This article has Open Peer Review reports available.
The effect of imposing a higher, uniform tobacco tax in Vietnam
© Van Kinh et al; licensee BioMed Central Ltd. 2006
Received: 05 October 2005
Accepted: 26 June 2006
Published: 26 June 2006
Tobacco use increases the risk of many fatal diseases such as cancer, emphysema, heart disease and other circulatory diseases. If trends in tobacco use continue, approximately 500 million people alive today will die from smoking, and, by 2030, tobacco is expected to be the leading cause of premature death, accounting for about 10 million deaths per year.
While smoking prevalence has been decreasing in many of the more developed nations, rates have been high and increasing in many of the poorer nations. For example, smoking rates among Vietnamese males 15 year of age and older was 50% in 1997–98, but rose to 56% in 2002.
The Government of Vietnam is aware of the impact of smoking on public health. It has launched a campaign against smoking by issuance of a Government Resolution on National Tobacco Control Policy in 2000 with target of reducing the tobacco use prevalence rate to 20% for males and 2% for females in 2010. Vietnam was one of the first Asian nations to sign the World Health Organization's Framework Convention for Tobacco Control in 2003.
Tobacco tax increase is among the measures suggested in the 2000 Resolution to achieve its goal of lower smoking prevalence. Other components of the tobacco control program include total ban on cigarette advertising, ban on distributing free cigarette samples, ban on vending machines selling cigarettes, and ban on smoking in all public places. Research evidence shows that imposing taxes on tobacco, as part of a comprehensive tobacco control program, is among the most effective methods of reducing tobacco use.[6, 7] Higher taxes create incentives for some regular smokers to quit smoking, help prevent the young from initiating smoking, and reduces consumption among continuing smokers.
The majority of evidence on the price responsiveness of tobacco demand is for high-income countries were data and the research capacity exists. Estimates of the price-elasticity for overall cigarette demand fall in a relatively wide range due to model specification, data issues and estimating methodology,[8, 9] but the majority center in the relatively narrow range from -0.25 to -0.5.[11, 12] Evidence does indicate that low income groups in the high income countries are more sensitive to cigarette prices compared to higher income groups.
Recent studies have begun to focus on tobacco consumption in low-income countries, including South East Asia. For Indonesia, Djutaharta et al. (2002) used time-series data to estimate cigarette price elasticities in Indonesia ranging from -0.32 to -0.57, Adioetomo et al. (2001) used household level survey data to obtain a conditional (i.e., on quantity smoked per smoker) price elasticity of -0.6, but the impact of price on smoking participation (i.e., the decision to smoke) was not significant. For Sri Lanka, Arunatilake (2002) used household level data and estimated that the price elasticity was -0.53 for the whole sample and between -0.68 and -0.29 for the poorest two quintiles, Arunatilake and Opatha (2003) used aggregate monthly data and estimated price elasticities ranging from -0.227 to -0.908. A price elasticity of tobacco demand in Thailand of -0.67 was estimated by Supakorn (1993) using aggregate tobacco consumption. Isra et al. (2003)  used a linear expenditure system and household level data, and found the price elasticity of the demand for tobacco products of -0.39. They also found that poorer smokers were more responsive to tobacco prices than their richer counterparts. Karki et al. (2003) estimated a conditional price elasticity of cigarette demand of -0.42 and a total price elasticity of -0.88 in Nepal using household data. For Myanmar, Nyo Nyo et al. (2003) obtained a total price elasticity of -1.62 using household data. A WHO study using time series data obtained an overall price elasticity in Vietnam of -0.53 for cigarettes, but did not take into account a possible substitution into other tobacco products. Also for Vietnam, Laxminarayan and Deolalikar (2004), controlling for use of other tobacco products, obtained a price elasticity of smoking initiation of -1.18, but did not find a significant impact of cigarette price on quitting. Guindon, et al. (2003) estimated the demand for cigarettes in South-East Asia using panel data. They obtained short-run price elasticity estimates ranging from -0.17 to -0.78, with most estimates clustering at around -0.74, and long-run elasticities ranging from -0.4 to -1.21. Few studies suggest a limited impact of price on smoking behavior, but a study of cigarette demand in China and Russia obtained elasticities ranging from 0 to -0.15 using micro-level data. Although variations exist in the elasticity estimates and in the quality of the studies, the evidence strongly confirms a negative relationship between smoking and cigarette prices.
Public policy makers can use tobacco taxes to manipulate cigarette price. Tobacco tax rates vary from country to country. In high-income countries, the tax component often accounts for at least two-thirds of the retail price of a pack of cigarettes. In low-income countries, on the other hand, it generally accounts for less than half of the retail price. In the countries attempting to reduce tobacco consumption, the tax component is typically between two-thirds and three-fourths of the retail price of a pack of cigarettes.
The effect of tobacco taxes may depend not only on the size, but also on the form of the tax. Specific tobacco taxes are added as a fixed amount to the price of cigarettes, while ad valorem taxes, such as value-added taxes or sales taxes, are a percentage of the base price. Ad valorem taxes may be imposed at the point of sale or, as in China, Vietnam and many African countries, on the wholesale price so that the retail price already includes the tax. Taxes may also vary according to the origin of the manufacturer or the type of product. For example, some governments impose higher taxes on cigarettes produced abroad than on domestically produced ones, or by type of cigarette such as non-filtered or filtered cigarettes. Among the 114 countries with available information in the Tobacco Control Country Profiles (WHO, 2000), 95 levy a uniform tax on tobacco, while 19 levy different tobacco tax rates based on types of products and sources of materials, 10 of which were part of the former Soviet Union and or the Soviet bloc. Vietnam is also one of the countries imposing differential ad valorem tax rates.
Prior to 1999, Vietnam imposed a tax rate of 70% on filtered cigarettes produced mainly from imported materials, of 52% on filtered cigarettes produced mainly from domestic materials, and a tariff rate of 70% on imported cigars. Since 1999, Vietnam has imposed separate tax rates on three types of cigarettes: a tax of 65% is imposed on filtered cigarettes made from imported materials; a tax of 45% is imposed on filtered cigarettes made from domestic materials; and a tax of 25% is imposed on non-filtered cigarettes made from domestic materials. Non-filtered cigarettes produced from domestic materials, besides having low production costs, are taxed least, and are thus sold at even lower prices relative to filtered cigarettes. Thus, the current three-tiered tax structure exacerbates price differences, with the price of many domestically produced cigarettes around US$0.07 per pack compared to a price of foreign brand cigarettes such as 555 and Marlboro that can reach up to US$1.
Although higher cigarette taxes can reduce smoking and the associated health problems, they have implications for the spending pattern among smokers and for tax revenue. Since low price cigarettes are more likely to be consumed by those with low incomes, the low price of unfiltered cigarettes potentially creates more incentive for those with low income (including youth) to smoke, those that can least afford it. Raising taxes on low priced cigarettes will raise price, which could increase that burden. However, if the low-price smokers are sufficiently responsive to price, they may actually reduce consumption through quitting or cutting back on quantity smoked, and have a lower financial burden. In that case, government revenue from cigarette taxes could actually fall.
This paper considers how smoking rates among different income classes is affected by the current three-tiered tax structure, and the potential effect on burdens of the poor, and then considers the effect of raising taxes on low priced cigarettes. Specifically, this paper examines how the multi-level tobacco tax policy affects low-price cigarette smokers, estimates the price elasticity of demand for cigarettes by income groups, and finally analyzes the effect of imposing a single uniform tax at the level of 65% on consumption and government revenue.
Description of the data
Due to the lack of a sufficiently detailed time series data, cross sectional data is employed to examine smoking behavior among various income groups and the financial burden that it imposes, and to estimate smoking elasticities of those that consume primarily the low price cigarettes. We use the second Vietnam Living Standard Survey (VLSS) which was conducted by the General Statistical Office from December 1997 to December 1998 with a sample size of 6000 households with 28,518 individuals. Quality control was implemented in several stages, from questionnaire design, data entry, random unexpected field visits, and consistency checks after the survey was completed.
The sample was selected using a three staged random stratified cluster sampling. Using sampling weights, the data yield unbiased population estimates at the national level and are disaggregated by urban and rural residence and the seven major regions. Adjustments for clustering and stratification were made to yield more accurate standard errors for hypothesis testing using the STATA statistical software.
The VLSS dataset contains variables on a wide range of socio-economic factors including education, employment, income, family structure, location, and living standards. The questionnaire has a separate section on smoking for people age 6 and older. Individual are first asked if they had ever smoked cigarettes for a period of 6 months or more. Those who responded in the affirmative were asked whether they currently smoke, how many cigarettes they smoke each day, and the amount of money spent on cigarettes over the past 12 months. A cigarette smoker is person who is at least 15 year or age, currently smokes cigarettes and has smoked at least for six months. It is assumed that all current smokers are daily smokers. The survey also asked about expenditures on pipe smoking. The answer to this question was used to identify a pipe smoker. There are respondents that smoke both cigarettes and pipe tobacco.
Besides the household questionnaire, a separate questionnaire collects information related to characteristics at the level of the commune, the lowest administrative unit in Vietnam with an average population of 7000. Three representative retail outlet owners/workers (who are familiar with commune prices) were interviewed about prices of goods and services in the commune. Information on two types of cigarettes (a high priced filtered cigarette known as 555 and a lower price filtered cigarette known as VINATABA) was obtained, yielding 6 price data points from each commune. This information was used to calculate an average commune price.
Analysis of the users of low price cigarettes
We first consider the percentage of smokers that smoke the different types of cigarettes. Because information is unavailable to specifically distinguish between types, we classify smokers by price of cigarettes. The average prices paid for each of the types is estimated using expenditure data from the VLSS. Average prices paid per pack were calculated by dividing tobacco expenditures by the number of cigarettes smoked and multiplying by 20.
In Vietnam, cigarette retail prices vary substantially across types and also vary within type. Based on our investigation of prices in the Hanoi market in June 2003, non-filtered cigarette prices are usually between 1000 and 2000 VND1 per pack (between $0.07 and $0.13 at an exchange rate of 15,000 VND per $1 USD in 2003). Prices of 555 or smuggled Marlboros are among the highest and range between 12,000 and 15,000 VND per pack (between $0.80 and $1.00). A low price is defined as less than or equal to 1/3 of the highest cigarette retail price, based on actual prices and expenditure patterns observed from the data. Since the highest retail price in 1998 was 15,000 VND, the low price is considered 5,000 VND ($0.33) and below.
The per capita income is calculated by dividing the total annual household income by the household size. Each individual is assigned to one of five quintiles according to his/her income.
Finally, using expenditures data from the 1998 VLSS, we compared expenditures on tobacco to other items such as education, health care, food, and with the total expenditure of households. We distinguish these comparisons by region, urban-rural and by income. The relative proportion of tobacco spending is important, since if the proportion is high, raising tobacco taxes will have more influence on the household budget.
Price elasticity estimates
To estimate demand elasticities using cross-sectional data, we employ a method that distinguished the decision to smoke from the quantity smoked. The first equation explains smoking status (indicated by a dummy variable) using the sample of all individuals, and corresponds to a smoking participation equation; the second equation is confined to smokers and uses the logarithm of the number of cigarettes smoked per smoker as the dependent variable, and corresponds to the quantity smoked per smoker or conditional demand equation.
The smoking status model is estimated using the linear probability model (LPM) defined as:
D Smoker = dummy variable of smoking status (smoker = 1, other while = 0)
p Cig = price of cigarettes at the commune level
p Pipe = price of pipe tobacco
y = annual per capita income
z k = variables relate to individual, household, geographic and commune characteristics.
β 0 = constant parameter
β 1 = percentage point change in the possibility of some one being smoker as price of cigarette change 1%
β 2 = percentage point change in the possibility of some one being smoker as price of pipe tobacco change 1%
β 3 = percentage point change in the possibility of some one being smoker as disposable income change 1%
β k Parameter indicating the relationship between individual characteristic and taste, household characteristics, geographic and commune characteristics and the possibility of being a smoker.
The equation design is used to estimate the probability of an individual's smoking status and quantity smoked as a function of cigarettes price while controlling for other variables. The price coefficient indicates the probability of smoking due to variations in cigarette prices. We use communal prices in the models. Since communal prices are provided for 555 and VINATABA brands, we estimated the models with the average price of these two brands, as well as using the prices of just VINATABA or 555. Out of a total of 187 communes, 50 are missing information on the price of VINATABA or 555 and are dropped when using average price. The price variables are converted into logarithmic terms.
The second independent variable is expenditures on pipe tobacco, as a proxy for its price. Pipe tobacco is used by 30.8% of smokers (National Health Survey 2002) and is a potential substitute for cigarettes. As the price of cigarettes increase, smokers may switch to the cheaper pipe tobacco. Since neither the commune nor the household survey collected the price of pipe tobacco or the amount of pipe tobacco consumed, we use information on household spending on pipe tobacco in place of the price of pipe tobacco. Due to potential endogeneity problems (when cigarette prices are high and smoking is reduced, pipe smoking can be expected to increase), we consider equations with and without the pipe variable.
Explanatory variables also include per capita income, characteristics of the individual such as sex, age, education, work experience, occupation; characteristics of the household such as household size and the sex, age, education, main occupation of household head; and commune and geographic characteristics2. The commune and geographic variables include urban status, region, and the ability to access employment, markets, transportation. Employment opportunities are captured by the existence of factories or traditional occupations in the commune, access to information is represented by the existence of a loudspeaker system, and access to markets is captured by the presence of regular markets. We used the logarithm of income. Because some rural households have negative incomes (greater investments into production than sales of final goods), we increased all household incomes by 200 thousand VND to make them all positive3. Additional File 1 summarizes the variables used in the models.
The conditional quantity model with only smokers is a double-log model with logarithm of number of cigarettes smoked by a smoker as the dependent variable, or:
Other variables are the same as in model (1)
To determine how price responses vary by income, we estimate separate equations by income group. Because income tends to be an unstable measure of overall average living standards of a household, we distinguish by quintiles to examine the low price and high price demand groups. To measure price elasticities of the overall population, the low income group, and the high income group, we estimate one equation for the entire population, one with the two low income quintiles and one with the two high income quintiles, respectively.
The equations were adjusted for clustering using the STATA software package.
Consumption and tax revenue estimation procedure
Using the information on price elasticities, we estimate the change in consumption and government tax revenue, assuming a uniform tax level of 65%. Because the highest tax rate currently imposed is 65% for filtered cigarette brands produced using imported materials, the price of this category is left unchanged and we only estimate the changes in consumption and tax revenues of unfiltered cigarette and filtered domestic material cigarette categories only. We begin with prices and quantities from the Ministry of Industry and Ministry of Finance.
Currently, taxes are imposed on the wholesale price – the price set by producers. Ignoring any mark-up at the retail level (for simplicity), the price paid by consumer is:
t 0 = original tax rate (25% or 45%)
t 1 = increasing the current tax rate to a uniform tax rate (65%)
% change in retail price = Δt P0 p/P0 c (5)
As price increases, consumption decreases by an amount that depends on the change in the price and the price elasticity of demand. To estimate the reduction in the quantity consumed (Q), we use the price elasticity of demand, , is:
In our model, the estimate of the overall price elasticity is the sum of the estimates of participation and conditional elasticities:
From equation (6) we obtain:
The magnitude of change in the total cigarette consumption is calculated using the initial quantities of cigarette consumed, the percentage change in prices (due to the change in tax rates) and the overall price elasticity estimates as indicated in (7).
Using equation (7) we obtain the change in consumption as:
Initial government revenue is calculated for each cigarette type as:
Tobacco consumption patterns
Male smoking prevalence and smoking intensity among 5 income quintiles in Vietnam (age 15+; based on VLSS 1998)
Smoking prevalence (any tobacco use)
Cigarette-only smoking prevalence
The distribution of smokers according to their purchasing price is presented in Figure 2. The dividing point between the low price and the rest is a price of 5000 VND. Based on the Vietnam Living Standards Survey 1997–1998, we found that under-5000 VND per pack smokers accounts for 78% of the market while the over-5000 VND-pack smokers accounts for the rest – 22%. Thus, Figure 2 reveals that the majority of smokers in Vietnam consume low priced cigarettes.
Characteristics of low-priced cigarette smokers
The low-priced cigarettes are particularly popular among households whose heads are agricultural workers, which includes about 60% of low-priced smokers. Further, about 11% of low-priced smokers live in households whose heads are jobless. Thus, the majority of low-priced cigarette smokers live in rural areas, where tobacco control activities are quite limited. In analyzing education level, low-priced cigarette smokers mostly live in households whose heads have low levels of education: 69% live in households where the head is illiterate or has only a primary education, and 16% live in households where the head has a lower secondary school education.
Tobacco expenditures by smokers
The comparison between tobacco spending with education, health care, food spending and total expenditure of households by quintile (%)
Tobacco spending/education spending
Tobacco spending/healthcare spending
Tobacco spending/food spending
Tobacco spending/total expenditure
Small city and town
North Mountain and Midland
Red River Delta
North Central Coast
South Central Coast
Very low Income
While it might be expected that low income households spend a larger part of their income on basic needs such as clothing, education, health care, housing, transport, and a smaller part on tobacco, we found that the tobacco spending of low income households represents a larger proportion of their expenditure than for higher income households. Low income households' tobacco spending is equal to one-and-a-half times their educational spending and is equivalent to health care spending. By contrast, tobacco expenditures for higher income households are 46% and 69%, of educational and health expenditures, respectively. The percentages of low income households and higher income households' tobacco spending are 5.3% and 3.6%, respectively.
Estimating the price elasticity of tobacco demand
The smoking participation model
Result from equation explaining the decision to participate to smoking, dependent variable is smoking participation (yes = 1, no = 0)
Two low quintiles
Two high quintiles
P > | t
P > | t
P > | t
Logarithm of VINATABA price
Expenditures on pipe
Year of education
Year of experience
Logarithm of per capita income
Red river delta (yes = 1)
North central coastal (yes = 1)
South central coastal (yes = 1)
South-east region (yes = 1)
Mekong delta (yes = 1)
Urban (yes = 1)
Existence of regular market (yes = 1)
Opportunity to find job (yes = 1)
Accessibility of car to commune (yes = 1)
Number of observation
The results also indicate that coefficients of smoking expenditures on pipe tobacco, household size, income and region are statistically significant, but the variables for access to markets, jobs and transportation are not significant. The coefficients on income, 2.380 for high income households and 2.421 for low income household, are high and significant in both models. As the income of Vietnamese rural households increases, smoking prevalence is on the rise in those rural areas. However, the magnitude of the income elasticity has to be viewed with a caution since Vietnamese households do not tend to report their true income.
The conditional quantity model
Estimated price elasticity of conditional demand based on different prices
Two low quintiles
Two high quintiles
Communal VINATABA price
Communal 555 price
Communal average prices
Average price elasticity of conditional demand using estimates for 555 and VINATABA (the significant estimtes)
The overall price elasticity of cigarette demand
The effect of imposing a uniform tobacco tax on consumption and tax revenues
Base Consumption, price and revenue with current tax
Quantity consumed (Million packs)
Retail price (1000 VND)
Tax rate based on whole sale price
Tobacco tax revenue (Billion VND)
Filtered cigarettes produced using domestic materials
Consumption, price and revenue after imposing a uniform high tobacco tax of 65%
Quantity to consume (Million packs)
Price after tax (1000 VND)
Tobacco tax revenue (Billion VND)
Filtered cigarettes produced using domestic materials
Change in price, consumption & government tobacco tax revenue from imposing a uniform high tobacco tax rate of 65%
Change in price (%)
Change in Tobacco Consumption
Change in Tobacco Tax Revenue(2)
Quantity (Million packs)
Value (Billion VND)
Filtered cigarettes produced using domestic materials
Currently, Vietnam imposes three different tax rates on tobacco according to the source of raw materials and whether filtered or non-filtered. The large discrepancies among these rates lead to large disparities among cigarette prices and wider availability of low-priced cigarettes in the market, which makes it easier for youth and the poor to get access to tobacco.
The evidence presented here indicates that most smokers smoke low priced cigarettes. Most consumers of low priced cigarettes are poor and live in rural areas or small towns; they tend to be employed in the agricultural sector. Low income households' tobacco spending is equal to one-and-a-half times their educational spending and is equivalent to health care spending. Higher income smokers also spend a large share of their income on tobacco, but the poor bear the largest relative economic burden. Because of their high smoking rates, the poor are also more likely to have higher risks of tobacco-related diseases such as cancer, heart and circulatory diseases, and emphysema. In addition, because these diseases can appear as early as age 40 (along with other smoking related medical conditions), the higher disease incidence is likely to further increase the economic burden by increasing the likelihood of not being able to continue to work and earn income. Thus, social goals of reducing poverty would dictate reducing tobacco spending by the poor.
The high participation elasticity indicates that raising the price to low income consumers will lead to substantial reductions in the number of cigarette smokers, enough to reduce overall expenditures on tobacco by the poor. Further reductions will occur through reductions in the quantity smoked by those continuing to smoke. As the price of cigarettes increases 10%, the quantity demanded of low income smokers that continue to smoke falls 5.9% while the quantity demanded of high income smokers falls by only 4.0%. The overall quantity demanded among smokers falls 5.0%. Thus, of those who continue to smoke, there will only be modest increases in tobacco expenditures, with a smaller increase among the poor compared to the higher income groups.
Imposing a uniform tax of 65% on tobacco will result in a 32% increase in the prices of low-priced cigarettes and a 16% increase for the domestic filtered category. Applying the overall price elasticity of cigarette demand of -1.44, we predict a decrease of tobacco consumption of about 25%, and an increase of more than 11% ($8.5 mil) in the tobacco tax revenue of the government. A study using a price elasticity of -0.53 predicted a 9.3% decrease in consumption and 33.5% (or $24.8 mil) increase in government revenue.
The results indicate that tax revenues are likely to increase as tax rates are increased for domestic unfiltered and domestic filtered to the rate for foreign filtered cigarettes. Although quantities consumed fall, the increase in tax rate more than compensates. Thus, government revenues increase despite the large reductions in consumption, suggesting that government could benefit from the moving to a uniform tax.
While we used a linear probability model for ease of interpretation and use, the error term is heteroskedastic and is not normally distributed, the predicted values are not constrained to be between 0 and 1. Nevertheless, the results reported here on price effects are consistent with the theory of cigarette demand and empirical evidence from other countries. Still, due to data limitations and simplifying assumptions, the results should be regarded with caution.
First, income tends to be reported incorrectly in Vietnam. Surveyed subjects tend to declare lower income because high income is often regarded as being linked to illegal activities and corruption. However, despite measurement error, the results indicate high sensitivity of cigarette demand to income. As the income is expected to grow in the near future in Vietnam, there is a danger that smoking prevalence will grow if no tobacco control measures are taken.
A second limitation is that the potential substitution into pipe smoking was taken into account by using expenditures on pipe tobacco, instead of price, creating potential endogeneity problems. When we omitted the pipe variable, we obtained consistent results, with slightly lower coefficients and levels of significance on some of the price variables. Nevertheless, in Vietnam, pipe tobacco is a likely an important substitute for cigarettes, especially among the poor. With an increase in the consumption of pipe tobacco, the price of pipe tobacco should increase, reducing some of the pipe use. However, any substitution into pipe use may still be preferable to smoking if pipe smoking is less expensive and imposes less economic burden on the poor and if pipe use has less harmful effects on health. Further work is needed on the role of substitution into pipe use and the health effects of pipe use relative to that of smoking.
A third limitation is that the price data is measured with error. This is, in part, because we were able to only include the price of medium and high price standard brands. We did, however, obtain relatively similar results with the various combinations of the measures, although the results were not always significant, as might be expected with measurement error. We also estimated equations which considered prices based on the expenditures on tobacco and quantity smoked (both undoubtedly measured with error), and obtained roughly consistent results. In computing the change in tax revenues, we did not consider substitution between cigarette types. As the price of the cheaper, non-filtered domestic increases, there may be substitution toward the filtered domestic brand, and, as the filtered domestic cigarettes increases, there may be more substitution towards the filtered cigarettes with foreign tobacco. However, these effects would tend to increase tax revenues as more cigarettes are purchased of the more expensive brands. For these reasons, we may have underestimated the increase in tax revenues, but overestimated the effect on consumption.
We do not consider smuggling or domestic tobacco growing. Imposing a uniform tax on cigarettes will increase the domestic market price of cigarettes and may widen the gap between domestic and the prices of cigarettes in neighboring nations providing more motivation for smuggling. In addition, higher market prices of cigarettes may encourage an increase in domestic tobacco growing. These changes in the tobacco market may reduce the tax revenues. However, a review of the literature indicates that smuggling depends on factors other than the gap between domestic and international prices of cigarettes. These factors include the strength and extent of law enforcement, activities of market control forces, and the living standards of poor people in border areas (poor, jobless people near some border entrance points tend to work for smugglers to move cigarettes into Vietnam). The Government of Vietnam currently pays considerable attention to smuggling in order to protect the domestic market. Nevertheless, the smuggling issue and the issue of domestic tobacco growing as a response to cigarette price changes merit further study.
We extrapolate future tax revenues from a single price elasticity estimate. In addition to smuggling, our estimates will depend on income, population and other tobacco control policies, among other factors. Rising income and population will increase tax revenues, but other tobacco control measures will reduce demand for cigarettes and, hence, tend to reduce revenues.
We focus on male participation. Tobacco use by women is much lower, with 1.8% of females smoking in 2002 and 50% of female smokers smoking pipes. Since smoking is prohibited by some parents, and may be considered improper for women (Do Hong Ngoc 1995), it is likely that there is some underreporting of cigarette and tobacco use among women, as well as youth. Nevertheless, it will be important to monitor female smoking in future years since these social norms may change with the rapid economic development experienced by Vietnam. The lack of awareness of the health risk associated with smoking may then result in a fast onset of female smoking.
Finally, the optimal tax rate that should be imposed is difficult to determine because it depends on many factors including the costs imposed by smokers on non-smokers. Another potentially important concern is employment in both the tobacco industry and tobacco cultivation. When imposing a high tax, the reduction in consumption may reduce employment in some areas. But the effect is also not clear since the tobacco industry can export both tobacco leaf and cigarettes. The choice of tax rate also depends on social values such as the protection of children and others from tobacco smoke, and on the purpose of taxation, i.e., to increase tax revenues or to reduce the burden of tobacco related diseases.
Worldwide, only few countries levy different tobacco tax rates on cigarettes. Most of them are former socialist economies in Eastern Europe and the former Soviet Union where this phenomenon was inherited from the planned economy structure. The differential treatment of various cigarette types can be justified neither on economic nor on public health grounds since all cigarettes are harmful to health.
An increase of tobacco taxes can have additional positive impact on public heath if part of tobacco tax revenue is earmarked for anti-smoking activities. Victoria, Australia was the first jurisdiction to establish a health promotion agency funded by tobacco taxes. Many others similar agencies were modeled upon the Victorian Health Promotion Foundation, such as ThaiHealth Foundation in Thailand. Thailand is one of the countries with the greatest success in tobacco control. The tobacco tax increased from 55% in 1992 to 75% in 2001, which is applied as a uniform tax for all tobacco products. Meanwhile the tobacco tax revenue doubled from 15,438 million Baht in 1992 to 31,247 million Baht in 2002 and consumption fell from 2,035 million packs in 1992 to 1,716 million packs in 2002. Thus, other tobacco control activities may further supplement the effect of a tax increase.
In this study, we found that the poor are more responsive to cigarette prices than those with higher incomes. Consequently, a tax increase would lead to relatively higher reduction of cigarette demand among the lowest income group and generally encourage that group to reduce the large economic burden that smoking imposes.
1 VND is an abbreviation for the local currency in Vietnam called Vietnamese Dong
2 The two poorest regions in Vietnam were used as a benchmark
3 200 thousand VND is the smallest amount that makes income for households positive when added to the reported income.
- Doll R: Uncovering the effects of smoking: historical perspective. Statistical Methods in Medical Research. 1998, 7 (2): 87-117. 10.1191/096228098668199908. (31)View ArticlePubMedGoogle Scholar
- World Bank, Development in Practice – Curbing the Epidemic, Government and the Economics of Tobacco Control, Washington, D.C. 1999Google Scholar
- 1997–8, Vietnam Living Standard Survey, as described below and in Bales, Sarah with Hoang Van Kinh. An Empirical Analysis of Smoking Using the Vietnam Living Standard Surveys, Report submitted to the World Bank. 2000Google Scholar
- 2001–02 Vietnam National Health Survey. This survey included information on smoking, household characteristics, health, and healthcare utilization. A total of 36,000 households – nearly 160,000 persons – were selected using stratified, random design.Google Scholar
- Government of Vietnam, Government Resolution on Tobacco Control Policy period 2000–No 12/NQ-CP dated 14th August. 2000Google Scholar
- Chaloupka FJ, Hu T, Warner KE, Jacobs R, Yurekli A: "The taxation of tobacco products,". Tobacco control in developing countries. Edited by: Jha P, Chaloupka F. 2000, 237-272.Google Scholar
- (U.S. Department of Health and Human Services. 2000. Reducing Tobacco Use: A Report of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health.Google Scholar
- Gallet CA, List JA: Cigarette demand: a meta-analysis of elasticities. Health Econ. 2003, 12 (10): 821-35. 10.1002/hec.765.View ArticlePubMedGoogle Scholar
- Wasserman J, Manning WG, Newhouse JP: The effects of excise taxes and regulations on cigarette smoking. Journal of Health Economics. 1991, 10 (1): 43-64. 10.1016/0167-6296(91)90016-G.View ArticlePubMedGoogle Scholar
- Farrelly MC, Bray JW: Responses to increases in cigarette prices by race/ethnicity, income, and age groups – United States, 1976–1993. Morb Mort Wkly Rpt. 1998, 47 (29): 605-9.Google Scholar
- The Economics of Smoking, Chaloupka FJ and Warner KE. The Handbook of Health Economics. Edited by: Joseph P Newhouse, Anthony J Cuyler. 2000, New York: North Holland; The Economics of SmokingGoogle Scholar
- Hu TW, Ren QF, Keeler TE: The demand for cigarettes in California and behavioural risk factors. Health Econ. 1995, 4 (1): 7-14.View ArticlePubMedGoogle Scholar
- Farrelly MC, Bray JW: Responses to increases in cigarette prices by race/ethnicity, income, and age groups – United States, 1976–1993. Morb Mort Wkly Rpt. 1998, 47 (29): 605-9.Google Scholar
- Djutaharta T: The impact of cigarette tax rate increase on consumption and government revenue: aggregate analysis. Un published paper commissioned by the World Bank. 2002Google Scholar
- Adioetomo SM: The economic aspects of tobacco consumption in Indonesia: a household analysis of the. 1999, Washington DC, Demographic Institute and Human Development Health, Nutrition and Population Division, World Bank, unpublished report, national socio-economic dataGoogle Scholar
- Arunatilake N: An Economic Analysis of Tobacco Demand in Sri Lanka. Sri Lanka Economic Journal. 3 (1): 96-120.Google Scholar
- Arunatilake N, Opatha M: The Economics of Tobacco in Sri Lanka. HNP Discussion Paper, Economics of Tobacco Control Paper 12, Washington DC, World Bank. 2003, [http://www.worldbank.org/tobacco]Google Scholar
- Supakorn B: Demand analysis of aggregate cigarette consumption in Thailand, 1976–81 [mimeograph]. Bangkok. 1993Google Scholar
- Isra S: The economics of tobacco in Thailand. HNP Discussion Paper, Economics of Tobacco Control Paper No 15. Washington DC, World Bank. 2003, [http://www.worldbank.org/tobacco]Google Scholar
- Karki YB, Pant KD, Pande BR: The economics of tobacco in Nepal. HNP Discussion Paper, Economics of Tobacco Control Paper No 13. Washington DC, World Bank. 2003-[http://www.worldbank.org/tobacco]
- Nyo Nyo K, Zaw KK, Naing KK: The economics of tobacco control in Myanmar. HNP Discussion Paper, Economics of Tobacco Control Paper No.14. Washington DC, World Bank. [http://www.worldbank.org/tobacco]
- Patrick Eozenou: Price Elasticity Estimations for Cigarette Demand in Vietnam. [http://pathcanada.org/vietnam/tobacco/research/docs/PriceElasticityEstimatesForCigaretteDemandInVietnamEN.pdf]
- Ramanan Laxminarayan and Anil Deolalikar Tobacco Initiation, Cessation and Change: Evidence from Vietnam, Health Economics. 2004, 13 (12): 1191-1201.Google Scholar
- Emmanuel Guindon G, Anne-Marie Perucic , David Boisclair : Higher Tobacco Prices And Taxes In South-East Asia. An Effective Tool to Reduce Tobacco Use, Save Lives and Generate Revenue. HNP Discussion Paper, Economics of Tobacco Control Paper No.11. Washington DC, World Bank. [http://www.worldbank.org/tobacco]
- Lance Peter M, Akin John S, Dow William H, Loh Chung-Ping: "Is Cigarette Smoking in Poorer Nations Highly Sensitive to Price? Evidence from Russia and China,". Journal of Health Economics, . 2004, 23: 173-189. 10.1016/j.jhealeco.2003.09.004.View ArticleGoogle Scholar
- Chaloupka FJ, Jha P: "Curbing the Epidemic:The Economics of Tobacco Control". World Bank Publication (Chapter 4). 1999Google Scholar
- Guindon GE, Tobin S, Yach D: "Trends and affordability of cigarette prices: ample room for tax increases and related health gains.". Tob Control. 2002, 11 (1): 35-43. 10.1136/tc.11.1.35.View ArticlePubMedPubMed CentralGoogle Scholar
- Blecher EH, van Walbeek CP: An international analysis of cigarette affordability, Tobacco Control. 2004, 13: 339-346.Google Scholar
- The Government of Vietnam, The Government document No3280/VPCP-KTTH dated 14th June 2002 on the base price to impose special tax on domestic produced cigarettes. WHO. 2002Google Scholar
- Tobacco Control Country Profiles. WHO. 2002Google Scholar
- Bach Thi Minh Huyen – Ministry of Finance, Tobacco Tax Policy in Vietnam, Tobacco Tax Workshop, Hanoi, July. 2003Google Scholar
- Author's investigation of Hanoi market in August. 2003Google Scholar
- Hoang Van Kinh: Financial burden of smoking on households in Vietnam, ThaiHealth Foundation. 1994Google Scholar
- Elisabeth Sadoulet, Alain de Janvry, Quantitative Development Policy Analysis. 1995, The Johns Hopkins University Press, 32-60.Google Scholar
- Ramu Ramanathan, Introductory Econometrics with Applications. 2002, Harcourt College Publishers, 255-257. FifthGoogle Scholar
- Bach Thi Minh Huyen, Share of Cigarette Consumption and Cigarette Prices. Ministry of Finance of Vietnam. Presentation at the Hanoi Tobacco Tax Workshop, Hanoi, April. 2003Google Scholar
- "Government Resolution on National Tobacco Control Policy period 2000-2010." No. 12/2000/NQ-CP(August 14th 2000).Google Scholar
- Hoang Van Kinh and Sarah Bales, "Tobacco in Vietnam: The industry, demand, control policies and employment," in "Economic Social and Health Issues in Tobacco Control", Report of a WHO international meeting, Kobe, Japan 3–4 December 2001, WHO. 2003, 144-Google Scholar
- Author's calculation from Vietnam Living Standard Survey. 1998Google Scholar
- Ministry of Health, Vietnam National Health Survey Medical Publishing House. 2004Google Scholar
- U.S. DHHS (1989). Reducing the health consequences of smoking: 25 years of progress: a report of the Surgeon General. Atlanta, GA, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health.Google Scholar
- Chaloupka FJ, Hu T, Warner KE, Jacobs R, Yurekli A: "The taxation of tobacco products,". Tobacco control in developing countries. Edited by: Jha P, Chaloupka F. 2000, Oxford University Press: New York, 237-272.Google Scholar
- Ramanan Laxminarayan and Anil Deolalikar: Tobacco Initiation, Cessation and Change: Evidence from Vietnam, Health Economics. 2004, 13 (12): 1191-1201.Google Scholar
- Chaloupka FJ, Jha P: (1999) "Curbing the Epidemic: The Economics of Tobacco Control". World Bank Publication & Tobacco Control in Developing Countries. Edited by: Chaloupka FJ, Jha P. 2000, Oxford University press, OxfordGoogle Scholar
- Merriman D, Yurekli A, Chaloupka F: How big is the worldwide cigarette smuggling problem. Tobacco control in developing countries. Edited by: Jha P, Chaloupka F. 2000, Oxford University Press: New York, 365-392.Google Scholar
- Do Hong Ngoc : "Smoking in women A survey in Ho Chi Minh City". Health Information and Educational Center, Ho Chi Minh City. 1995Google Scholar
- Morrow M, Barraclough S: Tobacco control and gender: A need for new approaches for the Asia-Pacific region. Development Bulletin, No. 54. Special issue on Tobacco Globalisation and Development, April. 2001Google Scholar
- Prakit Vathesatogkit, Thailand's Experience in Tobacco Control and Tobacco Tax Policy, Regional Workshop on Advancing Tobacco Tax Policy in Southeast Asia, 3–6 March. Hanoi. 2003Google Scholar
- Share of cigarette consumption and price based on the report of Ministry of Finance to submit to VINACOSH for Sida project on Tobacco control program. 2003Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.