Friday, June 23, 2017

Tobacco Taxes in Low- and Middle-Income Countries

The appropriate taxation and regulation of tobacco use is well-trodden ground in US policy: after all, the breakthrough announcement by the US Surgeon General that "smoking is hazardous to your health" happened back in 1964 (for a commemoration of the 50th anniversary, see "Smoking, 50 Years Later," January 28, 2014).  But there's a renewed stir about tobacco policy in the last few years, driven this time by an international focus.

For example, back in June 2015 William Savedoff and Albert Alwang wrote "The Single Best Health Policy in theWorld: Tobacco Taxes" (Center for Global Development Policy Paper 062). They write: "Tobacco use is the largest cause of preventable disease and death in the world. ... A 10 percent increase in the real price of cigarettes leads to an average 4 percent reduction in tobacco consumption. ... Tobacco taxes are one of the most cost-effective health interventions available. Using taxes to raise the price of tobacco products in low- and middle income countries costs between US$3 – US$70 for each additional year of life (as measured by Disability Adjusted Life Years or DALYs) (Ranson 2002), comparable to the cost of child immunization."

Last November, the Fiscal Affairs Group at the IMF put out a report called "How to Design and EnforceTobacco Excises?" In April 2017, the World Bank held a two-day event titled "Tobacco Taxation: Win-Win for Public Health and Domestic Resources Mobilization Conference" (the agenda and slides for some of the presentations are here).

But for those who want a genuine doorstop of a report summarizing the state of knowledge on these issues, the useful starting point is the nearly 700-page report from the National Cancer Institute and the World Health Organization, "The Economics of Tobacco and Tobacco Control" (December 2016).
It includes an early chapter on "The Economic Costs of Tobacco Use, With a Focus on Low- and Middle-Income Countries," as well as discussions of supply and demand factors affecting tobacco use and the range of public policy options. Here, I'll just mention a few main themes that caught my eye, some of which may be useful for generating classroom discussion and lecture examples.

Pretty much every report on the economics of tobacco taxation, and this one is no exception, quotes
Adam Smith's thoughts on the subject (The Wealth of Nations, Book V, Chapter III): "Sugar, rum, and tobacco are commodities which are nowhere necessaries of life, which are become objects of almost universal consumption, and which are therefore extremely proper subjects of taxation. ... In the meantime the people might be relieved from some of the most burdensome taxes; from those which are imposed either upon the necessaries of life, or upon the materials of manufacture."

In calculating the effects of a tobacco tax, a key parameter is how the higher price will affect the quantity of tobacco demanded. The NCI/WHO report notes (footnotes omitted):
"Price elasticity of demand is the key economic concept used to understand or measure changes in cigarette consumption resulting from changes in the excise tax and in the retail price of cigarettes. In an economic context, elasticity refers to the responsiveness of one variable to a change in another variable. The price elasticity of demand measures how responsive demand (or consumption) is to a change in the price of the product. Technically, the price elasticity of demand is the percentage change in the consumption of a product in response to a 1% change in the price of the product, with all else remaining constant. As will be discussed below, nearly all empirical studies have found that the price elasticity of demand for tobacco products lies between zero and minus one. Estimates for high-income countries (HICs) are clustered around –0.4; estimates for low- and middle-income countries (LMICs) are more variable and somewhat greater in absolute terms (further from zero), with estimates clustered around –0.5. In other words, for HICs, a 10% increase in the price of tobacco is expected to decrease tobacco consumption by 4%. For LMICs, a 10% increase in price would be expected to decrease tobacco consumption by 5%. Thus, tax and price increases are a potentially potent tobacco control tool in all countries.
"Many econometric studies have estimated price elasticities for other aspects of tobacco use beyond consumption, including prevalence, cessation, initiation, duration of smoking, frequency of smoking (e.g., daily vs. non-daily), and conditional demand (amount of the product consumed conditional on being a user of that product). Still others have estimated cross-price elasticities of the demand for tobacco products—that is, the impact of a change in the price of one tobacco product (e.g., cigarettes) on the use of another tobacco product (e.g., smokeless tobacco), or of a change in the price of a subcategory of one product (e.g., premium cigarette brands) on the use of a different subcategory of that product (e.g., discount cigarette brands)."
Of course, the responsiveness of tobacco use to tax policy varies by group. For example, the immediate effect of a tobacco tax increase over the short-run of a year or two will be smaller (a common finding is about half as large) than the long-run effect over time. Smoking by younger people, not yet as well-established in the habit, tends to be more responsive to higher tobacco taxes than smoking by older people. And people in lower- and middle-income countries (LMICs) tend to respond more strongly to higher tobacco taxes than those in higher income countries. The NCI/WHO  report states:

"In summary, the number of studies based on aggregate measures of tobacco use in HICs is growing. These studies are becoming increasingly sophisticated over time, and the resulting estimates of price elasticity are remarkably consistent. Regarding the short-run price elasticities for cigarette demand, most estimates have clustered around –0.4, with the majority ranging from –0.2 to –0.6. Early studies of tobacco use in LMICs produced wide estimates of price elasticity, with most suggesting that cigarette demand in LMICs is much more responsive to price than cigarette demand in HICs. The rapid expansion of research in LMICs in recent years indicates that the range of price elasticity estimates has narrowed somewhat, with the majority of short-run price elasticity estimates falling between –0.2 and –0.8, with many clustering around –0.5. In all countries, studies that model the addictive nature of tobacco use conclude that the long-run price elasticity of demand is greater than that estimated for the short run. Price elasticity estimates tend to be more inelastic in countries where low-priced, relatively affordable cigarettes are widely available.

"Since the early 1990s, many studies based on U.S. cross-sectional data have confirmed Lewit and colleagues’ 1981 conclusion that youth smoking is more responsive to price than adult smoking. For example, using data from the 1992–1994 Monitoring the Future surveys, Chaloupka and Grossman estimated an overall price elasticity of –1.31 for youth smoking. In a later study with similar findings, Lewit and colleagues examined the impact of cigarette prices on youth smoking prevalence and intentions to smoke. Data for this study came from cross-sectional surveys of 9th-grade students in 1990 and 1992 from the 22 U.S. and Canadian sites in the Community Intervention Trial for Smoking Cessation. This study estimated that the price elasticity of youth smoking prevalence was –0.87, and the price elasticity of intentions to smoke by nonsmoking youth was –0.95. These results indicate that youth are somewhat more sensitive to price than adults."
One difficulty for economists in thinking about tobacco policy is that many of the effects of smoking are felt by the smoker. If people choose to smoke, with knowledge of the health consequences, then who are economists or public health experts to say they are wrong? Of course, the issue of second-hand smoke or the effects of pregnant women smoking are different here--those are external effects on others. But for many smokers, the issue is one that economists sometimes refer to as "internalities," where your own decisions have an undesired and unexpected on yourself. If smokers are hooked by a habit whose power and effects they do not fully anticipate, then one can make a case for policies that make tobacco use more costly, or quitting easier. The report notes:
"Smokers tend to hold erroneous beliefs about smoking and health: They think they will be able to quit when they want to, that low-tar cigarettes are less harmful than other cigarettes, that they are in a lower risk group compared with other smokers, or that the general health risks do not apply to them as individuals. In fact, many adult tobacco users struggle with quitting, the great majority of smokers regret having started, and young people taking up tobacco use significantly underestimate the addictive potential of these products and overestimate their likelihood of quitting in the future."
One reason for the ongoing focus on smoking in low- and middle-income countries is that tobacco taxes are already relatively high in many high-income countries. As a result, the potential gains to public health (and the potential for using tobacco taxes to raise government revenue) are greater in many low- and middle-income countries. Moreover, tobacco taxes in low- and middle-income countries as a group have barely budged in the last couple of decades. The report notes (footnotes and references to tables omitted):
"Total tax burden is defined as the sum of all taxes—including general sales taxes, such as a value-added tax—expressed as a percentage of the retail price. According to the 1999 World Bank publication Curbing the Epidemic: Governments and the Economics of Tobacco Control, the total tax burden on cigarettes is highest in HICs and decreases as a country’s income level decreases. Using 1996 data for the sample of countries in this study, the average tax burden was 67% in HICs, 50% in upper middle-income countries, 46% in lower middle-income countries, and 40% in low-income countries. A similar analysis based on 180 countries was performed by WHO in 2014 using the World Bank’s income categories. Although the choice of descriptive statistics (i.e., unweighted/simple average, weighted average, and median) substantially influences the results, the 2014 WHO data confirm the earlier World Bank findings that the tax burden is higher for HICs and lower for LMICs. In fact, considering unweighted average tax burdens, the picture in 2014 is not different from that in 1996."
The rhetoric around tobacco taxes can become overheated. For example, the preface of the NCI/WHO report states at one point: "Globally, approximately six million people a year die from diseases caused by tobacco use, including 600,000 from exposure to secondhand smoke. This is six million too many. Every single death from tobacco is a preventable tragedy." But the notion that public policy should seek to prevent "every single death from tobacco" is an extreme form of prohibitionism.

People act in many ways affect health: diet, exercise, driving cars, not taking a multivitamin ever day. In all of these areas, public policy can provide some information and incentives for healthier behavior, while still stopping far short of invasive micromanagement of people's day-to-day activities. In the US, federal cigarette taxes were raised dramatically in 2009, and many states have additional tobacco taxes of their own. But in many low- and middle-income countries, raising their  low tobacco taxes to higher levels offers a substantial gains to public health--along with a government revenue source that does not discourage working, hiring, or saving.