The motives that lie behind increases in governmentally imposed burdens on smokers are no doubt many and varied. There are always a few well-meaning people who are burdened with superior wisdom about how others should like their lives. Those people have always found government force more convenient that persuasion as a means of sharing their wisdom with others. Politicians are constantly alert to opportunities to obtain more tax revenues at minimum political risk. Certainly, they have not been oblivious to the shift in public opinion against smoking, or to the fact that smokers have not only become a minority of the voting age population but a minority increasingly concentrated in the lower income classes that are less likely to vote than the more affluent.
Regardless of the real motives, however, political burdens are always easier to impose when they can be justified on grounds more noble than elitism, paternalism, and the desire for more of the public's money. For that reason, government policy recommendations are inevitably brought forth with a heavy veneer of public interest rhetoric. In the case of tougher restrictions and higher taxes on smokers, the argument is that those measures are needed to reduce the economic costs and inefficiencies that result from the use of cigarettes.
There has been a steady stream of studies in recent years that purport to show that smoking is costing the American public enormous sums. For example, Secretary of Health and Human Services Louis Sullivan recently released a study that claimed that the cost of smoking is over billion annually.[1] That figure is, for reasons that are discussed subsequently, a gross overstatement. But whether the HHS figure is accurate or not, in the absence of further information, the cost of smoking is entirely irrelevant as far as public policy is concerned. What is the cost of billiards, bowling, or sunbathing? While those may be interesting questions to merchants who sell pool tables, bowling balls, and suntan lotion, there is no public policy significance that attaches to the answers. Everything people do involves costs. As long as those who engage in an activity pay the costs, they can best assess the benefits and decide how much or how little to spend on the activity.
Only if it can be established that the full cost of a product is not being paid by those who consume it is it possible to make a reasonable case that the public interest requires imposing a differential tax on the product or restricting its use. So the relevant public policy question regarding smoking is not how much smoking costs but how much, if any, of the costs are paid by nonsmokers.
Not surprisingly, those who favor higher taxes on cigarettes and tougher restrictions on smoking claim that smokers are not paying their own way. Supposedly, smokers impose costs on nonsmokers, and indeed the whole of society, through lost productivity, greater demands on health care, and irritating side-stream smoke. The interesting things about those claims is that most people simply accept them as true. Little effort has been made to carefully examine the assertion that smokers fail to pay much of the costs of smoking or to question the automatic conclusion that that failure, even if it does exist, justifies a public policy to reduce smoking. In fact, when the evidence is carefully examined, no support can be found for the claim that smokers are not paying their own way.
Much of the confusion and misinformation that characterize the discussion of public policy and smoking can be cleared up by drawing on the insights of Ronald Coase. It was Coase who pointed out in his now-famous article on social cost (Coase 1960) that, if property rights are well specified and transaction costs are zero, the costs that result from an activity are fully paid by (or internalized to) those who are engaged in it.[2] Although the Coasian conditions for complete internalization of cost are not fully satisfied in the case of smoking, or any other economic activity for that matter, they are satisfied more completely than commonly recognized. In the case of smoking, cost internalization is sufficiently complete that government attempts to increase efficiency, by internalizing the cost of smoking through taxation and regulation, are extremely unlikely to be successful.
The problem with political attempts to reduce any externalities from smoking is that the political process itself is permeated with far greater externalities. Efficient government action, just as efficient action in the private sector, requires decisionmakers who take the full cost of their actions into account. And the fact that the Coasian conditions that lead to the internalization of costs are satisfied to a far greater degree in private settings where personal decisions on smoking are made than they are in political settings where public policy decisions on smoking are made.
Coase and Cost
The fundamental insight of Coase is surprisingly straightforward, given the impact it has had in economics,[3] and goes a long way in undermining the assertion that smokers do not pay their own way.
Coase pointed out that when property rights are well defined and transaction costs are zero, negotiations among the parties affected by the actions of any one of them will be complete and will result in everyone having to face the full cost of his actions. For example, even if I possess the property right to disturb my neighbors by playing loud music at any time of the day or night, I will face the full cost of doing so if my neighbors and I can negotiate with each other at no cost. Since my loud music imposes a cost on my neighbors, they are willing to pay me an amount that reflects that cost in return for my turning down the volume. If I refuse to do so, I am paying the full cost I impose on my neighbors by forgoing what they are willing to pay for peace and quiet.
Although the conditions that generate complete negotiations are not satisfied perfectly anywhere, they are satisfied more widely and completely than in commonly realized. It is difficult to imagine an activity in which we engage that does not impose costs on others. Yet in most cases we compensate fully those who bear those costs with the prices we pay in the marketplace.
In the case of cigarette smoking, it is obvious that Coasian negotiations through market exchange internalize all of the direct cost of smoking to the smoker. Those who incur the costs necessary to produce the ingredients that go into cigarettes, transport those ingredients to cigarette factories, manufacture and package the cigarettes, and then distribute and sell them to consumers are fully compensated by the smoker when he exchanges cash for his carton.
But what about the indirect costs that are claimed to result from smoking? It is widely believed that smokers impose costs on society in general by (1) placing disproportionate demands on the health care industry, (2) reducing the productivity of the economy with excessive absenteeism, and (3) imposing environmental tobacco smoke on others. Popular opinion notwithstanding, those supposed indirect costs of smoking are either nonexistent or they are costs that are largely, if not completely, internalized by Coasian considerations.
Costs That Go Up in Smoke
In studies that have attempted to estimate the indirect cost of smoking that are supposedly paid by society in general, rather than by smokers, by far the largest estimated costs are those associated with health care and worker absenteeism. The argument is that smokers demand more health care and miss more work than nonsmokers and, therefore, impose significant and uncompensated costs on all of us. Rarely has an argument so lacking in theoretical and empirical support been so widely and uncritically accepted. The claim that smokers create external cost in the form of medical care demands and worker absenteesim simply goes up in smoke when subjected to examination.
Over their lifetimes, do smokers demand more health care than nonsmokers? There is no compelling evidence that they do. Indeed, the evidence that does exist indicates that lifetime demand for health care is unaffected by smoking. Simulation studies that consider the health care demanded in countries under the counterfactual assumptions that no one smokes conclude that moving to a smokeless society would have no long-term effect on medical expenditures, even assuming that smoking is detrimental to health.[4] Upon reflection, that should not be surprising.
Whether we smoke or not, we all age, experience health problems, and eventually die; in the process, we demand health care. Smoking may affect the timing of the health care we demand, but there is no reason to believe that smoking affects the total amount of health care demanded.
But even if smokers do not make use of more health care over their lifetimes than nonsmokers, might it not be the case that the health care smokers demand is more likely to be subsidized than the health care that nonsmokers demand? Not if those who want to tax cigarettes and regulate smoking are correct in their belief that smoking reduces life expectancy. If it is true that smoking reduces life expectancy, then smokers are more likely than nonsmokers to demand the bulk of their medical care before the age of 65. And 65 is the age of eligibility for publicly subsidized medical treatment through the Medicare program.[5]
Worker absenteeism is the other major component of the cost that smokers are alleged to impose on others. There are statistical studies showing that smokers are absent from work more frequently than nonsmokers. For example, Rice and Hodgson (1985) conclude that smokers miss 32 percent more work than nonsmokers. Having found such statistically significant differences between the absentee rates of smokers and nonsmokers, the authors of those studies have typically concluded that smoking causes an increase in absenteeism. That conclusion is unwarranted, however, unless the study has controlled for all other factors that might affect absentee rates. For example, smokers are more likely to be found in blue-collar jobs than nonsmokers, and nonsmokers are more likely to be found in professional occupations than smokers. Since workers in blue-collar jobs miss more work on average than do those in the professions, it is not surprising that smokers miss more work on average than nonsmokers. But that information is meaningless if one wants to draw a conclusion about the effect of smoking on productivity. Consider the fact that pipe and cigar smokers (who happen to be disproportionately represented in the professions) miss less work, as a group, than do non-smokers Who would be so silly as to conclude on the basis of that information that requiring workers to smoke cigars or pipes is an effective way to reduce worker absenteeism?
In a recent study, Ault et al. (1991) have made a serious effort to examine the relationship between smoking and absenteeism by statistically controlling for such factors as age, drinking habits, sex, type of employment, and seniority. The conclusion of that study was that smoking, by itself, is not related in any statistically significant way to worker absenteeism. There is imply no empirical basis for concluding that smoking is imposing cost on society in general by causing an increase in the absentee rate of workers who smoke.
Coase in Action
For the sake of argument, we ignore the fact that empirical support is lacking for the assertions that smoking increases health care demands and worker absentee rates and ask whether those assertions, even if true, would justify government policies against smoking. As can be shown with straightforward economic analysis, the answer to that question is no. If smokers generate costs by demanding more health care or missing more work because they smoke, the Coasian conditions apply to the relevant market activity, and that activity is quite capable of internalizing the costs to those who smoke.
There is a very effective way of internalizing the cost of health care to those who make use of it, and that is to eliminate government programs that subsidize much of the health care demanded in the country. In the absence of public subsidies, the cost of providing health care would be reflected in the charges imposed on those receiving that care. There would, of course, be imperfections in a completely private health care market. But such a market would be characterized by well-defined property rights and relatively low transaction costs, so Coasian forces would ensure that the internalization of medical costs was nearly complete. Certainly, if the objective is to internalize the cost of health care, the private market performs far better than government programs.
The entire purpose of government-subsidized health care, of course, is to prevent market forces from internalizing medical costs by providing people with some minimal level of health care when they are ill, regardless of their ability to pay or of the reasons for their illness. Given that purpose, what justification can there be for singling out one group for a special tax because its members allegedly generate external costs by requiring more health care than others?
Next, assume for the sake of argument that smoking does cause a reduction in productivity. Would that mean that smokers are imposing external costs on society at large? The answer is a resounding no. The The Coasian conditions of well-specified property rights and low transaction costs are satisfied quite well in the labor market, and, therefore, there is no reason to fear that people can avoid the costs of their low productivity by imposing them on others.
Some people are more productive than others for reasons that range from natural abilities to personal preferences, and few, if any, are as productive as they could be if productivity were their sole objective. Indeed, most people would be far less productive than they are if productivity were not rewarded in the marketplace. If a person chooses to be less productive for whatever reason, that person can expect to pay for that choice with reduced chances for promotion and lower salary than realized by those who are more productive. Indeed, the reason society can be tolerant of the freedom people have in the workplace is explained by the fact that the costs and benefits resulting from the exercise of that freedom are internalized. Without Coasian internalization in labor markets, no economically viable society could tolerate freedom in the workplace. Because of the internalization, we live in a society that recognizes that individuals, not society, are the rightful owners of their labor. If a person decides to be less productive because he chooses to smoke, overeat, pursue a time-consuming hobby, or simply indulge a preference for a less stressful life, that decision is his, not society's, to make. The effect of smoking on productivity, whatever it may be, is simply not a legitimate public policy concern.
But what about the problem of environmental tobacco smoke (ETS)? Some argue that ETS can reduce the productivity of nonsmokers in the workplace, and no one can deny with credibility that many people are irritated by ETS in public establishments such as restaurants, theaters, and airplanes. But does that cost, which is seemingly a cost smokers impose on nonsmokers, warrant government policies aimed at restricting smoking? Despite the widespread view that the cost associated with ETS does justify government restrictions on smoking, the case for such restrictions is weak at best.
The problem of ETS occurs, almost by definition, where the interacting parties are in close proximity to each other. Given that proximity, people can be expected to accommodate diverse preferences in mutually advantageous ways, with subtle social pressures and common courtesy serving much the same function as explicit negotiation. In recent years, attitudes toward public smoking have changed in ways that have, in effect, shifted property rights to the ambient environment away from smokers and toward nonsmokers. That shift in property rights, along with the implicit (and sometimes explicit) negotiation that takes place between smokers and nonsmokers, has made smokers more sensitive to the ETS concerns of nonsmokers than in the past (in large part because few nonsmokers have had such concerns in the past). Certainly mutual accommodation and common courtesy are strong forces for moderating the potential frictions that exist among people who engage in a large number of potentially conflicting activities.
There are Coasian considerations, in addition to those of informal social pressures, operating to internalize the costs that may result from ETS. Consider the nuisance costs that may result from cigarette smoke in a privately owned establishment, such as restaurant. The owner of the restaurant is a residual claimant to the net proceeds generated by the restaurant. Those net proceeds are directly related to the value of the goods and services the restaurant provides to those who can, if they are not satisfied, take their patronage elsewhere. Some customers enjoy smoking while dining out, but ETS diminishes the dining experience of others. The owner of the restaurant has a strong incentive to assess the resulting benefits and costs associated with smoking in his restaurant and respond appropriately. In some cases, the appropriate response is to prohibit smoking entirely; in other cases, it is to do nothing to prohibit smoking; and in still others, it is to provide smoking and nonsmoking sections. The owner benefits from choosing the response that is best for his restaurant, and the costs of ETS are internalized to him as a residual claimant.
The same type of internalization takes place in working environments. The owner of a factory, for example, recognizes that frictions can exist among his workers for a large number of reasons, and he has every motivation to structure the work environment to accommodate those frictions at the least total cost, including the cost of the frictions. If some workers enjoy smoking at work while others object to ETS, the factors owner is the residual claimant to decisions that minimize the conflict. By appropriately weighing the costs and benefits of different policies regarding smoking in the workplace and implementing the policy that creates the greatest net value, the factory owner best serves both his interest and that of his workers.
It should be emphasized that the argument is not that residual claimants respond with perfect efficiency in internalizing the costs associated with ETS, or with anything else for that matter. But competitive pressures are constantly at work to eliminate those who are careless in accommodating the varied, changing, and often conflicting preferences of their customers and employees. And there can be little doubt that residual claimants in the marketplace do a far better job of responding efficiently to problems such as ETS than do remote government authorities who, when called into action, tend to impose uniform requirements on everyone. Simply put, private action does not have to hit the efficiency "bull's eye" in order to outperform government action.
Can Coase Perform in Public?
The economic case for public policies with regard to smoking is based on the argument that smoking generates external costs that should be internalized by government. That rationale rises a natural, though seldom considered, question: How effective is the government at internalizing cost?
When we attempt to answer that question, it is useful to consider how conducive the political setting is to internalizing the costs that result from public decisions. If the political process allows decisions to be made that impose uncompensated costs on some in order to provide benefits to others, then it is unlikely that public decisions will result in policy that efficiently internalizes the costs of private decisions. It may be possible at times to fight fire with fire, but it requires a tremendous leap of faith to believe that we can efficiently fight private externalities with public externalities.
We could expect the political processes to perform well at internalizing the cost of public decisions if there existed well-defined property rights in the political process that made it easy for all those affected by public decisions to negotiate with one another. Unfortunately, that is not the case. The right to political property (power) is not well defined; rather, it is allocated on the basis of competition for political influence. The advantages that come from political power have an ownership structure closer to that of common property resources than to that of private owned resources. Those who capture access to political power are able to exploit it to promote their objectives by imposing costs on those who lack access to that power.
Those who bear the cost imposed by the political influence of others can conceivably organize themselves for the purpose of political negotiation with those who are benefiting at their expense. Ideally, the group bearing the cost could either capture the relevant political property rights and extract compensation from those receiving the benefits or offer to pay the benefiting group to reduce its political demands. In either case, such negotiations would internalize the cost of the political benefits to those receiving them. Unfortunately, the cost that results from the exercise of political influence is commonly spread so thinly, and over such a large group, that organizing the group to take the type of collective action required for effective political negotiation is extremely difficult. Negotiation and compromise are, of course, ubiquitous features of politics and the political process. But, such negotiation and compromise generally occur among well-organized interest groups that decide just how much political advantage each will receive at the expense of a large, and unorganized, segment of the public. That is not Coasian negotiation of the type that internalizes costs. Rather, it is negotiation over who gets to benefit most from externalizing cost.
Although relatively small groups with an overriding purpose are far easier to organize than large groups with no common purpose, there are difficulties associated with organizing any group for political action. From the perspective of an individual member of a group, the connection between his or her political involvement in support of the group's objective and the benefits received from that involvement is tenuous, and political apathy is the usual response.[6]
The attraction of political apathy is strong and not overcome easily. People seldom become politically exercised over the prospect of obtaining an incremental increase in some commonplace advantage. The most effective way to galvanize people into an active political group is by appealing to their sense of passion and virtue. Noble causes, moral crusades, nonnegotiable demands, and strong emotions fuel successful political movements. Such movements typically require an enemy whose evil designs are seen as a threat to that which is righteous. Those who stand for virtue cannot think of cooperating with the enemies of virtue to advance the interests of both. The only acceptable objective is to vanquish evil regardless of the cost to either side. The greatest political mileage is achieved with arguments that minimize the acceptability of tradeoff, negotiation, and Coasian compromise.[7]
Certainly, smoking has been politicized by those who see it as an evil and government sanctions as the most convenient way to discourage it. That politicization has not served to promote the Coasian interaction between smokers and nonsmokers that could serve the interest of both. Smokers are a large and diverse group with no overriding common concern. The sheer number and diversity of smokers explain why they are far less organized, if organized at all, than such groups as the American Heart Association, the American Lung Association, and the American Medical Association. Those groups consist of relatively small numbers of people who are motivated by focused objectives that are promoted by political action against smoking.[8] The ability of such antismoking groups to obtain their objectives through their organized political influence, coupled with the political impotence of smokers, rules out any expectation that Coasian accommodation between antismokers and smokers can be realized through the political process.
It can be argued, of course, that smokers are represented by the political activities of the tobacco industry, which is able to organize for political action. Without embracing the proposition that the advantage of smokers would be advanced by the success of all political initiatives of the tobacco industry, it can be said that there is a significant amount of correspondence between the interests of smokers and those of the tobacco industry. The tobacco industry favors lower taxes on tobacco products, fewer governmental restrictions on where those products can be used, and more freedom to advertise (provide information on) them. But the ability of the tobacco industry to organize for political action, and the overlap between its interests and those of smokers, does little to facilitate Coasian interaction between smokes and antismokers in the political arena.
For the reasons previously discussed, the groups that oppose smoking realize the political advantage of characterizing their opposition as a noble crusade against evil. Groups whose interests are advanced by antismoking policies are best able to mobilize political support for those policies by presenting smoking as evil and tobacco companies as profiteering promoters of evil. The rhetoric of moral crusaders best describes the pronouncements of those who want "to make the dream of a smoke-free society come true" (Koop 1988, p. 11). Examples of such rhetoric are plentiful: The tobacco "industry has been fighting a war against the public" (White 1988, p. 232). "Like a cancer that has metastasized, the cigarette companies have spread throughout the American economy" (White 1988, p. 232). "How has such an unmitigated evil as tobacco become so entrenched in our society?" (Lee 1989). "[The tobacco industry] should be treated like the Medellin [drug] mafia" (Perschuk 1989). "Tobacco company executives are mini-Adolph Eichmanns" (Daynard 1987). "Any [favorable] advertising is inherently deceptive of a product [tobacco] ... which is guilty of all these sins" (Lukens 1990, p. 15). "People who work for tobacco companies have no reservations about killing people" (McNutt 1990).
Rhetoric such as that, and the emotional and moral certitude it evokes in the faithful, all but eliminates any possibility for mutually advantageous give and take. With little scope for Coasian compromises between competing interests, the political interaction between smokers (as represented by the tobacco industry) and antismokers becomes a negative-sum struggle. The only way the interests of committed antismokers can be advanced is at the expense of the tobacco industry and smokers. When a decision is made that goes against the antismokers, they feel not just disappointment and frustration but a righteous indignation at the evil that has been allowed to prevail. The process can, as demonstrated by the rhetoric of the most committed antismokers, degenerate into a malicious exercise in which the major preoccupation becomes inflicting harm and ridicule on the enemy.
Comparing the private interaction between smokers and nonsmokers in the marketplace with their public interaction in the political arena leads to the conclusion that Coase performs far better in private than in public. Indeed, in the case of smokers and nonsmokers, the reasonable conclusion is that Coase refuses to perform in public.
Conclusion
Individuals can be depended on to pursue their own advantages as they see them. Unfortunately, in our world of scarce resources and diverse preferences, the advantage of each is always in potential conflict with the advantage of others. The only hope for a productive and harmonious social order is in social institutions that reveal to people the costs their activities impose on others, and then motivates them to take those costs into consideration. Fortunately, the institutions of private property and markets generate exactly that type of revelation and motivation. With private property and markets, the Coasian condition of low transaction costs is satisfied sufficiently. The costs generated by activities are internalized to those who benefit from engaging in them. Perfection is, of course, an impossible dream, but for most activities, most of the time, the guidance provided by Coasian cost internalization generates private interaction, which is amazingly harmonious given the potential for conflict that exists in all human activity.
Regardless of how well private incentives internalize costs and harmonize social interaction, it is always possible for some to enhance their well-being by overriding private choices with political choices. Indeed, the advantage often seen in political action derives from the fact that private choices are being disciplined effectively by Coasian cost internalization. Such cost internalization, while promoting an efficient compromise among many competing interests, invariably leaves it possible to further advance the interests of some by allowing them to impose uncompensated costs on others. Often a group that has enough political influence can do just that by politicizing an issue. For example, by politicizing the issue of smoking, those who oppose smoking are able to promote their objectives by using the government's regulatory and taxation powers to impose uncompensated costs on those who smoke.
Private-sector incentives are doing a far better job of internalizing the costs that smokers generate than most people recognize. Furthermore, the same private-sector incentives also internalize the cots created by those who desire a reduction in smoking. In a Coasian setting, cost is reciprocal, and just as smokers have to accommodate the desires of nonsmokers, so nonsmokers have to accommodate the desires of smokers. Obviously, nonsmokers would prefer a less reciprocal arrangement in which all of the accommodation was required of smokers. And that is just the arrangement that the more adamant nonsmokers are attempting to realize by politicizing the issue of smoking and turning it into a moral political crusade for a "smokeless society." By pitting government power against Coasian incentives, that attempt decreases both the degree to which costs are internalized in social interaction and the civility and harmony with which that interaction takes place.
[1] See U.S. Department of Health and Human Services (1990).
[2] The typical statement of the Coase result (commonly referred to as the Coase theorem) is that when transaction costs are zero, efficiency is guaranteed regardless of how property rights are assigned. Since efficiency is guaranteed regardless of how property rights are assigned. Since efficiency requires that costs be internalized to the relevant economic decisionmakers, the Coase result can also be stated in terms of internalizing costs.
[3] The impact of Coase has been particularly important in the application of economic analysis to improve our understanding of the development and consequences of the law.
[4] See Leu and Schaub (1983) and Stoddart et al. (1986). Also see Tollison and Wagner (1988, chap. 4). It is true that there would be a gain in terms of present value if a reduction in smoking delayed medical expenditures. That, however, would be a transitional gain that would soon disappear.
[5] It can be argued that smokers impose costs on others through higher health insurance premiums. If that were a problem, however, insurance companies would have an incentive to charge higher rates for smokers than for nonsmokers. If they choose not to do so, then smoking is considered no different from any other category of behavior that some believe to be risky.
[6] The classic statement of that temptation to be a free-rider in a group setting is given by Olson (1965).
[7] For a more extended discussion of the contrast between political and market incentives as a means of avoiding such social hostility, see Lee (1982).
[8] Although groups politically organized around public health issues find it convenient to present their activities as noble quests for the common good, there are reasons for believing that narrow private advantage is an active motivating factor behind public health--and antismoking--lobbying. See Tollison and Wagner (1991) and Bennett (1990).
References
Ault, Richard, et al. "Smoking and Absenteeism." Applied Economics 23 (1991): 743-54.
Bennett, James T. Health Research Charities: Image and Reality. Washington, D.C.; Capital Research Center, 1990.
Coase, Ronald. "The Problem of Social Cost." Journal of Law and Economics 3 (October 1960): 1-44.
Daynard, Richard. Quoted in Lawrence Ingrassia, "Tobacco Industry Foe Daynard Wants to Snuff Out Cigarette Sales to Youth." Wall Street Journal, 20 April 1987, p. 27.
Koop, C. Everett. "Foreword." In Larry White, Merchants of Death: The American Tobacco Industry, pp. 9-11. New York: Beech Tree Books, 1988.
Lee, Dwight R. The Political Economy of Social Conflict of Malice in Plunderland. International Institute for Economic Research, Original Paper 36, February 1982.
Lee, J. D. Address to Association of Trial Lawyers. Reported in Andrew Blum, "Anti-Smoking Cause Gets Infusion of New Blood." National Law Journal (30 October 1989): 1, 38-39.
Leu, Robert E., and Schaub, Thomas. "Does Smoking Increase Medical Care Expenditure?" Social Science and Medicine 17 (1983): 1907-14.
Lukens, Thomas A. Hearing before the U.S. Senate Committee on Labor and Human Resources. Tobacco Product Education and Health Protection Art of 1990. 101st Cong., 2nd Sess., 3 April 1990.
McNutt, Kristen. "Tobacco Ownership of Food." Food Engineering, March 1990, p. 31.
Olson, Mancur. The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge: Harvard University Press, 1965.
Pertschuk, Michael. Quoted in Alix M. Freedman, "Philip Morris to Launch Image Ads." Wall Street Journal, 1 November 1989, p. B1.
Rice, D. P., and Hodgson, T. A. "Economic Cost of Smoking: An Analysis of Data for the United States." Paper presented at the Allied Social Science Association Annual Meetings. San Francisco, December 1985.
Stoddart, Greg L., et al. "Tobacco Taxes and Health Care Costs: Do Canadian Smokers Pay Their Way?" Journal of Health Economics 5 (1986): 63-80.
Tollison, Robert, and Wagner, Richard. Smoking and the State. Lexington, Mass: Lexington Books, 1988.
U.S. Department of Health and Human Services. "Smoking and Health: A National Status Report--A Report to Congress." DHHS Publication No. (CDC) 87-8396. Washington, D.C., February 1990.
White, Larry. Merchants of Death: The American Tobacco Industry. New York: Beech Tree Books, 1988.
