During the 1990s the prevalence of smoking and the prevalence of obesity in creased among U. S. adolescents. In the Monitoring the Future Study, the percent of twelfth-graders who report having smoked any cigarettes in the last thirty days rose from 28.3 percent in 1991 to 35.1 percent in 1998; similar rises in 30-day prevalence occurred among eighth-graders and tenth-graders [Johnston, 2002]. Data from the National Health and Nutrition Examination Surveys indicates that the prevalence of overweight among 12-19 year olds rose from 10.5 percent to 15.5 percent during the 1990s; a similar rise in prevalence occurred among children aged 6-11. Both of these trends are troubling, because smoking and obesity are among the top causes of preventable death in the United States.
Body weight and smoking are interrelated. Adult smokers weigh less than non smokers and smoking cessation by adults results in an average weight gain of 2-3 kilograms. In contrast, young smokers may be heavier than, or weigh roughly the same as, young non-smokers. This discrepancy may exist because any anorectic effects of smoking are slow to accumulate or because adolescents are more likely than adults to use smoking as a method of weight control.
The vast majority of people who will ever smoke begin smoking during adolescence[USDHHS,1994]. This may also be the stage of life when people, especially girls, are most sensitive to their body weight. The perception that smoking reduces levels or changes in weight is common among teenage girls.
This paper estimates models that measure the role of body weight in the decisions of adolescents to initiate smoking, controlling for cigarette prices and state tobacco control policies. Understanding the relative effects of weight and price may help to identify adolescents most likely to initiate smoking, and allow anti-smoking efforts to better target at-risk teens. We estimate our models using nationwide panel data of adolescents.
This paper relates to two previous literatures. The first is a medical and sociological literature on the relationship between girls’ body image and their probability of smoking initiation. The second is an economics literature that focuses on the relationship between cigarette prices, tobacco control policies, and smoking initiation. A few previous studies have examined the effects of concerns about weight on youth smoking. Tomeo et al. [1999], Wiseman et al. [1998], and French et al. [1994] find that concern about weight is correlated with current smoking or smoking initiation for female adolescents. Voorhees et al. [2002] show that teenage girls who are currently trying to lose weight, or who tried to lose weight in the past, are more likely to be daily smokers. Tucker [1983] finds that obese boys have stronger intentions of smoking than lighter boys. Cawley, Markowitz, and Tauras [2004] studied data from the National Longitudinal Survey of Youth 1997 Cohort and found that weight influenced the smoking initiation decisions of girls but not boys.
Most of the existing studies on smoking and weight suffer from the fact that their samples are a single cross-section. French et al. [1994] and Cawley et al. [2004] are the only studies to examine the probability of transition from non-smoker to smoker using a panel of respondents. Our study contributes to this literature by examining the transition from non-smoker to smoker using a large national panel data set of adolescents that covers a longer period than the data used in Cawley et al. [2004]. This paper also considers certain measures of body weight not considered by Cawley et al. [2004].
Another limitation is that all of these studies except Cawley et al. [2004] fail to control for cigarette prices and tobacco control policies such as youth access laws, which are potentially important predictors of adolescent smoking initiation. The omission of these economic variables will not bias the coefficient on weight unless they are correlated with weight; however, we believe that controlling for both is important because it yields information on the relative importance of the two factors in the smoking initiation decision.
In contrast, economic studies of adolescent smoking initiation have focused on the impact of price and tobacco control policies while ignoring the role of body weight. To date, six econometric studies have examined the impact of cigarette prices or taxes on smoking initiation, the results of which vary considerably. The earliest studies, Douglas and Hariharan [1994] and Douglas [1998], find that current cigarette prices are uncorrelated with smoking initiation. However, these results should be interpreted with caution because the smoking variables were created using retrospective data; incorrect recall, along with errors in matching historical price to past residence may bias the results. In another study using retrospective data, Forster and Jones [1999] find that higher taxes are associated with later initiation, although the magnitude of the impact is small.
The results from initiation studies that use longitudinal data are mixed. Tauras, Johnston, and O’Malley [2001] conclude that cigarette prices are strongly negatively correlated with the probability of transition to daily smoking. However, the effect of price on the probability of transition to smoking any quantity of cigarettes is not statistically significant. DeCicca, Kenkel, and Mathios [2002] control for state fixed effects and find that cigarette excise taxes are insignificant determinants of smoking onset. Neither of the aforementioned studies analyzes the smoking decision separately by gender. Cawley, Markowitz, and Tauras [2004] find significant gender differences; specifically, that price is negatively correlated with smoking initiation by boys but uncorrelated with smoking initiation by girls.
This paper is distinct from Cawley, Markowitz, and Tauras [2004] in that it uses a different dataset that spans a larger time period: the Children of the National Longitudinal Survey of Youth, 1979 Cohort.7 In addition, this paper is more concerned with objective measures of weight such as BMI and clinical weight classification (both levels and changes) while Cawley et al. [2004] focuses more on self reported body image and dieting.
Download
PDF Ebook Obesity, Cigarette Prices, Youth Access Laws And Adolescent Smoking Initiation
