83 292 130 382  93 001 589 582  2013  146 278 273  14.7  21 515 018  9.9  77 032 649 061  71 402 391 477  82 662 906 645  Chile  EDPG (12–65 y old)  2008  9 047 804  40.2  3 638 122  7.6  7 376 469 872  6 465 687 004  8 287 252 740  2010  9 426 076  35.3  3 324 577  8.5  7 630 861 705  6 773 205 398  8 488 518 012  2012  9 871 872  33.4  3 296 218  8.1  6 926 764 927  6 113 687 745  7 739 842 109  2014  9 683 210  33.8  3 268 955  8.6  7 286 892 666  6 201 028 754  8 372 756 579  Colombia  ECSP (12–65 y old)  2008  19 764 799  17.1  3 371 875  9.2  9 662 052 000  8 355 023 088  10 969 080 912  2013  23 301 463  13.0  3 017 539  8.0  7 225 791 000  6 490 067 368  7 961 514 632  Peru  DEVIDA (12–65 y old)  2006  10 748 960  17.5  1 875 854  4.2  2 856 799 389  2 349 876 077  3 363 722 701  2010  12 107 628  13.3  1 607 070  4.6  2 696 604 670  2 117 156 654  3 276 052 686  Country  Survey  Year  Total population  Month prevalence (%)  Number of smokers (past month)  Average intensity  Annual estimated consumption  Lower bound  Upper bound  Argentina  ENFR (18 y old and more)  2005  22 935 297  30.0  6 890 222  11.0  27 695 319 985  25 851 269 324  29 539 370 646  2009  24 434 595  28.5  6 975 173  11.2  28 005 396 000  26 342 428 288  29 668 363 712  2013  25 777 587  26.8  6 920 767  11.1  27 645 902 000  25 310 222 192  29 981 581 808  EnPreCosP (16–65 y old)  2008  21 669 915  31.0  6 714 423  11.2  25 506 871 000  23 371 745 672  27 641 996 328  2011  22 520 925  28.9  6 510 124  10.8  25 263 861 000  23 679 438 104  26 848 283 896  Brazil  GATS/PNS (15 y old and more)  2008  142 998 657  17.2  24 552 398  10.0  88 146 859 982  83 292 130 382  93 001 589 582  2013  146 278 273  14.7  21 515 018  9.9  77 032 649 061  71 402 391 477  82 662 906 645  Chile  EDPG (12–65 y old)  2008  9 047 804  40.2  3 638 122  7.6  7 376 469 872  6 465 687 004  8 287 252 740  2010  9 426 076  35.3  3 324 577  8.5  7 630 861 705  6 773 205 398  8 488 518 012  2012  9 871 872  33.4  3 296 218  8.1  6 926 764 927  6 113 687 745  7 739 842 109  2014  9 683 210  33.8  3 268 955  8.6  7 286 892 666  6 201 028 754  8 372 756 579  Colombia  ECSP (12–65 y old)  2008  19 764 799  17.1  3 371 875  9.2  9 662 052 000  8 355 023 088  10 969 080 912  2013  23 301 463  13.0  3 017 539  8.0  7 225 791 000  6 490 067 368  7 961 514 632  Peru  DEVIDA (12–65 y old)  2006  10 748 960  17.5  1 875 854  4.2  2 856 799 389  2 349 876 077  3 363 722 701  2010  12 107 628  13.3  1 607 070  4.6  2 696 604 670  2 117 156 654  3 276 052 686  ENFR = National Risk Factors Survey; GATS/PNS = Global Adult Tobacco Survey/ National Health Survey; EDPG = Survey of Drug Use in the General Population; ECSP = Psychoactive Substance Consumption Survey; DEVIDA = National Commission for Development and Life without Drugs. View Large Table 1. Descriptive Variables of National Surveys Country  Survey  Year  Total population  Month prevalence (%)  Number of smokers (past month)  Average intensity  Annual estimated consumption  Lower bound  Upper bound  Argentina  ENFR (18 y old and more)  2005  22 935 297  30.0  6 890 222  11.0  27 695 319 985  25 851 269 324  29 539 370 646  2009  24 434 595  28.5  6 975 173  11.2  28 005 396 000  26 342 428 288  29 668 363 712  2013  25 777 587  26.8  6 920 767  11.1  27 645 902 000  25 310 222 192  29 981 581 808  EnPreCosP (16–65 y old)  2008  21 669 915  31.0  6 714 423  11.2  25 506 871 000  23 371 745 672  27 641 996 328  2011  22 520 925  28.9  6 510 124  10.8  25 263 861 000  23 679 438 104  26 848 283 896  Brazil  GATS/PNS (15 y old and more)  2008  142 998 657  17.2  24 552 398  10.0  88 146 859 982  83 292 130 382  93 001 589 582  2013  146 278 273  14.7  21 515 018  9.9  77 032 649 061  71 402 391 477  82 662 906 645  Chile  EDPG (12–65 y old)  2008  9 047 804  40.2  3 638 122  7.6  7 376 469 872  6 465 687 004  8 287 252 740  2010  9 426 076  35.3  3 324 577  8.5  7 630 861 705  6 773 205 398  8 488 518 012  2012  9 871 872  33.4  3 296 218  8.1  6 926 764 927  6 113 687 745  7 739 842 109  2014  9 683 210  33.8  3 268 955  8.6  7 286 892 666  6 201 028 754  8 372 756 579  Colombia  ECSP (12–65 y old)  2008  19 764 799  17.1  3 371 875  9.2  9 662 052 000  8 355 023 088  10 969 080 912  2013  23 301 463  13.0  3 017 539  8.0  7 225 791 000  6 490 067 368  7 961 514 632  Peru  DEVIDA (12–65 y old)  2006  10 748 960  17.5  1 875 854  4.2  2 856 799 389  2 349 876 077  3 363 722 701  2010  12 107 628  13.3  1 607 070  4.6  2 696 604 670  2 117 156 654  3 276 052 686  Country  Survey  Year  Total population  Month prevalence (%)  Number of smokers (past month)  Average intensity  Annual estimated consumption  Lower bound  Upper bound  Argentina  ENFR (18 y old and more)  2005  22 935 297  30.0  6 890 222  11.0  27 695 319 985  25 851 269 324  29 539 370 646  2009  24 434 595  28.5  6 975 173  11.2  28 005 396 000  26 342 428 288  29 668 363 712  2013  25 777 587  26.8  6 920 767  11.1  27 645 902 000  25 310 222 192  29 981 581 808  EnPreCosP (16–65 y old)  2008  21 669 915  31.0  6 714 423  11.2  25 506 871 000  23 371 745 672  27 641 996 328  2011  22 520 925  28.9  6 510 124  10.8  25 263 861 000  23 679 438 104  26 848 283 896  Brazil  GATS/PNS (15 y old and more)  2008  142 998 657  17.2  24 552 398  10.0  88 146 859 982  83 292 130 382  93 001 589 582  2013  146 278 273  14.7  21 515 018  9.9  77 032 649 061  71 402 391 477  82 662 906 645  Chile  EDPG (12–65 y old)  2008  9 047 804  40.2  3 638 122  7.6  7 376 469 872  6 465 687 004  8 287 252 740  2010  9 426 076  35.3  3 324 577  8.5  7 630 861 705  6 773 205 398  8 488 518 012  2012  9 871 872  33.4  3 296 218  8.1  6 926 764 927  6 113 687 745  7 739 842 109  2014  9 683 210  33.8  3 268 955  8.6  7 286 892 666  6 201 028 754  8 372 756 579  Colombia  ECSP (12–65 y old)  2008  19 764 799  17.1  3 371 875  9.2  9 662 052 000  8 355 023 088  10 969 080 912  2013  23 301 463  13.0  3 017 539  8.0  7 225 791 000  6 490 067 368  7 961 514 632  Peru  DEVIDA (12–65 y old)  2006  10 748 960  17.5  1 875 854  4.2  2 856 799 389  2 349 876 077  3 363 722 701  2010  12 107 628  13.3  1 607 070  4.6  2 696 604 670  2 117 156 654  3 276 052 686  ENFR = National Risk Factors Survey; GATS/PNS = Global Adult Tobacco Survey/ National Health Survey; EDPG = Survey of Drug Use in the General Population; ECSP = Psychoactive Substance Consumption Survey; DEVIDA = National Commission for Development and Life without Drugs. View Large Regarding information on registered consumption, official information from cigarette sales is used whenever it is available. In the case of Argentina, the information on annual registered cigarette consumption comes from the Ministry of Agro-industry. According to this source, 36.9 billion, 43.3 billion, 42.4 billion, 43.1 billion, and 41.8 billion cigarettes were sold in 2005, 2008, 2009, 2011, and 2013, respectively.26 For Brazil, the information on registered cigarette sales comes from the Federal Revenues Service, which reports that 105.9 billion and 75.9 billion cigarettes were sold in 2008 and 2013, respectively.27,28 There is no official information on registered sales for Chile, Colombia, and Peru. Because of this, sales information from Euromonitor International (EI) is used. While there is evidence that EI’s illicit trade estimates are inconsistent over time and tend to overestimate real illicit trade,29 the same does not seem to be the case with the companies’ numbers on production, sales, or market share. These numbers are frequently used30–32 and there is no suspicion that they might be manipulated, basically because there are not significant incentives to do so. In the case of Chile, EI reports registered sales equal to 14.3 billion, 14.5 billion, 14.2 billion, and 12.6 billion cigarettes in 2008, 2010, 2012, and 2014, respectively.33 In the case of Colombia, EI reports registered sales equal to 18.7 billion and 13.1 billion cigarettes in 2008 and 2013, respectively.34 Lastly, in the case of Peru EI reports registered sales equal to 3.1 billion and 2.7 billion cigarettes in 2006 and 2010, respectively.35 Results Figures 1 and 2 show the results of the gap analyses carried out for the different countries. Given that what matters in this type of analysis are the deviations from the original trend, the initial data on registered sales and estimated for the initial year of each survey are equaled to 100. The consumption estimated in the surveys and their confidence intervals are estimated based on these initial values, meaning that the compared evolution of both trends provides information on the evolution of illicit trade. Figure 1. View largeDownload slide Gap analysis for Argentina. ENFR = National Risk Factors Survey. Figure 1. View largeDownload slide Gap analysis for Argentina. ENFR = National Risk Factors Survey. Figure 2. View largeDownload slide Gap analysis for Argentina. EnPreCosP = National Survey on the Prevalence of Psychoactive Substance Consumption. Figure 2. View largeDownload slide Gap analysis for Argentina. EnPreCosP = National Survey on the Prevalence of Psychoactive Substance Consumption. Figures 1 and 2 show the results for the two Argentine surveys considered. In the case of the ENFR, Figure 1 shows that registered sales increased more quickly than the ENFR’s estimated consumption between 2005 and 2009, meaning that there was a decline in the illicit cigarette trade. This gap is statistically significant and can be seen graphically, with the evolution in registered consumption outside the confidence interval built around estimated consumption. Both consumption trends are practically parallel between 2009 and 2013, leading to the assertion, under the assumptions made, that there was no change in the illicit cigarette trade between those two years. This is to a certain extent confirmed by the analysis of the EnPreCosp (Figure 2), which shows that there are no statistically significant differences between the evolution of registered and reported consumption between 2009 and 2011. This, in the context of a practically constant tobacco month prevalence, leads to the conclusion that illicit trade is likely to have remained constant in both absolute and relative terms (ie, total market share). These results are robust if EI figures for registered sales are used instead of official figures.8 In the case of Brazil, Figure 3 shows that the gap between registered cigarette sales and estimated consumption increases significantly over time (in the chart, registered sales are below the lower limit of the confidence interval of estimated consumption), suggesting that the proportion of illicit cigarettes consumed increased over time. This behavior is verified in the context of a statistically significant decline in cigarette consumption (in total quantities consumed). These results are robust if EI figures for registered sales are used, though in this case the increase in contraband is smaller.8 Figure 3. View largeDownload slide Gap analysis for Brazil. Figure 3. View largeDownload slide Gap analysis for Brazil. Figure 4 shows the results for Chile. The surveys’ consumption estimates in Figure 4 show an oscillating and slightly downward trend in the number of cigarettes consumed (assuming constant underreporting), while the consumption registered by EI shows a downward trend. However, the evolution of registered consumption is within the confidence interval for estimated consumption, meaning that one cannot reject the null hypothesis that they are equal and that illicit trade in this period had remained constant in relative terms during this period. Figure 4. View largeDownload slide Gap analysis for Chile. Figure 4. View largeDownload slide Gap analysis for Chile. Figure 5 shows the evolution of registered and estimated consumption for Colombia. Contrary to the previous case, here there is a clear downward trend in both consumptions. Similar to the previous case, the EI consumption estimate is within the confidence interval of the consumption observed, so the null hypothesis that they are equal cannot be rejected. In this case, one can also assert that there are no statistically significant relative variations in illicit trade during this period, despite claims of sharp increases made by the tobacco industry.36–38 Figure 5. View largeDownload slide Gap analysis for Colombia. Figure 5. View largeDownload slide Gap analysis for Colombia. Lastly, Figure 6 shows the evolution of registered consumption according to EI and the estimated consumption for Peru. In the case of registered consumption, according to EI, there were reportedly no significant differences between the consumption estimated in the surveys, meaning that the illicit cigarette trade in 2010 was statistically similar in relative terms to that of 2006. Figure 6. View largeDownload slide Gap analysis for Peru. Figure 6. View largeDownload slide Gap analysis for Peru. Discussion The lack of information on the illicit cigarette trade in South American countries has allowed the tobacco industry to directly present figures of its own on the problem (with its own studies) or indirectly through international consultancy firms with clear conflicts of interest. The presentation of these figures seems to have more to do with current debates on tax increases in countries than with real movements or concerns about the scale of illegal trade. The results confirm what was known in the case of Brazil, that the relative size of the illicit cigarette trade has grown recently, and they reveal that said increase is statistically significant, even if using EI figures on reported sales.14 This analysis does not tell us the reasons for this increase, which does not necessarily have to do with recent tax increases, but could also be because of a deficient institutional context and widespread corruption.5,39 In fact, there is evidence that tax increases have managed to lower total consumption and increase government revenues.14 Thus, it is clear that this policy must continue, but while strengthening oversight institutions and improving public practice related to transparency. It could otherwise be hard to obtain political support for subsequent tax increases. The results are novel in the case of the other countries analyzed. The result for Argentina should come as no surprise: this country has made little progress with tools to control tobacco, such as tax increases; it is one of the few in the region that has not ratified the FCTC and, thanks to the high levels of inflation, the real price of legal cigarettes shows a significant decline in the period analyzed. Thus, the real average price fell by around 23% between 2005 and 2009, while the decline was approximately 30% between 2005 and 2013, though there is controversy surrounding official inflation numbers.40 There were no tax increases in Chile between 1998 and 2010, though the real price of cigarettes increased by more than 50%. The increase in real price was 17% between 2008 and 2010 alone. These price increases were entirely decided by British American Tobacco Chile (BAT Chile), which controls over 95% of the market.33 During this time, the tobacco company showed no concern at all about the effects that these increases could have on illicit trade and it only appeared in the public debate when taxes began increasing moderately after 2010.41–43 Between 2010 and 2014, real prices of cigarette increased by about 68%. At any rate, the analysis presented shows that illicit trade did not change significantly between 2008 and 2014. EI, which takes BAT Chile as a source for its estimates, reported an illicit market of just 1.8% in 2008.44 There is no certainty about the way EI obtained this estimate (or any estimate of cigarette illicit trade), as the methodology is obscure, at times inconsistent, and has been subjected to criticism.29 According to the analysis carried out, there are no statistical reasons to assume that there were any significant changes in this between 2008 and 2014. In the case of Colombia, there was a tax change in 2010, as the specific excise tax was changed (from its 2006 level) and linked to inflation. In addition, a 10% surcharge was created to finance universal health coverage.45 These changes do not seem to have had any significant impact on the illicit trade trend. Moreover, the result presented here takes on greater significance if one relates it to the recently collected evidence that the relative size of the illegal cigarette market is reportedly just 3.5% on average in Colombia’s main cities.18 There is no evidence to suggest that there were structural changes in the cigarette market that could have altered this share between 2013, the last year analyzed here, and mid-2016. In fact, the real price of cigarettes increased by just 1.3% between January 2013 and January 2016. In the case of Peru, there was a tax change in early 2010, as the excise tax for tobacco went from an ad valorem rate of 30% to a specific amount of S/0.07 per cigarette, which implied an immediate change in nominal prices of about 8%. More importantly, the real price of cigarettes increased by slightly less than 6% between 2006 and 2010. Considering that real salaries grew by around 3% in this period,46 one finds that affordability remained constant or declined slightly. While the results presented here reveal previously unknown data for five South American countries, or which had not been systematized until now, there are limitations. The main limitation to consumption gap analyses is that they do not provide an estimate of the size of the market for illicit products, but rather a trend in its evolution. Thus, one cannot know the number of illicit cigarettes or the lost tax revenue. In addition, this analysis does not allow one to distinguish between the different components of non-registered consumption, such as smuggling, tax evasion, tax avoidance (bootlegging, duty-free shop purchases, etc.), or product counterfeiting.2,25 As with underreporting, one must assume that the proportional share of each of these components remains constant over time, meaning that deviations in the trend imply deviations in illicit trade. Another assumption that has to be made is that the proportion of illicit cigarette consumption for the population groups not covered by the usage surveys (typically, youths under 16 years and adults over the age of 65 years) is similar to what is registered for the rest of the population. It would be ideal to have several waves of surveys that allow this to be done, but only five countries in South America have two or more waves of population consumption surveys. It is to be hoped that the analysis conducted here can be extended as new surveys are collected. In summary, this study conducted with data of five South American countries found that illicit trade increased in only one country, decreased in another, and was largely unchanged in three others—despite the latter having substantial increases in tobacco excise levels or tobacco product prices during the study period. This adds to the evidence that the level of illicit trade is influenced by a myriad of factors and that there is no simple relationship, as claimed by the tobacco industry, that increasing tobacco excise levels results, inevitably, in increased illicit tobacco trade. Funding The author is grateful for the funding of the Pan American Health Organization (PAHO). The views and results presented here do not necessarily coincide with those of the PAHO. Declaration of Interests None declared. Acknowledgments The author thanks Rosa Sandoval, Itziar Belausteguigoitia, Roberto Iglesias, Mathieu Poirier, and Daniel Araya for their comments, along with those of the participants of seminars in Panama, Costa Rica, and a Congress in Cape Town, where preliminary results were presented. The usual disclaimer applies. References 1. World Health Organization, Framework Convention on Tobacco Control, 2005. Available from: http://www.who.int/fctc/en/ 2. Ross H. Understanding and measuring cigarette tax avoidance and evasion: A methodological guide . 2015. Available from: https:// tobacconomics.org/wp-content/uploads/2015/03/Ross_Methods_to_Measure_Illicit-Trade_03-17-15.pdf 3. British American Tobacco. Comercio Ilícito Argentina. Available from: http://www.batargentina.com/group/sites/BAT_9YXKEP.nsf/vwPagesWebLive/DO9T5K4G Accessed December 27, 2017. 4. La Tercera. Comercio Ilegal de Cigarrillos en Chile Creció un 386% en Cinco Años . La Tercera. 2017. 5. U.S. National Cancer Institute and World Health Organization. The Economics of Tobacco and Tobacco Control. National Cancer Institute Tobacco Control Monograph 21. NIH Publication No. 16-CA-8029A. Bethesda, MD: U.S. Department of Health and Human Services, National Institutes of Health, National Cancer Institute; and Geneva, CH: World Health Organization; 2016. 6. Lee K, Collin J. “Key to the Future”: British American tobacco and cigarette smuggling in China. PLoS Med . 2006; 3( 7): e228. Google Scholar CrossRef Search ADS PubMed  7. Collin J, LeGresley E, MacKenzie R, Lawrence S, Lee K. Complicity in contraband: British American tobacco and cigarette smuggling in Asia. Tob Control . 2004; 13( suppl 2): ii104. Google Scholar CrossRef Search ADS PubMed  8. Euromonitor International. Passport Data. 2017. Available from: https://www.portal.euromonitor.com/portal/magazine/homemain Accessed September 15, 2017. 9. González-Rozada M. Economía del Control del Tabaco en los Países del Mercosur y Estados Asociados: Argentina: 1996–2004 . Washington, DC: Organización Panamericana de la Salud; 2006. 10. Alcaraz V. Economía del Control del Tabaco en los Países del Mercosur y Estados Asociados: Bolivia . Washington, DC: Organización Panamericana de la Salud; 2006. 11. Iglesias RM, Szklo AS, Souza MC, et al.  . Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control . 2017; 26: 53– 9. Google Scholar CrossRef Search ADS PubMed  12. Debrott Sanchez D. Economía del Control del Tabaco en los Países del Mercosur y Estados Asociados: Chile . Washington, DC: Organización Panamericana de la Salud; 2006. 13. Jorratt M. Estimación de la evasión tributaria en los impuestos selectivos al consumo: El caso de Chile. Rev Adm Tribut . 2012( 34): 33– 45. 14. Iglesias RM, Szklo AS, Souza MCD, de Almeida LM. Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control . 2016. 15. Ramos A, Curti D. Economía del Control del Tabaco en los Países del Mercosur y Estados Asociados: Uruguay . Washington, DC: Organizacion Panamericana de la Salud; 2006. 16. Herrera Ballesteros VH. Análisis de la Demanda de Tabaco en Panamá y el Control del Efecto Asequibilidad con Medidas Fiscales y Control del Contrabando: Implicaciones Para Política Fiscal 2000–2011 . Panama: IDRC Digital Library; 2013. 17. Herrera Ballesteros VH, Zúñiga J, Gómez B, Roa R. Factores socioeconómicos asociados a la compra ilegal de productos del tabaco en Panamá. Salud Pública de México. 2016; 59( suppl 1):S88–S96. 18. Maldonado N, Llorente BA, Iglesias RM, et al.  . Measuring illicit cigarette trade in Colombia. Tob Control . 2018. doi:10.1136/tobaccocontrol-2017-053980 19. Chen J, McGhee SM, Townsend J, Lam TH, Hedley AJ. Did the tobacco industry inflate estimates of illicit cigarette consumption in Asia? An empirical analysis. Tob Control . 2015; 24( e2): e161– 167. Google Scholar CrossRef Search ADS PubMed  20. Stoklosa M. Is the illicit cigarette market really growing? The tobacco industry’s misleading math trick. Tob Control . 2016; 25( 3): 360. Google Scholar CrossRef Search ADS PubMed  21. Ibarra V.http://www.economiaynegocios.cl/noticias/noticias.asp?id=325219 El Mercurio. 2017. 22. Redacción Economía y Negocios. Coltabaco advierte que más impuestos al cigarrillo dispararán el contrabando . El Espectador. 2016. 23. Redacción Gestión. Contrabando de cigarrillos en el Perú concentra el 18% de venta total forma. Gestion . 2015. 24. Diario Financiero. Comercio Ilícito de Cigarrillos en Chile Creció Cuatro Veces en los Últimos Cinco Años . Accessed May 5, 2017. 25. Merriman D. Comprender, Medir y Combatir el Contrabando de Tabaco: Herramienta 7: Contrabando . The World Bank; 2005. 26. Ministerio de Agroindutstria de la República Argentina. 2017. Available from: http://www.agroindustria.gob.ar/sitio/areas/tabaco/estadisticas/_ archivos//000001-Volumen%20de%20Paquetes%20de%20Cigarrillos%20Vendidos%20por%20Rango%20de%20Precio%20(2004-2017).pdf 27. Receita Federal do Brasil. 2015. Available from: http://idg.receita.fazenda.gov.br/orientacao/tributaria/regimes-e-controles-especiais/producao-de-cigarros-no-brasil-anos-anteriores 28. Receita Federal do Brasil. 2016. Available from: http://idg.receita.fazenda.gov.br/orientacao/tributaria/regimes-e-controles-especiais/producao-de-cigarros-no-brasil-2013 29. Blecher E, Liber A, Ross H, Birckmayer J. Euromonitor data on the illicit trade in cigarettes. Tob Control . 2014; 24( 1): 100. Google Scholar CrossRef Search ADS   30. Linegar DJ, van Walbeek C. The effect of excise tax increases on cigarette prices in South Africa. Tob Control. 2018;27:65–71. 31. Seidenberg AB, Behm I, Rees VW, Connolly GN. Cigarette sales in pharmacies in the USA (2005–2009). Tob Control . 2012; 21( 5): 509. Google Scholar CrossRef Search ADS PubMed  32. Greenland SJ. Cigarette brand variant portfolio strategy and the use of colour in a darkening market. Tob Control . 2015; 24( e1): e65. Google Scholar CrossRef Search ADS PubMed  33. Euromonitor International. Tobacco in Chile. 2016. Available from: http://www.portal.euromonitor.com 34. Euromonitor International. Tobacco in Colombia. 2015. Updated August 2015. Available from: http://www.portal.euromonitor.com 35. Euromonitor International. Tobacco in Peru. 2015. Updated August 2015. Available from: http://www.portal.euromonitor.com 36. Caracol Radio. Precios de los Cigarrillos Subirían Ante Aumento de Impuestos. 2011. Available from: http://caracol.com.co/radio/2011/01/03/economia/1294047480_405801.html 37. El Espectador. Colombia Sigue Fumando Ilegal. 2011. Available from: http://www.elespectador.com/noticias/temadeldia/colombia-sigue- fumando-ilegal-articulo-292829 38. Portafolio. Contrabando de Cigarrillos Sigue en Aumento en el país. 2013 Available from: http://www.portafolio.co/economia/finanzas/contrabando-cigarrillos-sigue-aumento-pais-85638 39. Transparency International. Corruption Perception Index 2016: Transparency International. 2017. Available from: http://www.transparency.org/whatwedo/publication/corruption_perceptions_index_2016 40. Statement by the IMF Executive Board on Argentina. Press release. 2013. Available from: https://www.imf.org/en/News/Articles/2015/09/14/01/49/pr1333 41. Serrano G. Reforma tributaria: Mañana sube el precio de cigarrillos, alcohol y bebidas azucaradas . La Tercera. 2014. Accessed September 30, 2014. 42. 24horas.cl. Aumenta Contrabando de Cigarros en Nuestro País. 2013. Available from: http://www.24horas.cl/nacional/aumenta-contrabando- de-cigarros-en-nuestro-pais-927199] 43. La Segunda. El Gigantesco Negocio de los Cigarrillos “Ilegales”: Se Venden en Kioscos y Ferias. 2012. Available from: http://www.lasegunda.com/Noticias/Nacional/2012/07/768557/el-gigantesco-negocio-de-los-cigarrillos-ilegales-se-venden-en-kioscos-y-ferias] 44. Euromonitor International. Tobacco in Chile. 2010. Available from: http://www.portal.euromonitor.com 45. Ministerio de Salud de Colombia. Impuestos al Tabaco. In: Ministerio de Salud de Colombia, editor. Papeles en Salud . Bogotá, Colombia: Ministerio de Salud de Colombia; 2016. 46. Instituto Nacional de Estadísitca e Informática de Perú. Sistema de información económica: INEI, Peru; 2017. Available from: http://iinei.inei.gob.pe/iinei/siemweb/publico/ © The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Nicotine and Tobacco ResearchOxford University Press

Published: May 15, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off