Compensating behaviors, regret, and heterogeneity in the dynamics of smoking behavior
Introduction
If smokers are aware of addiction and the possible illness associated with smoking, and are hence ‘rational addicts’ (Lahiri & Song, 1998; Stigler & Becker, 1977; Becker & Murphy (1994), Becker, Grossman, & Murphy (1988); Becker & Murphy, 1991; Chaloupka, 1991; Orphanides & Zervos, 1995; Suranovic, Goldfarb, & Leonard, 1999), evidence of rationality may also be manifested in their other health behaviors such as regular physical exercise. In essence, unhealthy and healthy behaviors are two sides of a coin, which are linked via an individual's concerns about his or her health. A rational smoker may choose to quit and suffer the withdrawal symptoms, or as a compromise, may choose alternative measures to counteract the detrimental health consequences of smoking without giving up the habit.
Addiction follows a dynamic path, which is created by an individual's smoking decisions made since initiation. The health consequences of smoking can alter an individual's smoking pattern, but these may be (partially) alleviated via counter-measures such as medical intervention. With the development of modern medicine and technology, smokers may think that they do not have to quit in order to improve or maintain their health when they suffer certain smoking-related diseases. Instead, they might simply utilize medical intervention to deal with the illness caused by smoking, or choose to exercise more regularly to offset the detrimental effects of smoking, without altering their smoking patterns. The compensating behavior results from worse health, or an increased awareness of the consequences of smoking. It was found that better knowledge of the health consequences of smoking is related to less smoking and more exercise (Kenkel (1990), Kenkel (1991)).
Little empirical support has been found for the theoretical substitution among cessation, exercise and medical care. Instead, literature from the fields of medicine and public health suggest that smokers are less likely to exercise and have unhealthy lifestyles. (Patterson, Haines, & Popkin, 1994; Coulson, Eiser, & Eiser, 1997; Steptoe & Wardle, 1997) Because these studies are associational rather than analytical, however, it is inappropriate to conclude that there is a causal relationship between smoking and an individual's level of exercise. Evidence, especially evidence of causality, is not provided in these studies. Specifically, it is still not clear whether the observed association is the result of the fact that smokers choose not to exercise because they are busier (time allocation problem), that the smoker samples used are healthier, or that smokers’ psychological attributes are different. Additional research is needed to address questions concerning the direct and indirect effects of smoking in exercise behavior.
Similarly, the relationship between smoking and medical care has not been fully addressed by previous research. In particular, studies have failed to distinguish medical care resulting from poor health due to smoking from medical care voluntarily sought by smokers, which may not be related to smokers’ current health. In a behavioral model often used by health service researchers, three groups of factors are identified to affect an individual's behavior to seek medical care: predisposing, enabling and need factors. (Andersen & Newman, 1973; Andersen & Aday, 1978; Aday & Andersen, 1984; Andersen, 1995) Poor health resulting from smoking is a need factor, which increases the probability that an individual will seek care. However, as more and more smokers are exposed to health education concerning the detrimental effects of smoking, their predisposing factors may change accordingly. That is, smokers may be more inclined to seek medical care than nonsmokers, not necessarily because of poor health, but because of their altered perception that they are more likely to get sick.
In previous empirical work, researchers identified several factors that contribute to smoking behavior: age, gender, race, education, psychiatric impairment, income, cigarette price, excise taxes, as well as state and federal policies and regulations on smoking. (Conway, Ward, Vickers Jr., & Rahe, 1981; Gunn, 1983; Linn, Sandifer, & Stein, 1985; Kabat & Wynder, 1987; Kenkel (1989a), Waldron & Lye (1989b); Kenkel, 1990; Viscusi, 1990; Kenkel, 1991; Viscusi, 1991; Blaylock & Blisard, 1992; Nides & David, 1995; Hsieh, Yen, Liu, & Lin, 1996; Kinnunen, Doherty, Militello, & Garvey, 1996) In many studies using cross-sectional data, smoking status was treated as a binary dependent variable, with individuals’ socioeconomic, demographic, and some physiological characteristics included as explanatory variables. One of the most debated issues was the function of health status in determining smoking behavior. One economic theory suggests that smokers may experience regret because of health concerns. (Orphanides & Zervos, 1995) If the theory holds, empirical studies would indicate a significant impact of health on smoking and health would be endogenous to an individual's decision about whether or not to continue smoking.
Three empirical economic studies demonstrated the effect of health on smoking. (Jones, 1994,1996; Shmueli, 1996) They also pointed out that in modeling smoking behavior, lack of control for the endogeneity of health biases the results. However, the findings on how health affects smoking were not unanimous. That is, both “curative” and “preventive” cessation were found using different data and model specifications. “Curative” cessation means that a smoker quits when his or her health deteriorates. “Preventive” cessation indicates that a healthy smoker quits to maintain good health status.
Lastly, most of the previous studies treated smokers as homogenous in their smoking decision-making. That is, heavy smokers were considered to behave the same way as light smokers. To illustrate a complicated addictive behavior such as smoking, a more detailed model incorporating the potential heterogeneity of smokers is necessary. (Kassel, Shiffman, Gnys, Jean, & Zettler-Segal, 1994).
A theoretical model of smoking behavior based upon the assumption of rationality includes the proposition that individuals know the health consequences of smoking and that they can choose curative and preventive interventions other than smoking cessation to improve or maintain their health. Hence smokers allocate their time and resources accordingly. The main objective of this paper is to study an individual's smoking behavior based on a microeconomic theory while taking into account his or her other health behaviors including seeking medical intervention when illness occurs and preventive measures such as exercise.
This paper aims to examine the following aspects of smoking behavior: compensation, regret, and heterogeneity.
Section snippets
Data
The data used in estimation were extracted from the RAND Health Insurance Experiment (HIE), 1974–1982. Because there was insufficient information on medical expenditure for HMO enrollees to enable the determination of the volume of services received in terms of dollars, only the individuals who were assigned HIE insurance and used the Fee-For-Service mode were included in the sample for the estimation. Individuals younger than 14 years old were excluded because the main interest of this study
Results
The results obtained suggest that individuals take into consideration their health when making smoking decisions. Smokers project their health concerns to their smoking patterns, time spent in exercise and the utilization of medical care. The smoking pattern of light smokers is significantly different from that of heavy smokers.
Selected estimates of the parameters from six equations pertaining to an individual's choices concerning smoking, medical expenditure and exercise, and health status are
Conclusion
Smoking is one of many types of interrelated health behaviors that combine to form an overall constellation that is an individual's health. After all, there are many ways to improve or diminish one's health. Incorporating multiple health behaviors in a study of smoking provides a more realistic picture of smokers’ general health behavior.
Previous epidemiological studies found that smokers are more passive in other health behaviors such as exercise. However, these studies failed to address
Acknowledgements
Thanks especially to John Akin for many insightful suggestions and advice. Thanks also to Donna Gilleskie, David Guilkey, Tina Shih and Frank Sloan.
References (43)
Smoking clinic failures and recent life stress
Addictive Behavior
(1983)- et al.
Smoking, health knowledge, and anti-smoking campaignsAn empirical study in Taiwan
Journal of Health Economics
(1996) Health, addiction, social interaction and the decision to quit smoking
Journal of Health Economics
(1994)Smoking cessation and healthA response
Journal of Health Economics
(1996)- et al.
Psychosocial and personality differences in chippers and regular smokers
Addictive Behavior
(1994) - et al.
The demand for prescription drugs as a function of cost sharing
Social Science and Medicine
(1985) - et al.
Health lifestyle patterns of U.S. adults
Preventive Medicine
(1994) Smoking cessation and healthA comment
Journal of Health Economics
(1996)- et al.
An economic theory of cigarette addiction
Journal of Health Economics
(1999) - et al.
Employment, unemployment, occupation, and smoking
American Journal of Preventive Medicine
(1989)
Family roles and smoking
American Journal of Preventive Medicine
Addiction and backslidingA theory of compulsive consumption
Journal of Economic Behavior and Organization
The national profile of access to medical careWhere do we stand?
American Journal of Public Health
Revisiting the behavioral model and access to medical careDoes it matter?
Journal of Health and Social Behavior
Access to medical care in the U.SRealized and potential
Medical Care
Societal and individual determinants of medical care utilization in the United States
Milbank Memorial Fund Quarterly
Procrastination and obedience
American Economic Review, Papers and Proceedings
A theory of rational addiction
Journal of Politcal Economy
Rational addiction and the effects of price on consumption
American Economic Review, Papers and Proceeding
An empirical analysis of cigarette addiction
The American Economic Review
Self-evaluated health status and smoking behaviour
Applied Economics
Cited by (10)
The effect of tobacco policies on youth physical activity
2020, Economics and Human BiologyCitation Excerpt :A tax-induced reduction in smoking, for example, may increase one's capacity for and enjoyment from exercise; any weight gain that results could provide an additional motive to increase exercise. Conversely, if smokers are using physical activity to compensate for the harms caused by smoking (Xu, 2002; Radtke et al., 2011), they may exercise less if cigarette taxes cause them to smoke less. Cigarette taxes could also have income effects on exercise, even if smoking behavior is not affected.
Regret minimization, willingness-to-accept-losses and framing
2006, Applied Mathematics and ComputationFuture directions of multiple behavior change research
2017, Journal of Behavioral MedicineGerman-language version of the compensatory health belief scale : Assessing its psychometric properties
2013, Zeitschrift fur GesundheitspsychologieSmoking-specific compensatory health beliefs and the readiness to stop smoking in adolescents
2011, British Journal of Health PsychologyFactors related to smoking in college women
2008, Journal of Community Health Nursing