Understanding and Measuring Tobacco Tax Avoidance and Evasion: A Methodological Guide. Hana Ross, PhD School of Economics University of Cape Town

Transcription

1 Understanding and Measuring Tobacco Tax Avoidance and Evasion: A Methodological Guide Hana Ross, PhD School of Economics University of Cape Town

2 Overview Motivation for developing the Guide What the Guide is and what it is not Target audience Methods to estimate the scope of tax avoidance/evasion Understanding the existing evidence Conclusion / Recommendations

3 Motivation for the Guide Update and expand the World Bank Toolkit (2001) on how to estimate the scope of tobacco smuggling Summarize the lessons learned from applying the 5 methods described in the Toolkit Assess the validity of the old and new methods Provide a tool for assessing the value of the existing evidence

4 What the Guide Is and What It Is Not It describes the impact of tax avoidance/evasion on the supply of and demand for tobacco products, on tax revenue, and on the price elasticity of demand It provides a set of methods to estimate the scope of illicit tobacco trade It offers the most suitable methodological solutions for separating illegal tax evasion from legal tax avoidance It estimates resources needed to generate an estimate It provides a set of criteria for assessing studies of tax avoidance/ evasion It does not cover the topic of how to reduce tax avoidance and tax evasion

5 Target Audience Primary target Investigators with some level of experience Beginning investigators paired with a senior mentor/advisor Secondary target Policy makers Tobacco control partners

6 Methods Framework Intro: Main principles of conducting research Each method Background section (links to theory) Step-by-step guide how to apply the method Pros and cons Recommendation when to use Examples from published literature

7 11 Methods to Estimate the Scope of Tax Avoidance/Evasion Surveys of tobacco users (WB) Examination of cigarette packs obtained from smokers (WB) Examination of discarded cigarette packs Examination of cigarette packs obtained from retail Comparison of sales with consumption (Gap Analysis) (WB) Econometric modeling (WB) Comparison of tax paid sales with estimated consumption Comparison of actual and projected tobacco tax revenue Key informant interviews (WB) Monitoring tobacco trade (WB) Analyzing seizures of illegally transported tobacco

8 Examination of Cigarette Packs When to use: - Pack s features allow determination whether taxes were paid - Budget considerations: costs depend on the sources of packs and the representativeness of the sample - Estimates are not needed quickly

9 Example of Illegal and Legal Cigarettes

10 Example of illegal packs Contraband purchased in Egypt, Nov 2007 Contraband purchased in Egypt in Contraband purchased in with South Korean warnings and marked Nov 2007 with UK health warnings Jan in Pakistan, For duty- free in South Korea and marked on the side For duty- marked GCC Duty Free free sale only (Gulf CooperaRon Council) from Rob Cunningham July 2009

11 Examination of Cigarette Packs (coninuted) Observational study: classifying packs as low-tax or full-tax products Source of packs: - tobacco users; - retail outlets; - collected on the street and in trash Costs can be reduced if data collection added to an existing survey; packs from streets are cheaper compared to packs from users, but representativeness of the sample limited.

12 Examination of Cigarette Packs (continued) Features to examine: - the absence of the correct tax stamp, - an incorrect health warning, - markings of a duty free store, - missing price information (if required by the law), - low price - other features of a pack required by the law.

13 Examination of Cigarette Packs (continued) Steps: 1) develop a sampling frame (aiming for a representative sample); 2) train pack collectors; 3) collect data (possession of unopened packs preferred); 4) inspect packs and record data (standard questionnaire, double recording); 5) categorize the packs as low-tax, full-tax, uncertain (combine features; establish tax avoidance use information from users, statistics on tourists/commuters); 6) analyze the data (% of low-tax among all packs collected and classified; % not categorized; account for smoking intensity & share of low-tax purchases among users)

14 Research Protocol for Examination of Cigarette Packs - Example Select a random sample of 30 out of NYC s 2,216 Census Tracts Weight the population (weighted 3) and those employed (weighted 1) in those areas Train data collectors Have data collectors walk the perimeter of selected areas and pick up every littered pack they found. Stopping rule is minimum of: (a) 9 packs with cellophane or (b) 45 minutes 14

15 Maps of NYC Census Tracts 15

16 Typical Tract Map 16

17 Tax Stamp Sales, New York 17

18 Table 6 Taxed Sales, Estimated Consumption & Avoidance in NYC Cigarette Market Before and After June 3, 2008 Tax Increase Before tax increase Round 1 (Millions per month) Taxed sales (observed)2 Ratio consumption to taxed sales3 Implied Consumption4 Implied Avoidance5 After tax Increase1 Round 3 Round 4 Change Round 1 vs 3 Round 1 vs 4 May 15 to May Sep. 8 to Oct. Sep. 11 to Oct. 7, elapsed time ~3 elapsed time ~15 30, , months months (0.13) (1.57) 9.98 (1.57 ) (0.15) (1.46) (1.46) (0.16) (1.52) (1.52) (3.03) 0.72 (3.03) (3.09) 0.34 (3.09) Standard deviations in parentheses. 1We omit round 2 data since we are unable to derive reliable data on taxed sales for that round. 1 taxed sales are assumed to be average NYC tax stamp sales in March and April Round 3 taxed sales are assumed to be average in July and August Round 4 sales are assumed to be average January to December Ratio of consumption to taxed sales and the standard deviation of this ratio are estimated using the littered pack sample in Table 2. 4Implied consumption= taxed sales * (consumption/taxed sales). 5Implied avoidance is consumption minus taxed sales. 2Round Sources: Taxed sales (Retail NYC tax stamp sales) were obtained from Office of Tax Policy Analysis of the NYS Department of Taxation. Price per pack from Minimum Wholesale and Retail Cigarette Prices New York State Department of Taxation and Finance Publication 509 various dates. 18

19 Examination of Cigarette Packs (continued) Advantages: - Eliminates the validity problems associated with self-reported data Disadvantages: - Difficulty to separate tax evasion from tax avoidance without info from users, a lab, assumptions about tourists/commuters (with the exception of retail surveys) - The generalization of the results can be limited for packs from streets, trash bins, retail stores

20 Surveys of Tobacco Users Smokers can tell us what they smoke, where they bought it, how much it cost. When to use: - When the users purchasing behaviour and/or the self-reported features of a cigarette pack allow one to determine whether taxes were paid - Budget considerations: generating a representative sample can be costly - Estimates are not needed quickly

21 Surveys of Tobacco Users (continued) Steps: 1) Decide the survey mode face-to-face, telephone, mailing in a questionnaire, via Internet (impact on the accuracy, representativeness, the costs) 2) Develop a sampling frame and sample size (aiming for a representative sample); 3) Decide the unit of observation (e.g. household, tobacco user - selecting only those who consume low-tax products will generate biased results)

22 Surveys of Tobacco Users (continued) Steps: 4) Develop questionnaire - Objective questions preferable (e.g. quantity and the frequency of low-tax purchases; socio-econ characteristics); can also obtain subjective opinion on whether the full-tax has been collected 5) Collect data in relatively short period of time 6) Analyze the data (% of smokers who possessed low-tax cigarettes; % of low-tax cigarettes consumed by those who possessed them; account for smoking intensity among those consuming low-tax products; % of low-tax cigarettes among those surveyed, and among the entire population after applying the appropriate weights )

23 Surveys of Tobacco Users (continued) Advantages: - Can dis<nguish between tax avoidance and tax evasion. - One of the most direct methods of obtaining es<mates of the scope of tobacco tax avoidance and of availability of low- tax products via various supply channels Disadvantages: - Tent to understate tax evasion: Do users know if they are smoking illicit product?; Voluntary par<cipa<on may under- report using illicit products; It is recommended to combine it with an independent examina<on of cigarere packs. - Rela<vely expensive

24 Combining 2 Methods Collec<on of data in Warsaw from: Packs discarded on streets in Warsaw in September 2008 (N=761) and September 2011 (N=754) Packs in smokers personal possession in Warsaw in September 2011 (N=400)

25 Cigarette Prices in Eastern Europe Russia Poland Germany $2.81 $7.08 Czech Rep. $4.24 Slovakia $3.08 Hungary $3.43 Russia Lithuania $1.37 Belarus $0.24 Russia $1.15 Ukraine $0.94 Moldova Romania $0.14 $1.98 Lithuania $2.78 Belarus $0.15 Russia $1.33 Poland Germany $3.58 $6.75 Czech Rep. Ukraine $4.33 $1.80 Slovakia $3.24 Moldova Hungary Romania $0.13 Understanding and Measuring Tobacco$2.79 Tax Avoidance and Evasion: A Methodological Guide $3.52

26 Packs Collected in % Packs intended for the Polish market Pack not intended for the Polish market

27 Packs Collected in % Packs intended for the Polish market Pack not intended for the Polish market

28 Packs Collected in 2008 UKRAINE RUSSIA BELARUS ENGLISH LANGUAGE CROATIA ROMANIA OTHER

29 Packs Collected in 2011 UKRAINE RUSSIA BELARUS ENGLISH LANGUAGE MOLDOVA BULGARIA SLOVAKIA

30 Academic and Tobacco Industry Estimates, Warsaw, Poland (2011)

31 Gap Analysis Survey data capture the total market, not just legal market If we know the size of the legal market, we can calculate the size of the illegal market When to use: available reliable and consistent estimates of tobacco consumption over time and unbiased records of tax-paid sales during the same period esrmated illicit trade tax paid sales assumed underreported consumpron survey reported consumpron 10%

32 Gap Analysis (continued) Steps: 1) Obtain several years of reliable data on physical quantities of tax paid sales Collect data in relatively short period of time 2) Obtain data on cigarette consumption from national representative surveys for the same years (account for nondaily smokers and young smokers if they are reported separately) 3) Estimate smoking intensity for adults and the youth 4) Multiply the smoking population by smoking intensity 5) Compare the tax paid sales with the estimated consumption and study how the gap evolved over time

33 Gap Method Example If 20% of the population smoke, 1 million people, 200,000 smokers If each smoker smokes 3,000 cigarettes per year, (< 10/day) we know the total market is 600 million cigarettes If the legal market is 500 million, then the illicit market is 100 million or 17%

34 Gap Analysis (continued) Advantages: - transparency, replicability, and low costs. Disadvantages: - People may not report correctly their consump<on - Surveys may not be representa<ve - Does not provide the scope, only devia<on from the trend - Cannot dis<nguish between tax avoidance and tax evasion

35 Methods - Summary No single method will produce a definitive answer because each has pluses and minuses Weakness of a particular approach can be exacerbated by specific market conditions, so it is important to use local specific knowledge and creativity when applying these methods. It is important to triangulate the estimates of the scope of the problem using different methods Many studies apply the same method over time to capture changes in the scope of tax avoidance/evasion rather than generating a single point estimate (addresses methods weaknesses, evaluates the impact of policies and other factors)

36 A set of criteria to assess the existing evidence: Understanding Existing Evidence Peer review Funding source Grounded in theory Transparency and replicability Generalization of results Use of objective criteria to determine low-tax purchases Correct measurements (distinguishing between tax avoidance, tax evasion, counterfeits) Presentation of results (range, confidence interval, testing assumptions) Cross-validation of results using multiple methods/change over time Acknowledgement of limitations

37 Criteria Characteristics of studies that meet the criteria for Characteristics of studies that do not meet the good quality criteria for good quality Peer reviewed Funding Published in a peer-reviewed journal. Grounded in theory Funding from tobacco industry not disclosed/ acknowledged. Funding entity has a potential conflict of interest with respect to the subject of the study. Study describes various types of tax avoidance/evasion. Uses results of a study that cannot distinguish between tax avoidance/evasion. Takes into account various factors that could influence the scope of tax avoidance/evasion. Transparency and replicability Methods and data are adequately described so that the results can be replicated. Data is not publically available. Assumptions are clearly stated. Generalizability of results Sample size and sampling design are not adequately described; sample size might be too small to permit regional level estimates. The sample selection is biased toward urban Calderoni F (2014): A new method for estimating the illicit cigarette market areas. at the subnational level and its application to Italy, Global Crime, 26 There is no attempt to establish the February 2014 (Transcrime) representativeness of the sample. Objective criteria preferred over Low-tax products are identified based on a set of The identification criteria are defined by a subjective criteria objective criteria. stakeholder with a vested interest in the results. Measurements are defined The method cannot distinguish between various correctly types of tax avoidance and tax evasion categories. Identification of counterfeit Identification of counterfeit products is performed by products a party with a vested interest in in the results. Presentation of results Estimates are presented in a range that accounts for various assumptions used in generating the estimate. There are some indications that the size of It is not clear that the size of revenue loss is revenue loss due to illicit market is expressed expressed correctly, and it is not clear how it was as a share of potential revenue. calculated. Measures change over time using the same method or cross-validates a point estimate using multiple methods Estimates changes in tax avoidance/evasion over No cross-validation of results. time using the same method. Understandingofand Measuring Tobacco Tax Avoidance and Acknowledgement Points to possible weaknesses of the data. methodological weaknesses Evasion: A Methodological Guide

38 Conclusion / Recommendations Use multiple methods Execute methods according to the scientific principles described in this Guide Triangulate the results in order to: - cross-validate the estimates and - minimize the methodological limitations of individual methods

39

40 Thank you! Hana Ross