RedlrcirlLq Thmo USC It is likely, however, that these studies have un- derestimated the direct costs of smoking for a variety of reasons (Warner et al. 1999). For example, they ignore other significant economic costs, including the costs of transportation associated with obtaining medical care and the costs of nonmedical care associ- ated with accommodating a person ivith a smoking- related chronic illness. These estimates also generally fail to account for other medical care costs related to cigarette smoking, such as burn care from injuries in smoking-related fires and perinatal care for low-birth- weight infants of mothers \vho smoke. Few studies have attempted to include the direct costs for non- smokers of diseases related to exposure to ETS, and none of these studies has tried to estimate the intan- gible costs of smoking-related illnesses (i.e., the pain and suffering associated \vith the illness and the grief experienced by family and friends). A human capital approach is generally used to estimate the indirect morbiditv and mortalitv costs associated with cigarette smoking. This apbroach views an individual as producing a stream of output or earnings computed at market value or as the im- puted value of housekeeping services. Thus, the value of a person is reflected by his or her earnings, and the lifetime value for that person is equal to the discounted stream of future earnings (Max and Rice 1995). This approach places a relatively high value on morbidity and mortality among young adults, men, and the more educated because of the relatively higher earnings that would be lost by these smokers (Markandya and Pearce 1989); moreover, lost earnings may not be an accurate reflection of the value people place on their health or on their lives. Furthermore, the human capi- tal approach is in contrast to the "willingness-to-pay" approach, which tries to estimate the value a person assigns to reducing his or her risk of premature death. A more controversial component in the compu- tation of the lifetime costs of smoking concerns the treatment of transfer payments. These transfer pay- ments include the reduction in income taxes and in- surance premiums paid by smokers because of reduced earnings associated with smoking-related illnesses, the value of Social Security and private pensions foregone because of smoking-related premature deaths, higher health care costs associated with smoking-related illnesses and paid bv public and private insurance plans, and increased Asick pay and disability benefits paid during smoking-related illnesses. Particularly ob- jectionable to many people is the idea that foregone Social Security and private pension benefits from smok- (frs who die piematurely from smoking-related illnesses should be considered "benefits" to nonsmokers. As Harris (U.S. House of Representatives 1994) and oth- ers have noted, premature deaths are not considered a benefit when policymakers determine what levels of funded research are appropriate for reducing prema- ture deaths from other risk exposures (CSH 1994; Warner et al. 1995, 1999). Nevertheless, several recent estimates of the costs of smoking have considered these foregone benefits in their computations of the economic costs of cigarette smoking (Manning et al. 1989, 1991; Shoven et al. 1989). These studies aim to provide a complete accounting of the costs of smoking to answer the question of whether payments by those who have ever smoked into collectively financed systems such as Medicare and Social Security equal receipts by those \\,ho have ever smoked. Theoretically Optimal Cigarette Taxes As \vas just discussed, several estimates of the optimal or fair tax on cigarettes are based on the vari- ous studies of the costs of smoking. In the context of the preceding discussion, an optimal tax is one that equates the total revenues from these taxes to the net external costs of cigarette smoking. These estimates have ranged from those implying that current taxes more than cover the external costs of smoking (Man- ning et al. 1989) to those that have suggested that cur- rent taxes are far too low. For example, one such study that included the costs of the long-term intellectual and physical consequences resulting from smoking-related low birth weight among infants born to mothers who smoke indicated that $4.80 was an appropriate tax on a pack of cigarettes (Hay 1991). Another study (Pigou 1962) advanced a similar notion in providing a theoretical justification for taxes on goods with market prices not fully reflecting the social costs associated with their production and con- sumption. From that perspective, these taxes could be viewed as improving economic efficiency by rais- ing a smoker's marginal cost of smoking to a level nearer the social marginal cost. For some goods, taxes could generate revenues that exceed total external costs because the taxes would be based on marginal rather than average external costs (Cook and Moore 1993). Estimates of optimal taxes on cigarettes imply that smokers are fully informed about the risks associated with cigarette smoking (Cordes et al. 1990). If smokers underestimate these risks, then even higher taxes could be appropriate to discourage people from smoking. This issue may be particularly relevant for an addic- tive product such as cigarettes if, when people take up smoking, they do not fully understand the addictive properties oi consumption and the implications of addiction for future choices. Gruber and Koszegi (2000), for example, concluded that if these "internali- ties" are taken into account, they suggest sizable addi- tional taxes of one dollar or more per pack of cigarettes. Among the most widely cited recent estimates of the optimal tax are the studies of the economic costs of cigarette smoking by Manning and colleagues (1989, 1991). These incidence-based estimates used data from the RAND Corporation's Health Insurance Experiment and the 1983 National Health Interview Survey. To calculate the optimal tax on cigarettes, the analyses estimated both the lifetime external costs associated with cigarette smoking and the perceived "savings" that result from smokers' dying earlier and not realiz- ing their pension and Social Security benefits. Using their midrange estimates, Manning and col- leagues (1989, 1991) concluded that for a new smoker, the total external cost of smoking was 43 cents per pack of cigarettes in 1986. This estimate comprised 1 cent in extra costs for sick leave, 2 cents in costs for smoking- related fires, 5 cents in added costs for group life insur- ance, 9 cents in lost tax revenues (to finance retirement and health benefits), and 26 cents in spending on addi- tional medical care. These costs would be offset, how- ever, by an estimated 27 cents per pack in external savings resulting from smoking-related premature deaths. Converting these figures to 1995 dollars (based on the medical service price index and the gross national product deflator), the CRS estimated a net external cost of 33 cents per pack for cigarettes, which is approxi- mately two-thirds of the average federal, state, and lo- cal taxes on cigarettes of 30 cents per pack in late 1993 (Gravelle and Zimmerman 1994). The CRS thus con- cluded that smokers were more than paying their wav. Critics of the studies of Manning and colleagues (1989, 1991) contend that many of the assumptions made in obtaining the estimates are inappropriate. If the analyses had not included the effects of unrealized pension and Social Security benefits of smokers who die prematurely, the resulting external costs of smok- ing would have amounted to approximately 89 cents per pack in 1995 dollars. Moreover, the studies of Manning and colleagues (1989,1991) made a debatable distinction between in- ternal costs (those borne by the smoker) and external costs (those that smokers impose on nonsmokers). For example, the lost productivity costs described in those analyses were treated as internal costs, whereas only the higher, collectively financed, group premiums for health, life, and otherinsurance that nonsmokers paid to cover smoking-related costs not reflected in the pre- miums paid by smokers were considered external costs. More controversial, however, was these analy- ses' assumption that the cost of ETS was an internal cost. This assumption was based on the argument that the family is the economic unit involved in making smoking and other decisions and that the health con- sequences of ETS are largely confined to the nonsmok- ing spouses of smokers. As Manning and colleagues (1991) note, when this assumption is modified to treat the consequences of passive smoking as external costs, the estimated external costs of smoking rise signifi- cantly. For example, under the assumptions of Gravelle and Zimmerman (1994) concerning prices, the estimates of Manning and colleagues (1991) imply that including the relatively conservative estimate of 2,400 lung cancer deaths from ETS would add approximately 31 cents per pack (in 1995 dollars) to the external costs of smoking. Similarly, updating the researchers' esti- mates of the costs of neonatal care for smoking-related low birth weight would add more than 4 cents per pack. Doing the same for deaths from smoking-related fires would add 20 cents per pack and for smoking-related fetal deaths would add 31 cents per pack. These estimates probably understate the true costs of ETS. After reviewing the literature on the links between ETS and heart disease, Glantz and Parmley (1995) concluded that 30,000-60,000 persons die pre- maturely from heart disease related to ETS. Including these numbers in estimates by using the same assump- tions used in the CRS report would add at least an- other 70 cents to the estimate of the optimal tax. Moreover, the CRS report ignored the 150,000-300,000 cases of ETS-linked lower respiratory tract infections in children up to 18 months old and the ETS-linked worsening of asthma in 200,000 to 1 million children (Environmental Protection Agency [EPA] 1992). ln- eluding these costs \vould lead to an even larger opti- mal tax. Finally, the estimates excluded the long-term developmental consequences suffered by infants with smoking-related low birth weight (Hay 19911; were these costs included, the optimal cigarette tax would be nearlv SS per pack. Using the human capital approach, Manning and colleagues (1989,199I) estimated that the life of a nonsmoker who died prematurely from ETS exposure was worth $1.66 million. In a recent cost-benefit evaluation of the proposed Smoke-Free Environment Act of 1993 (introduced in the 103rd Congress but not passed), the EPA (Mudarri 1994) used the willingness- to-pay approach and obtained a $4.8 million baseline estimate of the value of a life. The EPA also used this approach to include the effects of ETS on heart dis- ease and children's health when calculating the value of benefits from reduced ETS exposure. By using the willingness-to-pay approach and making some relatively conservative assumptions, the EPA estimated that the total benefits from the reduced ETS exposure that Lvould result from a ban on smok- ing in all worksites ITas $39-71 billion per year. This estimate assumed that the ban would reduce the num- ber of current smokers by 3-6 percent, the number of future smokers by 5-10 percent, and consumption among continuing smokers by lo-15 percent; the re- sulting total long-run reduction in consumption would be la-22 percent. The combined effect of these reduc- tions in smoking and of the creation of designated smoking areas was predicted to reduce out-of-home exposures to ETS by 90 percent and in-home exposures by a midrange estimate of 6 percent. Estimates from the 1992 EPA report on ETS and lung cancer suggested that 73 percent of exposures to ETS occur outside the home and that 27 percent occur in the home. The total reduction in ETS exposure rvas thus predicted to be 66 percent; if it were applied to estimated total ETS costs of $58.7-106.9 billion, this reduction ~rould vield the EPA's estimated cost benefits of $39-71 billion. Gi\,en current cigarette sales of about 24 billion packs per year, this estimate implied that the per pack external costs of ETS lvere between $2.33 and $4.45. This estimate is likely to be lo\y, because the short-term and long-term costs of fetal and perinatal exposure to ETS lvere not included in the EPA's computations. Viscusi (1995), hoM-ever, reached a much different conclusion in analyzing the social costs of smoking. This investigator updated much of the analysis by Manning and colleagues (1989, lYYl), used a willingness-to-pav approach, and included the same ETS risks used in the EPA's analysis (Mudarri 1994). Viscusi, however, ar- gued that the EPA approach overestimated the risks of ETS by failing to account for the change in the tar con- tent of cigarettes and the changes in cigarette consump- tion per smoker. Noting that the average tar content of cigarettes declined from 46.1 mg per cigarette in 1944 to 12 mg per cigarette in 1994, Viscusi asserted that the health risks associated with cigarette smoking, as well as the risks from exposure to ETS, are linearly related to the tar content of cigarettes. Although presenting no evidence for either assertion, he contended that esti- mates of the health risks based on consumption of higher-tar cigarettes and exposure to ETS from higher- tar cigarettes need to be adjusted to reflect the decline in tar content. When not adjusting for tar, Viscusi ob- tained an estimate for the per pack external costs of ciga- rette smoking well above the average tax on a pack of cigarettes; when adjusting for tar, he concluded that current cigarette taxes exceed the external costs of smoking. A clear consensus is lacking regarding the opti- mal tax on cigarettes. Optimal tax calculations from prevalence-based estimates that include the direct and indirect costs of smoking-related morbidity and mor- tality are likely to be inappropriate, because the calcu- lations include lost productivity and other costs that should arguably be considered internal costs. Similarly, optimal tax calculations from the recent incidence- based estimates probably underestimate the optimal tax, because these calculations exclude many of the external costs of smoking. Nevertheless, because of the grooving evidence of the substantial health conse- quences of exposure to ETS (including fetal and peri- natal exposure), a tax that would generate sufficient revenues to cover all external costs from smoking is likely well abo1.e the current average of federal, state, and local taxes on cigarettes. Cigarette Taxes and Health As the review of studies on cigarette demand demonstrated, increases in cigarette prices lead to sub- stantial reductions in cigarette smoking by deterring smoking initiation among youth, prompting smoking cessation among adults, and reducing the average ciga- rette consumption among continuing smokers. Be- cause of the substantial health consequences of cigarette smoking and the health benefits of smoking cessation, these reductions in cigarette smoking would lead to significant improvements in health by reduc- ing smoking-related morbidity and mortality. Thus, increases in cigarette excise taxes, which would result in increases in cigarette prices, would be an effective policy tool in improving health. Several recent studies have provided some esti- mates of the health benefits resulting from cigarette tax increases. For example, Warner (1986) used pub- lished estimates of price elasticity (Lewit et al. 1981; Lewit and Coate 1982) to estimate the impact of higher cigarette excise taxes on smoking and health. The study predicted that a sustained, real 15 percent tax- induced increase in cigarette prices in 1984 (which would have been equivalent to restoring the federal tax to its real value in 1951-a nominal tax of 32 cents per pack) would deter 800,000 young people from smoking and encourage about 2.7 million adults to quit. Using the conservative assumption that one of every four lifelong smokers dies prematurely of a smois oftIll> ~V~~~7lfl~ ofNiltio,14,"are comrnodi- ties which are no where necessaries of life, \vhich are become objects of almost universal consumption, and which are therefore extremely proper subjects of taxa- tion" (1976, Book V, p. 474). As described earlier in this chapter (in "Ration- ales for Tobacco Taxation"), various lti\,els of go\rern- rnent have long used cigarette and other tobacco taxes to raise revenues. Such policy is supported by eco- nomic theory. An economically efficient way to raise revenues while minimizing the M.elfare losses associ- ated with the price distortions resulting from taxes is to impose relatively higher taxes on goods with more inelastic demand (one for which the percentage reduc- tion in demand is smaller than the percentage increase in price) (Ramsey 1927). As described earlier in this chapter (in "Effect of Price on Demand for Tobacco Products"), the numerous studies of cigarette demand and the limited studies of the demand for other to- bacco products have implied that overall demand, at least in the short run, is inelastic. Thus, large increases in tobacco taxes can generate substantial increases in revenues, particularly in the short run. Since 1960, the dollar amount of federal revenues generated by tobacco taxes has increased significantly, from $1.9 billion to nearly $5.9 billion in 1997. Over this same period, state revenues from tobacco ha1.e also increased significantly in nominal terms, from slightly less than $1 billion to more than $7.5 billion. As new sources of tax revenues have been identified, however, tobacco revenues ha\,e constituted a smaller propor- tion of total revenues. Tobacco taxes accounted for 3.36 percent of all federal revenues in 1950, but they rvere only 0.44 percent of revenues in 1989 (CBO 1990). Similarly, total federal tobacco tax revenues as a share of the gross national product fell from 0.55 percent in 1950 to 0.08 percent in 1989. Merriman (1994) considered whether cigarette excise taxes are set to maximize the revenues from these taxes. More specifically, Merriman tested the idea that elected officials, in an effort to maximize their own utility, mav increase taxes on some goods to the point lyhere re\:enues from these taxes begin to decline (Buchanan and Lee 1982). Using published estimates of cigarette demand (Becker et al. 19941, the study found that cigarette excise taxes in every state were Mel1 below, the revenue-maximizing level of these taxes, at least as of 1985. Furthermore, these estimates of the marginal revenue effects of higher taxes were loljrer-bound estimates, because they held constant other states' taxes (a consideration that allowed for increases in the casual and organized smuggling of cigarettes in response to a tax hike in a given state). Coordinated state tax increases, as a result, would gen- erate even higher re\`enues. Grossman (1993) considered this issue of maxi- mizing the federal excise tax on cigarettes. Using pub- lished estimates of cigarette demand (Chaloupka 1991; Becker et al. 19941, Grossman predicted that in the long run, a real federal tax rate of $1.26 would maximize federal tax revenues at S16 billion and would gener- ate even larger immediate increases in revenues. Like- xvise, Becker and Grossman (1994) suggested that the long-run revenue-maximizing value of the federal ciga- rette excise tax is 95 cents per pack in 1994 dollars. This tax would generate approximately $12 billion in total revenues and would raise considerably more than in the short run. These estimates were consistent with the prediction that a sustained real increase of 75 cents in the federal tax on cigarettes would in the long run lead to a net increase in cigarette tax revenues of just over $16 billion (Gravelle and Zimmerman 1994). Other studies, however, have predicted that higher federal taxes would generate much greater rev- enues (Harris 1994; Womach 1994a). For example, Harris has predicted that raising the federal tax to $2.00 per pack would have generated nearly $20 billion in additional revenues annually, on average, from 1995 through 1999, xvhereas Chaloupka (1998) estimates that a $1.50 increase lvould, in the short run, raise $22.5 billion annuallv. The differences among the predicted revenue ef- fects of higher cigarette taxes mav be attributed to dif- ferent assumptions used to obtain these estimates as well as to differences in the period for which the pre- dictions are made. For example, two studies (Gross- man 1993; Becker and Grossman 1994) have assumed a linear demand function for cigarettes. One of the implications of this function is that the price elasticity of demand rises as price rises. Thus, when the effects of a large increase in the cigarette excise tax are pre- dicted, cigarette demand is assumed to become more responsive to price. This assumption implies that there is an inverted U-shaped relationship between cigarette taxes and revenues: increasing cigarette taxes from relatively low levels will initially lead to increased rev- enues; beyond some point, further increases in taxes will lead to even larger reductions in demand, thereby causing revenues to fall. The same basic argument is implicit in the well-known Laffer curve, Lvhich relates income tax rates to income tax revenues. The assumption of a linear demand function for cigarettes is in contrast to the assumption made by some other analysts that the price elasticity of demand is constant over the range of prices under consider- ation. Because almost all of the studies described in this section found that the demand for cigarettes is inelastic, the assumption of a constant elasticity im- plies that even very large increases in taxes \~111 al- ways generate large increases in revenues. The differences in revenues predicted by these two assumptions, although only minor \vhen analv- ses predict the impact of relati\,ely small cigarette tax increases, grow with the size of the tax increase. Be- cause either assumption could be questioned, the rev- enue effects of a tax increase will likelv fall some\yhere between the predictions obtained from the t\vo (Grossman et al. 1993). The limited evidence from the behavioral economics literature suggests, hoL1-ev-er, that the effects of large increases in cigarette prices \vill lead to larger reductions in cigarette demand than pre- dicted by the assumption of a linear demand function (Bickel et al. 1991). A second key factor leading to the differences discussed here is the distinction between the short-run and long-run effects of the tax hikes. Economic theory implies that the demand for most consumer goods will be more responsive to price in the long run than in the short run. For cigarettes and other tobacco products, additional factors increase the likelihood that the long- run effects of an increase in price on cigarette demand will exceed the short-run effects-that is, price elastic- ity will increase in a manner similar to the increase for other, nonaddictive goods and services. Increased cigarette taxes will thus lead to smaller increases in revenues in the long run than in the short run. That adolescents and young adults are more re- sponsive to prices than older adults are and the fact that cigarette smoking is an addictive behavior are of particular importance when predicting the short-run and long-run revenue effects of higher cigarette taxes, Age difference in price elasticity implies that sustained real tax increases will lead to greater reductions in smoking prevalence and consumption as the number of adolescents and young adults who have not yet decided to smoke replaces the number of older adults who already smoke. The assumption of addiction implies that price has a cumulative effect on consump- tion: the price increase immediately reduces current consumption by discouraging young people from ex- perimenting or-continuing to experiment with smok- ing, as well as by encouraging current smokers to smoke less; futureconsumption is then reduced by the continuously fewer current smokers who also continue to smoke less in the face of a sustained real increase in price. The cumulative effect of price on consumption thus exceeds the immediate effect. This sequence ulti- matelv leads to reduced revenues. in summary, federal and most state excise taxes on cigarettes are undoubtedly well below their revenue- maximizing levels. Thus, relatively large increases in these taxes would lead to substantial gains in revenues, particularlv in the short run. Moreover, because of the relatively inelastic demand for cigarettes, increases in cigarette taxes are an economically efficient means of generating substantial revenues while imposing rela- tively small lvelfare losses. But if there is little argu- ment that large increases in cigarette taxes would generate substantial increases in tax revenues in the short run, there are some questions on the revenue- maximizing values of these taxes and the long-run stability of revenues generated by large increases in cigarette taxes. Part of the difficulty in estimating the effects of large taxes on cigarettes is that there is little experience in the United States with relatively large increases. Sim- larly, it is unlikely that the long-run effects of the more recent large tax increases have been fully played out. The short-term experience in Canada is of limited use in addressing these issues. Cigarette taxes in Canada increased more than 500 percent between 1982 and 1992, Mhich increased real cigarette prices by 170 percent, and total smoking fell by 38 percent (Sweanor and Martial 1994). Because of the effects of other, contemporane- ous activities to reduce tobacco use, the impact of the large price increases on smoking were consistent with the estimates from the studies of U.S. cigarette demand described in this chapter. Moreover, total federal and provincial revenues generated by Canadian cigarette taxes \vere 240 percent higher in 19Y2 than in 1981 e\ren M.ith the concomitant considerable black market in cigarettes. This experience suggests that large increases in cigarette taxes in the United States would generate sizable tax revenues for manv vears. . I Conclusions 1. The price of tobacco has an important influence 3. Policies that influence the supply of tobacco, par- on the demand for tobacco products, particularlv titularly those that regulate international com- among young people. merce, can have important effects on tobacco use. 3 i. Substantial increases in the excise taxes on ciga- 1. Although employment in the tobacco sector is rettes would have a considerable impact on the substantial, the importance of tobacco to the U.S. prevalence of smoking and, in the long term, re- economy has been overstated. Judicious policies duce the adverse health effects caused by tobacco. can be joined to higher tobacco taxes and stron- ger prevention policies to ease economic diver- sification in tobacco-producing areas. References Advisory Commission on Intergovernmental Rela- tions. Cignrrtte Bootl~~~~~i~lg: A Shte oral Fdcrnl RC~~OII- sibilif!y. A Cmt7r77issiorr Xcporf. Washington: Advisory Commission on Intergovernmental Relations, 1977. Advisory Commission on Intergovernmental Rela- tions. Cigarrtfe Tn.r- Emsio~: A Sccord Look. Washing- ton: Advisory Commission on Intergovernmental Relations, 1985. Advisory Commission on Intergovernmental Rela- tions. S~~prl~~~carrf Fcatllfw ofFiw7l F~dcmliwl. Volfiim~ 1: Budget Pvocessrs rltld T~s Syst~m.s. Washington: Advi- sory Commission on Intergovernmental Relations, i99i. American Economics Group, Inc. T/w U.S. Potwco 1w illrjtrJ/ ill 7994: Its Eco/lorlric I~II/x~c/ i/l f/l? .Stafia.i. Wash- ington: American Economics Group, Inc., 1996. Arthur Andersen Economic Consulting. fi~bnrro Irdrls- fry Enrp/oymcnf: A Xtvitw~ of fhr Pviw Wrrftvh~rst~ Et-r,- rmzic Zqmf Xtprt 171111 Tolmcco lllstiflltc Esfim7fcY of "~CCJ~lU~?liC ~OSS~`~fJ'fll7l ~llCJW75ii7<(' f/W F&l'fl~ E.YCiW -hX." Los Angeles: Arthur Andersen Economic Consulting, 1993. Ashenfelter 0, Sullivan D. Nonparametric tests of mar- ket structure: an application to the cigarette industry. ]ourml of Imfusfrin1 Ecarm~ic-s 1987;35(4):483-98. Baltagi BH, Goel RK. Quasi-experimental price elas- ticities of cigarette demand and the bootlegging effect. Arurricarz /ourrIo of A,qriculturn1 ET(J~IOVI~C~ 1987;69 (4):750-4. Baltagi BH, Levin D. Estimating dynamic demand for cigarettes using panel data: the effects of bootlegging, taxation, and advertising reconsidered. R&w of Et-o- rzomics nrzd Sfofisfics 1986;68(1):148-35. Baltagi BH, Levin D. Cigarette taxation: raising rev- enues and reducing consumption. Sfrllcfurnl Chnr~gc nr?d Ecormmic Dyrmmics 1992;3(2):321-35. Barnett PG, Keeler TE, Hu T-w. Oligopoly structure and the incidence of cigarette excise taxes. /01/r~1~7/ of Public tcormruics 1995;57(3):457-70. Barzel Y. An alternative approach to the analysis of taxation. \ozrrrrd ofhlificnl Eco/zo~7z~ 1976;84(6):1177-97. Becker GS, Grossman M. And cigarette revenues up in smoke. Wnll Sfwrf ]olrrrml, Aug 9, 1994;Sect A:12 (co1 3). Becker GS, Grossman M, Murphy KM. An empirical analysis of cigarette addiction. Amricmr Eco~zomic RF iliem 1?94;84(3):39-418. Becker GS, Murphy KM. A theory of rational addic- tion. jo~irnnl of Poljfrcill Eco~~omy 1988;96(4):675-700. Bickel WK, DeGrandpre RJ. Modeling drug abuse policy in the behavioral economics laboratory. In: Green L, Kagel JH, editors. Ada~rrccs iu Belmioral Eco- rm77ic5. Vol~rrm~ S: Subsfrr~m Use t~tzd Ablrse. Norwood (NJ): Ablex Publishing Corporation, 1996;69-95. Bickel WK, DeGrandpre RJ, Higgins ST. Behavioral economics: a novel experimental approach to the studv of drug dependence. Drug rzrzd A/t-oh/ Drptv~drr~ct~ 1993;33(2):173-92. Bickel WK, DeGrandpre RJ, Hughes JR, Higgins ST. Behavioral economics of drug self-administration. II. A unit-price analysis of cigarette smoking. ]ourm/ c~f fhc Espcrimrrzfn/ A/ml!/sis of Bchnvior 1991;55(2):145-54, Bickel WK, Hughes JR, DeGrandpre RJ, Higgins ST, Rizzuto I? Behavioral economics of drug self- administration. IV. The effects of response require- ment on the consumption of and interaction between concurrently available coffee and cigarettes. Psyc/w ~hl7Ui/MOkJ~~~ 1992;107(2-3):211-6. Bickel WK, Madden GJ. The behavioral economics of smoking. In: Chaloupka FJ, Grossman M, Bickel WK, Saffer H, editors. TIP Emrrorlric Adysis @Subsfame LIw nr7d Abuse: An I~?fep?fio?? c~f Economrfric (IH~ Belmiovnl Ecorzowic Resetrrch. Chicago: University of Chicago Press, 1999:31-66. Bishop JA, Yoo JH. "Health scare," excise taxes and advertising ban in the cigarette demand and supply. Souflrtwz Ecorzomic ]ourrml 1985;52(2):402-11. Bloom JL. International Interests in U.S. Tobacco Legis- lation, Health Science Analysis Project Policy Analysis Paper No. 3, 1998;; accessed: March 3,200O. British-American Tobacco Company Limited. firbncco Tiisafior~ Guide: A Guide fo Alterrzatiw Metlrodr of Tnrirrg Cigarettes olzrl Ofhcr Tobncco Prorll~cfs. Staines (UK): British-American Tobacco Company Limited, 1994. Buchanan JM, Lee DR. Tax rates and tax revenues in political equilibrium: some simple analytics. Ecorzorrzic hquiry 1982;20(3):344-54. Bulow JI, Pfleiderer I'. A note on the effect of cost changes on prices. /olrrrrfll of Polificnl Ecor~?rizy 1983; 91(1):182-5. Bureau of Alcohol, Tobacco and Firearms. Tax and Rate Fees;; accessed: July 6,200O. California Department of Health Services. Hcnlflr lzlz11 Eco~7omic frrz~~~cf of Snzokirzg: Cnljforuin, 1988. Technical Report No. 5. Berkeley (CA): California Department of Health Services, California Chronic and Sentinel Diseases Surveillance Program, 1992. Centers for Disease Control and Prevention. Medical- care expenditures attributable to cigarette smoking- United States, 1993. Morbidity a,zd Mortality Weekly Rr)?ort 1994;43(26):469-72. Centers for Disease Control and Prevention. Response to increases in cigarette prices by race/ethnicity, in- come, and age groups-United States, 1976-1993. Mor- bidify aud Mortality Weekly Report 1998;47(29):605-9. Chaloupka F. Clean indoor air laws, addiction and ciga- rette smoking. /@@ied Eco~~omirs 1992;24(2):193-205. Chaloupka FJ. Men, women, and addictiorz: the case of tip wftr smoking. Working paper no. 3267. Cambridge (MA): National Bureau of Economic Research, 1990. Chaloupka FJ. Rational addictive behavior and ciga- rette smoking. Journal of Political Economy 1991;99(4): 722-42. Chaloupka FJ. The Impact of Proposed Cigarette Price Increases, Health Science Analysis Project Policy Analy- sis Paper No. 9, 1998;; accessed: September 23,1999. Chaloupka F, Corbett M. Trade policy and tobacco: towards an optimal policy mix. In: Abedian I, van der Merwe R, Wilkins N, Jha P, editors. The Ecorromics of 7X~ncco Corztrol: Towards a17 Optima1 Policy Mix. Cape Town (South Africa): Applied Fiscal Research Centre, University of Cape Town, 1998129-45. Chaloupka FJ, Grossman M. Price, tobacco control poli- cirs nrzrl yolrtlr sruokirrg. Working paper no. 5740. Cam- bridge (MA): National Bureau of Economic Research, 1996. Chaloupka FJ, Hu T-W, Warner KE, van der Merwe R, Yurekli A. The taxation of tobacco products. In: Jha I', Chaloupka FJ, editors. Tobncca Corrtrol irz DeveIopirzg Colrlzfrics. Oxford: Oxford University Press, in press. Chaloupka FJ, Laixuthai A. U.S. trade policy and tip- rettc srrzokitz,g irz A~io. Working paper no. 5543. Cam- bridge (MA): National Bureau of Economic Research, 1996. Chaloupka FJ, Pacula RL. Sex and race differences in young people's responsiveness to price and tobacco control policies. Tobflcco Corztrol 1999;8(4):373-7, Chaloupka FJ, Pacula RL, Farrelly MC, Johnston LD, O'Malley PM, Bray JW. Do higher clprettc prices encour- age youth to use marijuana? Working paper no. 6939. Cambridge (MA): National Bureau of Economic Re- search, 1999. Chaloupka FJ, Saffer H. Clean indoor air laws and the demand for cigarettes. Contemporary Policy issues 1992;10(2):72-83. Chaloupka FJ, Tauras JA, Grossman M. Public policy and youth smokeless tobacco use. Sozrther>z Economic /our11a1 1997;64(2):503-16. Chaloupka FJ, Wechsler H. Price, tobacco control poli- cies and smoking among young adults. ]ournal of Health Ecorlomics 1997;16(3):359-73. Clarifeld KW. Cigarette sales impress and confound. Iildusfry Swwys. Vol. 1. New York: Standard & Poor's Corporation, Jan 1983:B93. Congressional Budget Office. Fr[fw~l Tmxfiorl of Tobncco, Alcoholic Bewrap, a~lrl Motor Flrrls. Washington: US Government Printing Office, 1990. Connolly G, Chen T. International health and tobacco use. In: Houston TP, editor. T&cc-o Llw: Al1 Americrrn Crisis. Final report of the conference; Jan 9-12, 1993; Washington, DC. Chicago: American Medical Associa- tion, 1993:72-4. Contracting Parties to the General Agreement on Tar- iffs and Trade. Thailand: restrictions on importation of and internal taxes on cigarettes. In: Basir Iusfnrrllents nmf Sdrcfed DoclfmJllt~. 5rpp1e"`C"f No. 37. Protoco/s, Derisio~~s, Reports 2989-7 990 nt~d Forf~pSi~fh Sessim. Geneva: Contracting Parties to the General Agreement on Tariffs and Trade, 1991:200-2X. Cook PJ, Moore MJ. Taxation of alcoholic beverages. In: Hilton ME, Bloss G, editors. Eca~~w~ics nrd th Prc- wlfj0\1 of Alcohol-Rehfed Pro1dems. Research Mono- graph No. 25. Rockville (MD): US Department of Health and Human Services, Public Health Service, National Institutes of Health, National Institute on Alcohol Abuse and Alcoholism, 1993:33-58. NIH Pub- lication No. 93-3313. Cordes JJ, Nicholson EM, Sammartino FJ. Raising l-e\`- enue by taxing activities with social costs. Nnfiorln/ Tos ~ourm~l 1990;43(3):343-56. DeGrandpre RJ, Bickel WK, Higgins ST, Hughes JR. A behavioral economic analysis of concurrently available money and cigarettes. ]our/m1 oft/w E.~perilmrf~zl A/m/y- sis of Behmior 1994;61(2): 191-201. DeGrandpre RJ, Bickel WK, Hughes JR, Higgins ST. Behavioral economics of drug self-administration. III. A reanalysis of the nicotine regulation hypothesis. Psy chophi7nnnra/oS?i 1992;108(1-2):1-10. Department of Finance, Canada. Trjbacc-o X7sc.s n/d COJI- slllflptiorr. Ottawa (Canada): Department of Finance, Canada, 1993. Doron G. Tllc S/7&if7~ Pnrados: Public Rqlllnfiot? irk fl7r Cignrctfc lmlusfry. Cambridge (MA): Abt Associates, 1979. Douglas S. The duration of the smoking habit. Eco17or71ic hpiry lYY8;36(1):49-64. Douglas S, Hariharan G. The hazard of starting smok- ing: estimates from a split population duration model. [ozl~11~! of Hrnltll Ecorlwzics 1994;13(2):213-30. Eckard EW Jr. Competition and the cigarette TV ad- vertising ban. ECO)XX)~~C I,qlliry 1991;29(1):119-33. Eddy I', Walden S. Invasion. Telcpzph Magazine, May 22, 1993:18-32. Elster J. U/ysjcs nrlrl tlrc Sirens: Studies it1 R~fio~mlify ~rzrl Irmtiomlify. Cambridge: Cambridge University Press, 1979. Environmental Protection Agency. Respiratory Hmltll Effects of Pmsiw Smokirlg: Lutz~ Cnrzccv ard Other Disor- ders. Washington: Environmental Protection Agency, Office of Research and Development, Office of Air and Radiation, 1992. EPA/600/6-90/006E Evans WN, Farrelly MC. The compensating behavior of smokers: taxes, tar, and nicotine. RAND jourtml of E~orm7ic-s 1998;29(3):578-95. Evans WN, Ringel JS. Can higher cigarette taxes in- prove birth outcomes? \olrr,m/ c$Puhlic Emmuics 1YYY; 72: 135-53. Farrelly MC, Bray JW, Zarkin GA, Wendling SW, Pac- ula RL. T/IL' ~`ff&-s of~~riws o/jlf /~o/ick 011 fhe d~v77auii for. r7rnrijrrfl,m: wiifcfrc-c from t/w Nntioru71 Ho~tsehold S~~rz~ys 0\1 DI,L~S AI~IIJC. Working paper no. 6940. Cambridge (M4): National Bureau of Economic Research, 1999. Federal Trade Commission. Conlpctifiorl n~zrl fire Fi,lali- ii~7l Iu1lu7ct of tlw Proposed Tobacco Illdusfry Seff2erm1f. Washington: Federal Trade Commission, Bureaus of Economics, Competition, and Consumer Protection, 1997. Flewelling RL, Kenney E, Elder JP, Pierce J, Johnson M, Bal DG. First-year impact of the 1989 California cigarette tax increase on cigarette consumption. Anwi- C-~II ]cjlrrlri7l of P~/blic H~ltl~ 1992;82(6):867-9. Food and Agriculture Organization of the United Nd- tions. Tobacco: Supply, Dtwnrd ntd Trnde Projccfiorrs, 7995 ~~1~12000. FAO Economic and Social Development Pa- per 86. Rome: Food and Agriculture Organization of the United Nations, 1990. General Accounting Office. Tw[lc o/1(/ He7lfl7 fssllcs: Diclrofony Befzwr77 U.S. Tohcm E.rprf Policy 17171f Ar7fi- 5777okI'77~~ fr7ifiofizvs. Washington: General Accounting Office, 1990. GAO/NSIAD-90-190. General Accounting Office. 177fw7nfio77n[ 'finds: Adwr- fi5ir7g nrrd Pro777oti77~ U.S. Cijywffcs i,7 .!?/cl-fd Asin Co7777frim. Washington: General Accounting Office, 1992. GAO/GGD-93-38. Glantz SA, Parmley WW. Passiv,e smoking and heart disease: mechanisms and risk. ]0~71.17171 of tl7c A~~rcv~icnrr MdicnI At;socinfiorr 1995;73(131:1017-53. Goel RK, Morey MJ. The interdependence of cigarette and liquor demand. SOII~I~L~T~I Ecor7m7ic \071~17l7l 1995; 62(2):451-9. Gravelle JG, Zimmerman D. Cipowffc Tosrs fo Fur7d Htvlth Caw R~'for777: A17 Ecorwwic A77ol!/sis. Washington: Library of Congress, Congressional Research Service, 1994. CR!? Publication No. 94-214 E. Grise VN. Tl7e World Tobncro Ma~krf-Goilcr~7777lcrlf /r7- fruzv77fior7 nnr-i M7rlfihfrrrrl Policy Rcfor777. Staff Report *No. AGES 9014. Washington: US Department of Agri- culture, Economic Research Service, Commodity Eco- nomics Division, 1990. Grise VN. Tohncco: Bmkp~w77d,for 1995 Fnrr71 Lqislnfio77. Agricultural Economic Report No. 709. Washington: US Department of Agriculture, Economic Research Service, 1995. Grossman M. For best revenue, tax cigarettes $1.26. hicw ynovk Tin?es, June 18, 1993;Sect A:26 (co1 4). Grossman M, Coate D, Lewit EM, Shakotko RA. Eccl- ~?o~lric arid Other Factors ir7 Xluth Snzokirlg. Washington: National Science Foundation, 1983. Grossman M, Sindelar JL, Mullahy J, Anderson R. Policy \t,atch: alcohol and cigarette taxes. jonrrral nf Eic177o~77ic- Pmprfiw 1993;7(4):211-22. Gruber J, \ri~uf/7 5177oki77~ ir7 fl7c U.S.: pr-icrs nrrrl plici~`s. Working paper no. 7506. Cambridge (MA): National Bureau of Economic Research, 2000. Gruber J, Koszegi B. Is nddicfiorr "mfio77n/"? Tl7enry 017~i wi&wcc. Working paper no. 7507. Cambridge (MA): National Bureau of Economic Research, 2000. Harris JE. Taxing tar and nicotine. At7rcrirnr7 Ec0770177ic RL'i'il'iI' 1 Y80;70~3):300-11. Harris JE. The 1983 increase in the federal cigarette excise tax. In: Summers LH, editor. Ti7.r Policy nr7d U7e Em770777.~~. Vol. 1. Cambridge (MA): MIT Press, 1987: 87-111. Harris JE. A rwr-X-ill:; 711otld for prctlicti17~~ t/w co77s1tr77pfior7 r7~7d ww~71ic irllpc-ts elf lnr;qc i77rrme5 irr fhc U.S. ff&m/ c-i~~rrr'ftc tw-iw tnr. bvorking paper no. 4803. Cambridge (MA!: National Bureau of Economic Research, 1994. Harris JE, Chan SW. The continuum-of-addiction: ciga- rette smoking in relation to price among Americans aged 15-29. Hcvlti7 Ecor7nr77ics 1999;8(1):81--6. Hay JW. The harm they do to others: a primer on the external costs of drug abuse. In: Krauss MB, Lazear EP, editors. Scw~lri77g for Alfcr77nfizvx: Dr7lg-Cor1trol Prjliry i/l f/w U77ifd Sfnfcs. Stanford (CA): Hoover In- stitution Press, 1991:200-25. Henningfield JE, Clayton R, Pollin W. Involvement of tobacco in alcoholism and illicit drug use. Brifisl7 Iour- 1717[ OfAddicficJ77 1990;85(2):279-91. Hodgson TA. Annual costs of illness versus lifetime costs of illness and implications of structural change. Drq I77fomnfim7 ]ow77nl 1988;22:323-41. Hodgson TA. Cigarette smoking and lifetime medical expenditures. Milhr7k Qurrrfeuly 1992;70(1):81-125. Holzman D. Tobacco abroad: infiltrating foreign mar- kets. E77zIirLtr7777rr7fi/ Hcnlfh Pospccfizm 1997;105(2): 178-83. Hsieh C-R, Lin Y-S. The economics of tobacco control in Taiwan. In: Abedian I, van der Merwe R, Wilkins N, Jha I', editors. 7%~ Ecorzo~zjc~ of Tobncco Co/lfrol: Toi(lorTIs 1111 O@nznl Pulicy Mis. Cape Town (South Africa): Ap- plied Fiscal Research Centre, University of Cape Town, 1998:306-29. Hu T-w, Bai J, Keeler TE, Barnett PG, Sung H-Y. The impact of California Proposition 99, a major anti- smoking law, on cigarette consumption. ]ollrrznl ujPzrh- lit Hen/f/z Policy 1994;15(1):26-36. Hu T-w, Ren Q-F, Keeler TE, Bartlett J. The demand for cigarettes in California and behavioural risk fac- tors. Health Ecormr~rics 1995a;4(1):7-14. Hu T-w, Sung H-Y, Keeler TE. Reducing cigarette consumption in California: tobacco taxes vs an anti- smoking media campaign. Amrvicnrz ]olrnmi of Public Hedth 1995b;85(9):1218-22. Hursh SR, Bauman RA. The behavioral analysis of demand. In: Green L, Kagel JH, editors. Ad~~~ucrs irk Behnoiorol Eco~onzics. Vol. 1. Noruood (NJ): Ablex Pub- lishing Corporation, 1987:117-65. Jackson JD, Saba RI? Some limits on taxing sin: ciga- rette taxation and health care finance. Solrfl~arl Eco- rzunzic ~uurruzl 1997;63(3):761-75. Johnson PR. 7%~ Ecar7or77ics oftl7c Tbhaccn lrlrf~l~try. Nell, York: Praeger Publishers, 1984. Johnson TR. Additional evidence on the effects of al- ternative taxes on cigarette prices. \olcrrtnl of Political Economy 1978;86(2 Pt 1):325-B. Joossens L, Chaloupka FJ, Merriman D, Yurekli A. Issues in the smuggling of tobacco products. In: Jha I', Chaloupka FJ, editors. Tubflcco Cuntrol irk Dcaelopirz~ Couutvies. Oxford: Oxford University Press, in press. Joossens L, Raw M. Smuggling and cross border shop- ping of tobacco in Europe. British Medicn/ ]oun7nl 1995; 310(6991)1393-7. Joossens L, Raw M. Cigarette smuggling in Europe: who really benefits? fi~hmx~ Corztrol 1998;7(1):66-71. Joossens L, van der Merwe R. Cipvette Trade and Sn~lr~~. $ir7,y. The Economics of Tobacco Control Project Update No. 7. Cape Town (South Africa): School of Economics, University of Cape Town, 1997. Kandel D. Stages in adolescent involvement in drug use. Science 1975;190(4217):912-4. Kandel D, Yamaguchi K. From beer to crack: develop- mental patterns of drug involvement. Anlericarl ]oL~,.- rml of P~iblic Heo1th 1993;83(6):851-5. Keeler TE, Hu T-W, Barnett I'G, Manning WG. Taxa- tion, regulation, and addiction: a demand function for cigarettes based on time-series evidence. louvrznl of Hcnltlr Econninics 1993;12(1):1-18. Keeler TE, Hu T-w, Barnett I'G, Manning WG, Sung H-Y, Do cigarette producers price-discriminate by state? An empirical analysis of local cigarette pricing and taxa- tion. ]onvrinl ofH&flz Eco~~omics 1996;15(4):499-512. Lewit EM, Coate D. The potential for using excise taxes to reduce smoking. lour17nl LJ,f Hmltlz Ecomnrics 1982; 1(2):121-45. Lewit EM, Coate D, Grossman M. The effects of gov- ernment regulation on teenage smoking. /o~IY/z~~ nflaila nrd ECCJI~CJW~CS 1981;24(3):545-69, Lewit EM, Hyland A, Kerrebrock N, Cummings KM. Price, public policy, and smoking in young people. fiJ/lKC(J CCJJ7tr-01 1997;6(Supp1 2):S17-S24. Lightwood J, Collins D, Lapsley H, Novotny T. Estimat- ing the costs of tobacco use. In: Jha P, Chaloupka FJ, editors. Tubncco Corltml ir7 Dcilelopirlg Courrfries. Oxford: Oxford University Press, in press. Lyon AB, Schwab RM. Consumption taxes in a life- cycle framework: are sin taxes regressive? Revieul of Ero!ronlicj ~7j~d Stntisfics 1995;77(3):389-406. Manning WG, Keeler EB, Newhouse JP, Sloss EM, Wasserman J. The taxes of sin: do smokers and drink- ers pay their way? ~ourrml of the Americnn Medical As- socintiorl 1989;261(11):1604-9. Manning WG, Keeler EB, Newhouse JP, Sloss EM, Wasserman J. T1Te Costs of Poor Herr/f/z Hnbifs. Cam- bridge (MA): Harvard University Press, 1991. llarkandya A, Pearce DW. The social costs of tobacco smoking. EI-itisll ]0717~77~7/ of Addic-tiorr 1989;84(10): 1139-50. Max W, Rice DP. The cost of smoking in California, 1993. Pbncco Cor7frd 1995;-1(Suppl l):S39-S46. Meier KJ, Licari MJ. The effect of cigarette taxes on ciga- rette consumption, 1955 through 1994. A777cricarr /olrr- r7nl of l'ublic- H~vltl7 1997;87(7):1126-30. Merriman D. Do cigarette excise tax rates maximize revenue? Eco770777ic I77177firy 1994;32(3):419-28. Merriman D, yurekli A, Chaloupka FJ. Ho\v big is the Lvorldwide cigarette-smuggling problem? In: Jha I', Chaloupka FJ. ~JfV?cCO Co77fKJ/ ill Dc2it'/(~/~i/7g Co7777fric,-. Oxford: Oxford Uni\.ersity Press, in press. Miles RH. Coffirl h'fli1.i tr77d Corprlrnfc Sfr~ufcrcyi~zi. Englewood Cliffs (NJ): Prentice-Hall, 1982. Miller RH. The no net cost assessments for tobacco. Remarks prepared for the Tobacco Marketing Cost Committee Meeting; Mar 23-24, 1994; Fletcher (NC). Miller LS, Zhang X, Rice Dl', Max W. State estimates of total medical expenditures attributable to cigarette smoking, 1993. Plrblic H~vlt!~ Rqmrfs 1998;113(5):447-58. Miller VP, Ernst C, Collin F. Smoking-attributable medi- cal care costs in the USA. Socii71 Sric77c~ 0 Mdir-ir7~ 1999;48(3):375-91. Mitchell ML, Mulherin JH. Finessing the political sys- tem: the cigarette advertising ban. Sol~flwrfr Ecor7omir ]mimal 1988;54(4):855-62. Moore MJ. Dent/l nnd fohncco tases. Working paper no. 95-2. Durham (NC): Duke University, The Fuqua School of Business, Center for the Study of Business, Regula- tion, and Economic Policy, 1995. Mudarri DH. Thr Costs ar7d Be~~cjits of Swokifz~ Restric- tiar7s: A77 Assrssnze7zf of the S777okFrk E7rvivor~r7zer7f Art of 1993 (H.R. 34341. Washington: Environmental Pro- tection Agency, Office of Radiation and Indoor Air, Indoor Air Division, 1994. Mullahy J. Cigarette smoking: habits, health concerns, and heterogeneous unobservables in a micro-econometric analysis of consumer demand [dissertation]. Char- lottesville (VA): University of Virginia, 1985. National Cancer Policy Board. Tdirrcq Acfior~ to Rrduw `T&ncco Urt~. Washington: Institute of Medicine and Commission on Life Sciences, National Research Council, 1998. Nguyen D. Advertising, random sales response, and brand competition: some theoretical and econometric implications. lo~~mnl of Blrsillrjs 1987;60(2):259-79. Ni\rola PS. Rq17lnti17~y Utlfnir Tvadr. Washington: Brook- ings Institution, 1993. Northup AM. U.S. agricultural policy on tobacco. In: Houston TP, editor. Tohncco USC: An Am~ic1777 Crisis. Final report of the conference; Jan 9-12, 1993; Wash- ington, DC. Chicago: American Medical Association, 1993:84-8. Ohsfeldt RL, Boyle RG. Tobacco excise taxes and rates of smokeless tobacco use in the US: an exploratory ecological analysis. 7'&ncco Corzfrol 1994;3(4):316-23. Ohsfeldt R, Boyle RG, Capilouto E. Effects of tobacco excise taxes on the use of smokeless tobacco products in the US. Ekt7~Jrric Hea/fl7 Ecor7onrics Lrffrrt; 1997; 1(3):10-9. Ohsfeldt RL, Boyle RG, Capilouto EL. Tobacco taxes, smoking restrictions, and tobacco use. In: Chaloupka FJ, Grossman M, Bickel WK, Saffer H, editors. Tlze Eco- ~70777;~ Aj7nlysis of S1rbstar7cr USC ilr7d Abuse: A17 Irzfegra- tio~7 CJ~ Ecor7o777cfuir 177711 Bchaoioml Ecor7oruic Research. Chicago: University of Chicago Press, 1999:15-29. Orzechowski and Walker. T/v TOS Burdtv7 017 Tobacco: Hisforicnl Co~~~l~i/nfiof~ 2 999. Arlington (VA): Orzech- olvski and Walker, 2000. Oster G, Colditz GA, Kelly NL. Tl7e Eco~zo771ic Costs uf S777okirlg 1777d Brrrt~fifs of Qzriffi778. Lexington (MA): D.C. Heath and Company, 1984. Overton J. Diversification and international expansion: the future of the American tobacco manufacturing in- dustry with corporate profiles of the "Big Six." In: Fin- ger WR, editor. The Tohncco l17d1lsfvy i77 Tmnsifio77. Lexington (MA): D.C. Heath and Company, 1981: 159-96. Facula RL. Addesc~r~t a2col1oI ar7d 7mrijuar7a cu71s7~771p- fio77: is fhcrc rrally n ~~f~wfl~. Working paper no. 6348. Cambridge (MA): National Bureau of Economic Re- search, 1998a. Ecor7o777ic Appomh~s 365 Pacula RL. Does increasing the beer tax reduce mari- juana consumption? lolrnlal ofHen1fl1 ETOIIOIII~C~ 19Y8b; 17(5):357-85. Pacula RL, Grossman M, Chaloupka FJ, O'Malley PM, Johnston LD, Farrelly MC. Mnrijunm aud ;/oufh. Work- ing paper no. 7703. Cambridge (MA): National Bureau of Economic Research, 2000. Peterson DE, Zeger SL, Remington I'L, Anderson HA. The effect of state cigarette tax increases on cigarette sales 1955-1988. Anzcrica~z Journal of Public Hcolth 1992;82(1):94-6. Pigou AC. A Study in Public Firmwe. 3rd rev. ed. London: Macmillan and Co., 1962. Porter RH. The impact of government policy on the U.S. cigarette industry. In: Ippolito PM, Scheffman DT, editors. Eurpiricnl Approdw~ to Cor~sur?lcr Protectiorz Ecorlonlics. Washington: US Government Printing Of- fice, 1986:447-81. Price Waterhouse. T/w Emlotuic Irnpct of the Tohcco lrzdzrsfry 011 t/w Uflifed Strifes irr 1990. Arlington (VA): Price Waterhouse, 1992. Ramsey FP. A contribution to the theory of taxation. Ecot~nr~~ic [o~lrrrnl 1927;37(Mar):47-61. Rice DP, Hodgson TA, Sinsheimer r, Brocvner W, Kopstein AN. The economic costs of the health effects of smoking, 1984. Milh17k Q~mrtcrly 1986;h4(3):489-547. Rice DP, Max W. T/w Cost of .Snmki~~:: irk Cnlifor-rlin, 7 989. Sacramento (CA): California State Department of Health Services, 1992. Robert JC. T1zr Star!/ uf Tohnccu ir7 A~~~cP~c~. Chapel Hill (NC): University of North Carolina Press, 1967. Roemer R. Lrgislatiw Actiw~ to Corilblrf flrc World Tobncco E@`enzic. 2nd ed. Geneva: World Health Organization, 1993. Rucker RR, Thurman WN, Sumner DA. Restricting the market for quota: an analysis of tobacco production rights with corroboration from congressional testi- mony. lo~lr,rol of Political Ecnr~~~y 1995;103(1):112-75. Saba RP, Beard TR, Ekelund RB Jr, Ressler RW. The demand for cigarette smuggling. Ecorlclnlir I/lquiry 1993;33(2):189-202. Saffer H, Chaloupka FJ. Alcohol tax equalization and social costs. Ensferrr Ecormr~~ic- ]oumu1 1994;20(1):33-43. Schelling TC. Clloiccl~rln Cor1sa7ue17cr. Cambridge (MA): Harvard University Press, 1984. Schelling TC. Economics and cigarettes. Prez~e~ztiz~e MedicifTe 1986;15(5):549-60. Schevitz 7. Discount cigarette stores ignite sales. Sari Fm~~cisc~ E.rflnliller, Dee 12, 1994;Sect 4:Bl. Seldon BJ, Boyd R. The stability of cigarette demand. Applied Ecormmics 1991;23(2):319-26. Seldon BJ, Doroodian K. A simultaneous model of ciga- rette advertising: effects on demand and industry response to public policy. Rwiezu of Ecomr~zics Ned Sta- tistics 1989;71(4):673-7. Sesser S. Opium war redux. New Yorker 1993;69(29): 78-89. Shoven JB, Sundberg JO, Bunker JI? The social secu- rity cost of smoking. In: Wise DA, editor. The ECUWIJI- its of Aging. Chicago: University of Chicago Press, 1989:231-54. Showalter MH. Essays in applied econometrics. Essay III: monopoly behavior with intertemporal demands [dissertation]. Cambridge (MA): Massachusetts Insti- tute of Technology, 1991. Simonich WL. Gnwr~~r~rer~~ Atltisrlrokillx Policies. Neu York: Peter Lang Publishing, 1991. Smith A. All Irlquiry flrto f/w Riu7fllre nrzd Ca~ws of the Wenlfh of Nl7tiwl.i. In: Cannan E, editor. Chicago: Uni- versity of Chicago Press, 1976. Standard & Poor's, Irrdustry Surw~/s. Vol. no. 1. New York: Standard & Poor's Corporation, Jan 1989. Standard & Poor's, I/ltl~lstr-y .Slrrzlcys. Vol. no. 1. New York: Standard & Poor's Corporation, Apr 1993. Sullivan D. Testing hypotheses about firm behavior in the cigarette industry. lourun/ uf Political ECOIIUIJIJI 1983;93(33:586-98. Sumner DA. Measurement of monopoly behavior: an application to the cigarette industry. /olir1m1 of Polifiral Emmy 1981;89(5):1010-9. Sumner DA, Wohlgenant MK. Effects of an increase in the federal excise tax on cigarettes. Auurirnl7 \ourrm/ of Apiclilfl/ml Ecorml7irs 1985;67(2):233-42. Sumner MT, Ward R. Tax changes and cigarette prices. ~or~r~znl of Polific~72 Eco??o/~?y 1981;89(6):1261-5. Sung H-Y, Hu T-W, Keeler TE. Cigarette taxation and demand: an empirical model. Cmltfwipovnrl/ Eco~mtric Policp 1994;12(3):91-100. Srz,eanor DT, Martial LR. Tl?c S????ISS/~??:< of 7olwcic~ Prod??cts: Lcssorrs f?w? Cor~ndn. Ottawa (Canada): Non- Smokers' RightsAssociation/Smoking and Health Ac- tion Foundation, 1994. Tauras JA. 71?c trnwitiol? to ~r77okirl~~ cr~.55~7firJrr: e~~id~v~cc 1'ro?j? ????lltj/~l~ failure dl~rntiojl [~\?f~/!/;is. Working papel no. 7412. Cambridge (MA): National Bureau of Eco- nomic Research, 1999. Tauras JA, Chaloupka FJ. DcTfc'rr??i\?l7r?t.i (jf .~,rrt)h-ij?~ cc+ %::iot?: nr? n??n/ysis ofl/o~rj~~y n[l111t J??c)J nr?d m~~wi~. Work- ing paper no. 7262. Cambridge (MA): National Bureau of Economic Research, 1999a. Tauras JA, Chaloupka FJ. Pric-c, clcwj? ij?tloor nir, u1d c-iga- rc'ffe sn7okil7g: ri~idcrlcr fm?? /oJJ~itJrdir?i7~ dflfi7 for yc~JrJ?~y adults. Working paper no. 6937. Cambridge (MA): National Bureau of Economic Research, 1999b. Taylor AL, Chaloupka FJ, Guindon E, Corbett M. The impact of trade liberalization on tobacco consumption. In: Jha I', Chaloupka FJ, editors. Tobacco Cojltrol i,z Dc- idoping Cou?ztries. Oxford: Oxford University Press, in press. Taylor P. The Smoke Ring: Tobncco, Mor7cy, md Multino- fior?al Politics. New York: Pantheon Books. 1984. Tegene A. Kalman filter and the demand for cigarettes. Applied Ecommirs 1991;23(7):1175-82. Tennant RB. Tl7c hlericar7 Cigarette Irzdusfry: A Study ij? Eco??o~~?lc Amlysis nl?rl Public- Policy. New Haven (CT): Yale University Press, 1950. Thursby MC, Thursby JG. fufersf&c cipwtfr bootlqgir7;:: rrfcr7f, IZ'TWIIIC Io~ws, a??n e,ffecfs. Working paper no. 4763. Cambridge (MA): National Bureau of Economic Research, 1994. Toxvnsend J, Roderick P, Cooper J. Cigarette smoking by socioeconomic group, sex, and age: effects of price, income, and health publicity. Britisl7 Medico1 ~ouvr7~l 1994;309(6959):923-6. Tremblav CH, Tremblay VJ. The impact of cigarette advertising on consumer surplus, profit, and social welfare. Cor7tenrpornry Eco,loiJ?ic Policy 1995;13(1): 113-24. US Congress. U?~; accessed: Au- gust 9, 1999. US Department of Agriculture. Tuhaccu. Sit7mtior7 (777d Ofrtlook Rtyorf. Springfield (VA): US Department of Agriculture, Economic Research Service, 199%. TBS-245. US Department of Agriculture. USDA announces pro- visions of the 2000 crop flue-cured tobacco program and no net cost assessments [press release]. Washing- ton: US Department of Agriculture, Farm Service Agency, Dee 15,1999b. Release No. 1672.94. US Department of Agriculture. Declining cigarette con- sumption follows price hikes. AC~7-ic7r/t7~rrrl O~~tlook/ /17rllrar1/-Ft,brf~~r?/ 2000. Springfield (VA): US Depart- ment of Agriculture, Economic Research Service, 2000. US Department of Health and Human Services. T/v Health Co77st~jutv7ccs of Su7oki~rg: Nicotir7e Addit-fiord. A Report t+ tl7c Sl~r;Ccor7 &rleral. Rockville (MD): US De- partment of Health and Human Services, Public Health Serkrice, Centers for Disease Control, Center for Health Promotion and Education, Office on Smoking and Health, 1988. DHHS Publication No. (CDC) 88-8406. US Department of Health and Human Services. Rr- during the Healflz Corzsequences of Smoking: 25 Years of Prqress. A Report of the Surgeon General. Atlanta: US Department of Health and Human Services, Public Health Service, Centers for Disease Control, National Center for Chronic Disease Prevention and Health Pro- motion, Office on Smoking and Health, 1989. DHHS Publication No. (CDC) 89-8411. US Department of Health and Human Services. Snook- irzg at7d Hedtl7 irz tl7r Americm: A Reporf of the Surcgeor7 Gtwcrd. Atlanta: US Department of Health and Human Services, Public Health Service, Centers for Disease Control, National Center for Chronic Disease Preven- tion and Health Promotion, Office on Smoking and Health, 1992. DHHS Publication No. (CDC) 92-8419. US Department of Health and Human Services. Prt,- zwtirzg Tobacco Use An7or7g Yburzg Peq7lc: A Rcyorl o,ff/w Sl,rgtwz Gtvrcra~. Atlanta: US Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Center fol Chronic Disease Prevention and Health Promotion, Office on Smoking and Health, 1994. US Department of Health, Education, and Welfare. Su7oki77;; r7r7d Hcalfh: Rcporf uf the Advisory Corr7ruiftcc it) the Sllrporz Grr7rrnl of the P~~hlic Hc~lth Service. Wash- ington: US Department of Health, Education, and Welfare, Public Health Service, 1964. PHS Publication No. 1103. US Department of Labor. Consumer Price Index-All Urban Consumers;[follow the links "Statistics and Data, " "Bureau of Labor Statistics official databases," " Most Requested Series," "Con- sumer Price Indexes"]; accessed: August 23, 1999. US House of Representatives. Fi77077ci77g ProzGsior7s of tl7c Arfr~ri77istrnticJrl's Htvlfl7 5kcurity Act 17~711 Ofher Hen/f/r Rt$lrw Pro;~cw?k: H crrrir7Ss Befort> fl7e House Co/r7rilittw 017 W7ys rrr7tl Mtwrs, 103rd Congress, 1st Sess. Wash- ington: US Go\,ernrnent Printing Office, 1994. Serial No. 103-61. Viscusi WK. Cigarette taxation and the social conse- quences of smoking. In: Poterba JM, editor. Tars Polit.,~ c777d f/w Eco77~11771/. Cambridge (MA): MIT Press, 1995: 51-101. Wagner S. Cignwffc Co~r77try: Tdncco i7z A/ne~icrrr7 His- trq nljd Politics. New York: Praeger Publishers, 1971. Warner KE. Possible increases in the underreporting of cigarette consumption. lo~rrrzal of t/w A7nericnrr Sta- fisticnl Associatinr7 1978;73(362):314-8. Warner KE. State legislation on smoking and health: a comparison of two policies. Policy Sciences 1981;13(2): 139-52. Warner KE. Smoking and health implications of a change in the Federal Cigarette Excise Tax. /o7l777al CJ,( fl7c Awrici777 Medical Associafiol7 1986;255(8):1028-32. Warner KE. The tobacco subsidy: does it matter? ]wr- 7701 of tl7c Nnlior7al Cr717w 177sfifufe 1988;80(2):81-3. Warner KE. The importance of tobacco to a country's economy. Paper presented at the 9th World Confer- ence on Tobacco and Health; Ott lo-14,1994; Paris. Warner KE, Chaloupka FJ, Cook PJ, Manning WG, Newhouse JP, Novotny TE, Schelling TC, Townsend J. Criteria for determining an optimal cigarette tax: the economist's perspective. Totwcro Corrtrnl 1995;4(4): 380-6. lVarner KE, Fulton GA. The economic implications of tobacco product sales in a nontobacco state. ]olivllnl of ii7r Aural-imrr Mrdicnl Associatiorz 1994;271(10):771-6. Warner KE, Fulton GA, Nicolas P, Grimes DR. Employ- ment implications of declining tobacco product sales for the regional economies of the United States. ]o~rr- )d of tlrc Anzericnr7 Medical Association 1996;275(16): 1241-6. Warner KE, Hodgson TA, Carroll CE. Medical costs of smoking in the United States: estimates, their validity, and their implications. Tobacco Coi7frol 1999;8(3): 290-300. Wasserman J, Manning WG, Newhouse Jl', Winkler JD. The effects of excise taxes and regulations on cigarette smoking. Jo~rrrwl crfHenlfl7 Ecor?on~ics 1991;10(1):43-64. Watkins BG III. The tobacco program: an econometric analysis of its benefits to farmers. ilnwrican Ecor7onrisf 1990;34(1):45-53. Winston GC. Addiction and backsliding: a theory of compulsive consumption. ]ourwl of Econonlic Brt7miou ad Orpr~i~atior~ 1980;1(4):295-324: Womach J. Tobncco Price S7qporf: AJI Oz~eruieu~ of tl7c Pw~rwl. Washington: Library of Congress, Congres- sional Research Service, 1994a. CRS Publication No. 94-243 ENR. Womach J. U.S. Tokncro Pmd7~tior7: Prospects for Iire Fu- tuw. Washington: Library of Congress, Congressional Research Service, 1994b. CRS Publication No. 94-672 ENR. Womach J. T~~JoccLI-RL~~~~ Prqra777s 1717d Acfivities of tl7c U.S. Drprf,i7cr7f of A<~`icfrlfure: Opfw?fiol7 17)Zd Cost. Washington: Library of Congress, Congressional Re- search Service, 1999. CRS Publication No. 97-417 ENR. Yurekli AA, Zhang P. The impact of clean indoor-air laws and cigarette smuggling on demand for cigarettes: an empirical model. Hefilth Em~70772ics 2000;9(2):159-70. Zhang I', Husten C. Impact of the tobacco price sup- port program on tobacco control in the United States. Tobacco Co17frol 1998;7(2):176-82. Zhang I', Husten C, Giovino G. Effect of the tobacco price support program on cigarette consumption in the United States: an updated model. An7ericarz ]ournal of Public Health 2000;90(5):746-50. Chapter 7 Comprehensive Programs Introduction 373 Conceptual Frameworks 373 Description of Comprehensive Programs 374 Program Goals for Reducing Tobacco Use Statewide 374 Program Components for Reducing Tobacco Use Statewide 375 Community Intervention Trials 376 Statewide Interventions 387 Community Mobilization 382 National Programs 383 ASSIST 383 IMPACT 384 SmokeLess States Program 384 National Programs to Reduce Youth Access to Tobacco 381 States Currently Funded in the Nationwide Program to Reduce Tobacco Use 385 Examples of Major State Programs 385 Minnesota 386 California 387 Massachusetts 389 Arizona 391 Oregon 392 Maine 393 Programs Funded by State Settlements With the Tobacco Industry 391 Mississippi 391 Florida 395 Texas 397 Minnesota Settlement Program 397 Programs Meeting the Needs of Special Populations 398 Programs for the African American Community 398 Programs for Women 400 Federal and State Programs 401 Religious Organizations 401 Special Efforts to Reduce Chewing Tobacco Use 402 Components of Community Programs 403 Community Advocacy and Mobilization 403 Electronic Networking 403 Direct Advocacy 405 Media Advocacy 407 Countermarketing 309 Summary 47 6 Conclusions 11 T References 318 Introduction A comprehensive approach to reducing tobacco use recognizes that individual behavioral choices occur in a larger, complex context: a social setting of family, community, and culture; a complex economic and physical environment; formal and informal gov- ernment policy; and the prevailing legal atmosphere (Green and Richard 1993). The specific programs re- vie\\Ted in prior chapters can be better understood as part of a general framework for health promotion (World Health Organization [WI-IO] 1986; Ht~lfll Pmnofiolr Illferrlntiorznl 1997). Using such a frame- work, this chapter will review- community-based intervention studies and the current models for con- prehensive tobacco prevention and control that are funded by specific excise taxes or by settlements with the tobacco industrv. The evaluation of multicomponent interwntions and socioecological models of health promotion poses a special problem (Green and Kreuter 1991; Sanson- Fisher et al. 1996; Nutbeam 1998). The most effective models of health promotion are social movements that evolve (Kickbusch 1989; Allison and Rootman 1996; Downie et al. 1996; Nutbeam 1998). Thus, the nature and complexity of health promotion interventions do not fit the tightly defined, controlled, and presumably reproducible research model that is more suitable for epidemiologic testing (Elder et al. 1993; Mittelmark et al. 1993; Baum 1995; Allison and Rootman 1996; Conceptual Frameworks Macdonald et al. 1996; Nutbeam 1996, 1998). None- theless, surveillance data, periodic surveys, and other administrative data from multiple sites permit these interventions, as well as "natural experiments," to be studied. Traditionally, per capita consumption data, adult prevalence surveys, and surveys of tobacco- related behaviors among young people have been the core of this surveillance approach. Recently, a broader arrav of legislative, economic, media, and program data-has emerged to enhance sur\,eillance of the social environments that influence the use of tobacco prod- ucts. For example, the WHO's Glliilrlirzes ,fw Co~ztrol- /i/r;; 1711d MoJrifoiir7~~ thr TohnCro E@?JJ7iC (WHO 1998) provides detailed recommendations on the types of data that should be monitored for both planning and e\,aluating tobacco control efforts. For the United States, the Centers for Disease Control and Prevention (CDC) has published background information on sources of national surveillance data (Giovino et al. lYY4). The Federal Trade Commission provides annual estimates of trends in the tobacco industry's advertising and promotion expenditures. Surveillance data on protobacco influences are not well monitored, however, particularly at the state level. Finally, Wakefield and Chaloupka (1999) have provided a con- ceptual framework for the monitoring of comprehen- sive tobacco control programs, particularly those that focus on preventing teenage smoking. From its formation in the mid-1970s, health pro- motion has emerged as an approach that offers greater potential for change in the health-related behavior of populations than does health education (Green and Richard 1993; Downie et al. 1996; Hcnltlz Pronto- fioil IJ7termfional 1997). Health promotion emphasizes social, economic, and other environmental influences as the primary determinants of health behavior change (WHO 1986; Downie et al. 1996; Hmltl7 PmJ~~ofioJl /i~f~~rr~~tiom/ 1997). Though such health promotion strategies have been characterized as a ne\v approach to public health, ecological and policy-oriented approaches are similar to the public health methods of the latter part of the 19th century and the early de- cades of the 20th century (Kickbusch 1989; Green and Richard 1993; Mullan 2000). As the role of individual risk behaviors, such as tobacco use, was increasingly understood in the middle of the 20th century, individu- ally focused educational strategies gained primacy (Green and Richard 1993). These strategies produced some important changes in health behaviors, but their limits \2-ere realized in the cardiovascular disease prevention programs that took place in the United States during the 1970s and lY8Os (see "Community Intervention Trials," later in this chapter) (Green and Richard 1993; Luepker 1994; Winklebv 1994; Fisher 1995; Schmid et al. 1995; Susser 1995). - The shift from a health education approach that targets the individual to a health promotion approach that uses social, policy, and environmental strategies has several advantages. First, by recognizing that many environmental determinants of health behavior Description of Comprehensive Programs are not under the direct control of the individual, the ecological focus avoids blaming persons who fail to modifv their behavior. Second, many educational strateiies are more effective with better-educated, wealthier persons and may thereby increase the dis- parities in health between population groups and fail to reach those in greatest need. Third, regulatory and policy interventions can be more cost-effective than multiple efforts to modify individual behavior. The importance of comprehensive economic, policy, and regulatory interventions to reduce tobacco use has long been recognized by international experts (WHO 1979). For example, the evolving WHO guide- lines for such interventions have increasingly empha- sized policy and legislative measures, stressing that these types of health promotion and health protection strategies are essential elements of any national effort to reduce tobacco use (WHO 1998). In an extension of the WHO's efforts, the National Cancer institute (NCI) released a blueprint for related public health ac- tion in the United States (NC1 1991). This monograph stressed that the application of social environmental approaches should not compete \vith individual ap- proaches but should be combined synergistically with them. Similarly, the Center for Substance Abuse Pre- vention (CSAP) of the Substance Abuse and Mental Health Services Administration (SAMHSA) published guidelines that provide the concept, structure, and operations of a community-based approach to reduce tobacco use among youth (SAMHSA 1998a,b). To fur- ther help states overcome common obstacles to enforcing youth access laws, CSAI' also has prol,ided a document that provides strategies to address prob- lems such as interagency and intraagency issues, in- sufficient or uncoordinated resources, or lack of data sources (U.S. Department of Health and Human Ser- vices [USDHHS] 1999). More recently, the CDC (1999a) has synthesized a comprehensive framework for state- wide programs to reduce tobacco use. This framework integrates four program goals with four program com- ponents; optimally, each of the goals lvould be fully addressed in the implementation of each of the com- ponents. The framervork, described in the next sec- tion of this chapter, recognizes that comprehensive pro- grams will continue to evolve, in response both to new information and to new circumstances. In addition, the framework represents a distillation of evidence and judgment that have been discussed in detail in the ear- lier chapters of this report and that have been tested in the community-based trials and the comprehensive programs discussed later in this chapter. Program Goals for Reducing Tobacco Use Statewide 1. Preuerrt ilritiatiorr amorlg yomg people. The hallmarks of this goal are o Decreasing young people's access to tobacco products. o Increasing prohealth messages. o Reducing protobacco messages. o Increasing the price of tobacco products. Some of the mechanisms for decreasing young people's susceptibility to tobacco use are promoting vouth empowerment activities, providing school health education, offering positive alternatives, deglamorizing tobacco use, and in\,olving parents and families. 2. Promotr quitting among adults and yo1411g people. An environment that supports efforts to quit using tobacco can be fostered by o Increasing access to culturally appropriate, effective cessation services (e.g., by expanding insurance coverage). 9 Increasing the price of tobacco products. o Increasing restrictions on environmental tobacco smoke (ETS). o Increasing prohealth messages. o Decreasing protobacco messages. 3. Eiimiuate exposure to ETS. The continued ex- pansion of policies to eliminate exposure to ETS can be achieved by o Developing support for implementation. o Enforcing voluntarv private policies. o Enforcing public pilicv and public regulation. o Expanding coverage o? public areas. 1. Iderztify and eliminate disparities nmoq popla- tim groups. Intrinsicallv linked to achieving the first three goals, elimina;ing disparities entails Increasing the price of tobacco products through culturally acceptable programs. Decreasing exposure to ETS. Increasing prohealth messages. Decreasing protobacco messages, particularly those aimed at population subgroups. Increasing the availabilitv of culturally acceptable cessation services. increasing protective factors among young people. Decreasing young people's access to tobacco products. Development, funding, and implementation the major elements-some of which appear in several of these goals-are critically linked to com- munity involvement and, as noted, to a culturallv appropriate approach. Program Components for Reducing Tobacco Use Statewide Community interventions. Working through social organizations, systems, and networks promotes an environment that facilitates indi- vidual health choices and establishes freedom from tobacco use as the norm. The term "commu- nity" encompasses a diverse set of entities, includ- ing medical societies; schools; school districts; departments of education; voluntary health agen- cies; civic, social, and recreational organizations; businesses and business associations; city and county governments; public health organizations; labor groups; managed care systems; faith com- munities; and organizations for racial and ethnic minority groups. Community-based activities can include sup- porting legislated removal or restriction of stimuli to use tobacco (such as advertising and promotion, easy access to tobacco products via self-service display and vending machines, and ongoing ex- posure to ETS) as well as providing positive alter- natives (such as promoting cessation, encouraging prevention advocacy, developing role modeling through parents and adults, and fostering youth empowerment). By changing the community set- ting and institutions with which adults and young people interact, community-based activities work to denormalize, deglamorize, and discourage to- bacco use and to provide access to resources that increase users' abilitv to control their addiction and use of tobacco. This approach has the potential to effect substantial, sustained, populationwide change in tobacco use behavior. Collrlterlnnrketing. Changing a social environ- ment that fosters a norm of tobacco use is an essential element of national, state, and local pro- grams. This change requires strategies to counter the billions of dollars spent in advertising and pro- motion that reach young people and adults with misleading images about tobacco. Countermarket- ing efforts can include using media advocacy, paid media, and counteradvertising; increasing prohealth promotions and sponsorships; and pro- viding information on the tobacco industry's mar- keting and promotional tactics. These public health messages should use a strategy that targets all age groups and populations. In a comprehen- sive strategy, education messages will be mutu- ally reinforcing: clean indoor air messages will provide added motivation for adults to quit smok- ing; cessation messages for adults will discourage tobacco use among young people and accentuate the problem of addiction; and youth prevention messages will increase the salience of the tobacco issue among parents and community leaders. Program policy ad regulatiou. Areas in which policy and regulation to reduce tobacco use have been applied include minors' access, tobacco pric- ing, advertising and promotion, clean indoor air, product regulation, product labeling, ingredient disclosure, and policies on insurance coverage for cessation services. Policies and regulations can be established at the federal, state, and local lev- els (see Chapter 5). Ideally, policies and regula- tions need to be implemented at both the community level and statewide. Educating the public about policies and regulation is crucial to acceptance, but such education must be sup- ported by adequate enforcement. 4. Surveillarlce aud evaluatiorz. Surveillance and evaluation efforts are necessary to make the ongoing refinements that lead to more effective pre- vention strategies. In addition to traditional surveillance methods, nontraditional approaches- such as monitoring the promotional activity of the tobacco industry at the state and local levels, moni- toring the economic impact of smoking laws and other ETS policies, and performing periodic surveys of public opinion on program interventions-are critical for reducing tobacco use. The conceptual framework for comprehensive efforts to reduce tobacco use has been used to develop the current generation of statewide programs. However, even the most comprehensive programs Community Intervention Trials currently in place have not been able to fully imple- ment all recommended components. Policy and regu- lation components are especially hampered, since many state and local actions are limited by federal mandates and preemptions (see "Preemption of Local Action by State Policy" in Chapter 5). Moreover, only two states, California and Massachusetts, have imple- mented comprehensive programs for a sufficient time to provide evaluation data on the overall efficacy of the emerging comprehensive model. The following sections summarize the history and development of community-based, statewide, and other large-scale efforts to reduce tobacco use and con- clude with a review of existing data on the efficacy of the comprehensive model. Large-scale trials to prevent cardiovascular disease have been a major source of data on population- based approaches to reducing tobacco use. An empha- sis on the importance of addressing social and cultural determinants of smoking behavior grew directly out of early work on cardiovascular disease epidemiology. The Seven Countries Study, which was started in the mid-1950s by Keys and colleagues (Aravanis et al. 1970; Blackburn et al. 1970; Buzina et al. 1970; Fidanza et al. 1970; Kimura and Keys 1970; Taylor et al. 1970a,b), examined risk factors for cardiovascular disease in populations around the world and documented that disease rates and risk factors differed markedly across cultural and social environments (WHO 1982). In that study, more than 12,500 men aged 40-59 years from Finland, Greece, Italy, Japan, the Netherlands, the United States, and Yugoslavia were recruited for a pro- spective study of the relationship between personal behaviors (e.g., diet, physical activity, smoking) and risk of cardiovascular disease (Aravanis et al. 1970; Blackburn et al. 1970; Buzina et al. 1970; Fidanza et al. 1970; Kimura and Keys 1970; Taylor et al. 1970a,b). Al- though the most striking differences in lifestyle across cultures were in the composition of the men's diet, smoking was found to be a significant risk factor. This study, and many other early studies of cardiovascular disease epidemiology, encouraged researchers to start community trials to modify the identified risk factors in whole population groups (WHO 1982). Two landmark community trials that began in 1972 grew directly out of the work of the Seven Coun- tries Study investigators: the Stanford Three-Commu- nity Study (Farquhar et al. 1977) and the Finnish North Karelia Study (Puska et al. 1985). A third, less directly tied to this early work, was the Israeli Community Syn- drome of Hypertension, Atherosclerosis and Diabetes (CHAD) program (Gofin et al. 1986) begun in 1971. In addition, two worksite trials focusing on population- level changes in cardiovascular disease risk factors developed out of the Seven Countries Study and from related early work on cardiovascular disease epidemi- ology: the Belgian Heart Disease Prevention Project (Kornitzer et al. 1980) and the United Kingdom Heart Disease Prevention Project (Rose et al. 1980). Though investigators in these initial studies recognized the im- portance of the social and cultural environment in modifying risk factors for cardiovascular disease, in- cluding smoking (Farquhar 1978; WHO 1979; Farquhar et al. 1981,1985; Rose 1981; McAlister et al. 1982; Puska et al. 1985), the smoking cessation techniques of the time were primarily individually oriented (McAlister et al. 1976; Meyer et al. 1980). The Stanford and North Karelia studies shared some community organizing and conceptual perspec- tives in their planning (WHO 1982). Logistical and cultural differences between the United States and Fin- land dictated significantly different implementation, however. In the Stanford study, an intervention that Rducirlg Tobacco Use primarily used mass media was compared with the same mass media intervention plus intensive face-to- face counseling for high-risk individuals and was also compared with a control community that received no intervention. In the initial results, the community cohort receiving both the mass media and the face-to- face counseling for high-risk smokers had a signifi- cantly greater decrease than the control community in the prevalence of smoking (-50 vs. -14.9 percent) and in the number of cigarettes smoked (percentage reduc- tion of 51.6 vs. 21 .O percent) (Farquhar et al. 1977,1985; Maccoby et al. 1977; Meyer et al. 1980). In the Finnish study, the people of North Karelia province requested the intervention because of concerns raised by the results of the Seven Countries Study, in which residents of their province had participated (Puska et al. 1985,1995). The intervention had a strong focus on community organizing and environmental modification, together with multiple educational com- ponents using mass media and other strategies (McAlister et al. 1982; Puska et al. 1985). Although the intervention's early efforts had a greater emphasis on increasing direct cessation services than on prevent- ing smoking, the importance of nonsmoking environ- ments and other environmental changes \yas clearly recognized and emphasized (Koskela 1981). The five- year follow-up results of the studv found no signifi- cant difference in smoking prevalence between the North Karelia province and Kuopio, a comparison province with similar baseline smoking rates (Puska et al. 1979). Ten years on, a significantly greater re- duction in smoking prevalence was observed among men in North Karelia than in Kuopio (Salonen et al. 1981; Puska et al. 1983a,b; Vartiainen et al. 1986). The intervention trial has been continued, and new pre- vention and population-based cessation strategies have been added (Vartiainen et al. 1986; Korhonen et al. 1992,1993). Analyses of 20-year trends (from 1972 to 1992) in smoking in the two provinces found a sig- nificantly greater decline in smoking prevalence for adult men in North Karelia (from 52 to 32 percent) than in Kuopio (50 to 37 percent) and in southwestern Fin- land. Smoking prevalence for adult women increased at similar rates in both provinces (increasing from 10 to 17 percent in North Karelia and from 11 to 19 per- cent in Kuopio) (Vartiainen et al. 1998). The 20-year difference in trends in men between the two provinces appeared to be primarily related to cessation during the first 10 years and to prevention during the last 10 years. The CHAD program had a somewhat more indi- vidually focused intervention model directed at reduc- ing the risk factors for cardiovascular disease among residents in Israeli housing projects (Abramson et al. 1981). The health care providers serving the interven- tion communities provided risk factor screening and counseling for families, couples, and individuals liv- ing in the four adjacent housing projects. The resi- dents of comparison housing areas received usual care from their providers. In the intervention communi- ties, group discussions were held to provide social support and increase group influences on individual lifestyle changes. Comparisons between community health surveys conducted at baseline (1969-1971) and after five years (1975-1976) showed a significantly greater decline in smoking prevalence among men but not among women in the intervention communities than in control communities (Gofin et al. 1986). At the lo-year follow-up (1981), the prevalence of smoking had declined significantly between 1976 and 1981 among both men and women in the CHAD follow-up cohort, whereas no change or a slight increase in smok- ing had occurred among adults in Israel overall (Gofin et al. 1986). The Belgian Heart Disease Prevention Project was a controlled, multifactorial trial involving men aged 40-59 years at baseline at Belgian worksites (Kornitzer et al. 1980). Thirty pairs of factories were studied, with one site from each pair randomly assigned to the in- tervention group and one site to the control group. At baseline screenings for risk factors for cardiovascular disease, individuals in the upper two deciles of risk were identified and received semiannual individual counseling from the medical staff. Medical advice to quit smoking w'as reinforced in the factories by anti- smoking posters, written messages, and health educa- tion conferences encouraging workers to quit smoking and to encourage the same to their friends who smoked. Changes in smoking prevalence at the intervention and control worksites were monitored among both the high-risk individuals and in random samples of the total worksite populations. After two years of inter- vention, a significantly greater percentage of the high- risk smokers quit in the intervention group than in the control group (18.7 vs. 12.2 percent), but no difference was observed in the random samples. The United Kingdom Heart Disease Prevention Project was started in 1971 with 24 pairs of English and Welsh factories. Each member of the pair was ran- domly assigned to intervention or control status (Rose et al. 1980; Bauer et al. 1985). At baseline and in 1977- 1978, risk factor screening for cardiovascular disease was conducted among men aged 40-59 years in the intervention sites and in a lo-percent random sample of similarly aged men at the control sites. Over a five- to six-year period, all men in the intervention sites