Testimony of


Before the Committee on Ways and Means
U.S. House of Representatives

In Public Hearings on the Financing Provisions
Of the Administration's Health Security Act

1100 Longworth House Office Building, Washington, D.C.
Thursday, November 18, 1993


I am a practicing primary-care internist at the Massachusetts General Hospital in Boston and an economics professor at the Massachusetts Institute of Technology. I last addressed the Committee on Ways and Means in Savannah, Georgia, in April, 1989, when I served as invited faculty to the Committee's Annual Issues Seminar on deficit reduction. My biography is attached.

I am here to testify on the Administration's proposed 75-cent-per-pack increase in the Federal excise tax on cigarettes. I am solely responsible for the contents of my testimony, including any errors or omissions.


Cigarette smoking is now responsible for twenty percent of all deaths in the United States annually [1]. As a physician, I have personally witnessed the tragedy of disability and death wrought by smoking. While I shall address the Committee concerning questions of economic cost, I emphasize that smoking is first and foremost a health issue. When we talk about disease in dollar terms, we should take care not to trivialize the human lives at stake.

From my review of past and ongoing research, I estimate that cigarette smoking accounts for 8 percent of all health- care spending in the United States. The range of uncertainty in my estimate is from 4.2 percent to 11.5 percent [2,3,4].

If the Administration's Health Security Act were enacted by the end of the 103rd Congress, then the States would most likely begin to establish health alliances in 1995. Accordingly, I have chosen 1995 as the benchmark year for my economic calculations below.

By 1995, national health expenditures are projected to reach $1.1 trillion, or 15.5 percent of GDP [5]. Accordingly, in 1995, by my estimate, the adverse health effects of cigarette smoking will be responsible for $88 billion in health-care spending, with an uncertainty range of $46 to $127 billion [6].

Cigarette smokers represent 18 percent of the entire U.S. population (including infants and children.) Former smokers make up another 19 percent of the population. Sixty-three percent of the population has never smoked [7]. Accordingly, under universal health coverage, I estimate that in 1995, people who never smoked will contribute $55 billion toward the health-care costs of cigarette smoking. (The uncertainty range is from $30 to $80 billion). Current and former smokers will pay the remaining $33 billion. (The uncertainty range is $17 to $47 billion.)

With no intervening increase in Federal cigarette taxes, I expect U.S. cigarette consumption in 1995 to be 23.7 billion packs [8]. At that level of cigarette consumption, the health-care financing burden imposed upon people who never smoked would amount to $2.32 per pack (with an uncertainty range from $1.27 to $3.38 per pack). The full health-care costs of smoking, including those costs borne by current and former smokers, would amount to $3.71 per pack (with an uncertainty range from $1.94 to $5.36 per pack).


I have estimated the health-care costs of smoking that are subsidized by persons who never smoked. These costs vastly understate the total burden of smoking imposed on our society. Many of these "external" or "social" costs are easy to describe but difficult to quantify. Some economists focus only on the easy-to-measure costs; they assume that all unquantifiable costs somehow cancel each other out. I call this the "cold approach." As a physician, I know that the cold, hard numbers don't tell the whole story, that one cannot dismiss injury and suffering merely because it cannot be simply calibrated. I prefer the alternative "warm approach."

The death and disease caused by smoking results in a loss of American productivity. According to the Centers for Disease Control, in 1990, the death toll from smoking caused an annual loss of 1.1 million person years of life before the age of 65 [9]. This loss of productivity has numerous macro-economic consequences-- for example, reduced international competitiveness-- that are real but difficult to quantify.

In a May 1993 report, the Office of Technology Assessment estimated that premature deaths from smoking (along with lost work-days and productivity) caused a loss of $47.2 billion in personal income in 1990 [10]. At current inflation rates, that amounts to $56 billion in 1995. At a 25 percent marginal tax rate, OTA's estimated productivity loss would mean foregone income taxes of $14 billion, which might otherwise help to pay for national defense, environmental protection, drug enforcement, crime control, and other needed Federal services. As a "warm" economist, I cannot brush aside these hard-to-quantify external costs [11].

"Cold" economists assume that smokers and their families privately, rationally, and voluntarily bear the costs from smoking-related disease and death. This is a fiction that ignores the dual reality of teenage initiation into cigarettes and adult addiction to cigarettes.

The average American smoker now starts regular cigarette use at age fifteen, and many Americans start before age ten. Teens and pre-teens typically believe that they can stop at will. Yet each year, at least 17 million adult smokers try to quit but fail. On any single attempt to quit, the smoker's long-term success rate may be as low as 8 percent [12]. Adult cigarette smokers have cumulatively paid billions of dollars for all sorts of over- the-counter and prescription smoking cessation aids, and most market analysts believe that the pent-up demand for such products is enormous [13]. A "warm" economist recognizes that current cigarette smokers would collectively be willing to pay billions of dollars to have their addiction taken away from them. This external cost is hard to quantify, but again, it is genuine.

"Cold" economists say that a person who dies upon retirement saves the Federal purse and private pension plans the costs of Social Security benefits and retirement annuities. Warm economists say that this is not the kind of calculation that a civilized society engages in. Two members of this Committee (Reps. Archer and Rostenkowski) turned 65 this past year. Three other members (Reps. Houghton, Gibbons, and Pickle) have past the standard retirement age. According to the "cold" approach, society incurs additional external costs for each and every extra day that they survive and serve our country.

When Congress considers the merits of increasing Federal funding for breast cancer prevention, diagnosis and treatment, it does not remind itself that most women who die from breast cancer have already passed their sixty-fifth birthdays. It does not consider whether an improvement in breast-cancer survival would impose a burden on Social Security or private pensions. Congress considers the funding of breast cancer research primarily a matter of health. The same standard should apply to the taxation of cigarettes.

Senator Moynihan has proposed a tax on ammunition to help finance the Administration's Health Security Act. When Congress considers this proposal, I hope that it does not consider the age distribution of the victims of fatal shootings, or the savings in external costs that might accrue if septagenarians were murdered. We should apply the same standard when we consider measures to reduce the death toll from smoking. No double standard for cigarette smoking should be applied [14].

During the fiscal year ending June 30, 1993, total governmental taxes on cigarettes-- including Federal, state and local excise taxes as well as applicable state sales taxes-- amounted to $0.58 per pack [15]. Of this amount $0.24 per pack represented the current Federal excise tax. Accordingly, even with an additional Federal tax of $0.75 per pack, I believe that the total tax burden on cigarettes would fall far short of its true social cost.


The Administration has proposed a 75-cent-per-pack increase in the Federal excise tax. If that tax were fully reflected in the retail price of cigarettes, then I estimate that U.S. cigarette consumption would decline about 12 percent [16]. Most of the resulting drop in smoking rates would represent adults quitting smoking and teenagers never starting. Altogether, there could be as many as 4 million fewer cigarette smokers. The adult quitters will experience immediate health benefits in terms of reduced rates of cardiovascular disease, and more long-term benefits in terms of reduced rates of cancer and chronic lung disease [17]. The teenagers who never started will add years to their life expectancy.

Some opponents of the Administration's proposal have argued that an increase in the Federal excise tax will cost the U.S. economy millions of jobs. These claims are markedly exaggerated. For a full discussion, I refer the Committee to a recent report by Arthur Andersen Economic Consulting [18]. I attach a one-page summary of the Arthur Andersen analysis that was prepared by the Coalition on Smoking OR Health, an organization representing the American Cancer Society, the American Heart Association, and the American Lung Association. (Not attached to current version.)

The Committee needs to understand that the primary, direct negative impact an increase in the Federal excise tax will be on American cigarette manufacturers and their shareholders-- not retailers or farmers. The adverse impact on cigarette manufacturers will be greater if the 75-cent tax is not fully passed on to consumers.

Cigarette manufacturers have known for months that the Federal tax on cigarettes would rise from its current level of $0.24 per pack to nearly $1.00 per pack. Temporary price reductions, announced last spring by Philip Morris and other companies, were intended partly to alleviate the impact of higher future taxes. During the first part of 1993, manufacturers' wholesale prices for king-size cigarettes were cut by 37 cents per pack [19]. The increasing market shares of discount and generic cigarettes will also blunt the price effect of a Federal tax increase.

The Treasury Department estimates that a 75-cent-per- pack tax would net $11 billion in additional Federal dollars in the first year alone. If the 75-cent tax increase were fully passed on to consumers, then I estimate the first-year impact to be closer to $12 billion. Still, the Treasury's estimate is within the margin of uncertainty of my own calculations.


Some members of the Committee may ask: If we tax cigarettes because they are detrimental to health, then why don't we also tax the saturated fat in tenderloin beef cuts, or extra salt in salted peanuts? But tobacco products are a unique and special case. As the First Lady has testified, they cause serious harm when used exactly as intended. What is more, cigarettes are toxic to all smokers at every dose.

By contrast, beef contains important nutrients including protein and essential amino acids. Peanuts contain Vitamin E, for one, and as some researchers note, eating nuts may help prevent heart disease. For many people, eating saturated fats does not raise blood cholesterol. For others, eating salt does not cause hypertension [20]. In short, I do not see the taxation of tobacco for health reasons as pushing our society down an inevitably slippery slope.

I have estimated that in 1995, under universal health insurance, people who never smoked will pay $55 billion toward the health-care costs of smoking. This is one of many important, but less quantifiable external costs of cigarette use. The $11 to $12 billion increase in net revenues in 1995-- to be derived from the Administration's proposed cigarette tax hike-- will not come close to covering these external costs.

Still, I must again emphasize as a physician that smoking is foremost a health problem, not a matter of cold economic calculation. Health-care reform is about saving lives. When I tell one patient that she has inoperable lung cancer, when I urge another to quit before he has a fatal heart attack, I don't ask myself whether their illnesses are raising or lowering the Federal deficit. I just think about getting them better.


1. "Cigarette Smoking-Attributable Mortality and Years of Potential Life Lost-- United States, 1990," Morbidity and Mortality Weekly Report 42 (Aug. 27, 1993): 645-49. (Return to text.)

2. On average, an adult cigarette smoker (current or former) spends 20 percent more on health care than an adult who has never smoked. This excess rate of spending varies from 10 to 30 percent, depending on the source of data and the methods used by researchers to compare smokers and nonsmokers, but in some studies it runs as high as 100 percent. See, for example: J. Paul Leigh and James F. Fries, "Health Habits, Health Care Use and Costs in a Sample of Retirees," Inquiry 29 (Spring 1992): 44-54; Thomas A. Hodgson, "Cigarette Smoking and Lifetime Medical Expenditures," Milbank Quarterly 70 (1992): 81-125; Willard G. Manning, Emmett B. Keeler, Joseph P. Newhouse, et al., The Costs of Poor Health Habits (Cambridge, Mass.: Harvard University Press, 1991); Dorothy P. Rice, Thomas A. Hodgson, et al., "The Economic Costs of the Health Effects of Smoking," Milbank Quarterly 64 (1986): 489-547; Gerald Oster, Graham A. Colditz, and N.L. Kelly, The Economic Costs of Smoking and Benefits of Quitting (Lexington, Mass.: Lexington Books, 1984). (Return to text.)

3. Current and former cigarette smokers together represent half of all adults. (See: "Cigarette Smoking Among Adults-- United States, 1991," Morbidity and Mortality Weekly Report 42 (Apr. 2, 1993): 230-33.) If the average adult smoker spends 20 percent more than the average nonsmoker (as explained in note 2), then overall smoking will be responsible for one-eleventh (or 9.1 percent) of all health-care spending by adults. Put differently, each smoker incurs six dollars in health-care costs for every five dollars spent by a nonsmoker. With equal numbers of smokers and nonsmokers, the extra dollar spent by the smoker thus constitutes one out of eleven dollars spent on health care. The same logic can be used to estimate an uncertainty range for the proportion of health-care dollars contributed by smoking. If the average smoker spent just 10 percent more than a nonsmoker, then smoking would account for 4.8 percent of all health-care spending among adults; and if the average smoker spent 30 percent more than a nonsmoker, then smoking would account for 13.0 percent of all health-care spending among adults. (Return to text.)

4. Health-care spending among persons aged 19 years or more accounts for an estimated 88.4 percent of all personal health care spending. (See: Daniel R. Waldo, Sally T. Sonnefeld, David R. McKusick, and Ross H. Arnett, II, "Health Expenditures by Age Group, 1977 and 1987," Health Care Financing Review 10 (Summer 1989): 111-120, Table 3.) Accordingly, if smoking accounts for 9.1 percent of health care costs in adults (as explained in note 3), then it accounts for 8.0 percent of health-care costs in the entire population. This estimate ignores the costs imposed on the unborn and on infants and children by mothers who smoke, and therefore understates the total costs attributable to smoking. (Return to text.)

5. Sally T. Burner, Daniel R. Waldo, and David R. McKusick, "National Health Expenditures Projections Through 2030," Health Care Financing Review 14 (Fall 1992): 1-29. (Return to text.)

6. The Office of Technology Assessment estimated that cigarette smoking was responsible for $20.8 billion in health-care costs in 1990. (See: "Statement of Roger Herdman, Maria Hewitt, and Mary Laschober on Smoking-Related Deaths and Financial Costs: Office of Technology Assessment Estimates for 1990, Before the Senate Special Committee on Aging, Hearing on Preventive Health" (Washington, D.C.: Office of Technology Assessment, U.S. Congress, May 6, 1993): page 4.) My extrapolation of OTA's estimates to 1995 would give a value of $34.4 billion. I believe that OTA's estimates are too low, and are inconsistent with other studies on excess health-care spending by smokers. (Return to text.)

7. According to the 1991 National Health Interview Survey, 26 percent of Americans aged 18 years or older are current smokers, while another 24 percent formerly smoked. (See: "Cigarette Smoking Among Adults-- United States, 1991," Morbidity and Mortality Weekly Report 42 (Apr. 2, 1993): 230-33.) In 1995, there are expected to be 194.1 million Americans aged 18 or more, out of a total population of 262.75 million. (See: U.S. Bureau of the Census, "Population Projections of the United States by Age, Sex, Race, and Hispanic Origin: 1992 to 2050," Current Population Reports, Series P-25 1092 (1992): Table 2.) If 1991 smoking prevalence rates remained unchanged, then in 1995 there will be 50.5 million current smokers and 46.6 million former smokers. If half of adults smoke, and if 73.9% of Americans are aged 18 or more, then 37 percent of Americans will be past or present smokers. (Return to text.)

8. In 1993, U.S. consumption is expected to be 24.6 billion packs. (See: U.S. Department of Agriculture, Economic Research Service, Tobacco Situation and Outlook Report TS-224 (Sep. 1993): Table 1.) I assume a continuing annual rate of decline of 1.8 percent. (Return to text.)

9. "Cigarette Smoking-Attributable Mortality and Years of Potential Life Lost," cited in note 1 above. (Return to text.)

10. See: "Statement of Roger Herdman, Maria Hewitt, and Mary Laschober on Smoking-Related Deaths and Financial Costs," cited in note 6 above. (Return to text.)

11. The following quotation (from page 41 of Manning et al., "The Costs of Poor Health Habits," cited in note 2 above) typifies "cold" economic analysis: "For any given level of national defense, the earlier mortality of smokers raises the tax burden to nonsmokers. We assumed that these effects were offset by nonsmokers' enjoyment of less pollution and less-crowded roads." (Return to text.)

12. Jeffrey E. Harris, Deadly Choices: Coping with Health Risks in Everyday Life (New York: Basic Books, 1993): Chapter 6, "Smoking and Nothingness." (Return to text.)

13. Eben Shapiro, "After Nicotine Patches: Sprays, Pills, Inhalers?" Wall Street Journal, 8 Nov. 1993, page B1. (Return to text.)

14. Surveys consistently show that the nonsmoking public would prefer not to be exposed to environmental tobacco smoke (ETS). Their preferences are reflected, at least in part, by statutes restricting smoking in public places. It is difficult to place a dollar value on 3,000 annual deaths caused by ETS, or on the respiratory irritation experienced by many nonsmokers. But a "warm" economist knows that they are not zero. (Return to text.)

15. Computed from Tobacco Situation and Outlook Report, as cited in note 8. (Return to text.)

16. My estimate is based upon a price elasticity of demand that equals -0.4. Such a price elasticity is consistent with the recent experience of Canada, in which cigarette consumption fell in conjunction with increases in federal and provincial excise taxes. See: Jeffrey E. Harris, "Two Bucks Will Finance Health Care for 10 Million," New York Times, 4 June 1993, Op-Ed page. (Return to text.)

17. Jeffrey E. Harris, Deadly Choices: Coping with Health Risks in Everyday Life, cited in note 12 above. (Return to text.)

18. Arthur Andersen Economic Consulting, "Tobacco Industry Employment: A Review of The Price Waterhouse Economic Impact Report and Tobacco Institute Estimates of 'Economic Losses from Increasing the Federal Excise Tax'," (Los Angeles: Arthur Andersen, Oct. 6, 1993). (Return to text.)

19. Calculated from Tobacco Situation and Outlook Report, as cited in note 8 above. (Return to text.)

20. Jeffrey E. Harris, Deadly Choices: Coping with Health Risks in Everyday Life, cited in note 12 above. (Return to text.)


Jeffrey E. Harris, M.D., Ph.D., is an economics professor at M.I.T. and a primary-care internist at the Massachusetts General Hospital in Boston.

His new book, DEADLY CHOICES: COPING WITH HEALTH RISKS IN EVERYDAY LIFE (Basic Books), helps consumers sort through the barrage of medical news about smoking, weight control, exercise, cholesterol, breast cancer, and sex and AIDS. His book chapter on cigarettes was the basis for a recent Wall Street Journal story (Nov. 8, 1993) about new technologies for managing nicotine withdrawal.

Dr. Harris last addressed members of the Ways and Means Committee in April, 1989, in Savannah, Georgia, where he was an invited faculty member to the Committee's annual Issues Seminar on deficit reduction.

Dr. Harris has served as a contributor and consulting editor to several Surgeon General's Reports on Smoking and Health. He authored Chapter 3 ("Changes in Smoking- Attributable Mortality") of the 1989 Report of the Surgeon General, in which 390,000 deaths were attributed to cigarette smoking in 1985.

Dr. Harris is a member of the National Research Resources Advisory Council of the N.I.H., an appointment of the Secretary of Health & Human Services. He recently advised the Consumer Product Safety Commission about toxicity testing of fire-safe cigarettes under the Fire Safe Cigarette Act of 1990.

Dr. Harris testified as an expert witness on behalf of the Attorney General of Canada in 1990 in a constitutional challenge to the Canadian government's ban on tobacco product advertising. He has advised other U.S. federal agencies, including the Congressional Budget Office and the E.P.A. He has served on committees of the National Academy of Sciences concerning AIDS, diesel emissions, and the prevention of low birthweight.

Dr. Harris has written on the economics of the tobacco industry, as well as many other topics in health policy. His articles have appeared in the American Economic Review, the Journal of the American Medical Association, the Journal of the American Statistical Association, the New York Times, the Boston Globe, and U.S. News & World Report.

Dr. Harris's favorite extracurricular activity is ice speedskating. During Nov. 6-7, 1993, he competed (along with his two children) in the Eastern States Short Track Cup in Saratoga Springs, NY. (He did not make it to the final heat in the Masters Division.)

Top of this document | Jeffrey E. Harris Home Page

Updated 02-Aug-96 by jeffrey@mit.edu