Health Systems & Reform ISSN: 2328-8604 (Print) 2328-8620 (Online) Journal homepage: http://www.tandfonline.com/loi/khsr20 The Health Gains, Financial Risk Protection Benefits, and Distributional Impact of Increased Tobacco Taxes in Armenia Iryna Postolovska, Rouselle Lavado, Gillian Tarr & Stéphane Verguet To cite this article: Iryna Postolovska, Rouselle Lavado, Gillian Tarr & Stéphane Verguet (2018) The Health Gains, Financial Risk Protection Benefits, and Distributional Impact of Increased Tobacco Taxes in Armenia, Health Systems & Reform, 4:1, 30-41, DOI: 10.1080/23288604.2017.1413494 To link to this article: https://doi.org/10.1080/23288604.2017.1413494 © 2018 The Author(s). Published with View supplementary material license by Taylor & Francis© Iryna Postolovska, Rouselle Lavado, Gillian Tarr, and Stéphane Verguet Accepted author version posted online: 18 Submit your article to this journal Dec 2017. Published online: 18 Dec 2017. Article views: 136 View related articles View Crossmark data Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=khsr20 Health Systems & Reform, 4(1):30–41, 2018 Published with license by Taylor & Francis ISSN: 2328-8604 print / 2328-8620 online DOI: 10.1080/23288604.2017.1413494 Research Article The Health Gains, Financial Risk Protection Benefits, and Distributional Impact of Increased Tobacco Taxes in Armenia Iryna Postolovska1,2, Rouselle Lavado2,3, Gillian Tarr4,5 and St ephane Verguet1,* 1 Department of Global Health and Population, Harvard T. H. Chan School of Public Health, Boston, MA, USA 2 The World Bank, Washington, DC, USA 3 Asian Development Bank, Mandaluyong City, Metro Manila, The Philippines 4 Department of Epidemiology, University of Washington, Seattle, WA, USA 5 Department of Pediatrics, University of Calgary, Calgary, AB, Canada CONTENTS Abstract—The majority of Armenian adult males smoke, yet tobacco Introduction taxes in Armenia are among the lowest in Europe and Central Asia. Methods Increasing taxes on tobacco is one of the most cost-effective public Results health interventions, but many opponents often cite regressivity as an Discussion argument against tobacco taxation. We use a mixed-methods approach Conclusion to study the potential regressivity of tobacco taxation and the extent to References which the regressivity argument hindered increases in tobacco taxation in Armenia. First, we pursued an extended cost-effectiveness analysis (ECEA) to assess the health, financial, and distributional consequences (by consumption quintile) of increases in the excise tax on cigarettes in Armenia. We simulated a hypothetical price hike leading to a tax rate of about 75% of the retail price of cigarettes, which would be fully passed on to consumers. Second, we conducted a series of stakeholder interviews to examine the importance of the regressivity argument and identify the factors that allowed tobacco tax increases to be adopted as public policy in Armenia. We show that increased excise taxes would bring large health and financial benefits to Armenian households. Half of tobacco-related premature deaths and 27% of associated poverty cases averted would be concentrated among the bottom 40% of the population. Though regressivity was raised as a concern at the initial stages of the policy adoption process, our qualitative stakeholder analysis indicates that the recent accession to the Eurasian Economic Union and the fiscal constraints faced by the Keywords: agenda setting, Armenia, extended cost-effectiveness analysis government created a window of opportunity for tobacco taxation to (ECEA), financial risk protection, tobacco taxation be placed on the policy agenda and adopted as government policy, and Received 18 May 2017; revised 27 November 2017; accepted 27 November 2017. the ECEA findings were an important input into the process. *Correspondence to: Stephane Verguet; Email: verguet@hsph.harvard.edu Supplemental data for this article can be accessed on the publisher’s website. Ó 2018 Iryna Postolovska, Rouselle Lavado, Gillian Tarr, and St ephane Verguet. INTRODUCTION This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), In Armenia, smoking prevalence is high and tobacco use is which permits unrestricted use, distribution, and reproduction in any one of the leading risk factors for premature mortality.1 medium, provided the original work is properly cited. 30 Postolovska et al.: Distributional Impact of Increased Tobacco Taxes in Armenia 31 About 25% of Armenian adults smoke, largely the men Armenia was the first among the former Soviet Union (53% smoking prevalence among males versus 2% among countries to ratify WHO’s Framework Convention on females).2 Smoking is also higher among the poorer socio- Tobacco Control (FCTC) and shortly after adopted a national economic groups: it is particularly high among men in the law in March 2005 on restrictions on the sale, consumption, second and third wealth quintiles (60% prevalence) com- and use of tobacco.20 However, the government subsequently pared to the bottom (49%) and top (42%) quintiles.3 failed to act,21 and Armenia now ranks behind many other Price measures are central to tobacco control.4-8 Price is a neighboring countries (e.g., Ukraine, Russia) on tobacco con- key determinant of smoking uptake and cessation, and price trol.5 Tax measures have remained inadequate, constituting increases on tobacco are highly effective in inducing smokers 34% (17% excise tax, 17% VAT) of the price of the most to quit and deterring nonsmokers from initiating.6 Higher pri- sold cigarette brand in 2014. ces can also result in current smokers reducing cigarette con- In 2015, the Armenian government approved a draft legis- sumption and preventing ex-smokers from returning to lation to revise its tax code in order to raise revenues.22 This smoking.6,7 Though tax hikes can generate additional reve- included higher tobacco taxes and it was the first time the nue for financing for development, as stated in the 2015 government had been able to successfully pass legislation to Addis Ababa Action Agenda,9 the main objective of tobacco increase tobacco taxes. However, concerns were raised about taxes is to discourage smoking and avert its adverse health the regressivity of such changes, hence prompting the need consequences. to analyze the distributional impact. Here, we apply a mixed- Tobacco taxes can often be applied as excise taxes of two methods approach to examine the validity of the regressivity types: specific and ad valorem. A specific excise tax is a fixed argument in Armenia by using Extended Cost Effectiveness monetary value per quantity (e.g., per cigarette pack), Analysis (ECEA) methods23-25 and identify factors that whereas an ad valorem excise tax is levied as a percentage allowed increased tobacco taxes to be adopted as public pol- per tobacco product (e.g., per retail price of cigarette pack).6 icy in Armenia. Unlike other types of taxes (e.g., value-added tax or VAT), large specific excise taxes can narrow the price gap between different types of cigarette brands and prevent substitution to METHODS lower-priced cigarettes as prices increase.7,10 However, only In the Armenian context, we applied a mixed-methods 33 countries so far have raised tobacco tax rates to the World approach using ECEA and stakeholder analysis to answer Health Organization’s recommended rate of 75% of the retail two questions: (1) Would higher tobacco taxes be regressive? price.5,11 Opponents, particularly the tobacco industry, have and (2) Did their perceived regressivity hinder the adoption used the potential “regressivity” of tobacco taxes as an argu- of enhanced tobacco control policy? The ECEA was used to ment to build coalitions against cigarette price increases.12-14 assess the health, financial, and distributional consequences According to this argument, because the poor spend a larger (by individual consumption quintile) among Armenian proportion of their income on smoking than the rich, smokers (males only) of increased cigarette taxes. Its results increased cigarette taxes could disproportionately hurt the were then used in Armenia’s policy dialogue, allowing us to poor.15 However, this argumentation is largely based on the conduct a series of interviews with key stakeholders to ana- immediate fiscal impact and not on a comprehensive analysis lyze the agenda setting and adoption processes of tobacco of all of the potential benefits of increased tobacco taxes.16 taxation and identify factors that enabled tobacco taxation to Tobacco use can lead to substantial societal costs, because be adopted as public policy in 2015. half of those who prematurely die of tobacco-related non- communicable diseases are in the prime of their productive years.17 It can result in high out-of-pocket (OOP) disease Modeling Approach treatment costs and ultimately push households into poverty. ECEA is used for health policy assessment23,24 and has been Despite an explicit publicly funded health benefits package, applied to a wide range of policies, including tobacco taxa- OOP health care expenditures represent almost 54% of tion.25,26 It examines multiple policy outcomes: the health Armenia’s total health spending,18 and 9% of households gains (e.g., premature mortality averted), the financial conse- incur catastrophic health expenditures (health expenditures quences for individuals and households (e.g., OOP expendi- > 25% of nonfood expenditures).19 Therefore, in addition to tures averted due to disease treatment averted) and the averting premature mortality, reducing the burden of tobacco corresponding financial risk protection (e.g., cases of could potentially avert high OOP expenditures in Armenia. medical impoverishment averted), and the distributional 32 Health Systems & Reform, Vol. 4 (2018), No. 1 consequences among the population (e.g., per socioeconomic We focused our analysis on the male population only: in group). In doing so, ECEA goes beyond traditional cost- Armenia, 53% of males smoke compared to 2% of females.3 effectiveness analysis in enabling quantification of the finan- The population was divided into five-year age groups (from cial protection and equity (distributional) benefits of policy.24 age 0 to 84 and a single group for men > 84) using popula- Few studies have investigated the distributional conse- tion estimates from the World Bank’s Health, Nutrition, and quences of increased tobacco taxes. Here we built on a previ- Population Statistics database.29 It was subsequently divided ously developed ECEA model25,27 to examine the premature into individual consumption quintiles, with maximum con- deaths averted, the OOP expenditures averted and financial sumption for each quintile estimated using the 2014 Inte- protection provided, and their distributions across socioeco- grated Living Conditions Survey (ILCS).19 We applied these nomic groups among Armenian male smokers, following consumption quintile cutoffs and the Gini index (estimated increased excise taxes on tobacco. from ILCS) to simulate a consumption distribution in the Using Armenia’s baseline excise tax rate of 17% of the population using a gamma distribution.30,31 Age- and quin- retail price and an average price per cigarette pack of 525 tile-specific smoking prevalence rates were used to calculate Armenian dram (AMD; approximately 1.25 USD),28 we sim- the total number of smokers per age group and consumption ulated a price increase leading to the WHO-recommended quintile. tax rate of 75%.5 The average price per cigarette pack would Due to lack of data, we could not obtain the price elastic- correspondingly increase by about 145% (an additional 1.80 ity of demand for tobacco products in Armenia. Instead, we USD). We assessed distributional impact in terms of averted used a price elasticity of ¡0.54, estimated from the 2015 premature tobacco-related deaths; averted OOP expenditures Kyrgyz Integrated Household Survey32 using the approach on tobacco-related disease treatment; government savings of Hu et al.33 and Adioetomo et al.34 This price elasticity resulting from tobacco-related disease treatment costs falls within the ¡0.40 to ¡0.80 elasticity range estimated for averted for those eligible for the publicly funded basic bene- developing countries.6,8 Consistent with other studies,6,7,26 fits package (BBP); averted cases of medical impoverishment the 2015 Kyrgyz Integrated Household Survey estimates (number of individuals falling below the national poverty indicated the poor to be more price responsive, with the price line as a result of OOP tobacco-related treatment costs); and elasticity ranging from ¡0.74 (poorest quintile) to ¡0.28 averted cases of catastrophic expenditures (number of indi- (richest quintile; see supplementary appendix). Although the viduals spending more than 10% of their individual con- Kyrgyz Republic exhibits a lower smoking prevalence sumption on tobacco-related treatment costs; Figure 1). among men (26%) and a lower average price per cigarette FIGURE 1. Conceptual Framework Capturing the Health, Financial, and Distributional Consequences of Increased Tobacco Taxes among Armenian Male Smokers Postolovska et al.: Distributional Impact of Increased Tobacco Taxes in Armenia 33 pack (0.60 USD) compared to Armenia, excise tax rates are consumption quintile as a reference and applied to the dis- similarly low (16% specific, 8% ad valorem excise tax ease-specific hospitalization rates. The average cost of treat- rates).5 We did not have the necessary data to estimate price ment per disease was obtained from Armenia’s BBP price elasticity by age group, yet, consistent with reviews,6,11 those list.39 The BBP fully funds services for socially vulnerable under age 25 were assumed to be twice as responsive to price groups, including the poor and those with disabilities.40 changes as those above age 24. According to the 2014 ILCS data,19 almost 28% of the popu- We updated a previously published static model27 follow- lation was eligible for the BBP. In our modeling, the govern- ing over time all Armenian men alive in 2015. The large ment was assumed to fully pay the treatment costs for those excise tax increase was assumed to be fully passed on to con- covered by the BBP; those individuals not eligible would pay sumers through an approximately 145% increase in tobacco the full BBP price out of pocket. The change in OOP spend- retail price, and half of price elasticity would be due to par- ing would be ticipation elasticity.6 By age group (a) and consumption À Á X X  quintile (q), the number of individuals who would quit (from DOOPq D 1 ¡ Bq DD a; q P d u d ;q Cd ; (3) a d the current adult male smoking population) or not initiate smoking (among males currently < 15 years), denoted DSa,q, where Bq is the fraction of population covered in quintile q, would depend on the initial number of smokers (Sa,q), the Pd is the contribution (in percentage) of disease d to tobacco- 2), price elasticity ea,q (per age group participation elasticity (1 related premature deaths, Cd is the treatment cost for disease and quintile), and relative price change (D p p ): d, and ud,q is the utilization of health services for disease d per quintile q. Likewise, government savings among those   1 Dp covered by BBP would be DSa;q D eq;a Sa;q : (1) 2 p X X  Govtsavings; q D Bq a DDa;q d P d u d ;q Cd : (4) For the number of premature deaths averted (DDa,q), we used estimates from Verguet et al.27 and Doll et al.35 for the We did not estimate potential health care costs incurred changes in expected mortality based on age at cessation (ra), resulting from increased survival among quitters. However, assuming that half of smokers would die from their addic- previous studies have suggested that quitting could be associ- tion.35-37 Hence, the premature deaths averted would amount ated with a reduction in overall health expenditures.41-43 to For the cases of medical impoverishment (poverty)   averted, we counted the number of individuals who would 1 Dp fall below the poverty line as a result of OOP tobacco-related DDa;q D eq;a ra Da;q : (2) 2 p disease treatment costs. Almost 30% of Armenia’s popula- tion lives below the national poverty line (41,700 AMD per Though higher prices would likely lower smoking intensity month or 100 USD).44 Given that the national poverty line among continuing smokers, we only estimated the mortality was estimated in per adult equivalent terms,45 we identified reduction associated with quitting and did not model any sub- an annual individual consumption cutoff in the simulated stitution effects of smokers switching to lower priced consumption distribution at the 30th percentile45 (about 1200 cigarettes. USD per year). Therefore, we calculated the number of indi- For OOP and government expenditures averted, we allo- viduals for whom the simulated annual consumption was cated the averted premature deaths (equation 2) to four main above this poverty line but whose annual net consumption causes: heart disease, neoplasms (lung cancer), stroke, and would decrease to less than 1200 USD after paying for chronic obstructive pulmonary disease.1 Health care utiliza- tobacco-related disease treatment. Likewise, for averted tion for each cause was derived using the total annual number cases of catastrophic expenditures, we calculated the number of hospitalizations by the International Statistical Classifica- of individuals for whom OOP expenditures on tobacco- tion of Diseases and Related Health Problems group in related disease treatment would be greater than 10% of Armenia’s Ministry of Health’s 2015 statistical yearbook38 annual individual consumption. and the prevalence rates of the four diseases.1 To derive hos- In addition to the base-case scenario, we conducted three pitalizations by quintile, we used data on quintile-specific sensitivity analyses. First, we applied a flat price elasticity of utilization rates for inpatient services from the 2014 ILCS.19 ¡0.54 to all quintiles. Second, we used two alternative pov- The utilization rates were normalized using the middle erty thresholds: a lower poverty line of about 80 USD per 34 Health Systems & Reform, Vol. 4 (2018), No. 1 month and a food poverty line of about 60 USD per month.45 Half of those deaths would be averted among the bottom two Around 10% and 2% of the population were classified as quintiles, with only 10% of them among the richest quintile. poor using the lower and food poverty lines, respectively.45 This is largely driven by the higher price elasticity among Third, we used two alternative thresholds for catastrophic the poor (almost 2.6 times higher among the poorest than the expenditures: 20% and 40% of individual consumption. richest quintile). As a sensitivity analysis, when assuming a Table 1 shows all of the input parameters used. All analyses flat elasticity across all quintiles, as expected, the total num- were conducted using R software (R 3.3.2, www.r-project. ber of deaths averted would remain similar (about 86,000) org). but the distribution would be more uniform following the quintile-specific smoking rates: 20% of deaths would be averted in the richest quintile versus 17% in the poorest quin- Stakeholder Analysis tile (supplementary appendix, Table A1). Following Bump and Reich,46 to gain a better understanding Large savings in OOP and public spending would also of the political circumstances in which the tax hike was pro- occur. With averted tobacco-related disease treatment costs posed, we conducted interviews with Armenian stakeholders, among those eligible for the BBP, the government would which focused on these stages of the policy cycle: the initial save approximately 26 million USD, and almost 63 million placement on the policy agenda or agenda setting, the techni- USD of OOP expenditures would be averted among those cal design of the reform proposal, and the adoption of the not covered by the BBP. Almost 37% of these OOP savings tobacco tax as public policy.47,48 would accrue to the bottom two quintiles and 30% to the Qualitative data were collected through semistructured middle quintile in which fewer individuals were eligible for interviews. We used a purposeful sampling approach to iden- the BBP. When we assumed a flat price elasticity, OOP sav- tify interviewees by constructing a preliminary list of stake- ings would be slightly greater (67 million USD), and almost holders prior to arriving in Armenia based on a literature 27% of those savings would incur among the richest quintile review of Armenia’s tobacco control efforts. The interviews versus 28% in the bottom two quintiles (Table A1). were conducted in Yerevan in June 2016. Interviewees With a hike to a 75% tax rate, almost 22,000 poverty cases included representatives from the Ministry of Health, inter- would be averted. Given that 30% of the population already national organizations, health professionals, local nongov- lived below the poverty line, no poverty cases would be ernmental organizations, and universities. We interviewed a averted among this population. Almost 27% of the averted total of 11 individuals (Table 2), using a semistructured poverty cases would accrue to the second poorest quintile interview guide, although stakeholders were encouraged to and 14% to the richest quintile. Using a lower poverty line talk generally about tobacco control. (about 980 USD per year), the number of poverty cases Contemporaneous notes were taken and content analysis averted would rise slightly to 23,000; using the food poverty was performed once all interviews were completed to iden- line (about 690 USD per year), 24,000 poverty cases would tify relevant themes. We identified reasons for failure to be averted (Table A2). Similarly, almost 33,000 cases of cat- strengthen tobacco control in the past, the degree to which astrophic expenditures (defined as health spending > 10% of distributional consequences of the policy were important, individual consumption) would be avoided. and changing factors that allowed the tobacco tax dialogue to The ECEA showed that tobacco taxes could be progres- commence in 2015. This work was supplemented by infor- sive in Armenia and that the poor could accrue large benefits. mation extracted from national surveys, news releases, and This analytical result served as an important element in refut- published research relating to tobacco control in Armenia. ing the regressivity argument. The stakeholder analysis sug- We reviewed the laws and initiatives, which were cited dur- gested that regressivity more so than any other argument ing interviews, to gain a detailed account of the historical against tobacco taxation (e.g., smuggling, revenue losses) evolution of Armenia’s tobacco control measures. The Har- was at the center of the policy debate over increasing the tax. vard Human Research Protection Program granted an exemp- In several interviews, stakeholders stated that Armenia had tion for this study. strong tax and customs administration systems. Tobacco products, as well as other goods and imports, have holo- graphic labels and unique identification codes, and tax offi- RESULTS cers commonly make sample purchases to test the In Armenia, increasing taxes to about 75% of the retail price information provided on the products. This was argued to be of cigarettes would avert about 88,000 premature deaths a strong deterrent to smuggling. In addition, two interviewees among current estimated quitters and noninitiators (Table 3). mentioned the ease of tobacco tax increases. Unlike other Postolovska et al.: Distributional Impact of Increased Tobacco Taxes in Armenia 35 Input Value Reference 29 Male population 1,419,370 Male population distribution, age group (years) <15 21% 29 15–24 16% 25–44 30% 45–64 25% 65 9% Individual annual consumption (USD 2014) Q1 (poorest) <1091 19 Q2 1092–1458 Q3 1459–1744 Q4 1745–2191 Q5 (richest) >2191 Male smoking prevalence, per age group (years) 15–24 38% Authors’ calculations based on ref. 3 25–44 67% 45–64 58% 65 31% Male smoking prevalence, by wealth quintile Q1 (poorest) 49% Authors’ calculations based on ref. 3 Q2 61% Q3 59% Q4 49% Q5 (richest) 42% Daily cigarette consumption 24 cigarettes Authors’ calculations based on ref. 3 5 Price per pack of cigarettes (USD 2014) before tax increase 1.25 39 Tobacco-related disease treatment costs (USD 2014) COPD 424 Stroke 350 Heart disease 1724 Neoplasm (lung 4781 cancer) 19, 61 Fraction of population eligible for the publically financed Q1 (poorest) 40 basic benefits package (%), by consumption quintile Q2 30 Q3 27 Q4 23 Q5 (richest) 19 Utilization (%) of health care services per tobacco-related Neoplasms 40 Authors’ calculations based on ref. 38 disease Circulatory system 75 diseases Respiratory system 27 diseases Relative use of health care services by consumption quintile Q1 (poorest) 0.72 Authors’ calculations based on ref. 19 (standardized to use Q3 as a reference) Q2 0.73 Q3 1 Q4 1.06 Q5 (richest) 1.17 35 Reduction in mortality risk by age (age group in years) at 15–24 97% quitting smoking 25–44 85% 45–64 75% 65 25% Price elasticity of demand for tobacco products, by Q1 (poorest) ¡0.74 Authors’ assumption based on estimates consumption quintile Q2 ¡0.65 from the Kyrgyz Republic32 Q3 ¡0.65 Q4 ¡0.46 Q5 (richest) ¡0.28 (Continued on next page ) 36 Health Systems & Reform, Vol. 4 (2018), No. 1 Input Value Reference 45 National monthly poverty line 41,698 AMD (100 USD) 45 National poverty rate (percentage of population) 30 COPD D chronic obstructive pulmonary disease. TABLE 1. Inputs for the Modeling of the Distributional Impact of Increased Tobacco Excise Taxes in Armenia proposed tax changes, they noted that tobacco and alcohol additional revenues.49 In addition, Armenia’s 2015 accession taxes were easier to enforce and did not require any addi- to the Eurasian Economic Union resulted in its own set of tax tional regulation. As a result, higher tobacco/alcohol taxes measures and regulations, including the mandated harmoni- were adopted as government policy in 2015 and entered into zation of rates of excise duties on alcohol and tobacco prod- force on January 1, 2017, and remaining changes to the bud- ucts over the next five years.52,53 Therefore, in October 2015, get code will be implemented in the following year.49 the government approved a package of draft laws on the tax Based on the interviews and a literature review, our analy- code including tobacco excise taxes. More specifically, the sis showed that, unlike previous unsuccessful attempts, two recently approved tax code mandates alcohol/tobacco excise important contextual factors helped to garner support for the taxes to increase by 15% per year over 2017–2021, leading inclusion of higher tobacco taxes in Armenia’s new tax code: to an excise tax rate of 44% by 2021.49 first, tobacco tax increases were included alongside tax The tax reform design was based on two key principles: increases on other goods and services, including labor the new tax system should enhance growth and equity and income tax, and, second, it was seen as an inevitable step for generate revenue to allow higher social and capital expendi- the harmonization of taxes in the Eurasian Economic Union. tures.49,54 Equity was particularly important: according to Economic pressures presented an opportunity for an over- our stakeholder analysis, concerns about regressivity were an haul of the existing tax system. In recent years, the small but important factor in delaying increased tobacco taxes after relatively open Armenian economy has been hard hit due to FCTC ratification. its sensitivity to regional factors and shocks. The escalation Despite FCTC’s recommendations, the government had of tensions between Russia and Ukraine in 2014 and the not strengthened tobacco control, particularly in relation to decline in international oil prices led to declines in Russia’s raising excise taxes, and tobacco control efforts in Armenia economic growth prospects. Combined with sanctions, this had diminished since FCTC ratification.21 All stakeholders led to the appreciation of the US dollar, less external funding emphasized that Armenia was the first among the former available for Russian companies, and depreciation of the Soviet countries to ratify the FCTC in 2004. The early push ruble. This significantly affected Armenia through the reduc- for tobacco control measures in Armenia was largely attrib- tion of remittances (which in 2014 accounted for 20% of its uted to former President Kocharyan, himself a nonsmoker, gross domestic product [GDP]) and exports and the deprecia- who strongly advocated for FCTC implementation. In the tion of the exchange rate at the end of 2014.50,51 absence of strong public supporters, the importance of With a public debt of about 55% of its GDP and fiscal rev- tobacco control measures subsided after Kocharyan left enues of only about 22% of GDP, Armenia was facing signif- office in 2008. Though the Ministry of Health was a propo- icant fiscal pressures.49 In 2015, the World Bank and the nent of stronger tobacco control, most interviewees sug- International Monetary Fund supported measures to raise gested that it was not a powerful player in fiscal policy discussions. In addition, the tobacco industry held a strong lobby in the Parliament, with several former tobacco industry Stakeholder Group Number of Interviews executives having served on the Parliamentary Standing Ministry of Health 3 Committees on Financial Credit and Budgetary Affairs and Health professionals 2 on Economic Affairs, leading to several draft laws on International organizations 3 tobacco control being recalled from the Parliament. Local nongovernmental organizations 2 Universities 1 Realizing that the regressivity argument had been an important roadblock in the adoption of increased tobacco TABLE 2. Number of Interviews Conducted with Key Stakeholders taxes, the World Bank and the International Monetary Fund (n D 11) in Armenia provided technical assistance to simulate scenarios of Postolovska et al.: Distributional Impact of Increased Tobacco Taxes in Armenia 37 (Poorest) (Richest) Total Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 Premature deaths averted (in 1000s) 88 21 23 22 13 9 (71–106) (17–25) (18–28) (18–27) (11–16) (7–10) Out-of-pocket expenditures related to 63 10 13 19 12 9 tobacco-related disease treatment (51–77) (8–12) (11–16) (15–22) (10–15) (7–11) averted (million USD) Government savings related to tobacco- 26 7 6 7 4 2 related disease treatment averted (20–30) (6–8) (5–7) (5–8) (3–4) (2–3) (million USD) Poverty cases averted (in 1000s) 22 0 6 8 5 3 (18–27) 0 (5–7) (7–10) (4–6) (2–3) Cases of catastrophic health expenditures 33 5 7 10 6 5 (>10% of consumption) averted (in (28–40) (4–6) (6–8) (8–12) (5–8) (4–6) 1000s) a No poverty cases are averted in the poorest consumption quintile given that 30% of the population is already below the poverty line. Lower and upper bounds (calculated by using §20% variations in price elasticities) are indicated in parentheses. TABLE 3. Estimated Distributional Impact of Increased Tobacco Taxes among the Current Male Population in Armenia, by Individual Con- sumption Quintile for a Shift to a 75% Tobacco Tax Ratea proposed tax increases on various products, including tobacco, male population, cumulatively into the future, 88,000 prema- to address the equity concerns.49,54 Experience from other ture deaths, 63 million USD of OOP medical expenditures, countries (e.g., the Philippines) played an important role in and 22,000 poverty cases could be averted. Because the poor addressing the regressivity concerns and allowing the Ministry are more sensitive to price increases, the health gains would of Finance to adopt the proposed changes as government be more concentrated among the bottom two consumption policy. quintiles of the population. Given that a larger share of the Though several interviewees cited examples from the poor is eligible for the BBP (exempt from OOP payments), Philippines and Thailand, where tobacco and alcohol taxes averted tobacco-related disease treatment costs would benefit are earmarked for health, earmarking was not discussed at the middle quintiles. Cumulative government savings on length and few interviewees supported it in Armenia. They tobacco-related disease treatment costs for those BBP-eligi- cited the danger of setting a precedent, which would result in ble would amount to 26 million USD, about 12% of the other ministries and government agencies requesting their country’s annual health budget.18 own earmarked sources. One stakeholder also cited the Our modeling findings about the distributional effects on unsuccessful attempt to earmark proceeds from a VAT on increased tobacco taxes were used as an input in the policy dia- medicines for health in 2001 as a reason why earmarking logue. Yet the extent to which regressivity concerns had hin- tobacco taxes was not thought to be viable in Armenia. dered prior adoption of increased tobacco taxes remained Tobacco tax was seen as an important measure to reduce unclear; hence, we conducted interviews to identify the factors consumption, but all stakeholders emphasized that other that allowed tobacco taxes to be placed on the policy agenda FCTC measures should be enforced. They stressed the impor- in 2015. Based on our stakeholder analysis, we identified sev- tance of raising public awareness and enforcing smoke-free eral reasons. First, the fiscal constraints faced by the govern- zones and indicated that though they supported further ment and the 2015 accession to the Eurasian Economic Union tobacco tax increases, they believed that national cessation initiated a comprehensive review of the existing tax policy. support services (currently not available in Armenia) should Though previous attempts to increase tobacco taxes were follow to fully realize the benefits of higher cigarette prices. unsuccessful in part due to lack of sufficient support in Parlia- ment, the comprehensive nature of the tax reform allowed tobacco measures to be included in the proposal. Second, DISCUSSION despite initial regressivity concerns, the stakeholder interviews The ECEA indicates that higher tobacco taxes in Armenia suggested that experience from other countries and our analy- would avert large numbers of premature deaths and poverty sis of simulations of the potential impact of such taxes on the cases. With a large hike to a 75% tax rate, among the current poor were strong arguments for agreeing to raise tobacco taxes 38 Health Systems & Reform, Vol. 4 (2018), No. 1 as part of the overall fiscal reform. Our case study presents evi- the consumer. Although this is a standard assumption in dence of a successful attempt to increase tobacco taxes as part modeling studies,4,6 the empirical evidence is mixed,6,56-59 of a broader governmental tax reform, yet the proposed excise and we may have overestimated the impact of increased tax of 44% (to be achieved by 2021) will still remain well excise taxes. Lastly, the stakeholder analysis we pursued was below WHO’s recommended level. simple: we conducted a limited number of discussions with participants from a small group of organizations and institu- tions, and some of these interviews may also have suffered Limitations from recall bias. Our analysis presents a number of limitations. First, we were not able to calculate the price elasticity of demand for CONCLUSION tobacco products in Armenia, and our model was based on elasticity estimates from the Kyrgyz Republic. Yet, the Kyr- Our study contributes to the literature on the distributional gyz elasticities fell within the range of elasticities previously impact of higher tobacco taxes. Though the regressivity argu- estimated in low- and middle-income countries.6 To test the ment has been commonly used against tobacco price increases sensitivity of our findings to elasticity assumptions, we also and was perceived to be a barrier in Armenia, similar to other simulated impact using a flat price elasticity across all quin- recent studies,25,26,60 we do not find that higher tobacco prices tiles. Second, we did not model substitution effects of indi- in Armenia would necessarily disproportionately burden the viduals switching to lower-priced cigarettes following price poor. As recent studies have found, higher price responsive- increases. However, unlike other types of taxes, high specific ness among the poor may shift the burden of additional taxes excise taxes would narrow the price gap between the most to the rich.25,26 Not only can higher taxes reduce premature and least expensive cigarettes and encourage cessation rather mortality through smoking cessation, but they can also than substitution to lower-priced cigarettes.7,10 Third, we decrease potential OOP expenditures on tobacco-related dis- assumed that reduced smoking intensity would not yield any ease treatment. Given the associated high costs, tobacco taxes health benefits: individuals smoking fewer cigarettes per day can bring substantial financial protection to individuals by pre- as a result of tax hike would not have better health outcomes venting such OOP medical expenditures altogether.25 in our model; we did not model secondhand smoking either. Though the health benefits associated with smoking cessa- Consequently, we are likely to underestimate the full impact tion have been well established, this has not necessarily been of higher tobacco taxes on avoiding premature mortality and enough to encourage countries to raise tobacco taxes. Our providing financial protection. Fourth, in the absence of data stakeholder analysis in Armenia suggests that identifying the on OOP expenditures per disease, we used the BBP price list potential windows of fiscal opportunities is important and as a proxy for incurred OOP expenditures. Although this is could enable a push for higher tobacco taxes. Because the the official price for services in all public facilities, there global health community encourages the use of fiscal policies may be informal payments.55 In addition, data on expendi- to change behavior (e.g., tobacco, alcohol, sugar-sweetened tures on medicines not covered by the BBP were not avail- beverage taxes), the sole public health argument might be able and hence could not be included. Our results are thus insufficient. In the case of Armenia, our qualitative interviews likely to underestimate the expenditures related to tobacco- indicate that governments could successfully increase tobacco related disease treatment and the number of poverty cases taxes by including them as part of broader fiscal reforms rather averted, because OOP medical expenditures are likely to be than pushing them through as stand-alone reforms. Though the higher than the established government fees for the BBP. proposal marks an important step forward, the proposed 44% Fifth, we only included the cost of inpatient care, because we rate in Armenia remains well below WHO’s recommended were not able to obtain detailed data on utilization for each level. Concerted efforts need to be taken to ensure that further disease and associated costs at the primary care level. How- tax increases are implemented and other tobacco control meas- ever, primary care is free for all citizens in Armenia. Sixth, ures are enacted following best practices. the health and financial benefits were modeled into the future (for the current Armenian male population), when individu- DISCLOSURE OF POTENTIAL CONFLICTS als are expected to face tobacco-related diseases. Hence, OF INTEREST there is large uncertainty in our assumptions, because we assume that key inputs (e.g., consumption, costs, utilization, Iryna Postolovska was a consultant to the World Bank for BBP coverage) would remain the same over time. Seventh, this study, and Rouselle Lavado was a staff member of the we assumed that the tax increases would be fully passed onto World Bank. The findings, interpretations, and conclusions Postolovska et al.: Distributional Impact of Increased Tobacco Taxes in Armenia 39 in this article are entirely those of the authors. Responsibility 7. Chaloupka FJ, Hu T, Warner KE, Jacobs R, Yurekli A. The tax- for the views and opinions expressed rests solely with the ation of tobacco products. In: Jha P, Chaloupka FJ, editors. authors; they are not endorsed by any member institution of Tobacco control in developing countries. New York (NY): Oxford University Press; 2000. p. 237-272. the World Bank Group, its Executive Directors, or the coun- 8. Jha P, Chaloupka FJ. Curbing the epidemic: governments and tries they represent. the economics of tobacco control. Washington (DC): World Bank Publications; 1999. 9. United Nations. Addis Ababa action agenda of the Third Inter- national Conference on Financing for Development; 2015 Jul ACKNOWLEDGMENTS 13–16. Addis Ababa (Ethiopia), New York (NY): United Nations; 2015 [cited 2016 Nov 16]. http://www.un.org/esa/ffd/ We appreciate the assistance of Samvel Kharazyan and Arpine wp-content/uploads/2015/08/AAAA_Outcome.pdf. Azaryan in arranging the stakeholder interviews and are grate- 10. Jha P, Peto R. Global effects of smoking, of quitting, and of ful to all interview respondents for their participation in this taxing tobacco. N Engl J Med. 2014;370(1):60-68. study. We thank Thomas Bossert, Volkan Cetinkaya, ¸ Alan doi:10.1056/NEJMra1308383. Fuchs, Margaret Kruk, Myl ene Lagarde, Patricio Marquez, and 11. World Health Organization. WHO technical manual on tobacco Moritz Meyer for helpful comments on an earlier version of tax administration. Geneva (Switzerland): World Health Orga- nization; 2010. this article. Parts of this paper have been reproduced or adapted 12. Koh HK. An analysis of the successful 1992 Massachusetts from: “Postolovska, Iryna; Lavado, Rouselle F.; Tarr, Gillian; tobacco tax initiative. Tob Control. 1996;5(3):220-225. Verguet, St ephane. 2017. Estimating the Distributional Impact doi:10.1136/tc.5.3.220. of Increasing Taxes on Tobacco Products in Armenia: Results 13. Balbach ED, Herzberg A, Barbeau EM. Political coalitions and from an Extended Cost-Effectiveness Analysis. World Bank, working women: how the tobacco industry built a relationship with the Coalition of Labor Union Women. J Epidemiol Com- Washington, DC. Ó World Bank. https://openknowledge. munity Health. 2006;60(Suppl 2):ii27¡ii32. worldbank.org/handle/10986/26386 License: CC BY 3.0 IGO” 14. Smith KE, Savell E, Gilmore AB. What is known about for which support was provided by the World Bank’s Global tobacco industry efforts to influence tobacco tax? A systematic Tobacco Program, co-financed by the Bill & Melinda Gates review of empirical studies. Tob Control. 2013;22(2):144-153. Foundation and the Bloomberg Foundation. Finally, we are doi:10.1136/tobaccocontrol-2011-050098. grateful to Michael Reich and two anonymous reviewers for 15. Remler DK. Poor smokers, poor quitters, and cigarette tax regressivity. Am J Public Health. 2004;94(2):225-229. valuable suggestions. doi:10.2105/AJPH.94.2.225. 16. Lavado R, Meyer M, Postolovska I, Sotomayar R. Beyond the perceived regressivity of higher tobacco taxes: turning short- REFERENCES term losers into long-term winners?” In: Marquez PV, Moreno- 1. Institute for Health Metrics and Evaluation. GBD Compare j Dodson B, editors. Tobacco tax reform: at the crossroad of IHME Viz Hub. 2017 [accessed 2016 Nov 18]. http://vizhub. health and development. Washington (DC): World Bank; 2017. healthdata.org/gbd-compare. p. 123-150. 2. National Institute of Health. Assessment of the health system in 17. Bloom DE, Cafiero ET, Jan e-Llopis E, Abrahams-Gessel S, 2016. Yerevan (Armenia): Ministry of Health Armenia; Bloom LR, Fathima S, Feigl AB, Gaziano T, Mowafi M, Pan- 2016 [accessed 2016 Dec 19]. http://moh.am/uploads/ dya A, et al. The global economic burden of noncommunicable HSPA_2016_ARM_2016.pdf. diseases. Geneva (Switzeland): World Economic Forum; 2011. 3. Ministry of Health. Armenian national health sector perfor- 18. World Bank. World development indicators j Data. 2016 [cited mance assessment. Dataset. Yerevan (Armenia): Ministry of 2016 Nov 20]. http://data.worldbank.org/data-catalog/world- Health of the Republic of Armenia; 2015. development-indicators. 4. U.S. National Cancer Institute, World Health Organization. 19. National Statistical Service of the Republic of Armenia. Arme- The economics of tobacco and tobacco control. Bethesda nia Household’s Integrated Living Conditions Survey. Dataset. (MD): U.S. Department of Health and Human Services, Yerevan (Armenia): National Statistical Service of the Repub- National Institutes of Health, National Cancer Institute; 2016. lic of Armenia; 2014. NIH Publication No. 16-CA-8029A. 20. Law of the Republic of Armenia on the sale, consumption 5. World Health Organization. WHO report on the global tobacco and limiting the use of tobacco. 2004 [accessed 2016 epidemic, 2015. Geneva (Switzerland): World Health Organi- Nov 20]. http://www.tobaccocontrollaws.org/files/live/Armenia/ zation; 2015. Armenia%20-%20Law%20on%20Restrictions%20of%20Sale% 6. International Agency for Research on Cancer. IARC hand- 2C%20Consumption%2C%20and%20Use.pdf. books of cancer prevention, tobacco control: Vol. 14. Effective- 21. Movsisyan NK, Connolly GN. Measuring Armenia’s prog- ness of tax and price policies in tobacco control. Lyon (France): ress on the Tobacco Control Scale: an evaluation of International Agency for Research on Cancer; 2011. tobacco control in an economy in transition, 2005–2009. 40 Health Systems & Reform, Vol. 4 (2018), No. 1 BMJ Open. 2014;4(2):e004410. doi:10.1136/bmjopen- 36. Jha P, Jacob B, Gajalakshmi V, Gupta PC, Dhingra N, Kumar 2013-004410. R, Sinha DN, Dikshit RP, Parida DK, Kamadod R, et al. A 22. The Government of the Republic of Armenia. Official news: nationally representative case–control study of smoking and Package of draft laws on tax code approved. 2015 [accessed death in India. N Engl J Med. 2008;358(11):1137-1147. 2016 Dec 19]. http://www.gov.am/en/news/item/8122/. doi:10.1056/NEJMsa0707719. 23. Verguet S, Laxminarayan R, Jamison DT. Universal public 37. Jha P, Ramasundarahettige C, Landsman V, Rostron B, finance of tuberculosis treatment in India: an extended cost- Thun M, Anderson RN, McAfee T, Peto R. 21st-Century effectiveness analysis. Health Econ. 2015;24(3):318-332. hazards of smoking and benefits of cessation in the United doi:10.1002/hec.3019. States. N Engl J Med. 2013;368(4):341-350. doi:10.1056/ 24. Verguet S, Kim JJ, Jamison DT. Extended cost-effectiveness NEJMsa1211128. analysis for health policy assessment: a tutorial. Pharmacoeco- 38. National Institute of Health of Armenia. Health and wellness nomics. 2016;34(9):913-923. doi:10.1007/s40273-016-0414-z. statistical yearbook 2015. Yerevan (Armenia): Ministry 25. Verguet S, Gauvreau CL, Mishra S, MacLennan M, Murphy of Health Armenia; 2016 [accessed 2016 Nov 20]. SM, Brouwer ED, Nugent RA, Zhao K, Jha P, Jamison DT. http://moh.am/uploads/Health_Health%20Care%20Year%20 The consequences of tobacco tax on household health and Book_2015_Arm.pdf. finances in rich and poor smokers in China: an extended cost- 39. Ministry of Health of the Republic of Armenia. Armenia basic effectiveness analysis. Lancet Glob Health. 2015;3(4): benefits package price list. Yerevan (Armenia): Ministry of e206¡e216. doi:10.1016/S2214-109X(15)70095-1. Health of the Republic of Armenia; 2016. 26. Salti N, Brouwer E, Verguet S. The health, financial and distri- 40. Richardson E. Armenia: health system review. Health Syst butional consequences of increases in the tobacco excise tax Transit. 2013;15(4):1-99. among smokers in Lebanon. Soc Sci Med. 2016;170:161-169. 41. Hodgson TA. Cigarette smoking and lifetime medical expendi- doi:10.1016/j.socscimed.2016.10.020. tures. Milbank Q. 1992;81-125. doi:10.2307/3350086. 27. Verguet S, Tarr G, Gauvreau C, Mishra S, Jha P, Liu L, 42. Fishman PA, Khan ZM, Thompson EE, Curry SJ. Health care Xiao Y, Qiu Y, Zhao K. Distributional benefits of tobacco costs among smokers, former smokers, and never smokers in tax and smoke-free workplaces in China: a modeling study. an HMO. Health Serv Res. 2003;38(2):733-749. doi:10.1111/ J Glob Health. 2017;7(2):020701. doi:10.7189/ 1475-6773.00142. jogh.07.020701. 43. Rasmussen SR, Prescott E, Sørensen TI, Søgaard J. The total 28. Ministry of Health of the Republic of Armenia, World Health lifetime health cost savings of smoking cessation to society. Organization. Armenia: reporting instrument of the WHO Eur J Public Health. 2005;15(6):601-606. doi:10.1093/eurpub/ Framework Convention on Tobacco Control. Geneva (Switzer- cki024. land): World Health Organization; 2014 [accessed 20 Novem- 44. World Bank, National Statistical Service of the Republic of ber 2016]. http://apps.who.int/fctc/implementation/database/ Armenia. Social snapshot and poverty in Armenia: based on sites/implementation/files/documents/reports/armenia_2014_re the results of the 2014 Integrated Living Conditions Survey of port_final.pdf. Households. Yerevan (Armenia): World Bank and National 29. World Bank. Health, nutrition and population data and statis- Statistical Service of the Republic of Armenia; 2015. tics. 2016 [accessed 2016 Nov 21]. http://datatopics.world 45. World Bank, National Statistical Service of the Republic of bank.org/hnp/. Armenia. Social snapshot and poverty in Armenia: based on 30. Kemp-Benedict E. Income distribution and poverty: methods the results of the 2015 Integrated Living Conditions Survey of for using available data in global analysis. 2001 [accessed Households. Yerevan (Armenia): World Bank and National 2016 Nov 16]. http://gdrs.sourceforge.net/docs/PoleStar_Tech Statistical Service of the Republic of Armenia; 2016 [accessed Note_4.pdf 2016 Dec 16]. 31. Salem A, Mount T. A convenient descriptive model of 46. Bump JB, Reich MR. Political economy analysis for tobacco income distribution: the gamma density. Econometrica. control in low- and middle-income countries. Health Policy 1974;42:1115-1127. Plan. 2013;28(2):123-133. doi:10.1093/heapol/czs049. 32. National Statistical Committee of the Kyrgyz Republic. Kyrgyz 47. Kaufman RR, Nelson JM, editors. Crucial needs, weak incen- Integrated Household Survey. Dataset. Bishkek (Kyrgyz tives: social sector reform, democratization, and globalization Republic): National Statistical Committee of the Kyrgyz in Latin America. Washington (DC): Woodrow Wilson Center Republic; 2015. Press; 2004. 33. Hu T-W, Ren Q-F, Keeler TE, Bartlett J. The demand for ciga- 48. Roberts MJ, Hsiao W, Berman P, Reich MR. Getting health rettes in California and behavioural risk factors. Health Econ. reform right: a guide to improving performance and equity. 1995;4(1):7-14. doi:10.1002/hec.4730040102. New York (NY): Oxford University Press; 2008. 34. Adioetomo SM, Djutaharta T, Hendratno. Cigarette consump- 49. International Monetary Fund. Republic of Armenia Third tion, taxation, and household income: Indonesia case study. review under the extended arrangement. IMF Country Report Washington (DC): World Bank; 2005. No. 16/246. Washington (DC): International Monetary Fund, 35. Doll R, Peto R, Boreham J, Sutherland I. Mortality in 2016 [accessed 2016 Dec 16]. https://www.imf.org/en/Publica relation to smoking: 50 years’ observations on male British doc- tions/CR/Issues/2016/12/31/Republic-of-Armenia-Third- tors. BMJ. 2004;328(7455):1519(9 pages). doi:10.1136/ Review-Under-the-Extended-Arrangement-and-Request-for- bmj.38142.554479.AE Waiver-and-44109. Postolovska et al.: Distributional Impact of Increased Tobacco Taxes in Armenia 41 50. Stepanyan A, Roitman A, Minasyan G, Ostojic D, Epstein N. 19]. http://www.worldbank.org/en/news/opinion/2016/07/01/ The spillover effects of Russia’s economic slowdown on neigh- interview-with-ms-gohar-gyulumyan-on-the-tax-code. boring countries. Washington (DC): International Monetary 55. Lewis M. Informal payments and the financing of health care in Fund; 2015. developing and transition countries. Health Aff (Millwood). 51. Ayvazyan K, Daban Sanchez T. Spillovers from global and 2007;26(4):984-997. doi:10.1377/hlthaff.26.4.984. regional shocks to Armenia. Washington (DC): International 56. Linegar DJ, van Walbeek C. The effect of excise tax increases Monetary Fund; 2015. IMF Working Paper. on cigarette prices in South Africa. Tob Control. 2018; 27:65- 52. Eurasian Economic Commission. On the draft agreement on the 71. doi:10.1136/tobaccocontrol-2016-053340 tax policy principles in respect of excise duties on tobacco 57. Hanson A, Sullivan R. The incidence of tobacco taxation: evi- products in the Member States of the Eurasian Economic dence from geographic micro-level data. Natl Tax J. 2009;62 Union. Order N 126. 2015 [accessed 2016 Nov 20]. http:// (4):677-698. doi:10.17310/ntj.2009.4.05. www.eurasiancommission.org/ru/Lists/Decisions/DispForm. 58. Sullivan RS, Dutkowsky DH. The effect of cigarette taxation on aspx?IDD531. prices: an empirical analysis using local-level data. Public Finance 53. Eurasian Economic Commission. Draft agreements on the tax Rev. 2012;40(6):687-711. doi:10.1177/1091142112442742. policy principles in respect of excise duties on alcohol and 59. Delipalla S, O’Donnell O. Estimating tax incidence, market tobacco products in the EAEU Member States are approved. power and market conduct: the European cigarette industry. Int Moscow (Russian Federation): Eurasian Economic Commis- J Ind Organ. 2001;19(6):885-908. doi:10.1016/S0167-7187(99) sion; 2015 [accessed 2016 Dec 19]. http://www.eurasiancom 00057-0. mission.org/en/nae/news/Pages/29-10-2015-5.aspx. 60. Fuchs A, Meneses F. Are tobacco taxes really regressive? 54. Gyulumyan G. Is it fair that 80–85 percent of personal income Washington (DC): World Bank; 2017. tax collections is formed through those earning less than AMD 61. Jamal S. Improvements in targeting of basic benefits package in 120 thousand income per month? 2016 [accessed 2016 Dec Armenia. Washington (DC): World Bank; 2016.