Lighting up and slimming down: the effects of body weight and cigarette prices on adolescent smoking initiation

https://doi.org/10.1016/j.jhealeco.2003.12.003Get rights and content

Abstract

This paper examines the influence of body weight, body image, and cigarette prices in determining adolescent smoking initiation. Adolescents who desire to lose weight may initiate smoking as a method of appetite control. Such behavior may undermine the goals of tobacco control policies that seek to prevent smoking initiation. Using a nationally representative panel of adolescents, we show that smoking initiation is more likely among females who are overweight, who report trying to lose weight, or who describe themselves as overweight. In contrast, neither objective nor subjective measures of weight predict smoking initiation by males. Higher cigarette prices decrease the probability of smoking initiation among males but have no impact on female smoking initiation. These gender-specific differences may help explain the mixed and inconclusive evidence of the impact of price on smoking initiation found in previous literature.

Introduction

Smoking and obesity are the two leading causes of preventable deaths in the United States. It is estimated that over 400,000 deaths per year are attributable to smoking (CDC, 1993), and 280,000–350,000 deaths per year are attributable to obesity (Allison et al., 1999). In December 2001, Surgeon General Dr. David Satcher warned that obesity may soon overtake tobacco as the number one cause of preventable death in the US (Associated Press, 2001).

Body weight and smoking status are interrelated. Adult smokers tend to weigh less than non-smokers (Nemery et al., 1983, Fehily et al., 1984, USDHHS, 1988, Klesges et al., 1989), and smoking cessation by adults results in an average gain of 2–3 kg in weight (Gordon et al., 1975, USDHHS, 1990, Klesges et al., 1989, Pinkowish, 1999).

The vast majority of smokers begin smoking during adolescence (USDHHS, 1994). This may also be the stage of life when people, especially females, are most sensitive to their body weight. The perception that smoking controls weight is common among youths, especially among teenage girls (Camp et al., 1993, Brooks, 1998). As a result, teenagers who want to lose weight may begin to smoke based on the belief that smoking will facilitate weight loss.

From a policy perspective, it is important to know the roles of body weight and body image in determining smoking initiation. Previous research has shown that tobacco control policies, such as higher taxes on cigarettes, can deter smoking. However, as the prevalence of obesity increases among children and the dangers associated with being overweight become well-known, smoking rates may rise if personal goals of weight loss or weight maintenance overshadow the deterrent effects of anti-smoking public policies.

This paper examines the roles of body weight and tobacco control policies in smoking initiation by adolescents. In particular, the impact of perceived body weight, actual body weight, cigarette prices, and socioeconomic factors on smoking initiation are examined for males and females. The results indicate that females who report trying to lose weight and those who describe themselves as overweight are more likely to initiate smoking. In contrast, neither objective nor subjective measures of weight predict smoking initiation for males. Higher cigarette prices deter smoking initiation by males, but have no impact on smoking by females.

Section snippets

Relevant literature

A few previous studies have examined the correlation between concern about weight and smoking status in youths. 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 adolescent girls are more likely to be daily smokers if they are currently trying, or previously tried, to lose weight. Tucker (1983) is the only study to look at the

Methods

In every time period, a teenager faces the decision whether or not to initiate smoking. This decision is likely to be affected by a number of factors, including the full price of cigarettes, the anticipated weight loss effects of smoking, knowledge about and concern for the detrimental health impact of smoking, along with individual taste factors. Based on the previous literature discussed above, we expect that price will to some extent deter smoking initiation, while weight concerns or the

Data

This paper uses data from the National Longitudinal Survey of Youth 1997 Cohort (NLSY97). The NLSY97 is a nationally representative sample of 9,022 youths aged 12–16 as of 31 December 1996. The first wave was conducted in 1997, with follow-ups in 1998, 1999 and 2000. The age range of this panel (12–21) covers the ages at which the vast majority of eventual smokers initiate.

Two measures of smoking initiation are constructed. The first indicates a transition from non-smoker to smoking any

Estimation and results

Estimates from the duration analysis are presented in Table 2, Table 3, Table 4. Table 2 contains results for models that include an indicator variable for BMI ≥85th percentile. Table 3 contains results for models that include an indicator variable for attempting to lose weight. Table 4 contains results for models that include indicator variables for self-perceived overweight and underweight.

The first three columns of each of Table 2, Table 3, Table 4 employ the less stringent measure of

Endogeneity

As discussed earlier, a correlation between the weight measure and the error term in the smoking initiation regression will bias our results. Two methods are used to test and potentially correct for the possible endogeneity of body weight and weight due to unobserved third factors. The first is the method of instrumental variables (IV), which identifies the causal effect using variables that are strongly correlated with objective weight and uncorrelated with the error term in the smoking

Summary

This paper examines the influence of body weight, body image, and cigarette prices on the smoking initiation decision of adolescents. Discrete time duration models are estimated using data from the National Longitudinal Survey of Youth, 1997 Cohort, for two definitions of smoking initiation: the first representing the transition from non-smoker to smoking any quantity, and the second representing the transition from non-smoker to frequent smoker.

We find that females who have a high body mass

Acknowledgements

The authors thank Jeff DeSimone, Michael Grossman, Judy Shinogle, and conference participants for helpful comments and suggestions. The authors gratefully acknowledge funding from the Robert Wood Johnson Foundation. All views expressed are those of the authors.

References (35)

  • S.M. Douglas et al.

    The hazard of starting smoking: estimates from a split sample duration model

    Journal of Health Economics

    (1994)
  • A.M. Fehily et al.

    Diet, smoking, social class and body mass index in the Caerphilly Heart Disease Study

    American Journal of Clinical Nutrition

    (1984)
  • T. Gordon et al.

    Changes associated with quitting cigarette smoking: the Framingham study

    American Heart Journal

    (1975)
  • M.H. Alciati et al.

    State laws on youth access to tobacco in the United States: measuring their extensiveness with a new rating system

    Tobacco Control

    (1998)
  • D.B. Allison et al.

    The heritability of body mass index among an international sample of monozygotic twins reared apart

    International Journal of Obesity

    (1996)
  • D.B. Allison et al.

    Annual deaths attributable to obesity in the United States

    Journal of the American Medical Association

    (1999)
  • Allison, P.D. Event History Analysis: Regression for Longitudinal Event Data. Sage Publications, Beverly Hills, CA,...
  • J.D. Angrist

    Estimation of limited-dependent variable models with dummy endogenous regressors: simple strategies for empirical practice

    Journal of Business and Economic Statistics

    (2001)
  • Associated Press. US Warning of Death Toll from Obesity. New York Times, 14 December...
  • Brooks, A., 1998. Teenage girls start smoking to lose weight. British Medical Journal (International) 317,...
  • J. Bound et al.

    Problems with instrumental variables estimation when the correlation between the instruments and the endogenous explanatory variables is weak

    Journal of the American Statistical Association

    (1995)
  • D.E. Camp et al.

    The relationship between body weight concerns and adolescent smoking

    Health Psychology

    (1993)
  • Centers for Disease Control and Prevention, 1993. Cigarette Smoking-Attributable Mortality and Years of Potential Life...
  • Davidson, R., MacKinnon, J.G., 1993. Estimation and Inference in Econometrics. Oxford University Press, New York,...
  • P. DeCicca et al.

    Putting out the fires: will higher taxes reduce youth smoking?

    Journal of Political Economy

    (2002)
  • Douglas, S.M., 1998. The Duration of the Smoking Habit. Economic Inquiry XXXVI,...
  • J. Durbin

    Errors in variables

    Review of the International Statistical Institute

    (1954)
  • Cited by (177)

    • The effect of tobacco policies on youth physical activity

      2020, Economics and Human Biology
      Citation Excerpt :

      Key exceptions include Dragone et al. (2016), which extends their theoretical model of obesity and smoking to include exercise in showing that cigarette taxes could lower obesity, and Conway and Niles (2017) which estimates the spillover effects of cigarette costs on adult exercise behavior. Spillover effects in teens have been studied far less and are limited to the use of smoking for weight loss efforts (Cawley et al., 2004, 2016). These studies help shape our conceptual framework and empirical approach.

    View all citing articles on Scopus
    View full text