Tobacco Documents Online

statement of Dwight R. Lee Ramsey Professor of Economics University of Georgia, Athens before the Committee on Ways and Means U.S. House of Representatives November 18, 1993 Mr. Chairman and distinguished members of the Com- mittee, I appreciate the opportunity to testify on issues related to the financing of health care reform. I am the Ramsey Professor of Economics at the University of Georgia. I am here today at the request of The Tobacco Institute. The Institute has asked me to address questions raised by the Committee's October 29 hearing announcement regarding the Administration's proposal to help finance health care reform by increasing the federal excise tax on cigarettes by 75 cents per pack -- a 312 percent hike. The views expressed in this statement are my own and do not necessarily reflect the views of The Tobacco Institute. It is important to note, at the outset, that the President has stated that his purpose in proposing to increase the federal excise tax on cigarettes is solely to help finance his health care program -- not to reduce smoking. This is what the President said in response to a question from Andrea Mitchell at his press conference last Friday: "I didn't want to raise any money from anybody to do anything other than to pay for the health care program, although I think that higher tobacco taxes [would] discourage use and that's a good thing. But that wasn't what was behind it." Given that the President's sole justification for increasing the federal cigarette excise tax is to help pay for his health care package, it is fitting that the Committee's questions focus on the economic justification for, and the fairness of, the Administration's proposed tax increase. The Committee's concerns are well-founded. The Committee has asked, first, whether smokers "inflict significant external costs on the rest of society" and, if so, whether the federal cigarette excise tax should be raised "to deal with this." As I will explain, if smokers do impose external costs on the rest of society, they already are more than paying their own way at current tax levels. It is unfair, therefore, to make them pay more, and it could not be justified economically. The Committee also has asked whether we can measure how much health care costs would be reduced by raising the ciga- rette excise tax to the level suggested by the Administration, with the decline in smoking that would be expected to result. I am aware of no hard data that would permit such a measurement. At the same time, though, I would point out that a recent government study has suggested that reducing smoking could increase, not decrease, health care costs in the long run. Finally, the Committee has asked for an assessment of the effects of the tax increase proposed by the Administration on tobacco production. The proposed tax increase undoubtedly would reduce consumption and thereby decrease production. This would result in increased unemployment nationwide, with the Southeastern states being especially hard hit. The proposed tax increase also would reduce state revenues and trigger additional federal spending that would substantially offset the $10.4

2 billion in "new" revenues that proponents of the proposed tax increase project. FAIRNESS ISSUES The Committee's questions force us to examine the basic purpose of health care reform legislation. Is it to establish a pay-as-you-go system, in which Americans are taxed according to the health care costs that they are thought to incur as individuals, by virtue of their particular behaviors and lifestyles? If so, it would be arbitrary and unfair to single out smokers. It would be similarly arbitrary and unfair to focus solely on smokers if the purpose of the legislation is to reduce health care costs by promoting "healthier" lifestyles and be- haviors. As I have noted, President Clinton has said that this is not the basis on which the Administration has proposed to increase the federal cigarette excise tax. If that were the basis, however, fairness and consistency would require the targeting of a long list of "risky" lifestyles. From the standpoint of fairness and economic policy, the Administration's -proposal to finance health care reform through increased cigarette excise taxes is particularly ironic. A principal aim of the President's health care reform program is to help lower-income families. But cigarette excise taxes, like all other excise taxes, are inherently regressive. The Congres- sional Budget Office reported in 1987 that a cigarette excise tax increase would hit lower-income families more than six times harder than higher income families. Indeed, the CBO, which studied the distributional effects of excise tax increases on beer, wine, liquor, tobacco, gasoline, airfare and telephone services, concluded that " [a] n increase in the excise tax on tobacco would be the most regressive of all."' A Congressional Black Caucus Task Force report released by Congressman Mervyn Dymally (D-Cal.) stated that even a modest increase in excise taxes would "considerably magnify the incidence, prevalence and the enormity of poverty in the United States."z Let me now address the Committee's three questions in greater detail, and from a purely economic standpoint, leaving aside these questions of fairness. EXTERNAL COSTS External costs refer to costs that are incurred by one person but paid by another. If you incur a cost and you pay it, it is an internal, not an external cost. Thus, if you miss work and lose a day's pay, that is an internal cost. If you get sick and a total stranger pays for your treatment, that is an external cost. The Committee has asked generally about the external costs of smoking. It has not limited its inquiry to external costs related to health care. I will consider the various types of possible external costs separately. m Ga As discussed below, smokers already pay at current tax ~ levels more than their fair share of any external costs borne by ~ the government as a supposed result of their smoking. It is ~ N ~ 1 Congressional Budget Office, "The Distributional Effects of an Increase in Selected Federal Excise Taxes," pp. 1-2 (Jan. 1987). N 2 Report for the Chairman of the Congressional Black Caucus, "Analyzing the Possible Impact of Federal Excise Taxes on the Poor, Including Blacks and Other Minorities," p. 4 (July 1987).

3 inappropriate, moreover, to view private health insurance pre- miums paid by nonsmokers as external costs. Finally, foregone wages and "lost" productivity are not properly viewed as "external" costs. It has not been established, in any event, that smokers, as a group, are less productive than nonsmokers when all relevant factors are taken into account. A. Health Care _ 1. Costs to Government Let's start with the external costs that smokers are said to impose on federal, state and local governments in health care expenditures attributed to smoking-related illnesses (i.e., illnesses that reportedly are suffered more commonly by smokers than by nonsmokers). In a report released in May of this year, the Office of Technology Assessment (OTA) estimated that smokers "cost" federal, state and local governments $8.9 billion in health care expenditures because of illnesses viewed as smoking-related.3 Assuming the validity of that estimate for the sake of discus- sion, the fact is that smokers currently pay federal, state and local governments $11.3 billion in cigarette excise taxes and another $2 billion in sales taxes -- a total of $13.3 billion. Only smokers pay this $13.3 billion. Nonsmokers do not. Thus, through excise and sales taxes, smokers cur- rently are paying $4.4 billion more to federal, state and local governments than the $8.9 billion that OTA claims smokers "cost" all levels of government in health care expenditures. OTA estimates the federal government's share of these government "costs" at $6.3 billion. This translates to 24 cents per pack of cigarettes sold -- the current level of the federal cigarette excise tax. Clearly, with respect to government costs, smokers are more than "paying their own way" at current tax levels. 2. Private Medical Costs OTA estimates that smokers also generate $11.9 billion in health care costs that are not borne by the government -- that is, health care costs that are paid by smokers individually or through private insurance. For purposes of accurate cal- culations, even this $11.9 billion estimate must be reduced to $7.5 billion by the $4.4 billion in excess taxes that smokers pay. The Committee also should recognize, however, that there are more fundamental problems with OTA's estimate. By definition, health care costs paid by smokers are not "external" costs. Such health care costs include co- payments, deductibles and other costs that are not covered by insurance. These out-of-pocket costs cannot properly be included in any tabulation of "external costs" that smokers are thought to impose. OTA never attempted to calculate or disag- gregate these costs that are paid by smokers. Thus, it has no basis to claim that smokers do not also pay their way in the private insurance market. It is, in any event, inappropriate to view private health insurance premiums paid by others as an "external" cost. Insurance by definition involves the pooling or sharing of risk: I promise to pay your bills if you get sick, and you promise to pay mine if I get sick. Because the obligation to pay is reci- procal, insureds are, in effect, one person. Since we are a "joint enterprise" for the purpose of sharing risk, the cost that each of us may impose on the other is therefore not genu- 3 "Smoking-Related Deaths and Financial Costs: Office of Technology Assessment Estimates for 1990" (May 1993).

4 inely external. Neither does the fact that an employer may pay for an employee's health insurance make the premium an external cost. Health insurance premiums paid by employers ultimately are paid by employees_through reduced wages. The premise of insurance is the spreading and sharing of risk. It would defy this premise to isolate smokers as a "high risk" group for purposes of financing health care reform. It also would perpetuate a discriminatory feature of our current health care system, a feature that the Administration's reform package seeks to eliminate. Ironically, Secretary Bentsen has suggested that an increase in the federal cigarette excise tax is justified as a substitute for differential premiums for smokers and nonsmokers currently offered by heath insurers. Such premium differentials generally would be disallowed by community rating under the Administration's proposal. The Surgeon General, however, has indicated that it is quite uncommon for private health insurers to offer such premium differentials for smokers and nonsmokers. Even more to the point, the Surgeon General has stressed the paucity of "actuarial data to document that nonsmokers incur fewer health care costs" than smokers." Under Secretary Bentsen's rationale, therefore, the proposed tax increase would "substitute" for a premium differential that is more imaginary than real and that is not justified actuarially in any event. B. Foregone Wages and "Lostff Productivity OTA suggests that smoking results in $40.3 billion in foregone wages and $6.9 billion in "lost" productivity. Even assuming for the sake of discussion that these estimates were accurate, they do not represent "external costs." The most important point to make here is that, since these "costs" are not related to health care, it is inappropriate to consider them in determining whether a proposed federal cigarette excise tax increase may be justified as a means of financing health care r m. These rather obvious points aside, foregone wages are, by definition, costs borne directly by the employee. They cannot be considered costs incurred by anyone else. Thus, the $40.3 billion that OTA assigned to foregone wages cannot be viewed as an external cost that justifies any increase in the cigarette excise tax. "Lost" productivity cannot be considered a cost at all unless one assumes that society somehow is entitled to maximum productivity from its members, so that anything less than maxi- mum effort is a social "loss." That is, of course, an absurd conception. When a person is absent from work for whatever reason -- to go on vacation, have a tooth pulled, serve on a jury, or attend a child's school play -- there is no "cost" to society. The fact that someone does something other than work does not represent a social loss unless we view ourselves as "owned" by society and society is viewed as having the power to determine how we spend our time based on its own criteria of value. This is not my vision of America or any other free society. It has not been established, in any event, that smokers, as a group, are less productive than nonsmokers when all relevant factors are taken into account. The large majority of studies that report an association between smoking status and increased employee absenteeism acknowledge that factors other than smoking may account for the apparent association. As James 4 See Reducing the Health Consequences of Smoking -- 25 Years of Progress: A Report of the Surgeon General 548-49 (1989).

. 5 Athanasou, an antismoking advocate, stated in an early review article: "Sickness absence is a complex behavioral phe- nomenon in which a multiplicity of health, social and psychological factors are involved. * * * Most investigators have implicitly assumed that the only difference between a non-smoking and a smoking group is their tobacco habit and that any other personal factors are equally distributed within these groups. * * * None of the reported studies has considered the additional effects on sickness absence of job satisfaction, attitudes to work, personality, other psychosocial or socioeconomic variables and the urban factor in conjunction with the effects of smoking."5 Professors Robert D. Tollison and Richard E. Wagner likewise commented in their 1992 study of the issue: "[SJmokers and nonsmokers are not identical in all respects other than smoking. Among other things, smokers have an above-average represen- tation in blue collar occupations, they also consume on average an above-average amount of alcohol, although there are many teetotaling smokers and nonsmoking alcoholics, and they generally exercise less than nonsmokers, although smoking bicyclists, swimmers, and joggers can be found. In assuming that people are identical except for their smoking, various diseases and their associated costs are improperly attributed to smoking."6 A recent study by Gabel and Colley-Niemeyer is instructive. These researchers investigated absenteeism among employees of a state public health department in the U.S. in a 1990 study. Although they found a higher rate of absenteeism among smokers than among non-smokers, they also suggested that "in addition to education level and sex, selected life-style behavior may be related to smoking [that] may significantly influence sick leave and absenteeism."7 As the editor of the journal in which the study appeared stated, "separating absenteeism differences in smokers and nonsmokers is difficult if these two groups of_ workers differ in other health-related behavior."8 5 Athanasou, James A., "Sickness Absence and Smoking Behavior and Its Consequences," Journal of Occupational Medicine (1975), vol. 17, p. 444. 6 Tollison, Robert D. and Wagner, Richard E., The Economics of Smoking (1992) at 64. 7 Gabel, Harold D. and Colley-Niemeyer, Brenda, "Smoking in a public health agency: Its relationship to sick leave and other lifestyle-behavior," Southern Medical Journal (1990), vol. 83, no. 1, p. 16. 8 Novotny, Thomas E., "Smoking policies and the healthy worker effect," Southern Medical Journal (1990), vol. 83, no. 1, pp. 11-12. See also, e.g., Carmichael, A. and Cockcroft, A., "Survey of Student Nurses' Smoking Habits in a London Teaching Hospital," Respiratory Medicine (1990), vol. 84, p. 280; Hawker, Rosalind and Hoitby, Ian, "Smoking and absence from work in a population of student nurses," Public Health (1988), vol. 102, pp. 161-67; Parkes, Katharine R., "Smoking as a moderator of the relationship between affective state and absence from work," Journal of Applied Psvcholoav (1983), vol. 68, no. 4, pp. 698-

6 Professor Richard Ault of Auburn University and several colleagues recently reviewed a series of studies sug- gesting that smokers miss more work than nonsmokers because they smoke. Ault and his colleagues found that the studies merely compared absenteeism rates for smokers and nonsmokers, without considering whether the observed difference was due to smoking per se or to other underlying determinants of absenteeism that are more prevalent among smokers than among nonsmokers. Accord- ing to Ault, the failure to consider such other determinants has resulted in "spurious conclusions about the relationship between smoking and absenteeism."9 'rAE EFFECT OF A TAX INCREASE ON DEMAND FOR HEALTH CARE The Committee also has asked whether we can measure how much health care costs would be reduced by raising the cigarette excise tax to the level suggested by the Adminis- tration, with the decline in smoking that would be expected to result. As noted, I am aware of no hard data on this point. However, a number of experts and government authorities assume that smokers would live longer and make greater demands on the health care system if they did not smoke, and thus believe that reducing smoking might well increase, not decrease, health care costs. OTA stated in its recent report, for example, that the reduction or elimination of smoking -- "may not lead to savings in health care costs. In fact, significant reductions in smoking pre- valence and the attendant increase in life expectancy could lead to future increases in total medical spending, in Medicare program outlays, and in the budgets of the Social Security and other government programs." As an economist, I cannot assess the validity of OTA's assumption that reducing smoking would increase life expectancy. But if OTA's assumption is accepted, its statement -- that reducing smoking could increase health care costs over the long run -- seems self-evident. OTA's statement likewise makes it self-evident that proponents of a federal cigarette excise tax increase cannot justify such a tax increase as a means of reducing the nation's escalating health care costs. THE EFFECTS OF THE TAX INCREASE ON PRODUCTION There is no dispute that the proposed tax increase would reduce consumption and would thereby decrease production. This and other effects of reduced consumption would dramatically increase unemployment and substantially offset the $10.4 billion in additional federal revenues that proponents of the tax increase project. According to Price Waterhouse, there are approximately 681,000 jobs in the U.S. tobacco sector of the economy.t0 The tobacco sector includes tobacco growing and manufacturing, the distribution and retailing of tobacco products, and the industries that supply these sectors. 808 ("[A] major methodological limitation of many studies relat- ing to sickness absence and smoking is the implicit assumption that smokers and nonsmokers differ only in smoking behavior."). 9 Ault, Richard W., et al., "Smoking and Absenteeism," Anolied Economics (1991), vol. 23, pp. 743, 751-52. 10 Price Waterhouse, "The Economic Impact of the Tobacco Industry on the United States in 1990" (Oct. 1992).

• 7 The Administration estimates that the proposed 75- cent-per-pack cigarette excise tax increase would result in a 12-15 percent reduction in demand. It is estimated that this would cost about 82,000 tobacco sector jobs. Along with these lost jobs would be a payroll loss of approximately $1.9 billion. Through an inevitable ripple effect, this payroll loss would generate a loss of nearly 192,000 jobs throughout the economy. The South would be particularly hard hit by these job losses. It is estimated that nearly 40,000 tobacco sector jobs would be eliminated in 12 Southeastern states. In the six major tobacco producing states of Georgia, Kentucky, North Carolina, South Carolina, Tennessee, and Virginia the tobacco sector job losses are estimated to come to approximately 33,500 jobs. The proposed tax increase would trigger a significant increase in required federal spending as well. Increases in federal spending would be required as tobacco workers_become unemployed because of decreased production. A reasonable esti- mate of these losses is $1.72 billion. In addition, the tobacco component of the Consumer Price Index (CPI) is 1.7458 percent. A 75-cent increase in the price of a pack of cigarettes (a 40 percent increase) would boost the CPI by .7 percent. That in turn would require a .7 percent increase in federal spending on all indexed federal pro- grams, such as the Social Security, food stamps, and federal pension programs. At 1993 spending levels, this would amount to $3.92 billion in additional obligated federal spending. The total quantitative losses come to $5.64 billion, which leaves only $4.76 billion net revenue from the proposed 75-cent excise tax increase. This is less than half the gross estimate proponents claim the proposed tax increase actually would raise. When the multiplier effect takes effect, the net revenue from the proposed tax increase is reduced even further. It also should be recognized that reduced cigarette sales from the proposed federal excise tax increase will reduce state cigarette excise tax revenues. These revenues are cur- rently around $6.7 billion per year. Given a reasonable price elasticity of demand for cigarettes, the proposed 75-cent increase in the price of cigarettes would reduce this revenue by approximately $800 million. Some of this loss to the states undoubtedly would have to be made up by the federal government. This would represent yet another source of erosion of the net revenue the federal government would receive from the proposed cigarette excise tax increase. * * * Increasing the federal cigarette excise tax is simply not justified from an economic standpoint. Smokers already more than "pay their way" at current levels of taxation. ..Making them pay more would be discriminatory and unfair. In addition, OTA's recent report suggests that reducing smoking actually could increase health care costs over the long run rather than reduce them. Finally, the proposed excise tax increase would trigger additional federal expenditures that would substantially offset the new tax revenues and impose significant unemployment costs on the economy, particularly in the South. For all of these reasons, the proposal to increase the federal cigarette excise tax to finance health care reform should be rejected.