On October 3, the Cato Instituters F. A. Hayek Auditorium was the setting for a seminar, "Cutting through the Smoke: The Science and Politics of Tobacco." Among the speakers were Jane Gravelle, an economist with the Congressional Research Service; Jacob Sullum, articles editor of National Review; Sheldon Richman, senior editor at the Cato Institute; and moderator William A. Niskanen, chairman of the Cato Institute. The views presented here do not necessarily reflect those of the Congressional Research Service.
Jane Gravelle: After the Clinton administration proposed a fairly substantial increase in the cigarette tax as a way of funding health care reform, my colleague Dennis Zimmerman and I wrote a paper entitled "Cigarette Taxes to Fund Health Care Reform and Economic Analysis."* The part of the paper I'd like to talk about is the justifications for increasing the cigarette tax. I'm an economist, so I start with the presumptions that people have subjective preferences about what they like to do and how they spend their money and that, in general, we want to allow people to enjoy their lifetime resources in accord with those preferences. We would intervene in those decisions only under certain kinds of circumstances that we try to delineate and measure.
The first one of those circumstances is the possibility of "spillover effects" from smoking not primarily secondhand smoke but another kind of potential spillover effect, which is the financial spillover costs from smoking that are incurred because we have social and medical insurance systems that charge uniform premiums and don't differentiate those premiums according to whether or not you smoke. We wanted to try to measure the magnitude of the spillover effects and then compare them with the existing level of cigarette taxes to determine whether an increase in cigarette taxes is justified.
We were surprised to find that there was really only one study that did that kind of analysis in a conceptually appropriate framework that could give a measure of the efficient level of the cigarette tax. When you buy a pack of cigarettes, you pay the price of the cigarettes. You also assume some implicit costs that you know about if you are aware of the health effects of smoking. But there might be another part of the cost that you don't pay, the cost that smokers impose on other people. That is the kind of cost that we were trying to examine. When we looked at the study, done by health economist Ray Manning and several associates (funded by the RAND Corporation), we found that the spillover effect per pack of cigarettes was 33 cents. At the time, the sum of federal, state, and local cigarette taxes was about 50 cents per pack. So the cigarette tax was already higher than the spillover cost.
Let me briefly mention some of the components of the spillover cost. The first were the increased health care costs of smokers, which are incurred earlier in life than are those of nonsmokers, that accounted for about 49 cents per pack. There was also another 1 cent per pack for the excess sick-leave cost of smokers relative to nonsmokers. There were 7 cents for excess life-insurance premiums, 3 cents for fire-insurance premiums, and about 12 cents in Social Security and Medicare taxes lost because of premature death.
However, there were also offsetting savings, mainly because smokers die earlier. The medical cost is actually a net of the cost of smoking-related deaths less the illnesses smokers do nor get later in life because they die earlier There is also a savings of 6 cents pet pack in nursing home cost and 33 cents per pack in reduced Social Security benefits.
The Center for Disease Control released an estimate that the health related costs of smoking sum to $50 billion a year, and someone at CDC too; that $50 billion and came up with a cost of $2.06 per pack of cigarettes. Then people asked, "Why do they say $2.06 and you say 33 cents?" I can use that to explain how important it is to think about this cost in the right conceptual terms. Fifty billion dollars a year is not the increase in medical costs due to smoking. That's true only if the alternative to dying from a smoking-related disease is perfect health and eternal life. The $50 billion should have netted out of it, if you're actually asking how much smoking adds to medical bills, the costs that we aren't paying for people who die before they get other diseases. That sounds very crass and coldhearted, but if you're looking at costs, that's the way you have to do it.
On the question of policy responses, we can talk about either taxes or regulation. As is the case for taxes, the case for regulation is dubious because while you do need some sort of regulatory decision about exposure when you've got a monopoly provision of goods and services if you have a large economy with many options for jobs and restaurants and bars, you can make a case that private businesses and indi- viduals can make their own decisions.
There is another issue that we discussed in the paper that would be a justification for an additional cigarette tax, and that is depending on the term you prefer helping individuals help themselves, government paternalism, or intervening to prevent problems of asymmetric information. You might argue that we need a tax if people really aren't seeing the health costs, but the evidence suggests that most individuals actually overestimate the risk of smoking. They think that smoking is more likely to kill them than it really is. That is not surprising, because people have a tendency to exaggerate in their own minds highly publicized risks for which they have no referent. A case in point is the risk that passive smoking may cause lung cancer. You hear that passive smoking might increase your risk 20 percent, 50 percent, 70 percent, but what you don't know is that lung cancer among nonsmoking individuals is a very, very rare disease. So you can have a big percentage increase, and it's still a very small risk.
Jacob Sullum: A few years ago when I was working on a story about the antismoking movement forReason magazine, I interviewed Scott Ballin, chairman of the Coalition on Smoking or Health. I raised the question of why people smoke. "There is no positive aspect to it," he assured me. "The product has no potential benefits." Not everyone concurs with that assessment; in a recent column in Vanity Fair, for example, Christopher Hitchens wrote that "cigarettes improve my short-term concentration, aid my digestion, make me a finer writer and a better dinner companion, and in several other ways prolong my life."
There seems to be substantial disagreement here, which illustrates two of the main problems with the cost/benefit analysis of smoking. How do we decide which costs and benefits to consider, and how do we measure them? Basic economics would tell us that smokers do, in fact, benefit from smoking. We know that, because they spend money to buy cigarettes and continue to smoke despite the health hazards involved. Of course, this conclusion assumes that smokers are both well informed and rational. Anti-smoking activists don't think they are. The activists cannot fathom how a well-informed, rational individual could voluntarily choose to smoke. Surely smokers don't realize how dangerous their behavior is. And if they do, surely they are pharmacologically compelled to smoke. But it does not appear that smokers are the nicotine slaves that the anti-smoking activists make them out to be.
In his book Smoking: Making the Risky Decision, economist W. Kip Viscusi reports that smokers are well aware of the risks associated with smoking; in fact, they tend to overestimate them. Furthermore, smokers are more apt to take risks in other areas of life than are nonsmokers. That indicates that the decision to smoke is just one aspect of a general attitude toward risk, not an aberration caused by the addictive properties of nicotine.
I think the issue of addiction in general has been overblown, and not just with respect to cigarettes. To say that someone is addicted to cigarettes simply means that there are significant costs involved in quitting. But as Viscusi notes, people make many decisions in life, including decisions about education, employment, marriage, and children, that are hard to reverse. That does not mean that those decisions are irrational. And to say that change is difficult is not to say that it's impossible. About as many Americans have quit smoking as continue to smoke. About 95 percent of them quit without formal treatment. Furthermore, about a quarter of the adult population continues to smoke despite the availability of nicotine gum and patches. That fact alone suggests that there's something more to smoking than chemical dependence. At the risk of seeming naive, let me suggest that the something else is simply this: smokers like to smoke. Whether they like it for the reasons Christopher Hitchens cites or for entirely different reasons, they get something out of smoking aside from an increased risk of lung cancer, heart disease, and emphysema. The Environmental Protection Agency's recent analysis of the costs and benefits of the Smoke-Free Environment Act, which would ban smoking in almost all nonresidential buildings, con- cedes that point. "Since people smoke despite the risks and costs," the report says, "one would presume that provided those people are rational, fully knowledgeable, and able to accurately assess the consequences of smoking, including potential addiction the benefits of smoking to them outweigh the risks and costs." Yet those benefits do not show up on the EPA's ledger. When they're doing their cost/benefit analysis of the nationwide smoking ban, they explicitly do not consider the forgone pleasure of smokers.
Admittedly, the benefits would be hard to measure; but without taking into account the reasons for which people smoke to begin with, how can you possibly pretend to analyze the net effect of smoking restrictions? The EPA also excludes from its analysis the inconvenience imposed on smokers who would have to go outside for a cigarette, the forgone pleasure of smokers who would eat out less often if smoking in restaurants were banned, and the loss to businesses whose employees would periodically leave the workplace for cigarette breaks. In fact, as far as the EPA is concerned, the cost of the nationwide smoking ban would be limited to enforcement and the construction of a few smoking lounges. In that cost/benefit analysis, smokers do not count. Smokers also do not count in some of the cost/benefit analyses that are used to justify increased cigarette taxes. One of the obvious effects of cigarette taxes is to increase the price of cigarettes. When the price of cigarettes goes up, a certain number of smokers will cut back and a certain number will quit. Is that good or bad? The answer depends on how you look at it. If you focus on the health care costs associated with smoking, you will conclude that a reduction in smoking saves money. But if you focus on smokers' desires, you will recognize that an increase in the cigarette tax means that some people will forgo pleasure that they would otherwise have enjoyed. That is a simplified model; we're looking at only two factors forgone pleasure and health care savings. But one of the important things to consider in weighing those two factors is, who is paying health care costs to begin with? If you had a situation in which smokers were paying the total bill for their health care out of pocket, then presumably smokers would be taking into account those costs when they chose to smoke and there really would be no issue of spillover cost.
More common is the situation in which smokers are insured by private carriers. Some insurers may charge smokers a higher premium; some may not, especially if it's group insurance provided by an employer. If smokers have to pay a higher premium, then I would argue, again, that the odds of incurring higher medical costs have already been taken into account and there really is no spillover effect. Furthermore, even if smokers are not charged more, I don't think there is any argument for government intervention, because the parties to the transaction the insurers, the employers, the smokers have all voluntarily agreed on the way they're going to handle medical costs. An insurer can choose to charge smokers more. (There has been some movement to prevent insurers from discriminating on that basis. I think that's a big mistake.)
A major problem is that the government pays some people's medical bills, so the costs of medical care do not figure into a person's decision to smoke. Anti-smoking activists argue that because it provides public health insurance, the government is justified in taxing cigarettes or using other measures to discourage smoking. Since the gov- ernment picks up the tab, it should be able to regulate behavior that generates the expense. If you follow the debates over motorcycle helmet laws or seatbelt laws or drug prohibition, you know that that sort of rationale is used for all kinds of things. It doesn't take more than a moment's reflection to see the totalitarian implications of that argument. If the government may regulate risky behavior to avoid future outlays under public health insurance, there is no end to the controls it may impose. I'm sure each and every one of you engages in behavior that could be thought to raise your risk of getting various kinds of diseases or injuries. And I'm sure that you wouldn't want to see your behavior either banned or taxed prohibitively.
The real problem here is not risky behavior per se but the decision to subsidize it with taxpayers' money. That's what people really should be resisting. Here is a drug addict, a smoker, a sky diver, a bungee jumper who has been injured or developed a disease because of behavior that she voluntarily engaged in. Why should we have to pick up the tab? If people are genuinely upset about that, they should campaign to get rid of the public health insurance programs that subsidize risky behavior.
Consider another example: antismoking activists cite studies that find that smokers miss more days at work than nonsmokers and that smokers increase maintenance costs by dropping ashes and leaving smoke everywhere. The crucial point is that employers are free to take both factors into account when they make personnel decisions. If the costs are high enough, some employers may refuse to hire smokers, or they may pay them less or give them lower benefits. There is no reason to believe that the government is in a better position to assess those costs than are individual employers. The same point, by the way, applies to employer provided health insurance. If employers believe that smokers will draw more on that benefit, they may rationally choose not to hire smokers or to make them pay more of their premiums. Many people would argue that secondhand smoke is different because others are forced to breathe it. Just as the government regulates the pollution that comes out of a factory's smokestack or a car's tailpipe, it should regulate the pollution that comes from the end of a cigarette. That analogy is fundamentally flawed. Unlike air pollution, secondhand smoke on private property is not imposed on people against their will. If you choose to eat in a restaurant, fly on an airplane, or work in an office where smoking is permitted, you thereby consent to exposure to secondhand smoke. You may not like it you may even worry that it will increase your risk of getting lung can- cer but you have implicitly decided that the annoyance and the possible risks are outweighed by the benefits of eating in a particular restaurant, flying in a particular airplane, or working in a particular office.
If, on the other hand, you refuse to enter any enclosed space in which smoking is allowed, you send a signal to airlines, restaurateurs, and employers. If there are enough people like you, some businesses will ban smoking, and other businesses will segregate smokers from nonsmokers or work out some kind of better ventilation system. Decisions about smoking rules should be left to the millions of private property owners. Do we really want the government to dictate the conditions of employment and of other contracts, or should those arrangements be left to individuals? We all lose something if we allow the government to penalize people for unfashionable habits. Personally, I don't perceive enough of a benefit from smoking to make it worth the risk. But I'm sure a lot of people, including many smokers, would have difficulty understanding why I enjoy bungee jumping. I'm not asking that other people share my tastes; I'm only asking that they tolerate them. And toleration is another casualty of the crusade against smoking.
Sheldon Richman: I am offended by people in government who think that it's their job to use television and the print media to tell me not to smoke. I have read the Constitution, and I cannot find authorization for government to hector us and pester us on issues of private conduct.
It's not as if people don't know that smoking entails some health risks. The term "coffin nail" was coined in the 19th century. People know that smoking carries risks. Those risks are not news to anyone, so I really can't see a case for government's hectoring us about it. I'm especially offended that kids are hounded in school from a very young age to take pledges that they won't smoke, to go home and bug their parents about smoking, and to report their parents' smoking and drinking habits to the social welfare agents who go to schools to talk to the kids. Government paternalism permeates society more than nicotine does, and it's much more toxic, much more pernicious.
When we talk about regulating smoking, or most other behavior, we are talking about rights and liberty and people's ability to conduct their lives the way they wish. In my view, civil liberties are always resolvable into property rights. All violations of civil liberties violate someone's right to use and dispose of his property. I'm starting with the most basic property your self, your person, your body, and your personal resourcesÚand then going on to the external things that you justly acquire. Free speech violations, for example, interfere with the right to use one's communications resources, starting with the mind and the larynx. Bona fide violations of privacy interfere with the use of one's home, or papers, or place of business. Thus, civil liberties are property rights. Government smoking regulations, by their very nature, therefore, violate civil liberties and property rights. All such regulations dictate what people may do with their property.
Now I want to say something about the word "public." The equivocation over the word is inexcusable and, I think, intentional in many cases. The Capitol is a public building, but the Cato Institute building is also a public building, in a sense, and a restaurant is a public building, in the sense that the public is able to come into it. A private home, however, is not "open to the public." But, obviously, the Capitol and a restaurant are not public in the same way. So I suggest that we use three terms for the three kinds of property we're going to be discussing: "government property," for property that's ulti- mately funded and maintained by the taxpayers; "commercial property," for property that's open to the public; and "private property," for homes. My remarks are going to be about commercial and private property.
Let's look at some violations of freedom of speech or the First Amendment in the name of making America smoke free. We've had since about 1970 or 1971 a ban on television and radio advertising, a clear violation of freedom of speech. In Baltimore, there's a proposal to restrict youth-oriented, whatever that is, billboard advertising of cigarettes. A clear violation of freedom of speech. Why is that not being immediately laughed out of court? It's compulsion.
In California and Massachusetts there are laws that require that cigarette taxes be used to fund anti-tobacco messages. The taxes sometimes go to nonprofit organizations that lobby for more restrictions. That is not so much a violation of free speech as it is an example of forced speech. People are being forced to finance speech with which they don't agree and that they may not wish to hear broadcast.
We are faced with an ever-growing list of infringements of our rights. The Occupational Safety and Health Administration is proposing a rule to ban smoking in all workplaces. Washington has become the first state to ban smoking in commercial and state buildings. In Maryland the regulator of occupational safety and licensing has imposed a workplace smoking ban that is currently being held up in the courts. Los Angeles has banned smoking in outdoor cafes which, it seems to me, is overkill. And Rep. Henry Waxman (D Calif.) has introduced the Smoke-Free Environment Act in Congress to compel owners of nonresidential buildings that are open to the public and regularly frequented by more than 10 people to ban smoking or restrict it to a separately ventilated room.
I would like to look ahead slightly. It seems to me that the nonresidential criterion is surely only a temporary distinction and that if we get a ban on smoking in workplaces, it won't be long before we begin to see regulation, perhaps creeping, of the home. Why not? Repairmen come into people's homes. Don't those workers have a right to be in a smoke-free environment? And what about children? They are more vulnerable than adults. The issue has already come up in custody cases, and John Banzhaf, head of Action on Smoking and Health, has said, "The same protection as in custody battles will eventually be extended to children in ongoing marriages through child neglect proceedings."
Our rights of association and contract are being violated as well. All-smoking flights are banned; in other words, the government won't even allow airlines to have designated flights for smokers, or set up an all-smoking airline. Now, let me point out a violation of freedom of contract on the other side. We're seeing actions against employers who want to discriminate against smokers, either refusing to hire smokers or requiring, as a condition of employment, that employees not smoke, even in their own homes. Smokers and the tobacco companies don't like that, but it is perfectly within the rights of an employer to specify terms of employment. I don't think I'd like to work under those conditions but the employer has the right to lay down that condition. Such bans have been passed, I think, in 28 states and the District of Columbia. Smokers have even gotten some protection from the Americans with Disabilities Act, because nicotine is being claimed to be addicting, and, therefore, smoking is a disability.
The Food and Drug Administration is considering regulating nicotine as a drug, which presumably will mean limits on nicotine content. That is a violation of freedom of contract between tobacco companies and customers. It's also an extension of the drug war and will have the predictable results. When we argue against drug prohibition, we like to say that people don't shoot each other over cigarettes; that happens only in the black market. Well, if the FDA gets its way, people will be shooting each other over cigarettes, the illegal high-nicotine cigarettes that you will have to buy on the street.
The bottom line in all these cases is that the government presumes to dictate the terms of association between adults. I'd like to point out that the validity of the alleged hazard of secondhand smoke is irrelevant to what I've been saying; even if secondhand smoke does present a health risk, that does not justify the regulations. The official in Maryland who presumed to ban smoking said that if nerve gas were wafting through a restaurant, it would certainly be appropriate for the government to protect people. Obviously, that's a very bad comparison. When you walk into a restaurant and take your first whiff of cigarette smoke, you are perfectly capable of leaving. Try that in a room full of nerve gas. Competition for customers and for employees may induce private no-smoking rules or less severe regulations, but that's a separate issue, having nothing to do with civil liberties except that it's within the rights of an employer to set such rules. The key point here is that there's no right to be on someone else's property on terms other than mose set by the property owner. That's really the bottom line.
In his book Cigarettes Are Sublime, Richard Klein said that life means choosing your poisons. I think that was a fancy way of saying that life is risk, that risk is inherent in life. It follows from that maxim that the rules we've been discussing interfere with a funda- mental right and civil liberty, namely, the right to choose your own poison. In 1990 the Tobacco Free America Coalition, consisting of the American Cancer Society, the American Heart Association, and the American Lung Association, published its "Blueprint for Success: Countdown 2000, 10 Years to a Tobacco-Free America." It was not a blueprint for a campaign of persuasion; no, it called for enactment of legislation to make America smoke free by the year 2000. All of that stands on the flimsy foundation of what the EPA itself calls "the a priori hypothesis that a positive association exists between exposure to environmental tobacco smoke and lung cancer."
Liberty is under assault. I am not a cigarette smoker, most people are not smokers, but it is perilous for us to ignore this assault merely because it is aimed at someone else. Let's not have to say years from now, "When they came for the smokers, I didn't speak out because I wasn't a smoker." How long before something you do is singled out by the health fascists? The cigarette may be an unfortunate symbol of today's struggle for freedom in the United States of America, but those of us who value liberty over security, and even over health, must now rally round it.
William Niskanen: The interaction of smokers and nonsmokers creates a real sodal problem. Some members of each group are offended by the behavior of some members of the other group. Many smokers are offended by the attempts of others to control their behavior. Many nonsmokers are offended by secondhand tobacco smoke. The basis for what is now traditional tobacco policy may be described as secular puritanism. Tobacco is not a new target of puritanism. At various times over the past several centuries, smokers have been subject to much more severe sanctions. What is new is that the present puritanism is supported by the secular left rather than the religious right. Contemporary secular puritans use a different public rationale and a different political coalition to promote favored policies.
The important question is whether this problem should be sorted out by social rules or by law. And I think that the real issue is not whether exposure to secondhand tobacco smoke somehow increases the probability of health problems for nonsmokers.
There would be a better case for regulation if a brief, casual exposure to secondhand smoke created severe health problems. But no one has made that claim. All of the studies that have been done are based on sustained exposure over long periods of time, within the home or within the workplace, or somewhere else.
When the per incident health risks are small, health does not trump all other dimensions of a sustained relationship. People make their choices about sustained relationships — within a workplace, within a family, within a commercial context — on the basis of a variety of considerations of which health is but one. And when the per incident health risks are small, the case for regulation disappears. If a brief, casual exposure to nerve gas can kill you, there's reason and presumption in favor of regulation. But that is not the case with tobacco smoke; nor is it the case for much of the behavior that we try to regulate.
May I conclude that courtesy, contract, consent, accommodation, and tolerance are the best guides to the rules for interaction between smokers and nonsmokers. No one has a universal right to smoke anywhere. Similarly, I contend, no one has a universal right to deny other people the opportunity to smoke. The best rule, I suggest, is that the owners of each physical space are the only people who have the right to set the smoking rules in that space.
Owners have the best incentives to set workable rules — interactions with employees, with consumers, with marriage partners, and with children or whomever. There is also no reason why the rules should be the same in all spaces. This approach necessarily involves some bargaining within the family, within the firm, within dubs, within the groups of shops in the same mall. But decentralization of the choice of smoking rules will lead to a much higher level of total consent than any government regulation that presumably reflects the preferences of the majority of the larger body of people.
The most important role for government in this approach is one that it has not performed well, and that is to provide succinct, unbiased information to guide decentralized choices among alternative rules. Let's try to sort out the rules for smoking in this way before some different majority tries to use the government to control more important matters such as chocolate, french fries, skiing, and sex.