State Regulations of Smoking in Public Places: Determinants and Implications on the Demand for Smoking and Consumers' Behavior

Transcription

1 Western Michigan University ScholarWorks at WMU Dissertations Graduate College State Regulations of Smoking in Public Places: Determinants and Implications on the Demand for Smoking and Consumers' Behavior Ioana Raluca Mazare Western Michigan University Follow this and additional works at: Part of the Behavioral Economics Commons, Health Economics Commons, and the Substance Abuse and Addiction Commons Recommended Citation Mazare, Ioana Raluca, "State Regulations of Smoking in Public Places: Determinants and Implications on the Demand for Smoking and Consumers' Behavior" (2001). Dissertations This Dissertation-Open Access is brought to you for free and open access by the Graduate College at ScholarWorks at WMU. It has been accepted for inclusion in Dissertations by an authorized administrator of ScholarWorks at WMU. For more information, please contact

2 S T A T E R E G U L A T IO N S O F S M O K IN G IN P U B L IC PLA C ES: D ETERM LNANTS A N D IM PL IC A T IO N S O N T H E D E M A N D FO R S M O K IN G A N D C O N S U M E R S B E H A V IO R by Ioana R aluca M azare A D issertation S ubm itted to the Faculty o f T he G raduate C ollege in partial fulfillm ent o f the requirem ents fo r the D egree o f D octor o f P hilosophy D epartm ent o f E conom ics W estern M ichigan U niversity K alam azoo, M ichigan D ecem ber 2001

3 INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, som e thesis and dissertation copies are in typewriter face, while others may be from any type of com puter printer. The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., m aps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. ProQ uest Information and Learning 300 North Z eeb Road, Ann Arbor, Ml USA

4

5 UMI Number: UMI UMI Microform Copyright 2002 by Bell & Howell Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United S tates Code. Bell & Howell Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml

6 C opyright by Ioana R aluca M azare 2001

7 ACKNOWLEDGEMENTS I w ould like to begin by expressing m y gratitude to all the m em bers o f my dissertation com m ittee. I w ould like to thank D onald A lexander for his useful com m ents, suggestions and corrections on this m anuscript. I am especially grateful to M atthew H iggins for his continuous help w ith the em pirical m ethodology for my research and his support throughout the year and the entire doctoral program. I also w ant to thank H uizhong Zhou and R obert F einberg for th eir com m ents at various stages o f this m anuscript. Secondly, I w ould like to acknow ledge the trem endous support o f my fam ily, in p articular m y m other, A ntoaneta, m y father, Spiru, and m y sister, A lina. T heir love and th eir b elief in m y success have given m e strength and m oral support throughout th ese years. L astly, I w ould like to dedicate m y research to the person w ho is always next to m e, inspires m e and show s me everyday how beautiful life is, m y husband, Florin. Ioana R aluca M azare ii

8 TABLE OF CONTENTS A C K N O W L E D G E M E N T S...ii L IS T O F T A B L E S...vi L IS T O F FIG U R E S...viii C H A P T E R I. IN T R O D U C T IO N... 1 II. H ISTO RY O F S M O K IN G R E G U L A T IO N A N D ST A T E R E G U L A T IO N O F SM O K IN G IN PU B L IC P L A C E S The Evolution o f S m oking R egulation State Regulations R egarding Sm oking in Public P laces State R egulations R egarding Sm oking in Public Places -C a te g o rie s Facts on C igarettes S m oking in the U nited S tates A dult Sm oking P revalence U nited States P o p u latio n State-Level C igarette Sm oking D ata Sm oking-a ttributable D iseases, M ortality and Y ears o f Potential Life L o st III. STA TE R EG U L A T IO N S O F SM O K IN G IN PU B LIC PLA C ES: D ECISIO N S O N T IM IN G A N D R E S T R IC T IO N S...44 Tim ing and R estrictiveness R egarding R egulations o f Sm oking in Public P laces...52 iii

9 T able o f C ontents C ontinued C H A PTER T im in g...52 R estrictiveness T heories o f R egulation...57 D iscussion o f the V arious Interest G ro u p s E m pirical M odels and R esu lts...64 T im e o f S tate N o-sm oking R egulations...64 R estrictiveness o f State N o-sm oking R eg u latio n s...95 C onclusion IV. D E M A N D FO R C IG A R E T T E S AND ST A T E R E G U L A T IO N S O F S M O K IN G IN PU B L IC P L A C E S T he Im pact o f P rice and A nti-sm oking Policies on the D em and for C ig arettes A nalytical F ram ew o rk T he D ata R esu lts E stim ation o f the P robit Equation for S tates D ecision to Regulate S m oking in Public P laces Estim ation R esults for the C igarette D em and E q u atio n C onclusion V. T H E IM PA C T O F S T A T E R E G U L A T IO N S O F SM O K IN G IN PU B L IC PLA C ES O N C O N SU M PT IO N O F A L C O H O L M ethodology iv

10 T able o f C ontents C ontinued C H A PTER T he D a ta D escriptive Statistics o f D ependent and Independent V ariab les R esu lts T h e Im pact o f N o-sm oking R egulation on A lcohol C o n su m p tio n T h e Im pact o f N o-sm oking R egulations on C igarette C o n su m p tio n C o n c lu s io n VI. SU M M A R Y A N D C O N C L U S IO N S E N D N O T E S A PPEN D ICES A. D escriptive Statistics for State-Level V ariables in R egression E q u atio n s 199 B. Sources o f S tate-l evel D ata C. D escriptive S tatistics for Independent V ariables from the B R FSS D a ta 205 B IB L IO G R A P H Y v

11 LIST OF TABLES I. State C lean Indoor A ir Law - Dates W hen the L aw s Have Been Enacted, as o f June 30, A.State R egulations o f Sm oking in Public Places, as o f June 30, 1995 R estrictiveness B.State R egulations o f Sm oking in O ther Places, as o f June 30, R estrictiveness Percentage o f A dults A ged 18 and O lder W ho R eported C igarette Sm oking, by State S elected Y ears D eaths R elated to Sm oking, E nacting D ates for the First State R egulation o f Sm oking in Public P laces C ategories o f S tate Legislation by D egrees o f R estriction E nforcem ent E xpectations A bout C oefficient Signs as P redicted by the Tw o T heories o f R egulation for the D uration M o d els C orrelations B etw een the R egression V ariab les W eibull D uration M odel for the First T im e W hen States R egulate Sm oking in a Public Place ( ) - M axim um L ikelihood E stim ates A.W eibull D uration M odels for State R egulations o f Sm oking in Public Places ( ) - M axim um Likelihood E stim ates I0B. W eibull D uration M odels for State R egulations o f Sm oking in Public Places ( ) M axim um Likelihood E stim ates C. W eibull D uration M odels for State R egulations o f Sm oking in Public Places ( ) M axim um Likelihood E stim a te s I I. E xpectations A bout C oefficient Signs as P redicted by the T w o T heories o f R egulation for the D uration M odels vi

12 List o f Tables - C ontinued 12. O rdered Probit M odel E stim ation for the O verall S everity o f State Regulation o f Sm oking in Public Places ( ) M axim um L ikelihood E stim ates O rdered Probit M odel E stim ation for the S everity o f R egulation o f Sm oking in Public Places ( ) M axim um L ikelihood E stim ates C orrelations B etw een R egression V ariables R esults from the Probit E stim ation of the Pooled D a ta R esults from the E stim ation o f the D em and for C igarettes E quation by Instrum ental V ariab les R esults from the E stim ation o f the D em and for C igarettes E quation by Instrum ental V ariables (W ith O ne Indicator V ariable for the R egulation o f S m oking in Public Places) A lcohol and C igarette C onsum ption - D escriptive S tatistic s A lcohol and C igarette C onsum ption O ver T im e Sm oking and D rinking Participation A lcohol and C igarette C onsum ption Instrum ental V ariables E stim ation of D em and for A lco h o l Instrum ental V ariables E stim ation of D em and for C ig a rette s vii

13 LIST OF FIGURES 1. Percent o f C urrent S m okers in the U.S. T otal P o p u latio n Percent o f C urrent S m okers by G en d er Percent o f C urrent S m okers by R ace Percent o f C urrent S m okers by A ge Percent o f C urrent S m okers by E ducatio n A verage N um ber o f C igarettes Sm oked by the C urrent A dult Sm okers - U.S. P opulation Percent o f C urrent S m okers W ho Sm oke 25 or M ore C igarettes per D ay - by G en d er Percent o f C urrent S m okers W ho Sm oke 25 or M ore C igarettes per D ay - by R a c e Percent o f C urrent S m okers W ho Sm oke 25 or M ore C igarettes per D ay - by A g e Percent o f C urrent S m okers W ho Sm oke 25 or M ore C igarettes per D ay - by E d u catio n Per C apita C onsum ption o f C igarettes in the U nited S tates, C om parative C auses o f A nnual D eaths in the U nited S tates, D eaths A ttributable to C igarette Sm oking in the U nited States, K aplan-m eier E stim ators o f Survival F unction and H azard Function for the O verall R egulation o f Sm oking in Public Places (A ny Location), viii

14 L ist o f Figures - C ontinued 15. K aplan-m eier E stim ators o f S urvival Function and H azard F unction for State R egulation o f S m oking in G overnm ent W o rk sites K aplan-m eier E stim ators o f S urvival Function and H azard Function for State R egulation o f S m oking in Private W orksites K aplan-m eier E stim ators o f S urvival Function and H azard F unction for State R egulation o f Sm oking in R esta u ra n ts K aplan-m eier E stim ators o f S urvival Function and H azard F unction for State R egulation o f S m oking in C om m ercial C hild D ay C a re K aplan-m eier E stim ators o f S urvival Function and H azard F unction for State R egulation o f S m oking in H om e-b ased C hild D ay C a re K aplan-m eier E stim ators o f S urvival Function and H azard F unction fo r State R egulation o f S m oking in O therp laces ix

15 C H A P T E R I IN T R O D U C T IO N O ver the past 40 years, policy m akers have been concerned w ith the health consequences of cigarette sm oking, and research at public and private institutions has increased the public s aw areness about the health hazards o f tobacco use. Today there is a consensus am ong health experts that cigarette sm oking is associated with deadly diseases, such as various cancers, cardiovascular and heart d iseases.1 M oreover, C halupka and W arner (1999) estim ate that tobacco products are responsible for over one-fifth o f the annual deaths in the U.S. during m iddle age.2 Since 1964, w hen the Surgeon G eneral m ade the first clear statem ent with regard to the health hazard o f cigarette sm oking, continuous efforts by state and federal governm ents have been taken in order to com bat sm oking and prevent tobacco u s e / Public policy actions have intensified in the past decades and have focused on specific issues, such as preventing teenagers from starting to sm oke and decreasing cigarette consum ption am ong sm okers. R esearch at public and private institutions has show n that sm okers are not th e o n ly ones at risk because o f cigarette sm oking. B y standers w ho inhale the sm oke o f cigarette m ay becom e ill as w ell, and secondhand sm oke represents an additional cause o f concern. T he Surgeon G eneral pointed out in its 1978 Report the potential dangers to w hich non-sm okers m ay be exposed 1

16 because o f cigarette sm oke.4 In 1986, the Surgeon G eneral R eport concentrated entirely on the health consequences o f environm ental tobacco sm oke (E T S ) on nonsm okers and acknow ledged that involuntary sm oking causes serious diseases, including cancer, in non-sm okers.3 T he grow ing aw areness o f the dan g er o f secondhand sm oke, to g eth er w ith the change in the public s attitude to w ard sm oking after 1964, has created pressure for the governm ents to control the n eg ativ e externality created by tobacco use and to protect non-sm okers. R egulations th a t restrict sm oking in public places represent the tool used by policy m akers to p ro tect non-sm okers from the health hazard o f secondhand sm oking. A lthough the federal governm ent restricted sm oking in federal w orkplaces and m ajor transportation facilities,6 state governm ents w ere granted considerable freedom in regulation o f sm oking in public places. T his has resulted in a w ide variation across states regarding the date when sm oking was regulated, the severity o f regulation, and the categories o f public locations subject to regulation. From an econom ic point a view, these regulations represent an opportunity to study the m echanism s behind s ta te s decision to pass social regulation and the im plications o f such legislation in society. T he present research focuses on these issues and analyzes regulations o f sm o k in g in public places, the factors that affect states regulatory behavior and the eco n o m ic im plications o f no-sm oking legislation. W hile the m ajority o f econom ic analysis has focused on the use o f taxation as a w ay to discourage consum ption o f cigarettes, very few studies have exam ined the effectiveness of no-sm oking restrictions in public places as an instrum ent used by

17 the governm ent to co rrect for a m arket failure and the negative effect o f ETS on nonsm okers health. T he interesting feature o f these regulations is that th ese laws are the product o f state legislatures, and each state has had considerable freedom in choosing w hether and w hen to regulate sm oking, w hich specific places w ere to be regulated, and how restrictive these regulations w ere. C hapter II o f this research presents a picture o f states regulations that restrict sm oking in public places, as o f Inform ation on state regulations o f sm oking in public places is available through the State T obacco Activities T rackin g A nd E valuation (STA TE) S ystem. T he S T A T E System sum m arizes regulation o f sm oking in public places for all 50 states and W ashington, D.C., and locations are grouped in six categories : (1) governm ent sites; (2) private-sector w ork sites; (3) restaurants; (4) com m ercial child d ay care; (5) hom e-based ch ild day care, and (6) o th er places (including bars, shoppin g m alls, g rocery stores, enclosed arenas, public transportation, hospitals, prisons, and hotels and m otels). T he regulatory package regarding sm oking in public places d iffers largely from one state to another. R egulations regarding sm oking in public places have been p assed in m ore than three decades, and there w ere still five states w ith no regulation at the end o f O nly a handful o f states regulate sm oking in all public places, and ban sm oking com pletely in som e public places. In C hapter III o f m y dissertation, I plan to exam ine em pirically, using the econom ic theories o f regulation, how specific econom ic and political factors interacted to explain the variation in the states sm oking legislation. T his is im portant 3

18 for several reasons. First, given the public concern and aw areness of the adverse health effects that sm oking has on both sm okers and nonsm okers, the fact that there are such w ide differences in sm oking legislation across states constitutes a concern for public policy. If such legislation is aim ed at controlling a negative externality (that o f secondhand sm oking), we should observe all states banning sm oking com pletely in all p ublic places to ensure the best protection o f all non-sm okers. The very different reality raises the question about how states decide in fact w hen they regulate sm oking, w hat places they regulate, and w hat restrictions they im pose by law. L o o k in g at the state-specific factors o f econom ic, political and social nature m ay help explain w hy each state takes a different approach to this m atter and what else, besides p u b lic s interest, interferes with the decision to regulate. Second, if the behavior at state level is understood and if it is know n w hat forces determ ine when and w heth er restrictions on sm oking are im posed, this co u ld be a good resource for policy m akers to find the m ost effective tools to induce changes in sm okers behavior and m inim ize the ex tern ality problem associated with secondhand sm oke. T h e strategy is to use the richness o f the data in order to em pirically test the tw o com p etin g theories o f regulation. T he traditional view on regulation is the public interest theory, according to w hich the public dem ands som e form o f regulation from the governm ent to correct for a real o r perceiv ed m arket failure. The basis o f m y approach is the econom ic theory o f regulation (ET). T h e ET predicts that regulation is the resu lt o f com peting interest groups th a t o ffer political support in exchange for legislation favorable to them. ET im plies th at regulation provides benefits to the 4

19 group that is better organized and dem ands regulation that favors it. There are several research questions that are addressed in C hapter III. First, w hat econom ic, political and social factors determ ine w hen a particular state restricts sm oking in public places? Second, w hich factors determ ine the specific places that are subject to regulation? A nd, third, w hat factors determ ine how restrictive the regulation will be? T he results o f the analysis reveal that the econom ic theory o f regulation is indeed helpful in explaining states regulatory behavior. There are specific factors that, w ithin each state, represent forces that policy m akers take into consideration regarding no-sm oking regulation. Specifically, tobacco com panies and restaurant ow ners represent tw o pow erful interest groups that delay regulation. S tates with higher divorce and unem ploym ent rates (which p roxy for the level o f stress in a state s population) regulate later, as w ell. Political factors are im portant also for state policy makers. States w here D em ocrats have more control regulate later. A lthough the result seems surprising, it captures the fact that D em ocrats control Southern states w here tobacco is produced and em ploym ent in tobacco industry is significant. Since D em ocrats prom ote a w orker-oriented agenda, they are also careful about the state legislation that affects tobacco sales. A nother explanation is the high state incom e that results from tobacco industry and influences state legislators in their decision to regulate sm oking in public places. W hile the econom ic theory explains the tim ing o f regulation, the public interest theory explains the severity o f regulation. O nce the decision to regulate sm oking is m ade, factors like incom e, the proportion o f children 5

20 in states population, and cigarette consum ption determ ine how restrictive regulations o f sm o k in g in public places are. A lthough the prim ary intent o f sm oking regulation in public places was to protect n o n sm o k ers heaith, the econom ic literature suggests that there m ight be unintended effects o f this legislation.7 In this case, by restricting sm oking in a num ber o f public places, sm okers need to change th eir behavior. For exam ple, sm okers m ay be forced to sm oke in restricted areas and only during break tim e. C onsequently, th e tim e for sm oking is reduced and discom fort due to the necessity to com ply with legislation is caused, w hich increases the cost o f smoking. Therefore, sm okers m ay p erceive cigarettes as becom ing m ore expensive and may change their cigarette consum ption. C hapter IV o f this research explores this question about the relationship between cigarette consum ption and regulations o f sm oking in public places. A new m ethodology to investigate this relationship is proposed. Previous work on this topic concludes that reg u latio n s o f sm oking in public places decrease the dem and for cigarettes. H ow ever, the results are likely to be unreliable because an im portant issue has been ignored, w hich w ill be discussed next. C onsum ption o f cigarettes and the regulatory package regarding sm oking in public places vary w idely across states. It m ay be the case that states w here sm oking is less prevalent are the states more likely to pass regulation against sm oking. In this scenario, legislation proxies the anti-sm oking sentim ent that exists in state s population and favors regulation o f sm oking. In those states, sm oking consum ption 6

21 w ould decrease any w ay, no m atter w hether legislation is passed or not. H eckm an (1978) develops a m odel w ith endogenous dum m y variables in a sim ultaneous equation system, w hich addresses this problem. T he dum m y variable indicates the existence o f legislation, and the endogeneity arises from the fact that the dum m y variable is generated by a laten t variable that crosses a certain threshold. The latent variable represents the sen tim en t tow ard sm oking, w hich w hen is strong enough leads to regulation in public places to be enacted. In C hapter IV, I ex ten d H eck m an s (1978) m odel to allow for m ultiple endogenous variables in a panel data set. I study the dem and fo r cigarettes for the fifty U.S. states from 1975 to I attem pt to construct a sentim ent variable in order to estim ate the attitu d e to w ard sm oking o f states population. I study w hether the change in the d em an d equation is due to regulation o f sm oking in public places, w hich is passed in a state o r to a strong anti-sm oking sentim ent. T he results show that regulations o f sm oking in public places have no effect on the dem and for cigarettes. Instead, the anti-sm oking sentim ent is a significant factor that causes the cig arette consum ption to decrease. T h e sentim ent tow ard sm oking is changing and g ro w in g in tim e because o f continuous inform ation that the public receives reg ard in g the d an g er o f cigarette sm oking. T herefore, sm oking is reduced because o f the attitu d e tow ard sm oking that is changing, w hich m akes people to sm oke less. T his is an im p o rtan t result and constitutes a contribution to the literature because it p ro v id es g reater insight into the exten t to w hich the no-sm oking regulations in public places are effective. T he conclusion is that these regulations are.7

22 effective in the protection o f non-sm okers, but they do not have a significant im pact on sm okers. The 1988 S urgeon G en eral s report provided evidence o f the strong correlation betw een the use o f cigarettes and use o f other licit and illicit drugs. Based on the data from the 1985 N ational H ousehold Survey on D rug A buse, the report show ed that a higher p ercentage o f current cigarette users w ere also using alcohol com pared with nonsm okers in all age groups.8 This evidence raises questions about the econom ic relationship betw een alcohol and cigarettes. T he literature has exam ined the cross-price effect in the dem and for alcohol and the dem and for cigarettes. H ow ever, the results are mixed and the nature o f the relationship rem ains unclear. In this part o f m y research I focus on estim ating the im pact that the cigarette price and the no-sm oking regulation have on the dem and for alcohol. If the nosm oking regulation is o m itte d, th e effect o f cig arette p ric e is over-estim ated. N o sm oking restrictions rep resen t an additional cost for non-sm okers w ho change their sm oking habits to com ply w ith legislation. Therefore, by co nsidering the no-sm oking regulation into the d em and for alcohol equation I estim ate m ore accurately the crossprice effect. M oreover, I am able to study w hether im posing restrictions on sm oking in public places has any effect on the alcohol consum ption. I use the m odel developed in C hapter IV, based on H eckm an (1978), and I investigate the im pact th a t no-sm oking regulations have on the dem and for alcohol. A sentim ent variable and probabilities that states regulate sm oking in public places are first estim ated, and then introduced in the dem and for alcohol. S im ilar to the analysis 8

23 conducted in C hapter IV, I investigate w hether the no-sm oking legislation affects in any w ay the alcohol consum ption, o r the public s sentim ent against sm oking is a m ore general sentim ent and expresses an attitude against any drug use, affecting the dem and for alcohol as w ell. Based on the cross-price effect cigarettes and alcohol are substitutes in consum ption. The resuits also reveal that the no-sm oking regulation in other public places have a negative and significant effect on the dem and for alcohol. R estricting or banning sm oking in other public places, such as bars, leads to a decline in the alcohol consum ption. I find that the anti-sm oking sentim ent has no im pact on the dem and for alcohol. T he estim ated sentim ent is characteristic to cigarette sm oking, and does not capture a more general attitude against drug use. C hapter VI concludes m y dissertation and sum m arizes the findings o f this research. 9

24 C H A PT E R II H ISTO R Y O F S M O K IN G R E G U L A T IO N A N D ST A T E R E G U L A T IO N O F S M O K IN G IN PU BLIC PLA C ES The E volution o f Sm oking R egulation Legislation regulating sm oking has had at least three functions o v er the years: to inform consum ers about the risk o f sm oking; to protect n o nsm okers from the adverse health effects o f environm ental tobacco sm oke (ETS); an d to prevent young people from sm oking. T he First Surgeon G eneral s report linking sm oking to various diseases w as published in 1964, and stated that cigarette sm o k in g w as a health hazard o f sufficient im portance in the U nited States to w arrant appropriate remedial action. 9 In the sam e year, the A m erican M edical A ssociation (A M A ) officially declared sm oking a serious health hazard. Federal, state, and local governm ents, as w ell as p riv ate organizations, concentrated their efforts and initiated a cam paign against sm oking. In the early years, the federal governm ent played the m ost active role, and the im m ediate m easures were im posed on tobacco m anufacturers.10 T he Federal C igarette Labeling and A dvertising A ct o f 1965 required package-w arning labels, say in g that cigarette sm oking m ay be hazardous to health. T he Public H ealth C ig arette S m oking A ct o f 1969 restricted the w arning labels to only inform ation that the S urgeon General 10

25 had determ ined that cigarette sm oking is dangerous to health, and to the ban of cigarette advertising on television and radio. T he C om prehensive Sm oking Education A ct o f 1984 institutes four rotating health-w arning labels, all listed as Surgeon G en eral s W arnings, on cigarette packages and ad v ertisem ents.11 By contrast, state governm ents had lim ited th eir actions to taxing cigarette purch ases.12 T he Federal governm ent charted a new course in its efforts to address the health concerns associated with cigarette sm oking w hen it unveiled its anti-sm oking cam paign in the Surgeon G eneral s 1972 R eport on Sm oking and H ealth.1'' Among other things, this w as the first report to address the potential negative-health effects of cigarette sm oking on nonsm okers. As a result o f intense research in this area, the Surgeon G en eral s report in 1986 expressed concern that involuntary sm oking is a cause o f disease, including lung cancer, in healthy nonsm okers. 14 T he 1986 report presented a detailed description of the health consequences o f exposure to environm ental tobacco sm oke (ETS) and proposed restricting sm oking in public places. T he grow ing aw areness o f the d an g er o f secondhand sm oking, together with the change in the p u b lic s attitude tow ard sm oking after 1964, allegedly created pressure for the governm ent to restrict sm oking in public places. Public health advocates like the A m erican L ung A ssociation have urged adoption o f laws and regulations m aking public places, w orkplaces, and schools sm oke free. R egulations that restricted sm oking in public places had been passed by state governm ents before the 1960s.1:5 T hese regulations w ere, how ever, largely aim ed at preventing fire and preventing the contam ination o f food being prepared or 11

26 packaged for public consum ption, and the second-hand sm oke health hazard was not a m ajor concern.16 This changed in the 1970s when policy m akers declared that the prim ary intent of these law s w as to insure the safety and com fort o f nonsm okers, because potential health hazards associated w ith second-hand sm oke w ere better understood.17 In 1986, the federal governm ent began to restrict sm oking in public places. These regulations covered transportation facilities (see the ban on sm oking on com m ercial airline flights in 1988) and governm ent w o rk sites.18 N onetheless, as the 1989 Surgeon G e n e ra l s report em phasized, restrictions on sm oking in public places at the state level w ere expected to be the norm by the end o f the century. 19 In 1973 A rizona becam e the first state to restrict sm oking in a num ber of public places and the first to do so explicitly because environm ental tobacco (secondhand) sm oke exposure w as co n sid ered a public hazard. In 1974 M innesota enacted the first com prehensive clean in d o o r air act, which restricted sm oking in m ost buildings open to the public. B etw een 1975 and 1984, tw elve states p assed nosm oking regulations in som e public places, w ith different degrees o f enforcem ent. Surgeon G en eral s reports focused m ore w ith every year on sm oking health effects and in 1984 it announced the goal o f a sm oke free society by the year State R egulations R eg ard in g Sm oking in Public Places T his study is generated by the w ide variation o f the legislation regarding sm oking in public places across states. T h e detailed inform ation on state regulations 12

27 is provided by the State T obacco A ctivities T rackin g and Evaluation (STA T E ) S ystem, w hich w as developed by the C en ter for D isease C ontrol and Prevention (C D C ) in the O ffice on Sm oking and H ealth, N ational C enter for C hronic D isease Prevention and H ealth Prom otion. The ST A T E S ystem sum m arizes the legislation in all 50 states and W ashington, D C, and id entifies four prim ary aspects o f tobacco control laws in each state: (i) sm oke-free in d o o r air, (ii) youth access to tobacco products, (iii) advertising o f tobacco products, and (iv) excise taxes on tobacco products. F o r the purpose o f this paper, the discussion will regard only sm oke-free indoor air laws, w hich for the ease o f exposition w ill be referred to as no-sm oking regulations. T he report includes all state no-sm oldng regulations effective as o f June 30, States define public places differently and im pose different restrictions on sm oking in these locations. Because the com p ariso n across laws based on public places, broadly defined, is difficult, locations w ere grouped in six categories: (1) governm ent sites; (2) private-sector w ork sites; (3) restaurants; (4) com m ercial child day care; (5) hom e-based child day care, and (6) other places (including bars, shopping m alls, grocery stores, enclosed arenas, public transportation, hospitals, prisons, and hotels and m otels). Seventeen states have laws that preem pt, in som e situations, m ore stringent policies at the local level.20 As o f June 30, 1995, forty-six states required sm oke-free indoor a ir to som e degree o r in som e public places (see Tables 1, 2A and 2B ). 21 T here are five states 13

28 Table 1 State C lean Indoor A ir Laws - D ates W hen the Law s W ere Enacted, as o f June 30, 1995 State G o v ern m ent W ork Sites Private W ork Sites R estaurant L aw C o m m ercial C hild D ay C are H om e- Based C hild D ay Care A labam a O th er Places A la sk a A rizona A rkansas C a lifo rn ia C olorado C onnecticut D elaw are D.C F lo rid a G eorgia H aw aii Id ah o Illin o is Indiana Io w a K an sas K entucky L o u isia n a M a in e M aryland

29 Table 1 - C ontinued State G o v ern m ent W ork Sites Private W ork Sites R estaurant Law C o m m ercial C hild D ay Care Home- Based Child D ay Care O th er Places M assachusetts L M ichigan M innesota M ississippi M issouri M ontana N ebraska N evada New H am pshire N ew Jersey New M exico New Y ork North Carolina N orth D ak o ta Ohio O klahom a O regon P en n sy lv an ia R hode Island South Carolina South D ak o ta

30 Table 1 - Continued State G overnm ent W ork Sites Private W ork Sites Restaurant Law C om m ercial C hild D ay C are H om e- B ased C hild D ay C are T ennessee O ther Places T exas U tah V erm ont V irg in ia W ashington W est Virginia W isco n sin W yom ing Source: The STA TE System. (A labam a, K entucky, M ississippi, N orth Carolina, and T ennessee) w ith no legislation o r legislation that preem pts localities from enacting any law to restrict sm oking in public places. State Regulations R egarding S m oking in Public Places - C ategories G overnm ent W ork Sites Forty-one states have law s restricting sm oking in state governm ent sites (Table 2A). T hirty-tw o states lim it sm oking to designated areas, w hile two states require either no sm oking o r designated sm oking areas with separate ventilation, and 16

31 State Table 2A State Regulations o f Sm oking in P ublic Places, as of June, R estrictiveness G overnm ent W ork Sites P riv ate W o rk S ites R estaurant Law C om - M ercial C hild D ay C are H om e-based C hild D ay C are A labam a A lask a A rizona A rkansas C alifornia C olorado C onnecticut D elaw are D.C Florida G eorgia H aw aii Idaho Illin o is Indiana Iow a K ansas K entucky L ouisiana M aine M aryland M assachusetts

32 Table 2A - Continued State G overnm ent W ork Sites P rivate W ork Sites R estaurant Law Com - M ercial Child D ay Care H om e-based C h ild D ay C are M ichigan M innesota M ississippi M issouri M ontana N ebraska N evada N ew H am pshire N ew Jersey N ew M exico N ew Y ork N orth C arolina N orth D akota O hio O klahom a O regon P ennsylvania R hode Island South 2 4 C arolina South D akota

33 Table 2A - Continued State G overnm ent W ork Sites Private W ork Sites R estaurant Law Com- M ercial Child Day H om e-based C hild D ay C are Care T ennessee Texas Utah V erm ont V irginia W ashington W est V irginia W isconsin W yom ing Note: 2=designated sm oking areas required o r allow ed; 3=no sm oking allow ed or designated sm oking areas allow ed if separately ventilated; 4=no sm oking allow ed (100% sm oke free). Source: STA TE System. seven states prohibit sm oking com pletely. T here are different specifications regarding the m inim um n um ber o f em ployees for restriction to be im plem ented, regarding the penalties, and enforcem ent authority. In K entucky and North C arolina, state governm ent w ork sites are perm itted, but not required, to develop policies on sm oking. Private W ork Sites In m ost o f the states, w orkers in private w orksites are unprotected against ETS. O nly tw enty-one state laws restrict sm oking in private work sites ( see T able 19

34 Table 2B State Regulations of Smoking in Public Places, Category-Other Places, as of June, Restrictiveness State Bars Shopping malls Grocery stores Enclosed arenas Public transportation Hospitals Prisons Hotels and motels Alabama Alaska Arizona Arkansas California Colorado Connecticut Delaware D.C Florida Georgia Hawaii Idaho Illinois Indiana Iowa Kansas Kentucky Louisiana Maine M aryland M assachusetts

35 Table 2B - Continued State Bars Shopping malls Grocery stores Enclosed arenas Public transportation Hospitals Prisons Hotels and motels Michigan ? M innesota Mississippi Missouri Montana Nebraska Nevada New Hampshire New Jersey New Mexico New York North Carolina North Dakota Ohio Oklahoma Oregon Pennsylvania Rhode Island South Carolina South Dakota Tennessee

36 to to Table 2B - Continued State Bars Shopping malls Grocery stores Enclosed arenas Public transportation Hospitals Prisons Hotels and Texas Utah Vermont Virginia Washington West Virginia Wisconsin Wyoming motels Note: 2=designated smoking areas or allowed; 3=no smoking allowed or designated smoking areas allowed if separately ventilated; 4=no smoking allowed (100% smoke free). Source: STA TE System.

37 3.A). Seven states m andate designated sm oking areas only in w ork sites that have a m inim um num ber o f em ployees, but none o f them provides a sm oke-free environm ent. O n ly one state requires a designated sm oking area if separately ventilated, and the rest o f them are less restrictive and do not im pose ventilation requirem ents. R estaurants T hirty-one states have laws that regulate sm oking in restaurants (Table 2A). O nly U tah s law com pletely prohibits sm oking in restaurant, and only C alifornia s law requires eith er no sm oking o r separate ventilation for sm o k in g areas. M any state laws exem pt sm all restaurants, generally those w ith a seating capacity o f few er that 50 persons, from sm oking regulation. C om m ercial C hild D ay Care T w enty-six states regulate sm oking in com m ercial child day care centers, and tw enty o f them are sm oke-free (Table 2A ). Six states allow o n ly designated sm oking areas. H om e-b ased C hild D ay C are C hildren are even less protected by law in hom e-base ch ild day care centers (Table 2A). Ten states regulate sm oking in hom e-based child d ay care centers. Six of these states prohibit sm oking, and four states allow designated sm oking areas. 23

38 E nforcem ent authority and penalties vary across states. O ther Sites Som e states have laws that regulate sm oking in o th er locations (Table 2B). Forty-tw o states restrict sm oking in hospitals, 42 on selected form s o f public transportation, 30 in grocery stores, and 23 in enclosed arenas. Few states have laws that restrict sm oking in bars, shopping m alls, prisons, and hotels and m otels. Facts on C igarette S m oking in the U nited States A dult S m oking Prevalence - U nited S tates Population In this section, I present som e historical trends and facts related to cigarette Figure P ercent o f C urrent Sm okers in the U.S. - Total Population Source: N ational Health Interview Surveys,

39 sm oking in the U.S. T he data for this analysis is taken from the N ational H ealth Interview Surveys, as presented in various C D C R eports.22 Figure 1 show s the tim e series trend in cigarette sm oking in total U.S. population, w hich includes persons o f ages 18 and older. T here has been a precipitous decline in the percentage o f current sm okers from 42.4% in 1965 to 24.7% in Since 1990 how ever the trend has leveled o ff at about 22.9% in T he dow nw ard historical trend in sm oking prevalence has changed in the 1990s, and the percentage o f current sm okers m aintains at alm ost the sam e level until T he only exception is the spike on the graph for 1992, when there has been an L egend. F igure 2. = m en, = w om en. Percent o f C urrent Sm okers by G ender Source: N ational H ealth Interview S urveys, increase in sm oking prevalence to 26.5%, up 7.2% com pared with the 1990 level. 25

40 M ore recent data continue to reveal a sluggish decline in sm oking prevalence in U.S. p o pulation, to 23.2% in 1997 and 22.9% in j Figure 2 com pares the percentage o f m en and w om en w ho are current sm okers, and tw o trends appear in these data. T he trend o f sm oking prevalence is declining for both genders, although the gap is closing in the past years. T here has been a larger drop o f 47% in the percentage o f m en w ho are sm okers, from 51.9% in 1965 to 27.0% in The reduction in sm oking prevalence am ong w om en has been sm aller, only 33.3%, from 33.9% in 1965 to 22.6% in In 1995, alm ost the sam e proportions o f m en and wom en are current sm okers. The narrow ing gap is explained by the m uch sm aller decrease in w o m en s sm oking. T hese data are consistent w ith the L egend. F igure 3. = black, = white. P ercent o f C urrent Sm okers by R ace Source: N ational H ealth Interview Surveys,

41 findings in the literature, that wom en are less responsive to changes in policies aim ed at disco u rag in g sm oking.24 C om parison o f sm oking prevalence by race is show n in Figure 3. Slightly m ore black people sm oke cigarettes com pared w ith w hite people, but the gap alm ost closed in the 1990s. The proportion o f w hite persons w ho are sm okers has gone dow n by 41 %, from 42.1 in 1965 to 24.8 in T he decline in the proportion o f sm okers am ong black persons has been larger, 43.8%, from 45.8% in 1965 to 25.7% in T he decline stopped in the 1990s for both w hites an d blacks. Figure 4 show s the sm oking prevalence for d ifferent age groups and it L egend = age 18 to 2 4, = age 25 to 4 4, = age 45 to 64, = above age o f 65. F igure 4. Percent o f C urrent Sm okers by A ge Source: N ational H ealth Interview Surveys,

42 illustrates a general d o w n w ard trend. T he highest proportion o f sm okers is adults betw een the ages 25 and 4 4 th ro u g h o u t the whole sam ple period; 51.2% in 1965 and 28.6% in Alm ost the sam e p ro p o rtio n s o f youths of ages betw een 18 and 24 and adults o f ages between 45 and 65 sm oke cigarettes. For the age group 65 and o v er there has been the sm allest proportion o f sm okers. The biggest declin e in sm oking prevalence occurred am ong young adults o f ages 18 to 24, from 45.5% in 1965 to 24.8 in 1995, and adults o f ages 25 to 4 4, from 51.2% in 1965 to 28.6% in It is interesting to observe that sm oking p rev alen ce has increased in the 1990s am ong the group age 18 to 24. The low est level o f sm oking am ong young adults has been registered in 1991, 22.9%, and after that is has increased w ith a peek o f 27.5 in T his new trend in youth sm oking, w hich is tru e fo r even younger age groups, represents the concern o f p o licy makers today and it the topic for the econom ic literature in this area. A lthough anti-sm oking policies have been im plem ented at federal and state level to discourage and prevent youth sm oking, th ere is yet m uch to be done. From Figure 5 w e can infer that education is inversely correlated with cigarette sm oking. People w ith 12 years o r less of education are m ore likely to sm oke than people w ith higher ed u cation. U ntil late 1970s there w as no difference in sm oking trends am ong peo p le w ith up to 15 years o f schooling. A fter 1979, the sm oking prevalence for p eo p le w ith 13 to 15 years o f schooling was m ore abrupt, w hile the proportions o f sm o k ers am ong people w ith high school o r less education continue to have sim ilar, less abrupt declining trends. F ew er people am ong those with 28

43 16 years or m ore o f education are cigarette sm okers over the entire sam ple period. T his group also registered the largest decline in sm oking prevalence o f 60%, from Legend. = less than 12 years o f educatio n, = 12 years o f education, = betw een 13 and 15 years o f ed u catio n, = 16 or m ore years o f education. Figure 5. P ercent o f C u rrent Sm okers by Education Source: N ational H ealth Interview Surveys, % in 1966 to 14.0% in T he sm allest decrease in sm oking prevalence is am ong less educated people, w ith less than 12 years of education. T he decline for this group has been 27%, from 41.7% in 1966 to 30.4% in A m ong the sm okers, it is interesting to determ ine the level o f cigarette consum ption and w hether it varies through time. Figure 6 show s that there is a decline in the average n u m b er o f cigarettes sm oked daily. T h ere w as no real decline until 1988, w hen the daily n um ber o f cigarettes sm oked fell from 20.2 cigarettes per 29

44 Figure Average N um ber o f C igarettes Sm oked by the C urren t A dult Sm okers Source: National H ealth Interview Surveys, Legend = m e n, = w om en. F igure 7. Percent o f C urrent S m okers W ho Sm oke 25 o r M ore C igarettes Per D ay - By G ender Source: National H ealth Interview Surveys,

45 Legend = w h ite, = black. Figure 8. Percent o f C urrent Sm okers W ho Sm oke 25 or M ore C igarettes per D ay - By R ace Source: N ational H ealth Interview Surveys, day to 16.7 cigarettes per day in But data for 1995 show s a 11.3% increase to 18.6 cigarettes per day in T he health consequences are m ore serious for those w ho sm oke high quantities o f cigarettes. T herefore Figures 7 to 10 illustrate the proportion o f current sm okers w ho consum e 25 o r m ore cigarettes per day am ong different dem ographic groups. M en are m ore likely to consum e large quantities o f cigarettes than w om en, and there is a larger decline in cigarette consum ption am ong w om en in recent years. M ore w hite sm okers consum e 25 o r m ore cigarettes per day than black sm okers. People o f ages betw een 25 and 64 are m ore likely to sm oke large quantities 31

46 o f cigarettes, and o n ly a small proportion o f young sm okers 18 to 25 years old use 25 o r m ore cigarettes p er day. O nly am ong sm okers 25 to 44 years old the proportion o f those sm oking 25 o r m ore cigarettes p er day is declining. T here is no clear distinction am ong the proportion o f heavy sm okers by education in the first part of the sam ple. It is only after 1990 that few er people with higher education tend to sm oke 25 o r m ore cigarettes per day. For sm okers with less than a high school education the proportion o f heavy sm okers increased L egend. = age 18 to 2 4, = age 25 to 4 4, = age 45 to 64, = above age o f 65. Figure 9. P ercent o f C urrent Sm okers W ho Sm oke 25 or M ore C igarettes per D ay B y A ge Source: N ational H ealth Interview S urveys, Figure 11 sum m arizes in a w ay the inform ation contained in F igures 6 to 10, 32