CIGARETTE AFFORDABILITY IN INDONESIA: 2002 2017 Rong Zheng Patricio V. Marquez Abdillah Ahsan Yang Wang Xiao Hu CIGARETTE AFFORDABILITY IN INDONESIA: 2002-2017 Rong Zheng Patricio V. Marquez Abdillah Ahsan Yang Wang Xiao Hu The World Bank Group SMOKING HAS Cigarette Affordability in Indonesia: 2002-2017 BEEN A MAJOR CONTRIBUTOR O DISEASE BURDEN INDONESIA. SMO PREVALENCE AMONG WORKIN AGE INDIVIDUAL HAS EXCEEDED 3 PERCENT SINCE 2 II // Executive Summary ACKNOWLEDGMENTS This report was prepared by a team led by Patricio V. Marquez, Lead Public Health Special- ist, World Bank Group, comprised of Rong Zheng (principal researcher and writer), Yang Wang, and Xiao Hu, WHO Collaborating Center for Tobacco and Economics, University of International Business and Economics, Beijing, China; and Abdillah Ahsan, Faculty of Economics and Business, University of Indonesia. The authors would like to thank the following peer reviewers for their insightful comments, inputs, and valuable suggestions that helped to improve earlier drafts of the report: Alan Fuchs, Senior Economist, Poverty and Equity Global Practice, World Bank Group Jaffar Al Rikabi, World Bank Group Frederico Gil Sander, Lead Country Economist, Economic Policy and Debt Department, World Bank Group Paul Isenman, former senior manager with the OECD/DAC Secretariat, the World Bank Group, and USAID Nigar Nargis, Assistant Professor, Department of Economics of the University of Dhaka, Bangladesh; Director, Economic and Health Policy Research, American Cancer Society, United States Corne van Walbeek, Project Director and Professor, School of Economics, University of Cape Town, South Africa Alexander Irwin edited the brief. Spaeth Hill LLC prepared the layout design. The preparation of the report was carried out under the World Bank Group Global Tobacco Control Program, coordinated by Patricio V. Marquez, with the support of the Bill & Melinda Gates Foundation and the Bloomberg Philanthropies. Washington, D.C., May 2018 III SMOKING HAS Cigarette Affordability in Indonesia: 2002-2017 BEEN A MAJOR CONTRIBUTOR O DISEASE BURDEN INDONESIA. SMO PREVALENCE AMONG WORKIN AGE INDIVIDUAL HAS EXCEEDED 3 PERCENT SINCE 2 IV // Executive Summary TABLE OF CONTENTS Abbreviations IX Executive Summary XI 1. Introduction 1 Smoking Prevalence in Indonesia: Health and Economic Consequences 1 Cigarette Production, Consumption, and Market Share in Indonesia 4 Tackling Affordability: A Critical Task for Tobacco Control in Indonesia 7 2. Literature Review 10 3. Methodology 12 Cigarette Affordability Definition 12 Price Data 13 Income Data 13 4. Results 14 Cigarette Affordability: Level, Trend, and Magnitude of Change 14 Indonesia’s Position in Global Rankings of Cigarette Affordability 18 5. Discussion: Clarifying Results, Limitations, and Policy Implications 22 6. Conclusions 30 References 32 V Cigarette Affordability in Indonesia: 2002-2017 TABLES Table 1. Cigarette Excise Tax Structure in Indonesia, 2011-2017 5 Table 2. Cigarette Affordability: RIP Average and CAI Average 14 Table 3. Retail Price for a Pack of 20 Cigarettes - Premium Brand and Cheapest Brand – Indonesia and Selected Countries, 2016, in US$ 21 Table 4. Cross-Test RIP: Expenditure vs. GDP Per Capita 22 Table 5. Adult Consumption Per Capita 24 Table 6. Total Number of Smokers and Consumption Per Smoker 26 Table 7. Growth in Number of Smokers and Cigarette Consumption Growth 27 VI // Table of Contents FIGURES Figure 1. Smoking Prevalence in the Adult Population (15+), Indonesia (1995-2016) 1 Figure 2. Market Share by Type of Cigarette, Indonesia (2011-2017) 4 Figure 3. Evolution of Cigarette Market Share by Tax Tiers, Indonesia (2010-2017) 6 Figure 4. Cigarette Production and Consumption, Indonesia (2011-2016) 6 Figure 5. Percentage Change in Cigarette Production by Type of Cigarette, Indonesia (2011-2016) 7 Figure 6. Smoking Prevalence, Cigarette Price, and Cigarette Consumption, Indonesia (2002-2016) 8 Figure 7. Cigarette Affordability, Indonesia (2002-2016) 15 Figure 8. Cigarette Retail Sales, Illicit Trade, and CAI Average 16 Figure 9. Cigarette Affordability: RIP Average (Euromonitor) vs. RIP Average (MoF) (2011-2017) 16 Figure 10. Correlation Between Cigarette Affordability, Price Growth, and Growth in GDP Per Capita, Indonesia (2011-2017) 17 Figure 11. Annual and Fixed-Base Growth Rate of Cigarette Affordability Using Average Price (MoF) 18 Figure 12. Cost of 100 Packs of the Most-Sold Cigarettes as Percentage of GDP Per Capita, 2016. 20 Figure 13. Cross-Test RIP: Expenditure vs. GDP Per Capita 23 Figure 14. Adult Cigarette Consumption Per Capita and Cigarette Affordability, Indonesia (2002-2016) 25 Figure 15. Change in Total Cigarette Consumption and Cigarette Consumption by Additional Smokers 27 VII SMOKING HAS Cigarette The Economics Affordability of Tobacco in Indonesia: 2002-2017 Taxation and Employment in Indonesia BEEN A MAJOR CONTRIBUTOR O DISEASE BURDEN INDONESIA. SMO PREVALENCE AMONG WORKIN AGE INDIVIDUAL HAS EXCEEDED 3 PERCENT SINCE 2 Executive VIII // Table Summary of Contents ABBREVIATIONS BPS: Central Statistics Board CAI: Cigarette Affordability Index CPDIR: Cigarette Price/Daily Income Ratio CPI: Consumer Price Index CTFK: Campaign for Tobacco-Free Kids GBD: Global Burden of Disease GDP: Gross Domestic Product HIC: High-income country IAKMI: (Ikatan Ahli Kesehatan Masyarakat Indonesia/Indonesian Public Health Association) IHME: Institute for Health Metrics and Evaluation LIC: Low-income country LMIC: Lower middle-income country MoF: Ministry of Finance MoL: Minutes of Labor NCD: Noncommunicable disease NCI: National Cancer Institute RIP: Relative Income Price SDGs: Sustainable Development Goals SKM: Machine-manufactured kreteks SKT: Hand-rolled kreteks SPM: Machine-made white cigarettes SUSENAS: National Socio-Economic Survey UMIC: Upper middle-income country US: United States WHO: World Health Organization IX SMOKING HAS The Economics of Tobacco Taxation and Employment in Indonesia BEEN A MAJOR CONTRIBUTOR O DISEASE BURDEN INDONESIA. SMO PREVALENCE AMONG WORKIN AGE INDIVIDUAL HAS EXCEEDED 3 PERCENT SINCE 2 EXECUTIVE SUMMARY Why Affordability? Cigarette affordability analysis is an important input to guide tobacco control policy, in particular tobacco taxation. Global evidence shows that when cigarettes are made sub- stantially less affordable for consumers, many people will quit or reduce smoking, despite the addictive power of nicotine, while many who would have smoked will never start. This study investigates trends in cigarette affordability in Indonesia from 2002 to 2017. Affordability analysis is critical for Indonesia today, because the government has accelerated tobacco control measures, notably by raising the country’s cigarette excise tax six times between 2011 and 2017. Affordability is key to understanding these measures’ successes and shortfalls, and to honing future strategies that can build on the momentum achieved. Background: Indonesia’s Tobacco Epidemic Health impacts. Estimated at 68.1 percent in 2016, Indonesia’s male smoking prevalence is among the highest in the world. The country’s five leading causes of death are all tobacco- related, including ischemic heart disease, cerebrovascular disease, tuberculosis, diabetes, and chronic respiratory diseases (IHME, 2017). Morbidity from smoking-related diseases accounts for more than 21 percent of all cases of chronic disease in Indonesia. Economic damage. Along with its health impacts, tobacco consumption imposes a heavy economic burden, primarily on Indonesia’s poor, who risk lost income, reduced labor productivity, and deeper impoverishment from out-of-pocket payments for the treatment of tobacco-related diseases. While the poor suffer most, all of Indonesian society is affected by the tobacco epidemic. Tobacco-related diseases increasingly burden the country’s national health insurance program. Indonesia’s smoking-attributable health expenditure is estimated at some US$ 1.2 billion per year (Barber et al., 2008; Goodchild et al., 2017), equivalent to about 8 percent of Indonesia’s total public expenditure on health (World Bank, 2016). Direct and indirect costs of tobacco-related disease divert resources needed for Indonesia’s development—posing a growing challenge to the country. Harm to children. In 2013, 56 percent of Indonesian children aged 0-4 years were exposed to secondhand smoke in their homes (IAKMI, 2014). Paternal smoking predicts an increased probability of short-term and chronic malnutrition among Indonesian children (Barber et al., 2008; Pradono et al., 2002). In Indonesian households where the father was a smoker, tobacco accounted for 22 percent of weekly per capita household expenditures, XI Cigarette Affordability in Indonesia: 2002-2017 with less money spent on food compared to households in which the father was a non- smoker (Semba et al., 2007). This is troubling, as Indonesian children suffer from high rates of malnutrition and stunting, compromising their neurological development, educational outcomes, and future economic productivity (World Bank, 2016). Key Findings from the Affordability Analysis Our analysis of cigarette affordability trends in Indonesia yields the following results: • Distinct chronological stages. In general, cigarette affordability in Indonesia presents three stages over the period 2002 to 2017: a fluctuating stage between 2002 and 2005; a stage of increasing affordability between 2005 and 2012; and a stage of decreasing affordability from 2012 to 2017. • The bad news: a long-range upward trend. Looking at the whole period 2002-2016, on average, cigarette affordability in Indonesia increased by 50 percent, while aggregate cigarette consumption increased by 50.6 percent. This increase in affordability followed a rise of similar magnitude from the early 1980s to 2000. Thus, in 2016, cigarettes were twice as affordable for Indonesian consumers as they had been in the early 1980s. • The good news: a recent drop in affordability, spurred by higher taxes. Focusing specifically on the years from 2011 onward, the analysis shows that ciga- rette affordability decreased by 10.2 percent from 2011 to 2017, using price data from Indonesia’s Ministry of Finance. Price data from Euromonitor show a similar afford- ability drop of 9.7 percent from 2011 to 2016. These welcome, though still modest, reductions in affordability have mainly been driven by the government’s successive tobacco excise tax increases. • An affordability-consumption disconnect, largely due to additional smokers. Even as cigarette affordability dropped from 2011 to 2016, aggregate cigarette consumption in Indonesia increased by 14.5 percent during this period, a seeming paradox. Further analysis shows, however, that increased consumption mainly stemmed from additional smokers. • Indonesia’s cigarette affordability in global perspective. Studies comparing cigarette affordability across countries based on average price or the price of the most-sold brand have tended to place Indonesia at a medium level of affordability, relative to other lower middle-income countries. However, because of the wide range of retail cigarette prices in Indonesia, some kinds of cigarettes are available there at prices that are very low in international comparison (WHO, 2017). Indonesia’s moderate global affordability rankings and simultaneous high prevalence show that XII // Executive Summary the availability of a variety of very cheap brands, together with non-price factors, are helping to drive the country’s epidemic. Non-price factors include cultural norms, as well as weak restrictions on smoking in public places and on tobacco advertising. The Bottom Line for Policy Progress has recently been made to begin reducing tobacco affordability in Indonesia, but much remains to be achieved. Recent tobacco tax reforms in Indonesia have boosted retail cigarette prices. The nominal average cigarette price rose by 65 per- cent between 2011 and 2016, from IDR 11,578.5 to IDR 19,116.3 per pack. The real average cigarette price climbed by 27 percent, and cigarettes were 10 percent less affordable in 2016 than in 2011. These gains have begun to reverse the long trend of increasing ciga- rette affordability seen in Indonesia since the turn of the century, and indeed for the nearly 40 years for which data are available. Indonesia’s tobacco epidemic continues to threaten the country’s future. Despite this progress, Indonesia’s tobacco epidemic remains one of the world’s most serious. It places the nation’s health, human capital, and economic dynamism at risk. The potential harms to children and youth, from active smoking and secondhand smoke exposures, are especially concerning. To reap the demographic dividend from its young and growing population, Indonesia must ensure that young people stay healthy. Cigarettes in Indonesia are still too cheap. Indonesian consumers can still buy ciga- rettes more cheaply than can smokers in most other middle- and high-income countries. A package of cigarettes can be bought for as little as US$ 0.45, among the lowest prices in the world (Indonesian Investments, 2018; WHO, 2017). The sale of single cigarette sticks remains a common practice in Indonesia, making cigarettes even more cheaply acces- sible. This mode of purchase may especially encourage adolescents to experiment with smoking. On the other hand, expensive brand cigarettes maintain a large market share, suggesting that many Indonesian consumers still find these products affordable. To get full benefit from affordability reductions, Indonesia’s tobacco excise system must be simplified. The country’s complex, multi-tiered tobacco tax structure might theoretically have favored small-scale domestic cigarette manufacturers. In practice, these small firms’ market share has sharply declined. The current fragmented tax model also produces deleterious public-health consequences by encouraging smokers to switch to cheaper brands when tobacco taxes rise, rather than quitting altogether. This undermines the effectiveness of using tax policy to cut tobacco consumption. XIII Cigarette Affordability in Indonesia: 2002-2017 Understanding this inconsistency provides an opportunity. Indonesian policy makers can resolve the conflict by simplifying the tobacco excise tax structure to reduce cigarette affordability aggressively across the board, for all cigarette types. Cigarette affordability cuts can only achieve full impact if all tobacco products become less affordable simultaneously and stay that way over time. A bold streamlining of the tiered tax system is crucial to achieve this goal. Tackling non-price factors. While continuing to lower affordability, tobacco control policy in Indonesia can further improve results by aggressively restricting tobacco advertising, enforcing smoke-free areas in public spaces, expanding the use of pictorial health warnings, and similar measures. The pervasive cultural perception of smoking as normal for adult men must and can be changed. “Go big, go fast.” International comparisons suggest that Indonesia’s current tobacco tax rates remain far below what is feasible in terms of revenue potential. Thus, tobacco tax tier consolidation and further bold tax increases could serve revenue purposes as well as fighting tobacco-related death and disease in the years ahead. A growing number of countries have achieved success with a tobacco tax strategy of “Go big, go fast”: large, rapid increases in tobacco tax rates, joined to impact-boosting measures including the swift merger of tax tiers. Together, these actions can permanently change consumers’ expectations about how much smoking costs and durably alter their behavior (Marquez and Moreno-Dodson, 2017). This is a promising path for tobacco control in Indonesia. XIV // Executive Summary XV SMOKING HAS The Economics of Tobacco Taxation and Employment in Indonesia BEEN A MAJOR CONTRIBUTOR O DISEASE BURDEN INDONESIA. SMO PREVALENCE AMONG WORKIN AGE INDIVIDUAL HAS EXCEEDED 3 PERCENT SINCE 2 Introduction XVI // Executive Summary 1. INTRODUCTION This study analyzes the recent evolution of cigarette affordability in Indonesia and weighs implications for the country’s tobacco control policy. This introductory section discusses Indonesia’s elevated smoking prevalence and quantifies destructive impacts on the nation’s public health and economy. We underscore disproportionate consequences for the poor and other vulnerable groups, including Indonesia’s children. Section 2 documents the principal types of cigarettes consumed in Indonesia and their market shares, assessing recent market changes. Section 3 considers the concept of affordability, including its defi- nition, measurement, and potential salience for policy. Subsequent sections summarize the results of a literature review and discuss research methods, then present the results of our analysis of cigarette affordability in Indonesia from 2002-2017. The analysis tracks affordabil- ity changes over the period, relating them to consumption levels and key policy actions, notably recent tobacco tax hikes. We also compare Indonesia’s cigarette affordability levels with those of other countries. A concluding section examines policy implications. Smoking Prevalence in Indonesia: Health and Economic Consequences Prevalence patterns. Smoking prevalence among Indonesian women is low, yet volatile (Figure 1). Prevalence among adult men, however, has followed a decades-long increasing trend. In 2016, an estimated 68.1 percent of Indonesian men smoked. This is among the highest rates in the world. Figure 1. Smoking Prevalence in the Adult Population (15+), Indonesia (1995-2016) 80% 70% 60% 50% 40% 30% 20% 10% 0% 1995 2001 2004 2007 2010 2011 2013 2016 Total 27% 31.5% 34.4% 34.2% 34.3% 36.1% 36.3% 32.8% Male 53.4% 62.2% 63.1% 65.6% 65.8% 67.4% 66% 68.1% Female 1.7% 1.3% 4.5% 5.2% 4.1% 4.5% 6.7% 2.5% Note: Tobacco use includes daily and occasional cigarette smoking and tobacco chewing. Source: SKRT (1995), SURKESNAS (2001, 2004), RISKESDAS (2007, 2010, 2013), GATS (2011), SIRKESNAS (2016) Indonesia NIHRD-Litbangkes. 1 Cigarette Affordability in Indonesia: 2002-2017 A crisis that demands action. The United Nations has called for global action to reduce tobacco use (United Nations, 2015). In the past decade, while many other countries have enforced tough tobacco controls, including tax and price measures, Indonesia has made little progress in curbing cigarette consumption. On the contrary, the number of Indonesian smokers almost tripled, from 25 million in 1980 to an estimated 73.6 million as of 2015 (World Bank, 2017). Within a single decade, from 2007-2017, Indonesia jumped from being the fifth-largest consumer of cigarettes after China, the United States, Russia, and Japan to being the world’s third-largest cigarette-consuming nation after China and Russia (CTFK, 2017; Euromonitor International, 2017; Hidayat and Surjono, 2016). In addition, more than 97 million nonsmokers are regularly exposed to secondhand smoke in Indonesia, including 70 percent of all children under the age of 15 (Barber et al., 2008; Pradono et al., 2002). Global and regional impacts. Tobacco use is an important contributor to the global burden of morbidity and mortality, accounting for 11.5 percent of deaths worldwide (GBD 2015 Tobacco Collaborators, 2017). Tobacco kills about one-half of its long-term consum- ers, ending their lives more than a decade prematurely, on average (Jha and Peto, 2014). So it is hardly surprising that tobacco also accounts for the world’s highest substance-at- tributable mortality rates. At 110.7 deaths per 100,000 population, tobacco-related death rates are three times higher than those attributable to alcohol (33.0 per 100,000) and 17 times higher than those for illicit drugs (6.9 per 100,000). In East Asia, including Indonesia, the tobacco attributable mortality rate, at 145.9 deaths per 100,000, is the second highest in the world, exceeded only by Oceania (Papua New Guinea, Kiribati, Federated States of Micronesia, and Solomon Islands, excluding Australia and New Zealand) at 269.3 deaths per 100,000 (Peacock, Leung, Larney, et al., 2018). Public health burden. Indonesia’s five leading causes of death are all tobacco-related. They include ischemic heart disease, cerebrovascular disease, tuberculosis, diabetes, and chronic respiratory diseases (IHME, 2017). In 2015, total deaths attributed to tobacco use in Indonesia reached 230,862 (Kosen et al., 2017). Morbidity from smoking-related diseases accounts for more than 21 percent of all cases of chronic disease in the country. Global evidence shows that tobacco use also helps to fuel the global epidemic of tuberculosis, and that it worsens problems such as mental illness, HIV infection, and alcohol abuse (Drope et al., 2018; NCI/WHO, 2016; WHO, 2017). Harm to children. Paternal smoking is a predictor of an increased probability of short- term and chronic child malnutrition in Indonesia. In households where the father was a smoker, tobacco accounted for 22 percent of weekly per capita household expenditures, with less money spent on food than in households in which the father was a non-smoker 2 // Introduction (Semba et al., 2007). This is troubling, as Indonesian children suffer from high rates of mal- nutrition with a prevalence of stunting at 37 percent and of wasting at 12 percent (World Bank, 2016). With over 8 million children affected, Indonesia has the fifth-highest number of stunted children in the world (Millennium Challenge Account-Indonesia, 2015). Stunting in the first two years of life can lead to irreversible damage, including shorter adult height, lower school attainment, reduced adult income, and increased incidence of morbidity in later life, which undermines human capital development. Economic damage. Along with its health impacts, tobacco consumption imposes a heavy economic burden, primarily on Indonesia’s poor. The economic burden connected with smoking falls hardest on low-income smokers who risk lost income, reduced labor productivity, and impoverishment due to out-of-pocket payments for the treatment of tobacco-related diseases. While the poor suffer most, all of Indonesian society is affected. The costs of tobacco use include illness, disability, premature death, and forgone con- sumption and investment. The smoking-attributable health expenditure in Indonesia is estimated at about US$ 1.2 billion per year (Goodchild et al., 2017; Barber et al, 2008). This represents about 8 percent of total public expenditures on health (including government budgetary and social insurance expenditures), and about 3.3 percent of total health expenditures (including government budgetary and social insurance expenditures, out-of- pocket, and other private spending) (World Bank, 2016). This figure is consistent with the observed experience in other countries. For example, using data from medical spending surveys in the United States, researchers calculated that 8.7 percent of all health care spending, or US$170 billion a year, is for illness caused by tobacco smoke, and that public programs like Medicare and Medicaid paid for most of these costs (Xu et al., 2015). Indirect costs. In terms of indirect economic costs from smoking, Kosen (2009) used a variation of the Peto method (Doll and Peto, 1981) developed for the Global Burden of Disease Project (various, cited at National Cancer Institute (NCI)/WHO, 2016) to estimate the cross-sectional direct health costs for 11 tobacco-related diseases, morbidity, and the present value of mortality from smoking in Indonesia for 2005. The study reported that the indirect costs from tobacco-attributable morbidity and mortality in Indonesia amounted to about US$ 1.9 billion and US$ 4.9 billion, respectively, for a total indirect cost of US$ 6.8 billion. A decade later, in 2015, the indirect morbidity, disability, and premature mortality costs associated with tobacco in Indonesia had reached an estimated US$ 28.78 billion (IDR 374,06 trillion), or 3.3 percent of Indonesia’s GDP (Kosen et al., 2017). Cumulative health and economic losses: the coming “tsunami.” If the tobacco use pattern now observed in Indonesia is not controlled, the country risks what some experts term a future “public health and fiscal tsunami,” due to increasing rates of tobacco-related 3 Cigarette Affordability in Indonesia: 2002-2017 noncommunicable diseases (NCDs) that are costly to treat and that will place growing strain on government and household budgets (World Bank, 2016). This in turn will under- mine human capital development and the total wealth of the country. Decisive policy action is needed to avert this threat and drive rapid reductions in smoking rates. Cigarette Production, Consumption, and Market Share in Indonesia Major cigarette types. Three types of cigarettes are produced in Indonesia: machine- made kretek (Sigaret Kretek Mesin, SKM), handmade kretek (Sigaret Kretek Tangan, SKT), and white cigarettes (Sigaret Putih Mesin, SPM). Figure 2. Market Share by Type of Cigarette, Indonesia (2011-2017) 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2011 2012 2013 2014 2015 2016 2017 White Cigarette (SPM) 5.87% 6.08% 5.99% 5.71% 5.77% 5.47% 4.90% Hand Made Kretek (SKT) 30.37% 28.63% 25.43% 21.67% 20.88% 20.72% 20.23% Machine Made Kretek (SKM) 63.75% 65.29% 68.58% 72.62% 73.35% 73.82% 74.79% Source: Ministry of Finance, Republic of Indonesia Shifting market shares. Kretek is a domestically produced cigarette type which consists of tobacco, cloves, and other ingredients. The period 2011-2017 saw a shift in market share towards SKM. Market share for SKM rose from 63.75 percent in 2011 to 74.79 percent in 2017. At the same time, market shares for the SKT and SPM formats were decreasing. Specifically, SKT market share decreased from 30.37 percent in 2011 to 20.23 percent in 2017, while SPM (mostly international brands) also lost ground slightly in the same period, falling from 5.87 percent to 4.98 percent (Figure 2). Dominant players. For machine-manufactured kreteks (SKM), the market is dominated (to 63 percent) by cigarette companies classed as Category I that produce over 3 billion sticks each year. The markets for hand-rolled kretek cigarettes (SKT) and machine-made white cigarettes are dominated by the same Category I companies. 4 // Introduction A complex tobacco tax structure. Within each of the three broad types of cigarettes, multiple tiers are defined for tax purposes, based on type, number of cigarettes produced, and per-unit maximum retail price (Table 1 and Figure 3). In 2017, for example, the tobacco tax structure included 12 such tiers, though these will be reduced to 10 in 2018. Table 1. Cigarette Excise Tax Structure in Indonesia, 2011-2017 Volume of produc- Range of retail price (Rp/stick) Tariff (Rp/Stick) tion Type Group (Billions 2013 2013 2011 2012 - 2015 2016 2017 2011 2012 - 2015 2016 2017 of sticks) 2014 2014 >= 2 660 660 669 325 355 375 630- 630- 315 345 I 660 660 631- 800 1000 1120 415 480 530 355 669 SKM 600- 600- 295 325 (machine- 630 630 made <2 430 430 549 588 740 820 245 270 285 305 340 365 kreteks) 380- 374- 440- 511- 590- 655 210 235 245 265 300 335 II 430 430 549 588 740 374- 170 380 >= 2 600 375 680 820 930 1030 325 365 380 425 495 555 450- 295 I 600 SPM 375- 245 (machine- 450 made <2 300 300 444 520 800 900 215 235 245 270 305 330 white) 254- 254- 585 175 190 290 II 300 300 345- 425- 505- 195 220 255 217- 217- 444 520 800 110 125 254 254 >= 2 590 590 749 825 1115 1215 235 255 275 290 320 345 550- 520- 550- 606- 775- 860 180 195 205 220 245 265 I 590 590 749 825 1115 520- SKT 155 550 (hand- 0.3 - 2 379 379 379 417 605 730 110 125 130 140 155 165 made kreteks) 349- 349- 349- 470 100 115 120 155 II 379 379 379 385- 403- 125 140 336- 336- 417 605 336- 90 105 110 349 349 349 < 0.3 234 234 250 286 400 465 65 75 80 85 90 100 III 370 400 80 80 Note: In 2017, for SKM and SPM, Production group 1 = output > 3 billion; Production group 2 = output < 3 billion. For SKT, Production group 1 = output > 2 billion; Production group 2= output 500m - 2 billion; Production group 3A: output > 10m and < 500m; Production group 3B < 10m Source: Ministry of Finance, Republic of Indonesia 5 Cigarette Affordability in Indonesia: 2002-2017 Figure 3. Evolution of Cigarette Market Share by Tax Tiers, Indonesia (2010-2017) 100.00 90.00 80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00 2010 2011 2012 2013 2014 2015 2016 2017 SKM1.1 SKM1.2 SKM1.3 SKM2.1 SKM2.2 SKM2.3 SKT1.1 SKT1.2 SKT1.3 SKT2.1 SKT2.2 SKT2.3 SKT3.1 SKT1.2 SPM1.1 SPM1.2 SPM1.3 SPM2.1 SPM2.2 SPM2.3 Source: Ministry of Finance, Republic of Indonesia Production and consumption trends. Both cigarette production and consumption rose dramatically in Indonesia between 2011 and 2016. Cigarette production climbed from 279 to 342 billion sticks, while consumption increased from 296 to 339 billion sticks over the period. Both measures saw a slight downturn in 2016 (Figure 4). Figure 4. Cigarette Production and Consumption, Indonesia (2011-2016) 360 360 340 340 320 320 300 300 Billion Sticks Billion Sticks 280 280 260 260 240 240 220 220 200 200 Production 279.4 301.0 345.0 344.0 348.0 342.0 Consumption 296.5 320.4 327.7 335.0 342.5 339.4 Figure 4-1: Cigarette production 2011-2016, Figure 4-2: Cigarette consumption 2011-2016, sourced from Euromonitor sourced from Euromonitor Source: Euromonitor International 2017. 6 // Introduction Varied results by cigarette type. In percentage terms, Indonesia’s overall cigarette production increased by 22.41 percent from 2011 to 2016, while consumption increased by 14.46 percent. Behind these aggregate figures, cigarette types fared differently. Production of machine-made kretek cigarettes (SKM) increased by nearly 25 percent, while handmade kretek cigarette (SKT) production decreased by 27 percent. White cigarettes (SPM) maintained a basically constant production level over the period (Figure 5). The market share change measured by production reflects Indonesian consumers’ apparent increasing preference for machine-made kretek over other types of cigarettes. Figure 5. Percentage Change in Cigarette Production by Type of Cigarette, Indonesia (2011-2016) 30.00% 20.00% 10.00% 0% -10.00% -20.00% -30.00% Machine Made Hand Made Kretek White Cigarette Kretek (SKM) (SKT) (SPM) 2011-2016 24.50% -26.65% 0.04% Source: Ministry of Finance, Republic of Indonesia Tackling Affordability: A Critical Task for Tobacco Control in Indonesia The power of tobacco taxes. Evidence from many countries points to tobacco taxation as the single most effective tool to reduce tobacco consumption and health care costs, while providing a potential revenue stream to finance country-level development. Higher tobacco taxes help raise cigarette prices, which can contribute to reducing the prevalence and intensity of smoking significantly, despite the addictive nature of tobacco. Dedicated tobacco taxes can also provide sustainable funding for social insurance programs, health promotion, and other programs targeting the Sustainable Development Goals (SDGs). An opportunity to be seized. Indonesia, however, has not yet fully harnessed this poten- tial. Price and tax incidence in the country has generally been low in absolute and relative terms. In 2016, total taxes (i.e., specific excise tax, local tax, and VAT) as a percentage of the retail price of Indonesia’s most-sold cigarette brand stood at 57.4 percent, far below the WHO-recommended standard of 75 percent (WHO, 2017). 7 Cigarette Affordability in Indonesia: 2002-2017 Beginning to turn the tide? Indonesia’s government raised its cigarette excise tax six times during the seven-year period 2011-2017, with hikes in every year but 2014. These regular excise tax hikes pushed up both nominal and real retail cigarette prices. Indonesia’s cigarette sales volume, however, continued to show increasing trends from 2011 to 2015, before dropping slightly in 2015-2016. As shown in Figure 6, cigarette consumption kept growing until 2016, when it also declined (slightly) for the first time in the period. The 2016 consumption drop appears due mainly to Indonesia’s unfavorable economic situation at the time, marked by a significant increase in the cost of basic necessities and reduced household purchasing power (Euromonitor International, 2017). Figure 6. Smoking Prevalence, Cigarette Price, and Cigarette Consumption, Indonesia (2002-2016) 350 40% 300 35% 30% 250 25% 200 20% 150 15% 100 10% 50 5% 0 0% 2010 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2004 2007 2011 2013 2016 Cigarette consumption (billion sticks) Smoking prevalence (% of adult population) 21000 9000 18000 8000 7000 15000 6000 12000 5000 9000 4000 3000 6000 2000 3000 1000 0 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Nominal retail price (IDR/20 sticks) Real retail price (IDR/20 sticks) Source: Cigarette consumption data combine registered retail sales and illicit trade volume. Data on both were sourced from Euromonitor International, along with data on smoking prevalence and retail price.1 1 Euromonitor data often draw on industry sources. Depending on the specific case, this could raise questions of reliability. However, there is no comparable substitute for the breadth and frequency of Euromonitor data. 8 // Introduction The real challenge: affordability. While tobacco product price conveys how expensive the product is, it does not measure individuals’ ability to purchase tobacco products. Over the past decade a clear dichotomy has emerged between developed and developing countries. Cigarettes are becoming less affordable in developed countries and much more affordable in many developing countries. This increase in the affordability of tobacco hinders positive health impacts from non-tax tobacco control measures. To strengthen tobacco control results, it is vital to monitor and reduce tobacco affordability. Understanding cigarette affordability in Indonesia: key for tobacco control policy making. An earlier study found that cigarettes became more affordable in Indonesia between 2000 and 2010 (Ahsan et al., 2011). This explains the rapid growth of tobacco production, consumption, and smoking prevalence in the past decade. In recent years, Indonesia’s government increased the tobacco excise tax on a regular basis. This paper investigates the impact that these tax increases have had on cigarette prices and con- sumption, along with the increases' potential public health consequences. To make tax and price policy an effective weapon against tobacco use in Indonesia, it is necessary to examine the current trend of cigarette affordability, especially before and after the tax increases implemented during the past few years. To understand changing patterns of cigarette affordability, this study examines tobacco taxes and cigarette prices vis-à-vis Indonesia’s economic growth and consumers’ increased purchasing power. The findings will provide supportive evidence to Indonesian policy makers for a shift from price-based tobacco policy solutions to affordability-based solu- tions, with the aim of making cigarettes less affordable, as well as optimizing the country’s tobacco excise tax structure. 9 Cigarette Affordability in Indonesia: 2002-2017 2. LITERATURE REVIEW Affordability refers to an individual’s purchasing power with regard to a product. Over the past decades, various methodologies have been developed defining cigarette affordability as a function of cigarette price and individuals’ income levels, with reference to the quantity or share of resources required to buy a pack of cigarettes. Guindon et al. (2002) defined affordability by Minutes of Labor (MoL), meaning how many working minutes are needed to buy a pack of locally produced or Marlboro (or equivalent) cigarettes. Guindon’s study showed that an Indonesian needed to work 62 minutes to pur- chase a pack of Marlboros or local-brand cigarettes in 2000, suggesting that cigarettes were not very affordable in Indonesia at that time, relative to the 56 countries in the study. In subsequent research examining cigarette affordability in Southeast Asia, Guindon et al. (2003) applied a different methodology by dividing relative prices of tobacco products by a country’s per capita gross domestic product (GDP). Most importantly, they looked at changes in affordability over time, rather than just across countries. The study found that tobacco products in Indonesia became 50 percent more affordable between 1980 and 1998. A similar trend was seen in Sri Lanka and India. Blecher et al. (2004) established the Relative Income Price (RIP), which has become the most widely used measurement indicator of cigarette affordability. RIP calculates the percentage of GDP per capita required to buy 100 packs of cigarettes. In this study of 70 countries, Indo- nesia was ranked as the third most affordable in the low-income country (LIC) group during the period 1999-2001. Kan (2007) developed an alternative instrument, the cigarette price/daily income ratio (CPDIR), to evaluate cigarette affordability by calculating the percentage of daily income needed to buy a pack of lowest-priced cigarettes. The CPDIR in Jakarta, one of the cit- ies in Kan’s study, was 0.14. This means that a pack of cigarettes cost 14 percent of the daily income in Jakarta in 2006, a medium affordability level compared with the 60 cities included in Kan’s study. Barber et al. (2008) extended Guindon’s (2003) analysis to the period from 2001 to 2005. In contrast to the rising affordability trend in the 1980s and 1990s, Barber’s study found that Indonesia’s affordability index remained nearly unchanged from 2001 to 2005. This suggests that tobacco tax increases during this period were sufficient to keep affordability from further increasing, but not to reduce it. Rumbogo and Ahsan (2011) examined cigarette affordability in Indonesia using two methods. Using the RIP method, they found that cigarettes became 50 percent more 10 // Literature Review affordable in Indonesia from 2003 to 2010. As an alternative metric, the researchers also used the percentage of Indonesia’s national average nominal daily wage needed to buy a pack of cigarettes. This approach showed that the nominal price of cigarettes represented only a small portion of an average Indonesian’s daily wages, and it confirmed that cigarettes became increasingly affordable between 2003 and 2010 (Rumbogo and Ahsan, 2011). WHO (2015) compared changing rates of cigarette affordability (defined by RIP) between 2008 and 2014 for 15 countries. Indonesia was one of the four countries that experienced an increase in cigarette affordability over the observed period. A recent study by the U.S. National Cancer Institute (NCI) and WHO (2016) found that Indonesia’s rate of cigarette affordability increase was the second fastest among ten lower middle-income countries studied, from 2000-2013. The examination of affordability change rate explains why, although cigarettes remain less affordable in LMICs than in high-income countries (HICs) overall, changes in affordability over time have led to a decrease in consumption of ciga- rettes in HICs, but an increase in LMICs. The above studies examined tobacco/cigarette affordability in Indonesia in different peri- ods and applied different methods. The diverse studies reached a common conclusion: that tobacco/cigarette affordability has been increasing in Indonesia during the last three decades. Specifically: 1. Tobacco affordability rose by 50 percent from the early 1980s to the end of the 1990s; 2. Affordability increased again by 50 percent from 2003 to 2010, thus reaching more than double its level in the early 1980s; 3. In the 2000s, cigarettes in Indonesia have been more affordable than in the median LMIC. However, as in LMICs in general, cigarettes have remained less affordable in Indonesia than in HICs; 4. The rate of increase of cigarette affordability in Indonesia has been very rapid, ranking second-fastest among ten lower middle-income countries from 2000-2013. Based on the extensive literature on the relationship between consumption and tobacco affordability, these findings could explain the steady increase of cigarette consumption in Indonesia in the last two decades. We conducted the present study to further investigate cigarette affordability in Indonesia with the following aims: (i) to examine cigarette affordability in the short (2011-2017) and longer (2002-2016) terms; (ii) to compare changes in cigarette affordability in Indonesia with those in other countries; (iii) to assess how effectively tobacco tax hikes in the past decade have reduced affordability; and (iv) to explore tobacco tax policy implications pertaining to tobacco control. 11 Cigarette Affordability in Indonesia: 2002-2017 3. METHODOLOGY Cigarette Affordability Definition A two-pronged approach. In this study, we use two complementary methods of defin- ing cigarette affordability. One is the RIP method. The other is the Cigarette Affordability Index (CAI) method. The strengths of RIP. Among the methods for measuring cigarette affordability applied in previous studies, the RIP method has several important advantages and has been widely adopted. First, GDP per capita is a good indicator of living standard and income. Second, this data is commonly available and therefore makes it easier to calculate cigarette affordability, compared with other approaches (Blecher et al., 2004). Third, every country calculates GDP per capita annually using a consistent methodology, thus enabling global comparisons of tobacco affordability. The World Health Organization (WHO) has adopted the RIP method, calculating cigarette affordability for all countries with available data. This makes it possible for our study to identify the current level of cigarette affordability in Indonesia compared with other countries. The RIP method defines cigarette affordability as the percentage of GDP per capita required to buy 100 packs of cigarettes. Equation (1) provides the formula for calculating the RIP. The higher the RIP, the less affordable cigarettes are. where RIP represents the relative income price of cigarettes, and P is the retail price of a pack of cigarettes with 20 individual pieces.2 A helpful complement: CAI. The Cigarette Affordability Index (CAI) method measures the magnitude of cigarette-affordability change compared with the base year. Equation (2) demonstrates the rationale of the CAI method. The CAI has been created as a useful way to present changes over time using the RIP. Since it is a measure of affordability, it has the desirable characteristic, as compared to the RIP, that a higher value means more rather than less affordability. This in turn means that the higher the CAI, the higher the expected smoking prevalence and cigarette consumption. 2 The RIP can be measured in nominal terms for both the cigarette price and the per capita GDP. This removes “translation issues” when working in constant prices or in a foreign currency, e.g. USD. 12 // Methodology Price Data For our affordability calculations, we adopt two sets of price data obtained from different sources. We term the first average price (Euromonitor). Sourced from Euromonitor International, it calculates per unit average price by dividing sales value over sales volume for each year between 2002 and 2016. While Euromonitor’s data are valuable and widely used, the concern exists that Euromonitor’s reliance on tobacco industry intelligence and an opaque modelling process may lead to biased estimates. This may especially be the case, if information provided by industry sources is influenced by firms’ common narrative that increases in tobacco excise taxes cause increases in illicit trade (Blecher et al. 2010; Skafida et al., 2014). To protect against potential distortions in the Euromonitor data, we calculate a second value termed average price (MoF). This is calculated from the transac- tion price of each cigarette type (SKT/SKM/SPM), weighted by the respective market share for each year from 2011 to 2017. Both the transaction price of each cigarette type and the respective market shares are provided by the Ministry of Finance of Indonesia. Transaction price data comes from a regular market survey conducted by the Ministry. All prices in this study are recomputed for a pack of cigarettes containing 20 individual pieces. To define real prices, the nominal retail price is adjusted by the consumer price index (CPI), sourced from the World Bank Database. Income Data The GDP per capita of Indonesia for the period 2002- 2016 is sourced from World Bank national accounts data.3 The GDP per capita of Indonesia for 2017 is sourced from the Central Statistics Board (BPS) of Indonesia.4 We use the nominal value here to calculate the RIP, combining it with the average nominal retail price. Based on the price and income data described, we have conducted two sets of analyses: (i) we examine average cigarette affordability using GDP per capita and the average price (Euromonitor) over the 2002-2016 period; (ii) we examine average cigarette affordability using per capital GDP and the average price (MoF) over the 2011-2017 period. 3 https://data.worldbank.org/indicator/NY.GDP.PCAP.CN?locations=ID&view=chart 4 https://www.bps.go.id/pressrelease/2018/02/05/1519/ekonomi-indonesia-triwulan-iv-2017--tumbuh-5-19-persen.html 13 The Economics of Tobacco Taxation and Employment in Indonesia 4. RESULTS Cigarette Affordability: Level, Trend, and Magnitude of Change Using the GDP per capita and average cigarette price, we examined cigarette affordabil- ity on average in Indonesia with the following aims: (i) to estimate the levels of cigarette affordability in each year; (ii) to present the trends of cigarette affordability through the observed period; (iii) to calculate the magnitudes of cigarette affordability change; and (iv) to examine the annual growth rates, as well as the fixed-base growth rates of cigarette affordability over the observed period. Table 2 lists the GDP per capita, two sets of average price data, and cigarette affordability measured by RIP and CAI respectively. Table 2. Cigarette Affordability: RIP Average and CAI Average Average affordability (Euromonitor) (1) Average affordability (MoF) (2) GDP per Year Average Price RIP Average CAI Average Average Price RIP Average CAI Average capita (Euromonitor) (Euromonitor) (Euromonitor) (MoF) (MoF) (MoF) Unit Million IDR IDR/20 sticks % IDR/20 sticks % 2002 8.92 5381.3 6.03% 1.00 2003 9.72 6012.3 6.18% 0.98 2004 10.93 6430.8 5.88% 1.03 2005 13.03 8222.3 6.31% 0.96 2006 15.47 8683.3 5.61% 1.07 Data not available 2007 18.06 9110.2 5.05% 1.20 2008 22.31 9316.2 4.18% 1.45 2009 24.94 9800.8 3.93% 1.54 2010 28.30 10413.5 3.68% 1.64 2011 31.87 11578.5 3.63% 1.66 13462.3 4.22% 1.00 2012 34.62 12421.2 3.59% 1.68 13924.6 4.02% 1.05 2013 37.88 13789.0 3.64% 1.66 15412.3 4.07% 1.04 2014 41.43 15446.4 3.73% 1.62 17745.3 4.28% 0.99 2015 44.67 17291.5 3.87% 1.56 19642.2 4.40% 0.96 2016 47.51 19116.3 4.02% 1.50 22302.2 4.69% 0.90 2017 51.89 Data not available 24418.5 4.71% 0.85 Data source: GDP per capita (2002-2016) is sourced from World Bank national accounts data. GDP per capita (2017) is sourced from the Central Statistics Board (BPS) Indonesia. Average price (Euromonitor) data are sourced from Euromonitor International. Average price (MoF) data are calculated by the authors from the average transaction price of each cigarette type (SKT/SKM/SPM) weighted by the market share. Data for the latter calculation are obtained from the Indonesian Ministry of Finance. Three phases of affordability. Figure 7 presents the affordability results using the Euromonitor data, which cover the years 2002-2016. The average cigarette affordability presents three stages during the observed period: a fluctuating stage between 2002 and 2005, a stage of increasing affordability between 2005 and 2012, and a stage of decreasing affordability from 2012 to 2016. During the whole period, cigarettes were most affordable in 2012 and least affordable in 2005. In 2016, cigarettes were 1.5 times more affordable than in 2002. The 2016 RIP (4.02 percent) remains well below the value in 2005, the year of lowest affordability, when the RIP stood at 6.31 percent. Figure 7. Cigarette Affordability, Indonesia (2002-2016) 1.8 7% 6.18% 6.31% 6.03% 5.88% 1.6 1.66 1.68 1.66 5.61% 1.64 1.62 6% 1.54 1.56 1.4 5.05% 1.5 1.45 CAI average (Euromonitor) RIP average (Euromonitor) 5% 1.2 4.18% 4.02% 1.2 3.93% 3.68% 3.63% 3.87% 3.59% 3.64% 3.73% 1 4% 1.07 1 0.98 1.03 0.96 0.8 3% 0.6 2% 0.4 1% 0.2 0 0% 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 CAI average (Euromonitor) RIP average (Euromonitor) 2016 Data source: RIP and CAI are calculated by the authors based on the data in Table 2. Relating affordability and consumption. Figure 8 describes the correlation between cigarette consumption and affordability. From 2002 to 2012, cigarette consumption steadily increased, in association with an increase in affordability. In the period 2012-2016, cigarette affordability presents a descending trend reflected in both the Euromonitor and MoF data. However, it is notable that the decreased affordability over this period did not lead to reduced consumption. The volume of cigarette sales continued to increase until 2016, when it declined—slightly—for the first time in the observational period 15 Cigarette Affordability in Indonesia: 2002-2017 Figure 8. Cigarette Retail Sales, Illicit Trade, and CAI Average 1.80 350.00 1.60 300.00 1.40 250.00 1.20 1.00 200.00 0.80 150.00 0.60 100.00 0.40 0.20 50.00 0.00 0.00 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Registered retailing sales (billion sticks) CAI average (Euromonitor) Illicit trade (billion sticks) CAI average (MoF) Data source: Both the registered retail sales and illicit trade are sourced from Euromonitor International. Illicit trade figures are difficult to estimate and should be treated with caution. The CAI average (Euromonitor) and CAI average (MoF) are calculated by the authors based on the data listed in Table 2. Figure 9 compares the RIP measured by average price (Euromonitor) and average price (MoF) between 2011 and 2017. Both present similar trends, indicating that cigarettes became less affordable over this period. RIP average (MoF) has a higher level than RIP average (Euromon- itor) in each year during this period because of a higher average retail price (MoF) compared with the average retail price (Euromonitor). The trends of a modest increase in the RIP – i.e., a modest decrease in affordability – are consistent between the two. Figure 9. Cigarette Affordability: RIP Average (Euromonitor) vs. RIP Average (MoF) (2011-2017) 7% 6% 5% 4.69% 4.71% 4.22% 4.28% 4.40% 4.02% 4.07% 4% 4.02% 3.63% 3.64% 3.73% 3.87% 3.59% 3% 2% 1% 0% 2011 2012 2013 2014 2015 2016 2017 RIP average (Euromonitor) RIP average (MoF) Data source: Calculated by the authors based on data in Table 2. 16 // Results Price changes vs. income changes. Indonesia’s government has raised the country’s cigarette excise on an annual basis since 2011. This has driven the average cigarette price steadily upward. Cigarette affordability since 2012 has decreased accordingly (Figure 10-1). The trend in cigarette affordability depends on the relative magnitude of the income change and the price change. As illustrated in Figure 10-1 and Figure 10-2, cigarette affordability increased in 2012, when the annual growth rate of nominal GDP per capita exceeded the annual growth rate of the nominal average cigarette price (MoF). On the other hand, when the annual growth rate of nominal average price (MoF) outpaces the annual growth rate of nominal GDP per capita, cigarette affordability decreases, as hap- pened in 2012-2017. Figure 10. Correlation Between Cigarette Affordability, Price Growth, and Growth in GDP Per Capita, Indonesia (2011-2017) CAI average (MoF) Average price (MoF) Annual growth rate of GDP per capita Annual growth rate of average price (MoF) 1.50 30000 20% Average price (MoF) CAI average (MoF) (IDR/20 sticks) 1.00 20000 10% 0.50 10000 0 0 0 2011- 2015- 2016- 2014- 2011 2012 2013 2014 2015 2016 2017 2012 2012- 2013 2013- 2014 2015 2016 2017 Figure 10-1: Cigarette a ordability and average Figure 10-2: Annual growth rate of income price (MoF) 2011-2017 and average price (MoF) 2011-2017 Data source: Calculated by the authors based on data listed in Table 2. Figure 11 illustrates the annual growth rate and fixed-base growth rates (with 2011 as the base year) of cigarette affordability. On average, cigarette affordability decreased by 10.2 percent between 2011 and 2017. 2012 was the only fiscal year when cigarettes became more affordable than in the previous year. Cigarette affordability decreased most sharply (by 6.3 percent) in 2016. 17 Cigarette Affordability in Indonesia: 2002-2017 Figure 11. Annual and Fixed-Base Growth Rate of Cigarette Affordability Using Average Price (MoF) 2011-2012 5.0% 2012-2013 -1.1% 2013-2014 -5.0% 2014-2015 -2.6% 2015-2016 -6.3% 2016-2017 -0.3% 2011-2017 -10.2% -15.0% -10.0% -5.0% 0.0% 5.0% 10.0% 15.0% Annual growth rate Fixed-base growth rate Data source: Calculated by the authors based on data in Table 2. Indonesia’s Position in Global Rankings of Cigarette Affordability To assess the importance of changes in cigarette affordability in Indonesia, it is helpful to compare Indonesia’s affordability levels and trends to those reported in other countries, particularly countries that can be seen as Indonesia’s economic peers. 1999-2001: low relative affordability at baseline. Blecher’s 2004 study ranks cigarette affordability, measured by RIP (the percentage of per capita GDP required to buy 100 packs of cigarettes), in 70 countries between 1999 and 2001. Indonesia appeared near the bottom of the affordability ranking, 59th of the 70 countries examined, indicating that, at that time, cigarettes were not very affordable in Indonesia compared to other countries (Blecher et al., 2004). Extending the analysis. Based on WHO’s published RIP data for the most-sold cigarette brand (WHO, 2015), we picked 70 countries to rank for cigarette affordability in 2008 and 2016 respectively. (Sixty-six of these countries overlapped with Blecher’s original study.) This allows us to see how Indonesia’s cigarette affordability has evolved over time, compared to patterns in other countries. 2008-2016: Moderate affordability, relative to peer countries. Our analysis shows that in 2008 and 2016, as expected, Indonesia’s RIP remained high (and thus its tobacco afford- ability remained low) relative to high-income countries (Figure 12). Indonesia’s affordability level was moderate compared to those of other lower middle-income countries. Specifically, in 2016, Indonesia’s RIP was the 8th highest among the 17 lower middle-income countries included in the study, placing it solidly in the middle of this group (Figure 12). 18 // Results Very low retail prices for some types of cigarettes. WHO’s global ranking of cigarette affordability is based on each country’s most-sold brand. This approach works well for countries where a limited variety of cigarette brands are available. It is less effective for countries with a large variety of brands whose retail prices are separated by wide gaps, as is the case in Indonesia. In such countries, the most-sold-brand approach may mask the collective importance of a large set of low-end brands. For example, WHO reports that the most sold cigarette brand in Indonesia in 2016 was Gudang Garam, priced at IDR 21,666.67 per 20 sticks. Indonesia’s cheapest brand, meanwhile, sold for IDR 5,833.33 per 20 sticks, one-quarter of the Gudang Garam price. While Indonesia’s overall affordability falls in a medium range for economic peer coun- tries, some of its cigarettes are retailed at prices that are very low in international perspec- tive, including compared to other lower middle-income countries (Table 3). 19 Cigarette Affordability in Indonesia: 2002-2017 Figure 12. Cost of 100 Packs of the Most-Sold Cigarettes as Percentage of GDP Per Capita, 2016. Zimbabwe (LI) Sri Lanka (LMI) Venezuela (UMI) India (LMI) Senegal (LI) Ecuador (UMI) Bangladesh (LMI) Kenya (LMI) Morocco (LMI) Cameroon (LMI) Guatemala (LMI) Peru (UMI) Indonesia (LMI) Pakistan (LMI) Malaysia (UMI) Mongolia (LMI) South Africa (UMI) Turkey (UMI) Romania (UMI) Thailand (UMI) Viet Nam (LMI) Nigeria (LMI) New Zealand (HI) Jordan (LMI) Egypt (LMI) Hungary (HI) Tunisia (LMI) Panama (UMI) Poland (HI) Australia (HI) Mexico (UMI) Philippines (LMI) United Kingdom (HI) Portugal (HI) Costa Rica (UMI) Ukraine (LMI) Israel (HI) Greece (HI) Uruguay (HI) Chile (HI) Gabon (UMI) Argentina (UMI) Brazil (UMI) France (HI) Spain (HI) Ireland (HI) Italy (HI) Czech Republic (HI) China (UMI) Norway (HI) Canada (HI) Singapore (HI) Russian Federation (UMI) Belgium (HI) Bahrain (HI) Netherlands (HI) Saudi Arabia (HI) Germany (HI) Finland (HI) Colombia (UMI) Sweden (HI) Austria (HI) Denmark (HI) United States of America (HI) Iran (UMI) Switzerland (HI) Japan (HI) Kuwait (HI) Belarus (UMI) RIP of most sold brand Paraguay (UMI) 0% 5% 10% 15% 20% Data source: WHO Report on the Global Tobacco Epidemic, 2017. Country income group classification is based on World Bank Analytical Classifications for 2016 (World Bank 2016).5 5 World Bank Analytical Classifications for 2016. https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country- and-lending-groups. 20 // Results Table 3. Retail Price for a Pack of 20 Cigarettes - Premium Brand and Cheapest Brand – Indonesia and Selected Countries, 2016, in US$ Country Price of Premium Brand Price of Cheapest Brand Indonesia $2.16 $0.45 Bangladesh $2.81 $0.64 Brazil $2.28 $1.53 Canada $8.56 $6.55 China $6.76 $0.38 Germany $6.67 $5.56 India $3.67 $0.87 Malaysia $4.19 $2.22 Mexico $2.86 $1.27 Philippines $1.26 $0.86 Singapore $9.65 $7.20 Sri Lanka $7.27 $3.84 Rep. of Korea $4.46 $3.57 Russian Federation $1.86 $0.72 Thailand $2.06 $1.15 United Kingdom $13.31 $8.17 Vietnam $1.17 $0.27 Source: WHO Report on the Global Tobacco Epidemic, 2017 (Appendix IX: Tobacco taxes, prices and affordability). 21 Cigarette Affordability in Indonesia: 2002-2017 5. DISCUSSION: CLARIFYING RESULTS, LIMITATIONS, AND POLICY IMPLICATIONS Cross-test analysis with household expenditure data validates the reliability of the cigarette affordability trend measured by GDP per capita. We have conducted additional statistical tests to confirm the soundness of the results presented above. One potential concern relates to the use of GDP per capita as the proxy for income in RIP meth- odology. While this approach is widely practiced and accepted, it is recognized that GDP per capita is an imperfect marker of income, particularly in the context of high levels of inequality. To address the concern, we conducted a cross-test analysis using self-reported per capita expenditure data as the income proxy. Expenditure data was processed and provided by the World Bank, based on the National Socio-Economic Survey (SUSENAS). Expenditure is sometimes a preferable measure of income, as it captures the disposable income in families. Expenditure may be an especially useful marker when a country’s informal labor market is large, as in Indonesia. Of individuals employed in Indonesia in August 2017, 39.7 percent held wage jobs (excluding casual work), leaving 60.3 percent of workers in non-wage jobs. Under the definition of formality used for official purposes in Indonesia, the formal sector comprises the wage-employed and self-employed individuals who employ perma- nent workers. In August 2017, 43.0 percent of Indonesian workers were formal under this definition, while 57.0 percent were informal.6 Table 4. Cross-Test RIP: Expenditure vs. GDP Per Capita RIP average RIP average (Euromonitor)_expenditure (Euromonitor) 2011 18.16% 3.63% 2012 18.22% 3.59% 2013 18.20% 3.64% 2014 18.42% 3.73% 2015 18.69% 3.87% 2016 18.77% 4.02% Note: RIP_ average (Euromonitor)_expenditure is calculated based on the average price sourced from Euromonitor and expenditure per capita sourced from the World Bank Group. RIP average_(Euromonitor) is calculated using the average price sourced from Euromonitor and GDP per capita sourced from the World Bank Group. 6 Data provided by World Bank Indonesia staff. 22 // Discussion: Clarifying Results, Limitations, and Policy Implications Figure 13. Cross-Test RIP: Expenditure vs. GDP Per Capita 20.00% 18.00% 18.77% 18.20% 18.42% 18.69% 18.16% 18.22% 16.00% 14.00% 12.00% 10.00% 8.00% 6.00% 4.00% 3.73% 3.87% 4.02% 3.63% 3.59% 3.64% 2.00% 0.00% 2011 2012 2013 2014 2015 2016 RIP average (Euromonitor) RIP average (Euromonitor)_Expenditure Data source: Based on data in Table 4. As shown in Figure 13, over the period 2011-2016, the affordability trend illustrated by expenditure data is generally similar to the affordability trend illustrated by GDP per capita, though slight differences are seen. The value of RIP measured by expenditure, however, is almost six times larger than RIP measured by GDP per capita (Table 4). This reflects the fact that expenditure per capita is lower than GDP per capita. Importantly, this relationship also suggests that the burden of tobacco expenditure on a family’s ability to purchase necessi- ties is much larger than it appears when RIP is calculated using per capita GDP. Additional smokers have contributed significantly to total aggregate cigarette consumption in Indonesia since 2011, partially off-setting the decreased afford- ability achieved by higher tobacco taxes. Over the last two decades, the Indonesian population has grown substantially, increasing the number of smokers and total cigarette consumption (Table 6 and Figure 8). Figure 14 takes account of that population growth. It shows a clear relationship between population growth and Indonesia’s cigarette afford- ability. Both were broadly rising through most of the period considered. 23 Cigarette Affordability in Indonesia: 2002-2017 Table 5. Adult Cigarette Consumption Per Capita Total Adult Population Adult Consumption Per Capita Consumption (Billion Packs) (Million) (Sticks) 2002 225.36 151.58 1487 2003 225.81 153.92 1467 2004 225.81 156.25 1445 2005 226.09 158.59 1426 2006 246.66 161.49 1527 2007 241.95 164.33 1472 2008 263.29 167.09 1576 2009 276.43 169.73 1629 2010 286.87 172.27 1665 2011 296.48 175.20 1692 2012 320.44 177.94 1801 2013 327.67 180.63 1814 2014 335.04 183.38 1827 2015 342.48 186.24 1839 2016 339.37 188.91 1796 Data source: Total cigarette consumption is sourced from Euromonitor International. The number of adults (15+) is sourced from the World Bank Group. The adult cigarette consumption per capita is calculated by the authors based on the above data. 24 // Discussion: Clarifying Results, Limitations, and Policy Implications Figure 14. Adult Cigarette Consumption Per Capita and Cigarette Affordability, Indonesia (2002-2016) 2000 1.80 1800 1.60 Per capita consumption by adult 1.68 1.66 CAI average (Euromonitor) 1.62 1.66 1600 1.56 1.64 1.40 1.54 1.50 1.45 1400 1.20 1200 1.20 1.00 1.07 1.03 1.00 1000 0.98 0.96 0.80 800 0.60 600 0.40 400 0.20 200 0.00 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Per capita consumption by adult (Sticks) CAI average (Euromonitor) Data source: Adult cigarette consumption per capita by is listed in Table 5. The CAI average (Euromonitor) is calculated by the authors based on data listed in Table 2. Overall, cigarette affordability increased by half from 2002 to 2016, and consumption per capita rose by 21 percent. However, we can delineate three distinct phases during these years: 2002-2012: Cigarette affordability and consumption evolve in step. From 2002 to 2005, both cigarette affordability and adult cigarette consumption per capita trended downward. Cigarettes were least affordable in 2005, and in that year adult cigarette consumption per capita fell to its lowest level during the whole observed period. Subsequently, from 2005 to 2012, cigarette affordability increased steadily, and adult per capita cigarette consumption likewise rose. Cigarette affordability reached its highest point in 2012. 2012-2015: An affordability-consumption disconnect, largely driven by additional smokers. From 2012 to 2015, cigarette affordability in Indonesia declined by about 7 percent, largely due to the regular tax increases that began in 2011. However, consumption per capita increased, rather than falling, during these years. Table 7 shows the growth in the number of smokers, cigarette consumption growth, and the percentage of increased consumption contributed by additional smokers over the period 2011-2016. As illustrated in Table 7 and Figure 15, around 85 percent of the increased total consumption in 2014 and 2015 was contributed by additional smokers. For 2013, the corresponding figure was 71 percent. In contrast, in 2012, an increased number of smokers had accounted for just 21 percent of the annual cigarette consumption rise. The balance was primarily contributed by higher affordability, together with non-price factors. In each year of rising aggregate cigarette consumption, the increased consumption not accounted for by additional smok- ers was due to greater affordability and other variables. These variables included changes 25 Cigarette Affordability in Indonesia: 2002-2017 in overall tobacco control programs and in group norms, as well as cumulative changes in affordability and expectations of future changes in affordability. The question persists whether reported cigarette consumption data are accurate enough to permit definitive decomposition of annual changes. A small remaining disconnect is noted between the 7.3 percent decline (measured by Euromonitor data) in affordability from 2012-2015 and the essentially flat consumption per capita -- about 1800 cigarettes per year, with a slight increase above 1830 in 2015. 2016: A turning point? One cannot place too much faith in the results of a single year, particularly because of lagged effects. However, it is encouraging that, in 2016, cigarette consumption declined in both per capita and overall terms. 2016 also saw the largest drop in cigarette affordability achieved during the period— 6.3 percent (Figure 11). Table 6. Total Number of Smokers and Consumption Per Smoker Total Consumption Total Number of Smokers Consumption Per Smoker Billion Sticks Million Sticks 2011 296 57.96 5115 2012 320 58.89 5441 2013 328 59.84 5476 2014 335 60.98 5495 2015 342 62.14 5511 2016 339 63.33 5358 Data source: Both total cigarette consumption and the number of smokers are sourced from Euromonitor International. The consump- tion per smoker is calculated by the authors based on data in the table. 26 // Discussion: Clarifying Results, Limitations, and Policy Implications Table 7. Growth in Number of Smokers and Cigarette Consumption Growth (2011-2016) (3) (1) (2) (4) = (2)/(3) ∆ Total consump- ∆ Smokers com- Cigarette consump- Consumption of ∆ tion compared with pared with previous tion of ∆ smokers smokers as % of ∆ previous year year (thousands) (million sticks) total consumption (million sticks) 2011-2012 932.40 5073 23963 21.17% 2012-2013 947.50 5189 7235 71.71% 2013-2014 1139.40 6260 7368 84.97% 2014-2015 1165.30 6422 7436 86.37% 2015-2016 1191.60 6385 -3111 -205.23% Data source: Calculated by authors based on data listed in Table 6. Figure 15. Change in Total Cigarette Consumption and Cigarette Consumption by Additional Smokers 30000 23963 25000 20000 Million sticks 15000 7436 7368 10000 6422 7235 6385 6260 5189 5073 5000 0 -3111 -5000 2011-2012 2012-2013 2013-2014 2014-2015 2015-2016 Cigarette consumption by an increased number of smokers Change of total cigarette consumption Data source: Data in Table 7. 27 Cigarette Affordability in Indonesia: 2002-2017 Indonesia’s multi-tiered tobacco tax system has weakened the positive impact of tax hikes to date. Indonesia applies an elaborate, multi-tiered tobacco excise system based on cigarette type, the number of cigarettes produced, and per-unit maximum retail price. Reforms between 2009 and 2017 reduced the number of tiers from 19 to the current 12. However, this still remains the most complex excise tax system in the world. In theory, such a system might be expected to benefit small-scale domestic cigarette firms, especially hand-rolled kretek (SKT) firms, by incentivizing consumers to choose these comparatively inexpensive products. In practice, such a benefit for small firms has not materialized. Between 2011 and 2016, market share for SKT dropped from 30.37 percent to 20.72 percent. Meanwhile, the fragmented tax structure blunts the impact of tobacco tax hikes by providing smokers with “escape routes” to shift to alternate brands when taxes and prices rise, rather than quitting altogether.7 Methodological lessons and paths for future work. By testing different approaches to affordability using different data sources, this study has generated methodological insights that may inform future work. (1) We have found significant drawbacks with the widely applied approach of using per capita GDP as a proxy for income. Particularly in a country such as Indonesia, with substantial income inequality and widespread informal employ- ment, household expenditure would be a better measure to capture actual disposable income in families. (2) Focusing on average cigarette prices or the price of a country’s most-sold brand can mask important features of affordability as it plays out in the daily lives and choices of consumers. For example, if the average cigarette price in a country is US$ 2.00, but the realistic range of prices extends from US$ 0.45 to US$ 6.00, there is clearly much scope for downtrading and substitution. This especially needs to be taken into account for countries like Indonesia, whose cigarette markets encompass a wide range of brand and price varieties. (3) The drawbacks of measuring affordability by per capita GDP and average price or most-sold brand price explain the paradox between Indonesia’s static global ranking in cigarette affordability and its notable rise in the global ranks of tobacco consumption (from the world’s fifth-largest to third-largest consumer nation) during the past decade. (4) To address the above drawbacks, household surveys compris- ing both price data and income data would be preferable. Such instruments would enable researchers to assess the affordability of different cigarette price categories consumed by different income groups. This nuanced approach would contribute important knowledge for tobacco tax policy. 7 Indonesia Tobacco Tax Policy Note: Evaluating options to reform the tobacco excise tax system, November 2014. 28 // Discussion: Clarifying Results, Limitations, and Policy Implications 29 Cigarette Affordability in Indonesia: 2002-2017 6. CONCLUSIONS Progress has recently been made to begin reducing tobacco affordability in Indonesia, but much remains to be achieved. Recent tobacco tax reforms in Indonesia have boosted retail cigarette prices. The nominal average cigarette price rose by 65 per- cent between 2011 and 2016, from IDR 11,578.5 to IDR 19,116.3 per pack. The real average cigarette price climbed by 27 percent, and cigarettes were 10 percent less affordable in 2016 than in 2011. These gains have begun to reverse the long trend of increasing cigarette affordability seen in Indonesia since the turn of the century, and indeed for the nearly four decades for which data are available. Indonesia’s tobacco epidemic continues to threaten the country’s future. Despite this progress, Indonesia’s tobacco epidemic remains one of the world’s most serious. It places the nation’s health, human capital, and economic dynamism at risk. The potential harms to children and youth, from active smoking and secondhand smoke exposures, are especially concerning. To reap the demographic dividend from its young and growing population, Indonesia must ensure that young people stay healthy. Cigarettes in Indonesia are still too cheap. Indonesian consumers can still buy ciga- rettes more cheaply than can smokers in most other middle-and high-income countries. A package of cigarettes can be bought for as little as US$ 0.45, among the lowest prices in the world (Indonesian Investments, 2018; WHO, 2017). The sale of single cigarette sticks remains a common practice in Indonesia, making cigarettes even more cheaply acces- sible. This mode of purchase may especially encourage adolescents to experiment with smoking. On the other hand, expensive brand cigarettes maintain a large market share, suggesting that many Indonesian consumers still find these products affordable. To get full benefit from affordability reductions, Indonesia’s tobacco excise system must be simplified. The country’s complex, multi-tiered tobacco tax structure might theoretically have favored small-scale domestic cigarette manufacturers. In practice, these small firms’ market share has sharply declined. The current fragmented tax model also produces deleterious public-health consequences by encouraging smokers to switch to cheaper brands when tobacco taxes rise, rather than quitting altogether. This undermines the effectiveness of using tax policy to cut tobacco consumption. Understanding this inconsistency provides an opportunity. Indonesian policy makers can resolve the conflict by simplifying the tobacco excise tax structure to reduce cigarette affordability aggressively across the board, for all cigarette types. Cigarette affordability 30 // Conclusions cuts can only achieve full impact if all tobacco products become less affordable simultane- ously and stay that way over time. A bold streamlining of the tiered tax system is crucial to achieve this goal. Tackling non-price factors. While continuing to lower affordability, tobacco control policy in Indonesia can further improve results by aggressively restricting tobacco adver- tising, enforcing smoke-free areas in public spaces, expanding the use of pictorial health warnings, and similar measures. The pervasive cultural perception of smoking as normal for adult men must and can be changed. “Go big, go fast.” International comparisons suggest that Indonesia’s current tobacco tax rates remain far below what is feasible in terms of revenue potential. Thus, tobacco tax tier consolidation and further bold tax increases could serve revenue purposes as well as fighting tobacco-related death and disease in the years ahead. A growing number of countries have achieved success with a tobacco tax strategy of “Go big, go fast”: large, rapid increases in tobacco tax rates, joined to impact-boosting measures including the swift merger of tax tiers. Together, these actions can permanently change consumers’ expectations about how much smoking costs and durably alter their behavior (Marquez and Moreno-Dodson, 2017). This is a promising path for tobacco control in Indonesia. 31 Cigarette Affordability in Indonesia: 2002-2017 REFERENCES Ahsan, Abdillah, and Titissari Rumbogo. 2011. Indonesia: Cigarette Tax and Price: Affordabil- ity and Impacts on Consumption and Revenue. Technical Report. Barber S, Adioetomo SM, Ahsan A, Setyonaluri D. 2008. Tobacco Economics in Indonesia. Paris: International Union Against Tuberculosis and Lung Disease. Blecher EH. 2010. “A mountain or a molehill: is the illicit trade in cigarettes undermining tobacco control policy in South Africa?” Trends Organ Crime 2010; 13: 299–315. Blecher, E.H., and van Walbeek, C.P. 2004. “An International Analysis of Cigarette Affordabil- ity.” Tobacco Control 13: 339–346. Campaign for Tobacco-Free Kids. 2017. The Global Cigarette Industry. https://www.tobaccofreek- ids.org/assets/global/pdfs/en/Global_Cigarette_Industry_pdf.pdf Doll R and Peto R. 1981. The causes of cancer. Oxford, England: Oxford University Press. Drope J, Schluger N, Cahn Z, Drope J, Hamill S, Islami F, Liber A, Nargis N, Stoklosa M. 2018. The Tobacco Atlas. Atlanta: American Cancer Society and Vital Strategies. Euromonitor International. 2017. Cigarettes in Indonesia. GBD 2015 Tobacco Collaborators. 2017. “Smoking prevalence and attributable disease burden in 195 countries and territories, 1990–2015: a systematic analysis from the Global Burden of Disease Study 2015.” The Lancet 389: 1885–906. Goodchild, Mark, Nigar Nargis, and Edouard Tursan d'Espaignet. 2017. “Global economic cost of smoking-attributable diseases.” Tobacco Control 0:1-7. Guindon G.E., Perucic A-M, and Boisclair D. 2003. Higher tobacco prices and taxes in South- east Asia: an effective tool to reduce tobacco use, save lives and generate revenue. World Bank Health, Nutrition and Population Discussion Paper. Economics of Tobacco Control Paper No. 11. Guindon G.E., Tobin S., and Yach D. 2002. “Trends and Affordability of Cigarette Prices: Ample Room for Tax Increases and Related Health Gains.” Tobacco Control 11: 35-43. Hidayat, Ade, and Nasruddin Djoko Surjono. 2016. “Impact of specific excise rate simplifi- cation on cigarette consumption and government revenue in Indonesia.” World Customs Journal 10:73-100. 32 // References IHME (Institute for Health Metrics and Evaluation). 2017. Indonesia Country Profile. Seattle: Institute for Health Metrics and Evaluation. Available at: http://www.healthdata.org/indonesia. IAKMI (Ikatan Ahli Kesehatan Masyarakat Indonesia/Indonesian Public Health Association). 2014. Fakta Tembakau dan Permasalahannya di Indonesia, Edisi V. Indonesia Investment. 2018. https://www.indonesia-investments.com Jha, Prabhat, and Richard Peto. 2014. “Global Effects of Smoking, of Quitting, and of Taxing Tobacco.” The New England Journal of Medicine 370:60-68. Kan, M.Y. 2007. “Investigating Cigarette Affordability in 60 Cities Using the Cigarette Price- Daily Income Ratio.” Tobacco Control 16: 429-432. Kosen S. 2009. Economic impact of tobacco use in Indonesia. Paper 593. Presented at the 14th World Conference on Tobacco or Health, Mumbai, India. Kosen S., Thabrany H., Kusumawardani N., Martini S. 2017. Health and Economic Costs of Tobacco in Indonesia: Review of Evidence Series. Jakarta. Marquez, P.V., and Moreno-Dodson, B. (eds.) 2017. Tobacco Tax Reform: At the Crossroads of Health and Development. Washington, D.C.: World Bank Group. Millennium Challenge Account - Indonesia. 2015. “Stunting and the Future of Indonesia.” Jakarta: MCA Indonesia. National Cancer Institute (NCI), in collaboration with World Health Organization (WHO). 2016. Monograph 21: The Economics of Tobacco and Tobacco Control. Bethesda, Md.: U.S. Department of Health and Human Services, National Institutes of Health. Available at: https://cancercontrol.cancer.gov/brp/tcrb/monographs/21/docs/m21_exec_sum.pdf39 Peacock, P., Leung, J., Larney, S., Colledge, S., Hickman, M., Rehm, J., Giovino, G.A., West, R., Hall, W., Griffiths, P., Ali, R., Gowing, L., Marsden, J., Ferrari, A.J., Grebely, J., Farrell, M., Degen- hardt, L. 2018. “Global statistics on alcohol, tobacco and illicit drug use: 2017 status report.” Addiction: doi: 10.1111/add.14234. Pradono K. 2002. Passive smokers, the forgotten disaster. Institute of Health Research and Development, Ministry of Health, Republic of Indonesia. Rumbogo, Titissari and Abdillah Ahsan. 2011. “Cigarette Affordability and Impact of Tobacco Taxes in Indonesia.” Bangkok: SEATCA. Semba, R.D., Kalm, L.M., de Pee, S., Ricks, M.O., Sari, M., and Bloem, M.W. 2007. “Paternal smoking is associated with increased risk of child malnutrition among poor urban families in Indonesia.” Public Health Nutrition: 10(1), 7–15. 33 Cigarette Affordability in Indonesia: 2002-2017 Skafida V, Silver KE, Rechel BP, et al. 2014. “Change in tobacco excise policy in Bulgaria: the role of tobacco industry lobbying and smuggling.” Tob Control 2014;23: e75–84. United Nations. 2015. Transforming our World: the 2030 Agenda for Sustainable Develop- ment. New York, United National General Assembly. World Bank. 2016. Indonesia Health Financing System Assessment: Spend More, Right and Better. Washington, D.C. World Bank Group. Available at: https://openknowledge.worldbank. org/ handle/10986/25363 World Bank. 2017. The Economics of Tobacco Taxation and Employment in Indonesia. World Bank, Washington, DC. World Health Organization. 2015. WHO Report on the Global Tobacco Epidemic, 2015: Rais- ing Taxes on Tobacco. Geneva: WHO. World Health Organization. 2017. WHO Report on the Global Tobacco Epidemic, 2017: Moni- toring Tobacco Use and Prevention Policies. Geneva: WHO. Xu, X., Bishop, E.E., Kennedy, S.M., Simpson, S.A., Pechacek, T.F., 2015. “Annual Healthcare Spending Attributable to Cigarette Smoking. An Update.” Am J Prev Med 48(3):326–333. 34 // References 35 Cigarette Affordability in Indonesia: 2002-2017