Tobacco Documents Online

ANDREWS OFFICE PRODUCTS CAPITOL HEIGHTS, MD (K)

DOCKET OFFICER DATE AUG 12 1994 TIME 41: ' " n AA - Comments on U.S. Department of Labor Proposed Rule on Indoor Air Quality in Federal Register, Vol. 59, No. 65, April 5, 1994, pp. 15968-16039 Docket No. H-123 August 12, 1994 By W. Kip Viscusi George O. Allen Professor of Economics Duke University Department of Economics Box 90097 Durham, NC 27708-0097 Telephone: (919) 660-1833 Faxi (919) 684-8974 TIBR 0007358

Professional Backaround I have been retained by Covington & Burling to provide an analysis of the proposed OSHA rule for indoor air quality. My professional career has been devoted to an analysis of the economics of risk-related issues. I am currently the George G. Allen Professor of Economics at Duke University and have published 15 books and over 150 articles, many of which deal with the issues addressed in this rulemaking. For example, many of these books focus on job safety. One of my recent books and several recent articles focus on cigarette smoking. I am the founding editor of the Journal of Risk and Uncertainty and currently serve on seven other editorial boards, including the American Economic Review. I am also quite familiar with the rulemaking process. During the Carter Administration I served as the Deputy Director of the President's Council on Wage and Price Stability, which was responsible for all White House oversight of major new government regulations. In that position, I also served as a member of the Regulatory Analysis Review Group. During the 1980s I had several contracts and purchase orders with the U.S. Department of Labor's occupational Safety and Health Administration and with the Office of the Secretary of Labor, in which I prepared studies focused on job safety issues. One of my efforts in this regard involved being called in to settle a dispute between the Occupational Safety and Health Administration and the U.S. Office of Management and Budget over 2 TIBR 0007359

the OSHA hazard communication rule. My analysis of this rule, which supported issuance of the regulation, was credited by both the New York Times and the washington Post as being instrumental in leading to the promulgation of the OSHA hazard communication standard. I also served as a consultant under contract in the preparation of several analyses that formed the substantive basis for the regulatory impact assessments submitted by OSHA as part of the rulemaking process. I have also served as a consultant or as a contractor to a wide variety of Federal regulatory agencies and to the U.S. Office of Management and Budget. I currently serve on two EPA Science Advisory Boards. For further information on my background, I have attached a copy of my C.V. and bibliography. Overall Comment I have been retained by Covington & Burling to focus primarily on the environmental tobacco smoke (ETS) aspect of the proposed indoor air quality standards. Although I will focus on this aspect of indoor air pollution, I will also offer some comments that are more general in nature. Overall, I believe that OSHA has not prepared a sufficiently thorough analysis of the benefits and the costs of the proposed regulation. The benefits are overstated, and many of the critical cost components have been ignored. Perhaps most important is that even from the rather narrow perspective of whether this regulation addresses a significant risk, there is no rationale for the regulation. The potential ETS risks are highly 3 TIBR 0007360

speculative and, even if they are at the levels claimed by OSHA, they may not even be sufficiently significant to warrant regulation. The proposal also has a number of serous methodological problems. 1. OSHA did not quantify the benefits and costs of the reaulation in a comprehensive manner. In effect, the supporting analysis treated detailed assessment of the benefits and costs as being beyond the scope of what was necessary for a proposed rule. The discussion of the OSHA Benzene case (see page 16000) suggested that all that must be shown is that there is a°significant risk of harm" in the workplace and that "the new standard is reasonably necessary to reduce or eliminate that risk." OSHA is required to quantify the benefits and costs of regulation under Executive Order 12866 governing the regulatory oversight process. Moreover, the existence of a significant risk alone cannot provide a sensible basis for policy. Ultimately some balancing is required. Taking the logic of OSHA's policy approach to another context, it would be desirable to abolish the construction industry because doing so would eliminate truly significant risks. There would, of course, be costs associated with this, but few could argue, given the inherently dangerous nature of construction activity, that banning such activity would be "reasonably necessary" to eliminate these risks. OSHA does not apply this narrow logic to risks throughout the economy because doing so would not be a sensible approach to 4 TIBR 0007361

regulation. Throughout its history, OSHA has attempted to achieve some kind of informal balancing of the costs of a regulation and the benefits derived from it. In some cases this balancing comes in the form of discussions of "affordability." In other instances, the focus may be on employment effects, which may serve as a signal that the costs involved are too large. The position that I will take here is that what is "reasonably necessary" depends on the costs associated with the regulation and not simply the risk. The ultimate test for any policy is that the benefits that it provides to society exceed the costs that it imposes. These effects may not all be quantifiable in dollar terms, but unless the regulation is in fact improving societal welfare, then one would be hard-pressed to argue that it is °reasonably necessary" to pursue such a course of action. 2. ;s the ETS risk significant? In justifying the ETS indoor air regulations, OSBA focuses primarily on whether the risk is significant. The significance of the risk provides the primary impetus for issuing the regulation. The discussion supporting the significance of the risk draws upon the OSHA Benzene decision in which the Supreme Court indicated that a one in a billion risk from drinking chlorinated water would not be considered significant, but a 1/1,000 risk from gasoline vapors would be significant. In that regard, OSHA 5 TIBR 0007362

has calculated that the risk from ETS is 1 in a 1,000 (page 16001) and consequently qualifies as being significant. It is helpful to consider this example in detail since it should not be taken at face value. Because OSHA has used a lifetime risk perspective, I will adopt a similar approach here. The amount of water people drink per day from different sources ranges from 2.1 - 2.9 quarts.' To be conservative, I will assume that people drink nine glasses of chlorinated water per day where this may come, for example, from sodas or other products. An individual who drinks 9 glasses per day each year for 70 years will drink 229,950 glasses in his or her lifetime. If the risk per glass is 1 in a billion, as hypothesized by the Court, the lifetime risk is 2/10,000. Now let us consider ETS. OSHA estimates that between 144 and 722 people will die from lung cancer (page 16000) each year because of ETS. If the base population of 74 million nonsmoking American workers exposed to ETS (page 16001) are exposed over their entire 40 year worklife expectancy,' the lifetime risk ranges from 1/10,000 to 4/10,000. Thus, the risk of drinking chlorinated water falls between the two bounds of the risk range estimated by OSHA for ETS. Overall, when translated into lifetime risks as opposed to risks from a particular exposure so that both the ETS risks and the risks being discussed by the 'See Food and Nutrition Encyclopedia (1994). ZThe worklife expectancy is 39.1 years. See BLS Bulletin 2254. Even for a 70 year worklife, the ETS risk range claimed by OSHA is not much greater, as it is 1/10,000 - 7/10,000. 6 TIBR 0007363

Court in the Benzene decision would be on the same time frame, we find that the risks are quite comparable and are of the same general magnitude. Even if we take OSHA's risk assessments at face value, the assessed risks are in the range for which the Court has indicated that "the risk clearly could not be considered significant." As I will note below, the risks associated with ETS appear to be much smaller than OSHA has claimed, if indeed there are any risks at all. 3. The ETS risk assessments are hiahly soeculative. What is relevant for the purposes of the expected effects of the regulation are the expected mean effects of ETS, not the upper bound. Moreover, it is the consensus estimate, not the worst case, that is pertinent. If one is applying classical statistical methods, as has been the case in past analyses of ETS, the standard procedure is to use 95 percent confidence intervals for these tests. All but one ETS study fails to meet this test of statistical significance. Although there continues to be considerable debate about the magnitude of the ETS risks, it is noteworthy that a recent report by the Congressional Research Service concluded that the risks involved remain highly speculative.' Indeed, even the comments by OSHA in the Federal Reaister indicate that our knowledge of ETS risks is highly imprecise and the apparent magnitude of these risks appears to be quite small. For example, in the Federal 'See Appendix A of Jane Gravelle and Dennis Zimmerman (1994). 7 TIBR 0007364

Register discussion OSHA notes that "...in passive smokers these risks are not so easily demonstrated" (page 15976). The discussion continues, as there is later the claim that "recent evidence suggests that passive smoking has subtle but statistically significant affects on the respiratory health of adults" (page 15977). These statements reflect both the imprecision of our knowledge as well as the absence of firm statistical evidence at the usual confidence levels for judging statistical significance. Other submissions no doubt will explore these issues in much greater detail, but for the purposes of obtaining a comprehensive economic assessment one should note that OSHA's risk assessments cannot be accepted at face value because of the absence of reliable data. 4. The overall costs of the indoor air quality standard will lead to adverse mortality effects. OSHA estimates that the annual cost of compliance with the indoor air quality standard will be $8.1 billion. For reasons that will be detailed below, this estimate is likely to be low, but for the purposes of the discussion here this estimate provides a useful reference point for discussion. Considerable recent economic literature` has documented the link between regulatory expenditures and health. In effect, regulatory expenditures make society poorer because these °See, for example, the entirety of vol. 8, no. 1(January 1994) of the Journal of Risk and Uncertaintv. 8 TIBR 0007365

expenditures represent a real opportunity cost to society. Society could allocate the cost of regulation, including lower wages, higher prices, and related impacts to other uses, such as more expenditures on medical care, food, housing, and other mortality-related expenditures. Based on the direct relationship between mortality and income, the U.S. Office of Management and Budget's economists Randall Lutter and John Morrall have estimated that for every regulatory expenditure of $9 million-$12 million there is a loss of one statistical life.s Using a quite different methodology linked to the labor market estimates of the value of life, I have estimated that for every regulatory expenditure there is the loss of one statistical life for every $50 million in regulatory costs.s For the purpose of this assessment, I will consequently use $10 million per statistical life as the lower bound and $50 million per statistical life as the upper bound for the analysis. Using these figures and the $8.1 billion cost estimate provided by OSHA, the indoor air quality regulation will lead to the loss of between 162 and 810 lives per year. By almost any standard, this is a substantial impact. Moreover, if one adjusts the risk assessments OSHA prepared for the proposed regulation to reflect the expected effects that the regulation is likely to have, as opposed to a more speculative worst case scenario, then the regulation may lead to more lives lost than will be saved. SSee Randall Lutter and John F. Morrall (1993). 6See W. Kip Viscusi (1994). 9 TIBR 0007366

Although a careful analysis may suggest that the regulation will not have counter-productive effects on mortality, what these estimates do suggest is that the assessment of the benefits and the costs of regulation must be more thorough and accurate than those that have been prepared. It is essential to obtain unbiased assessment of the true expected risks, the value of the risk reduction to society, and the associated costs of the regulation. The current analysis provides only a sketchy treatment of these components. In effect, this analysis takes as a given that society should proceed with the regulation provided that the agency can show that there is some significant risk reduction. What such a calculus neglects is that there may be other risk effects of the regulation that are adverse and may in fact outweigh the purported beneficial impacts. 5. The cost estimates defy economic logic and are not credible. In its assessment of the indoor air quality regulation, 03HA breaks out three different classes of effects. First, there will be health benefits to workers, which presumably should have some value to them and would be reflected in compensating wage differentials. Second, the regulation will have direct costs, which the agency estimates to be $8.1 billion annually. Third, the regulation will have cost savings, which the agency estimates to be $15 billion annually (page 16002). For the purposes of this discussion, let us neglect the health effects and focus on the cost savings. What this analysis 10 . TIBR 0007367

suggests is that the regulation would cost businesses $8.1 billion, but would save them $15.1 billion annually. Firms' failure to take advantage of this opportunity to make almost $2 for every dollar expended cannot be attributed to the market imperfections alleged by OSHA. These market imperfections, such as an assumed lack of worker information about the risks of ETS, pertain to the value of the health effects and their transmittal through the compensating differential mechanism. Instead, what we have is an assumed market failure on the part of firms, which fail to realize that they could make $15.1 billion more every year in return for an annual expenditure of $8.1 billion. There is no reason why firms would not take advantage of such an opportunity, if it in fact existed. The major implication of these numbers is not that there is a large market failure that merits government intervention. Rather, it is that the components of the OSAA regulatory analysis simply do not make economic sense. This lack of internal consistency calls into question the overall legitimacy of all the cost estimates that have been put forth. 6. Market forces will lead to the adoption of smoking restrictions and smoking policies that are desirable. Individuals who smoke in the workplace will generate a perceived environmental risk to their nonsmoking workers. If nonsmokers believe that there is a risk associated with ETS, they will demand a compensating differential to work in such environments. Firms will have an incentive to institute policies " TIBR 0007368

to limit workplace smoking or to provide smoking areas in much the same manner as they would to control other public aspects of the workplace environment, such as the level of noise and noxious fumes. In effect, the compensating differential mechanism gives firms a financial incentive to carry out the bargains that are appropriate given the perceived risks that are being generated. The OSHA proposal recognizes that these mechanisms are pertinent (page 16008) but discusses various market forces that could impede these mechanisms. Much of this discussion of market imperfections draws upon analyses that I first advanced in the economic literature. For example, the discussion of marginal workers and worker mobility on page 16008 first appeared in my book, Emnlovment Hazards: An Investigation of Market Performance (Cambridge: Harvard University Press, 1979), and in my article, "Unions, Labor Market Structure, and the Welfare Implications of the Quality of Work," Journal of Labor Research, Vol. 1, No. 1, Spring 1980, pp. 175-192. However, this discussion of market imperfections is much more pertinent to dimly understood risks than to smoking. The critical link in all these arguments is information. The discussion of the proposed rule correctly notes that in general workers do not always have "perfect knowledge of job risks." In this case, the risk involved is that of smoking. My past research on the perception of smoking risks indicates that throughout society there is not an accurate understanding of these risks. However, the nature of the bias is the opposite of 12 TI$g 0007369

the direction that would be needed to have a market failure that warrants government intervention. As I document in my book, Smoking: Makina the Riskv Decision (New York: Oxford University Press, 1992), in 1985 the average adult in the population assessed the risk of lung cancer from smoking to be 0.43, whereas estimates based on reports of the Surgeon General at that time placed the risk between .05 and .10. Even more recent official estimates of the lung cancer mortality risk are only as high as .06-.13, so that the perceived risk of lung cancer greatly exceeds the risk value that is assessed by the U.S. Surgeon General. To test the sensitivity of this result, I have explored other measures of the risk as well. In particular, I have examined the overall total smoking- mortality risk perc=ption in society, which I found to be 0.54. In contrast, the overall mortality risk to the smoker based on estimates by the U.S. Surgeon General ranges from 0.18 to 0.36. Thus, even in the case of the total mortality risks of smoking, there is substantial overestimation of these risks on the part of the population. A third measure of the risk perception pertains to the loss of life expectancy involved. In particular, do people fully understand the extent of the loss of life that the Surgeon General assesses based on smoking behavior? In the case of public perception regarding loss of life, one finds a result that parallels attitudes about the alleged risks of smoking: the overall assessed life expectancy loss is 11.5 years, whereas the 13 TIBR 0007370

estimated life expectancy loss based on reports by the u.s. Surgeon General range from 3.6 to 7.2 years.' What distinguishes the potential risks of smoking from other health hazards in the workplace is that they have been highly publicized. Indeed, the potential hazards of smoking are among. the most widely publicized risks in society. I have documented the substantial publicity given to smoking in Viscusi (1992). The current public debate over smoking and ETS involves prominent officials from the U.S. EPA, the U.S. Occupational Safety and Health Administration, and the FDA. Most workers would certainly be acutely aware of this debate. Indeed, OSHA cites evidence indicating that "88 percent of nonsmokers are aware of the negative health consequences of ETS" (p. 16007). These results are inconsistent with OSHA's view that the ETS risks are ignored. The opposite problem may in fact be the case. Unfortu- nately, the practical effect of the substantial publicity given to ETS is not necessarily the fostering of accurate risk perceptions. Literature on the economics and psychology of risk perception clearly documents that highly publicized risks tend to be overestimated. The potential hazards of smoking are among the most highly publicized and widely discussed risks in our society; in consequence, it is not surprising that the net result is that these risks are currently overassessed. What this overassessment implies from the standpoint of compensating differentials is that the market response to the 'See page 80 of W. Kip Viscusi, Smoking. 14 TIBR 0007371

risks of smoking will in fact be excessive. Rather than leading to too little market accommodation of the preferences of nonsmokers, this over-perception of the risk will lead to excessive restrictions on smoking in the workplace. Thus, the actions that have been undertaken may already be too stringent from the standpoint of their overall social desirability. As is indicated in the discussion of the proposed rule, the private market agreements have already been quite extensive in this area. OSHA summarizes the nature of these actions (page 16007): "A 1991 survey of company smoking policies shows that of the 85 percent of firms with smoking policies, 34 percent have complete bans and another 34 percent prohibit smoking in all open work areas. Over 90 percent of non-manufacturing establishment have smoking policies [H-030EX.77]. "Workplace smoking policies are more common in larger [than in smaller] businesses. In a survey of personnel managers, 63 percent of those with 1,000 or more employees report having a smoking policy compared with 52 percent of companies with fewer employees." Thus, we have not only a substantial role of smoking policies but also substantial heterogeneity of smoking policies across the labor market. One would expect, for example, that smoking policies would be more prevalent for large workplaces. There should be economies of scale in providing smoking areas in such work environments and also a greater need to standardize smoking policies as opposed to letting the voluntary discussions of small worker groups address the appropriate smoking policy on a decentralized basis. 15 TIBR 0007372

Another class of market imperfection that is discussed in the OSHA proposal (page 16008) pertains to the lack of "perfect employee mobility between jobs." This limitation pertains not only to potential ETS risks but also to every workplace hazard. If there were in fact worker immobility, one could use this rationale to justify the regulation of any type of workplace risk in any industrial situation. However, even though worker mobility is not perfect it is indeed substantial. There is little lifetime commitment to jobs in the U.S. economy. Moreover, I have documented in my past work that job hazards alone account for 1/3 of all manufacturing quit rates.e The textbook discussions of a coal mine worker trapped at a single firm in a company town are apocryphal and do not pertain to the 1990s. Moreover, even if there were a problem of worker mobility in which a worker began work on the job, subsequently discovered that there were hazards associated with it, and felt unable to quit because of the transactions cost involved, this stylized scenario would not fit the potential risks associated with ETS. The presence of ETS and the ventilation conditions of the workplace are readily observable characteristics that can be monitored by the worker. They are not the kind of dimly understood health risks for which there is associated the usual type of market failure. BSee W. Kip Viscusi, Risk By Choice: Regulating Health and Safety in the Worknlace (Cambridge: Harvard University Press, 1983). 16 TIBR 0007373

7. The discussion of societal externalities of smoking claims that smokers on balance impose a burden on societv, whereas in fact the opoosite is the case. Smokers pay substantial taxes specifically linked to their smoking behavior. For the fiscal year ending in June 30, 1993, federal taxes on cigarettes were $5.5 billion and state taxes were $6.2 billion. Municipal taxes were a total of $187 million, so that the total tax amount was $11.9 billion. Potentially offsetting these taxes are the effects claimed by OSHA on Social Security, welfare, Medicare, and related programs (page 16008). However, even taking the estimated hazards of smoking at face value, on balance the savings to society from the estimated early mortality of smokers offset the added cost to society in terms of higher health costs. At reasonable rates of interest, such as a 3 percent real rate of return, on balance smokers more than pay their own way. Indeed, the tax payments they provide is an additional bonus since even without taxes smokers on balance save society money. This result has been documented in a recent study by the Congressional Research Service as well as in a recent book published by the RAND Corporation.' Much of this discussion of societal externalities pertains not to the risks of ETS but rather to the risk to smokers who will be associated with the dominant external costs. This class of issues involves a much larger debate over the societal effects 9See Jane Gravelle and Dennis Zimmerman (1994) and Willard Manning, et. al., (1991). 17 TIBR 0007374

and, in particular, the role of smoking outside the workplace. Because the OSHA rule is not primarily directed toward that issue but rather toward workplace smoking, this aspect of OSHA's analysis is quite appropriately not well-developed. 8. The most important cost component, the decreased welfare of smokers and the societal loss from smoking restrictions, has been completely ignored in this analysis. As in other consumer contexts, individuals who smoke spend money on cigarettes and other tobacco products because the economic value they derive from these products exceeds their cost. In particular, they reap a welfare gain from smoking behavior. Similarly, there is a benefit to companies as well, who earn profits from producing tobacco products. There are also widespread benefits to society in terms of increased employment and tax revenues. None of these effects on either cigarette smokers or the tobacco industry has been recognized in the analysis. If smoking is banned in the workplace, for eight hours of the day, or approximately half of their waking hours, smokers will be unable to smoke. If smokers are relegated to a specific smoking area, their welfare will also be decreased and their productivity may be affected as well. The decreased consumption of cigarettes not only will impose a welfare loss on smokers, but would also lead to a reduction in the consumption of tobacco products overall. From the standpoint of a comprehensive economic analysis, it is important to 18 TIBR 0007375

calculate the lost producers' surplus (loss to companies) as weli as the loss in consumer surplus. The first matter that will be considered is the magnitude of the lost consumers' surplus from a decrease in the demand for cigarettes stemming from limitations on smoking in the workplace. From an economic standpoint, the consumer surplus that an individual consumer reaps from consuming a commodity is the difference between the price the consumer would be willing to pay for the good and the price that the consumer is actually charged. To estimate the amount of the consumers' surplus for the market as a whole, the main critical component is the shape of the consumer demand curve, or more specifically the demand elasticity, which gives the percentage change in the quantity of the good purchased that will result from a unit percentage change in its price. Table 1 presents estimates of the lost consumer surplus for five different demand elasticities ranging from -0.2 to -1.4. This range of estimates spans the estimated range in the literature, which I have surveyed in my book, Smoking: Making the Risky Decision (see especially pages 102-105). Most of the demand elasticities tend to be clustered in the range from -0.4 to -1.0 so that the elasticity of -0.7 can be viewed as the midpoint of the estimated range. The 1994 Congressional Research Service study (p. CRS-27) focused on a population-wide elasticity of -0.31. Teenagers appear to be most responsive to cigarette prices, as they exhibit demand elasticities of -1.4. What is 19 TIBR 0007376

TABLE I LOSSES FROM SMOKING RESTRICTIONS Panel A: Loss of Consumers' Surplus Post-Ban Consumers' Surplus Loss of Consumers' Surplus (billion $) from ban (billion S) Assumed Smoking Reduction Assumed Smoking Reduction Pre-Ban Billions of Average Consumers Packs Sold Price Surplus Elasticity (1993) (1993) (billion S) 10Yu 20% 30"/0 10°/a 20% 30% -0.2 25.1644 1.693 $106.51 $86.27 $68.17 $52.19 $20.24 $38.34 55432 -0.4 25.1644 1.693 $53.25 $43.14 $34.08 $26.09 $10.12 519.)7 $27.16 -0.7 25.1644 1.693 $30.43 $24.65 $19.48 $14.91 $5.78 $10.96 $15.52 -1 Li 25.1644 1.693 $21.30 $1725 $13.63 $10.44 $4.05 $7.67 $10.86 -1.4 25.1644 1.693 51522 $1232 $9.74 $7.46 $2.89 $5.48 57.76 Panel B: Loss of Producers' Surplus Fortune 500 Forbes SW Note: Profits are for U.S. sales only. Assumed Smoking Reduction Loss of Producers' Surplus 20% $0.18 $0.23 30% $0.27 $0.35

Panel C; Loss of Tax Revenue Loss of Revenues Assumed Smoking Reduction Tax Revenue (Billion 1993) IO°/a 20X 30% Federal $5.53 $0.55 $1.11 $1.66 State $6.18 $0.62 $1.24 $1.95 Local $0.19 $0.02 $0.04 $0.06 Total $11.89 S1.19 $2.38 $3.57

pertinent for the purposes of our calculations is the overall demand elasticity from the standpoint of the market, for which most studies yield results closer to the middle of the table. For illustrative purposes, I will focus on the demand elasticity of -0.4. At that demand elasticity, before the enactment of the OSHA regulations consumers would reap a surplus of $53 billion annually. In effect, smokers would be willing to pay $S3 billion more for cigarettes than they are actually charged. The loss in consumers' surplus depends on the effect of the smoking restrictions on the level of smoking. The estimates in Table 1 pertain to three different scenarios, where restrictions reduce the total consumption of cigarettes by 10 percent, 20 percent, and 30 percent. Focusing on the midpoint of this range, one has a consumers' surplus after the smoking reduction of $34 billion, leading to a total consumers' surplus loss of $19 billion annually. For demand elasticities that indicate less responsiveness to price, such as -0.2, there will be a larger estimated consumer surplus loss, and for demand elasticities that indicate greater responsiveness to price, such as the demand elasticity of -1.4, there will be a smaller welfare loss. The second party that loses because of the decrease in smoking is the tobacco industry or, more specifically, the stock- holders in that industry. Panel B of Table 1 reports on a series of calculations that assume for simplicity that profits are proportional to sales. A 20 percent reduction in cigarette 20 TIBR 0007379

consumption would lead to a loss in profits of approximately $0.2 billion per year. This calculation excludes the lost profits to tobacco farmers and other groups whose economic well-being is dependent on the tobacco industry. The final component of the societal loss that I have calculated pertains to the lost tax revenue from a reduction in smoking. If there is a 20 percent reduction in cigarette consumption, the total loss of tax revenues will be $2.38 billion, with the loss being roughly evenly split between Federal and State governments. What these calculations suggest is that the effect of the proposed regulation in reducing smoking at the workplace and smoking overall will be nontrivial. Moreover, the group with the greatest amount to lose will not be the tobacco industry but rather the individual smokers who will suffer an annual welfare loss on the order of $11 billion. The group that will suffer the second greatest loss will also not be the tobacco industry but rather the recipients of the cigarette taxes. The Federal and State tax loss will exceed the loss in profits to companies by a factor of 10. None of these effects of the proposed OSHA regulation were addressed in the OSHA analysis, and it is clear from these estimates that the effects on society are truly substantial. OSHA's analysis of the employment effects of the proposed regulation should also take into account the effects that will occur because of the reduced consumption of tobacco products, 21 TIBR 0007380

rather than limiting the employment effects discussion to the effect of higher costs of introducing ventilation equipment. In short, OSHA has ignored the central cost component and the adverse employment consequences associated with the ETS segment of the proposed rule. 9. The cost estimates of smoking restrictions fail to include the caoital costs. OSHA estimates the cost for eliminating ETS exposures as ranging from $0 to $68 million (page 16002). Such cost estimates appear to be strikingly low and should serve as a red flag for anyone considering the reasonableness of the cost estimates that have been prepared. Most important, OSHA completely neglects the capital costs for creating nonsmoking areas. Although the cost of setting up the appropriate ventilation system is included, the cost of the office space is not (page 16017). In effect, OSHA treats office space as being a free good in excess supply. This is certainly not the case generally. The cost for a 150 square foot room will go beyond that of the $4,000 cost of retro-fitting an HVAC system. Office space is not a free good and this cost, which will certainly exceed that of the HVAC system, should be included in a comprehensive analysis. 10. The indoor air quality standard does not pass an economic feasibility test. Even though OSHA did not calculate all the costs of the proposed regulation, it concluded that the costs imposed would be 22 TIBR 0007381

substantial. If demand curves are elastic, these costs could have an effect on profits of 0.94 percent overall, and as high as 4.5 percent for the fishing industry (page 16019). The effect on profits of close to 1 percent from a single regulation is not trivial. To put such effects in perspective, one should realize that the indoor air pollution regulation is not the only regulatory cost resulting from OSHA activities. Moreover, OSHA is not the only regulatory agency issuing regulations. If every agency in the federal government treated individual regulations as being inconsequential if they only decreased profits by 1 percent, then very few private businesses would remain solvent. Rather than presenting the two extreme cases (completely elastic and completely inelastic demand curves) as OSHA did in its analysis, it would have been more instructive to focus on the effects for demand elasticities for the particular industries. Moreover, it is inappropriate to dismiss as being unimportant a decline in profits of 1 percent. A more pertinent question would be how many firms will be forced to close their operations if their profitability rates would decline by almost 1 percent. In its analysis OSHA should assess the number of firms that will be forced to cease operation because of this regulation and the number of jobs that will be lost because of it. This job loss should include the adverse effects resulting from decreased tobacco consumption as well. 23 TIBR 0007382

11. The discount rate used is too low. Throughout the analysis OSHA utilizes a discount rate of 10 percent. As a real rate of return, this rate is certainly too high. Over the past decade, the real rate of return (e.a., the Treasury bill rate minus the rate of growth in the GDP deflator) has been in the range of 2 to 3 percent. The use of a real rate of return of 10 percent may be required for illustrative purposes, but in annualizing the capital costs of the regulation a more appropriate rate would be a real rate of return of 2 to 3 percent. 12. There is too much missing information to move forward with the reaulation. The overall sense that one derives from reading this proposed regulation is that it was rushed into print because of a believed risk associated with ETS. This risk has not been precisely documented in a manner that would pass the usual tests of scientific validity. Moreover, as was noted above, there are many other missing elements to the OSHA analysis. Zndeed, in the course of the analysis prepared by OSHA some of these missing data elements are identified. For example, it is noted that "OSHA has no data on the number of establishments currently permitting smoking in designated smoking areas" (page 16017). A careful regulatory analysis that undertook a survey of firms could readily establish this number, but this was not done. Similarly, it is noted that "OSHA estimated preliminary costs of complying with the proposed standard. OSHA's cost 24 TIBR 0007383

assumptions and methodologies are based on information available from the rulemaking record." Reliance on the rulemaking record can potentially provide biased estimates of compliance costs to the extent that these compliance costs are based on past actions by firms which have chosen to institute smoking policies. There will tend to be a self-selection process whereby the costs incurred by firms that have instituted HVAC systems in smoking areas, for example, will be less than those firms that have not. OSHA's discussion, which places substantial reliance on past practices, has in no way corrected for this self-selection bias. Many of the cost estimates in the OSHA analysis are also based on speculation. The entire sick building syndrome and the causes of it remain highly controversial. In particular, data pertaining to the role of ETS in contributing to these outcomes are not sufficiently precise to indicate a clear-cut linkage. The role of smoking in causing indoor air contamination and in leading to adverse health effects has not been precisely documented. The OSHA regulation runs the risk of addressing a "problem" that involves uncertain health effects and for which the contributory role of ETS has not been established. 25 TIBR 0007384

REFERENCES Ensminger, Audrey, M.E. Ensminger, James Konlande, and John Robson, (1994). Food and Nutrition Encvclonedia. 2nd Edition, Ann Arbor: CRC Press. Gravelle, Jane and Dennis Zimmerman, (1994). "Cigarette Taxes to Fund Health Care Reform: An Economic Analysis," Congressional Research Service Reoort to Congress Washington: Congressional Research Service-The Library of Congress. Journal of Risk and Uncertainty, Vol. 8, No. 1, (1994). Lutter, Randall, and John Morrall. (1993). "Health-Health Analysis: A New Way To Evaluate Health and Safety Regulation," Journal of Risk and Uncertainty, Vol. 8, No. 1, pp. 43-66. Manning, Willard, Emmett Keeler, Joseph Newhouse, Elizabeth Sloss, and Jeffrey Wasserman. (1989). The Costs of Poor Health Habits, Cambridge: Harvard University Press. Viscusi, W. Kip. (1979). Emplovment Hazards: An Investigation of Market Performance, Cambridge: Harvard University Press. (1983). Risk By Choice: Regulatina Health and Safety in the Workolace, Cambridge: Harvard University Press. (1992). Smokina: Making the Risky Decision, New York: Oxford University Press. (1994). "Mortality Effects of Regulatory Costs and Policy Evaluation Criteria," RAND Journal of Economics, Vol. 25, No. 1, pp. 94-109. 26 TIBR 0007385