We selected four states in India, namely Karnataka, Assam, Uttar Pradesh, and Maharashtra, based on recommendations from the Indian Ministry of Health and Family Welfare. These four states are also characterized by diverse socioeconomic demographic characteristics, tobacco use, and health insurance coverage. We focussed on male smokers aged 15 years and older, as they constitute more than 90% of all cigarette smokers in India ⁵ . To estimate the number of smokers by age and income groups in each state, we applied the age-specific smoking prevalence for males in each state from the second round of GATS survey in India in 2016–17 to the male population in each state ⁴ . The male population in each age group for each state was estimated demographic data from the 2011 Census of India, applied to the Indian male population and projected by the United Nations Population Division for 2018 ^{18, 19} . As the GATS survey did not report prevalence based on income, we used education level as a proxy for income group, such that those with no formal education were classified as the poorest 20% of the population and those with college education or higher were classified as the richest 20% of the population.

We calculated the base price of the different lengths of filtered cigarettes sold in India (less than 65 mm, 65–70 mm, 70–75 mm and longer than 75mm) using data from Market price A.C. Nielson for 2014 ²⁰ , and inflated the price using the inflation rate in India for 2018 obtained from the World Bank Development Indicators ²¹ . We focussed only on filtered cigarettes as they constitute more than 92% of the market shares in India. Applying the Goods and Services Tax (GST), GST Compensation Cess and the National Calamity Contingency Duty (NCCD) in accordance with the tax structure for cigarettes in India in 2018-19 to the base price ²² , we obtained the 2018-19 market price of a pack of 20 cigarettes. Using the proposed tax of INR 10 (which includes GST, GST Compensation Cess, and NCCD) plus 10% ad valorem, we calculated the percentage increase in market price. We assumed that the tax increase will be passed on to consumer prices, as recent analyses in India showed that in nearly all cases, most of the tax hikes were passed onto smokers, but tax decreases did not reduce consumer prices ²³ . The market price was assumed to be uniform in all of the four states studied.

For this study, we considered a conservative scenario and used the average price elasticity for cigarette demand in both high and low- and middle-income countries of -0.40 ^{12, 24, 25} . The price elasticity in India is likely to be higher, as John (2008) reported a price elasticity of -0.34 ²⁶ . We used this elasticity in our sensitivity analysis (see Sensitivity Analysis section below). Research suggests that young people and smokers who are on low income are more price sensitive than older people and smokers who are on high income ^{24, 27} . Hence, we used two times the price elasticity (-0.80) for young smokers aged 15–24 years, as reported by the U.S. Department of Health and Human Services and International Research Agency for Cancer ^{24, 28} . For smokers in the lowest income quintile and those in the highest income quintile, we used the price elasticity of -0.64 and -0.12, as used by the Global Tobacco Economics Consortium (GTEC) ¹² . We also assumed that the price elasticity of quitting is half of the price elasticity of demand.

We used the ECEA model to estimate the impact of the cigarette price increase on quitting, number of deaths attributable to four major tobacco-attributable diseases (chronic obstructive respiratory disease (COPD), stroke, heart disease and cancer) averted, life-years gained, treatment cost averted (all treatment cost, most of which are paid out of pocket in India) due to the four tobacco-attributable diseases, number of men avoiding catastrophic health expenditures and extreme poverty, as defined by the World Bank as income of under $1.90 per day in purchasing power parity, and additional tax revenues collected ^{12, 14} . To estimate the cost savings to AB-PMJAY─ the National Health Protection Scheme in India which provides a health insurance cover of INR 5 lakh to families living below the poverty line─ we added the cost savings in AB-PMJAY due to men prevented from falling below the poverty line to the treatment cost for those living below the poverty line before the tax increase. To obtain the former cost, we applied a 50% risk of dying from tobacco-attributable diseases, the proportion of smoking-related deaths due to COPD, stroke, heart disease, and cancer obtained from the Indian Million Death Study ² , and the treatment cost of each disease to the number of men who would otherwise have continued to smoke and hence, fall below the poverty line. The data inputs and sources of data are available as Underlying data ¹⁶ . For indicators where state-level estimates could not be obtained, national estimates were used. All treatment costs obtained from the literature were converted to INR and inflated to the costs in 2018 using the exchange rate and consumer price index obtained from the World Bank Development Indicators ²¹ .

We conducted sensitivity analyses to examine the impact of a 25% and 100% price increase with the price elasticity of demand for cigarettes of -0.40, and that of INR 10 plus 10% ad valorem with the cigarette price elasticity of -0.34 in India ²⁶ . For the lowest income group, we used a price elasticity of -0.635, as done by GTEC ¹² .

All analyses were performed using Stata version 15.1 ²⁹ .

Among the four states studied, Assam has the highest male cigarette smoking prevalence of 9.7% ( Table 1), and Uttar Pradesh has the highest absolute number of cigarette smokers of about 6.5 million men. Maharashtra has the lowest smoking prevalence overall as well as for almost all age groups. In Assam and Maharashtra, cigarette smoking prevalence is higher among the higher income groups, whereas in Uttar Pradesh, the prevalence is higher among the lower income groups. In Karnataka, the prevalence was comparable across all income groups.

Before the price increase, an estimated total of about 10.7 million males older than 15 years smoked cigarettes across all of the four states studied ( Table 2). Men in the bottom income group (poorest 20% of the population) accounted for 21%, while men in the top income group (richest 20% of the population) accounted for 15% of the total number of male smokers.

In 2018-19, the average market price of filtered cigarettes for a pack of 20 is about INR 224.49. With a tax increase to INR 10 plus 10% ad valorem, the price would increase to INR 344.04. This represents a 53.25% price increase. This increase in cigarette price would lead to about 1.5 million men quitting smoking across the four states, with the bottom income group having 7.4 times as many quitters as the top income group (485,725 vs 65,762). An estimated total of 665 thousand deaths due to COPD, stroke, heart disease, and cancer would be averted among current smokers due to quitting. The number of averted deaths in the bottom income group would be 7.4 times that in the top income group (210,289 vs 28,610). The deaths averted due to quitting would yield an estimated 11.9 million life-years, with the bottom income group gaining 7.3 times more life-years than those in the top income group (3,751,930 vs 513,319).

The cost averted for treating the four major tobacco-attributable diseases would amount to more than INR 1,729 crore ($Int 955 million). The treatment cost averted in the bottom income group would be 7.4 times higher than in the top income group (INR 467 crore vs 64 crore, or $Int 258 million vs 35 million). As a result of the total treatment cost averted, about 454 thousand men would avoid catastrophic health expenditures and about 75 thousand men would avoid falling into extreme poverty. The treatment cost and impoverishment averted would save about INR 672 crore ($Int 371 million) in AB-PMJAY. The increase in excise tax would generate an additional tax revenue of about INR 4,385 crore ($Int 2,422 million). In contrast to the distribution of health benefits, the extra revenue generated from men in the top income group would be about 3.1 times that from the bottom income group (INR 1,440 crore vs 460 crore, or $Int 759 million vs 254 million).

In Karnataka, there are about 1.7 million male cigarette smokers aged 15 and above in 2016–17. Men in the bottom and the top income groups each account for about 21% of the total number of smokers. Smoking prevalence declined modestly with age in those aged 15–29 to 60–69 but increased in those aged 70 and above.

With a 53.25% increase in cigarette price through the excise tax increase, about 215 thousand men would quit smoking, with the bottom income group having five times as many quitters as the top income group (75,743 vs 15,017). Quitting as a result of the price increase would avert about 92 thousand deaths due to COPD, stroke, heart disease, and cancer among male smokers. The number of deaths averted in the bottom income group would be five times that in the top income group (32,353 vs 6,414). As a result of the deaths averted, Karnataka would gain about 1.6 million life-years and avert about INR 294 crore ($Int 162 million) in treatment cost for treating the four tobacco-attributable diseases. The averted treatment cost in the bottom income group would be 4.8 times that in the top income group (INR 84 crore vs 18 crore, $Int 47 million vs 10 million). About 72 thousand men would avoid catastrophic health expenditures, with the bottom income group avoiding 6.7 times that of the top income group (22,618 vs 3,355). The tax increase would generate about INR 1,433 crore ($Int 792 million), of which the top income group would contribute 1.7 times that from bottom income group (INR 297 crore vs 228 crore, $Int 220 million vs 126 million).

Of the four states studied, Assam has the highest male cigarette smoking prevalence of 9.7% or about 1.2 million male cigarette smokers. Cigarette smoking prevalence vary considerably across age groups, with the highest prevalence among males aged 15–29 and lowest among males aged 60–69. The prevalence increases with income.

With a 53.25% increase in cigarette price, about 161 thousand men would quit smoking, with the bottom income group having twice as many quitters as the top income group (31,866 vs 15,706). Quitting as a result of the price increase would avert about 73 thousand deaths due to COPD, stroke, heart disease, and cancer among male smokes. The number of deaths averted in the bottom income group would be twice that in the top income group (14,467 vs 7,131). As a result of the deaths averted, Assam would gain about 1.3 million life-years and avert about INR 172 crore ($Int 95 million) in treatment cost for treating the four tobacco-attributable diseases. The averted treatment cost in the bottom income group would be 2.8 times that in the top income group (INR 37 crore vs 13 crore, $Int 20 million vs 7 million). As a result of the treatment cost averted, about 46 thousand men would avoid catastrophic health expenditures, and about 2,635 men would avoid falling into extreme poverty. The treatment cost and impoverishment averted would save about INR 55 crore ($Int 30 million) in AB-PMJAY. The tax increase would generate more than INR 638 crore ($Int 352 million), of which the top income group would contribute 6.7 times that from bottom income group (INR 240 crores vs 36 crore, $Int 132 million vs 20 million).

Uttar Pradesh has an overall male cigarette smoking prevalence of 8.6%, but has the highest number of male cigarette smokers among the four study states of about 6.5 million. The prevalence is highest among men aged 30–44 and declines with age to 4.3% among those aged 70 and above. In contrast to Assam, smoking prevalence decreases with income.

With a 53.25% increase in cigarette price, more than 1 million men would quit smoking, with the bottom income group having 18.8 times as many quitters as the top income group (322,707 vs 17,150). Quitting as a result of the price increase would avert about 439 thousand deaths due to COPD, stroke, heart disease, and cancer among male smokers. The number of deaths averted in the bottom income group would be 18.8 times that in the top income group (139,821 vs 7,431). As a result of the deaths averted, Uttar Pradesh would gain about 7.8 million life-years and avert about INR 1,112 crore ($Int 614 million) in treatment cost for treating the four tobacco-attributable diseases. The averted treatment cost in the bottom income group would be 15.4 times that in the top income group (INR 289 crore vs 19 crore, $Int 159 million vs 10 million). Due to the treatment cost averted, about 299 thousand men would avoid catastrophic health expenditure, 72 thousand men would avoid falling into extreme poverty, and about INR 618 crore ($Int 341 million) would be saved in AM-PMJAY. The tax increase would generate about INR 1,465 crore ($Int 809 million), of which the top income group would contribute 5 times that from bottom income group (INR 483 crore vs 96 crore, $Int 267 million vs 53 million).

Among the four study states, Maharashtra has the lowest male cigarette smoking prevalence of 2.7%. Smoking prevalence varies across age groups with the highest prevalence among men 30–44 years (3.4%) and lowest among men aged 70 and above (0.47%). Similar to Assam and in contrast to Uttar Pradesh, the prevalence increases with income.

With a 53.25% increase in cigarette price, about 143 thousand men would quit smoking, with the bottom income group having more than three times as many quitters as the top income group (55,410 vs 17,889). Quitting as a result of the price increase would avert about 61 thousand deaths due to COPD, stroke, heart disease, and cancer among male smokers. The number of deaths averted in the bottom income group would be 3.1 times that in the top income group (23,647 vs 7,635). As a result of the deaths averted, Maharashtra would gain about 1 million life-years and avert about INR 151 crore ($Int 83 million) in treatment cost for treating the four tobacco-attributable diseases. The averted treatment cost in the bottom income group would be about 4.2 times that in the top income group (INR 57 crore vs 14 crore, $Int 32 million vs 8 million). About 37 thousand men would avoid catastrophic health expenditures, with the bottom income group avoiding more than five times that of the top income group (15,409 vs 2,775). The tax increase would generate about INR 848 crore ($Int 468 million), of which the top income group would contribute 3.2 times that from bottom income group (INR 319 crore vs 100 crore, $Int 176 million vs 55 million).

Our sensitivity analysis yielded similar results. Across all three scenarios (25% price increase, 100% price increase, and price increase to INR 10 plus 10% ad valorem with a price elasticity of -0.34), the ratio of the additional life-years gained, treatment cost averted, and number of men avoiding catastrophic health expenditure between the bottom and the top income groups is similar to that in the baseline scenario ( Figure 1a–c). Compared to the baseline scenario, with a 25% cigarette price increase, the ratio of the additional tax revenue collected from the top income group to the bottom income group increases in all four states, while with a 100% price increase, the bottom income group would accrue tax savings as a result of quitting and reduced consumption ( Figure 1d). The cost savings to AB-PMJAY, in Assam and Uttar Pradesh, from a 25% price increase would be less than half that from the baseline scenario (INR 26 crore 55 crore in Assam, and INR 290 crore vs 618 crore in Uttar Pradesh), while that from a 100% price increase would be almost double (INR 103 crore in Assam and INR 1,160 crore in Uttar Pradesh) ( Figure 1e). The findings using the Indian price elasticity of -0.34 are similar to that of the baseline scenario.

Figshare: Data inputs and data sources for impact of cigarette tax increase on health and financing outcomes in four Indian states. https://doi.org/10.6084/m9.figshare.12043074 ¹⁶

This project contains the following underlying data:

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).