Cigarette smoking and self-control

https://doi.org/10.1016/j.jhealeco.2006.07.002Get rights and content

Abstract

This paper empirically studies time inconsistent preferences in the context of cigarette smoking behavior. With hyperbolic discounting, an individual has time inconsistent preferences, which give rise to a lack of self-control, i.e., she may perpetually postpone the execution of a plan. This implies that a smoker who wants to quit has a demand for control devices, e.g., a smoking ban in public areas or a hike in cigarette excise taxes. This paper empirically tests this implication, using a sample that is based on survey data from Taiwan. The estimation results indicate that a smoker's intention to quit has a positive effect on the smoker's support for smoking bans and a cigarette excise tax increase. These results lend support to the validity of the time inconsistent preferences in the context of cigarette smoking behavior. This casts doubt on the validity of the assumption that individuals have time consistent preferences in Becker and Murphy's [Becker, G.S., Murphy, K.M., 1988. A theory of rational addiction. Journal of Political Economy 96 (4), 675–700] rational addiction model.

Introduction

For several decades, cigarette smoking behavior has been studied by social scientists from a variety of disciplines. There are two aspects of the behavior which arouse researchers’ interest. Firstly, cigarette smoking is harmful to one's health and society has to bear enormous costs arising from the productivity lost and from the medical expenses associated with smoking-induced diseases. An understanding of cigarette smoking behavior thus has important implications for intervention policies. Secondly, the behavior per se is interesting. It is puzzling why people develop and maintain a seemingly destructive habit even though most of them are aware of the harmful consequences and its addictive nature.

Due to their inability to explain smokers’ behavior by the neoclassical framework and by assuming rationality, early researchers largely regarded cigarette smoking as myopic (e.g., Houthakker and Taylor, 1966, Houthakker and Taylor, 1970). The assumption that smokers are myopic were maintained either implicitly or explicitly, e.g., Lewit and Coate (1982), Mullahy (1985), Baltagi and Levin (1986), and Jones, 1989, Jones, 1994. These studies treat cigarette addiction as habit forming and model the reinforcement effect by allowing lagged consumption to have an impact on current consumption.

Later on, by allowing agents to recognize the dependence of the current consumption level on past consumption, some researchers were able to reconcile cigarette smoking with rationality. A notable example is the rational addiction model of Becker and Murphy (1988), where people have time consistent preferences, and are forward-looking. That is, in Becker and Murphy's (1988) model people are aware of the addictive nature of cigarette smoking and they choose to smoke simply because the lifetime benefits are greater than the costs.

Several hypotheses are derived in Becker and Murphy (1988). The most important one is that current consumption positively depends on future consumption. This implies that current consumption is inversely related to all prices (past, current and future), or equivalently current consumption is positively related to past and future consumption. Becker and Murphy's (1988) rational addiction model has given rise to a series of studies that empirically test the rationality of cigarette smoking and other addictive substances. For example, Chaloupka, 1990, Chaloupka, 1991 uses micro data and Perkurinen (1991), Keeler et al. (1993), Becker et al. (1994) and Bardsley and Olekalns (1998) use time series data to study cigarette smoking behavior in the context of the Becker–Murphy (1988) model. Their empirical findings are consistent with Becker and Murphy's (1988) key prediction that current consumption depends on future consumption.

Recently, researchers have begun to question Becker and Murphy's rational addiction model on several grounds. First of all, the supporting evidence obtained by previous studies that test the implications of the Becker–Murphy model may not be conclusive in regard to the validity of all the assumptions underlying the rational addiction model. As pointed out by Gruber and Köszegi (2001), what is required to generate the positive relationship between current cigarette consumption and future consumption is the “forward looking” assumption only. That is, the time consistency assumption of the rational addiction model is not really tested. In modifying Becker and Murphy's (1988) rational addiction model by allowing an individual's preferences to be time inconsistent, Gruber and Köszegi (2001) are able to obtain the same relationship between current consumption and future consumption.1 This means that while we can conclude from previous empirical evidence that individuals are forward looking, this evidence does not allow us to infer whether individuals’ preferences are time consistent or not.2

Dissatisfaction with the ability of the rational addiction model of Becker and Murphy (1988) to explain individuals’ struggles with harmful addictions has led some researchers to consider alternative approaches to analyzing individuals’ decision-making processes. A plausible approach is to view individuals as having a lack of self-control, which arises from having time inconsistent preferences.

There is ample evidence suggesting that individuals have time inconsistent preferences and face self-control problems. Empirical and experimental findings in the economics and psychology literature (see, e.g., Thaler, 1981, Benzion et al., 1989, Angeletos et al., 2001, Laibson et al., 2004) suggest that individuals’ discount rates for a fixed time interval are not constant, but decline when the time interval is further away from the present.

Moreover, Gruber and Mullainathan (2005) provide the first piece of evidence indicating that time inconsistent models are preferable to the rational addiction model. Based on individual-level cross-sectional data from Canada and the U.S., Gruber and Mullainathan (2005) find that smokers are happier after an increase in cigarette taxes. If smokers’ happiness rises with cigarette taxes through a reduction in cigarette consumption, rational smokers can achieve the same increase in happiness by reducing cigarette consumption even in the absence of an increase in cigarette taxes. Thus, Gruber and Mullainathan's (2005) findings challenge the validity of the rational addiction model. By contrast, these findings suggest that smokers are better described as having time inconsistent preferences such that an increase in cigarette taxes will help solve their self-control problem.

This study empirically investigates the implications of time inconsistent preferences in the context of cigarette smoking. When smokers have time inconsistent preferences which give rise to self-control problems, they have a demand for self-control devices if they want to quit. Accordingly, smokers who have an intention to quit smoking will support for public policies, which impose costs on smoking. From this implication, we obtain a testable hypothesis that a smoker's intention to quit smoking has a positive effect on her support for anti-smoking public policies, e.g., an increase in cigarette excise taxes, a ban on smoking in all public areas, and a ban on workplace smoking. We test this hypothesis based on individual survey data from Taiwan. Our hypothesis testing relies on an empirical model, which accounts for the possible endogeneity of a smoker's intention to quit.

The remainder of the paper proceeds as follows. We briefly illustrate the time inconsistent preferences associated with hyperbolic discounting and their implications in Section 2. The data source and variables used in the empirical investigation are described in Section 3. Our empirical model is illustrated in Section 4. The estimation results are reported and discussed in Section 5. A conclusion is given in Section 6.

Section snippets

A synopsis

Time inconsistent preferences are proposed by Stroz (1956), and further developed by Phelps and Pollak (1968), Pollak (1968), Peleg and Yaari (1973), Yaari (1977), and Goldman, 1979, Goldman, 1980. Motivated by findings from laboratory and field studies indicating that individuals are time inconsistent (e.g., Prelec and Loewenstein, 1991, Ainslie, 1992), there has been a recent revival in interest in time inconsistent preferences in the economic literature.

Notable recent theoretical studies of

Data and variables

The empirical work of our study is based on survey data obtained from the Panel Study of Family Dynamics (PSFD). The aim of the PSFD is to understand the structure and evolution of the Chinese family in Taiwan, as well as the mode of interaction of its members.6

The PSFD comprises multi-year panel surveys starting from 1999. The survey adopts a three-stage random sampling procedure. In the

Empirical model

The objective of the empirical analysis is to examine the effect of a smoker's intention to quit on her demand for self-control devices, which is proxied by her attitude toward the three anti-smoking public policies (i.e., a hike in cigarette excise taxes, and public area and workplace smoking bans). In our empirical specification, the strength of individual i's support for policy k, denoted by Πki*, is a function of her socio-economic characteristics, denoted by Xi and her intention to quit,

Results

Our estimation results are reported in Table 4, Table 5, Table 6, Table 7, Table 8, Table 9. Before examining whether our conjecture concerning self-control is supported by the estimation results, we first examine the specification of our empirical model. The likelihood ratio test for the explanatory power of our instruments WControl and HeardRisk yields a χ2-statistic of 39.1985, which is asymptotically equivalent to an F-statistic of 19.5993. This F-statistic is larger than the Staiger-Stock

Conclusion

If individuals have time inconsistent preferences, as created by hyperbolic discounting, they have self-control problems. An implication of time inconsistent preferences in the context of cigarette smoking is that smokers who want to quit will have a demand for self-control devices in order to avoid perpetual procrastination.

This study tests a hypothesis derived from this implication. According to this hypothesis a smoker's intention to quit smoking has a positive effect on her support for

Acknowledgements

Part of this paper's research was carried out during my visit to the Department of Economics and Finance, City University of Hong Kong, May 3–June 30, 2005. Financial support provided by the City University of Hong Kong, and valuable comments and suggestions from seminar participants at the City University of Hong Kong are gratefully acknowledged. I would also like to thank Professor Anthony Culyer and the two anonymous referees for their constructive comments and suggestions, which have led to

References (40)

  • Bardsley, P., Olekalns, N., 1998. Cigarette and Tobacco Consumption: Have Anti-Smoking Policies Made a Difference,...
  • G.S. Becker et al.

    A theory of rational addiction

    Journal of Political Economy

    (1988)
  • G.S. Becker et al.

    An empirical analysis of cigarette addiction

    American Economic Review

    (1994)
  • U. Benzion et al.

    Discount rates inferred from decisions: an experimental study

    Management Science

    (1989)
  • Chaloupka, F.J., 1990. Men, Women, and Addiction: The Case of Cigarette Smoking, Working Paper Number 3267. National...
  • F.J. Chaloupka

    Rational addiction behavior and cigarette smoking

    Journal of Political Economy

    (1991)
  • S.M. Goldman

    Intertemporally inconsistent preferences and the rate of consumption

    Econometrica

    (1979)
  • S.M. Goldman

    Consistent plans

    Review of Economic Studies

    (1980)
  • J. Grabowski et al.

    Tobacco use, treatment strategies, and pharmacological adjuncts: an overview

  • J. Gruber et al.

    Is addiction ‘rational?’ Theory and evidence

    Quarterly Journal of Economics

    (2001)
  • Cited by (62)

    • Does time preference affect smoking behavior? A dynamic panel analysis

      2019, Journal of Behavioral and Experimental Economics
      Citation Excerpt :

      However, Grignon (2009) reports that 57% of current smokers fail to quit smoking. Kan (2007) also shows that current smokers who want to quit are more likely to support tobacco control policy. In addition, Athamneh et al. (2017) find a significant association between the intention to quit smoking and time preference, although their analysis is not separated by gender.

    • Hyperbolic discounting can be good for your health

      2018, Journal of Economic Psychology
    • The welfare implications of addictive substances: A longitudinal study of life satisfaction of drug users

      2018, Journal of Economic Behavior and Organization
      Citation Excerpt :

      These consistent findings across studies have led to the acceptance of the RA model as the standard framework for modelling addiction (Gruber and Köszegi, 2001). By contrast, studies in psychology have argued that people are not forward-looking but rather have preferences that are, or at least appear to be, time-inconsistent (Laibson, 1997; O’Donoghue and Rabin, 1999; Kan, 2007). More specifically, the psychology literature argues that because of focusing illusion – i.e., the tendency to focus too much attention on certain aspects of an event while ignoring other factors (Schkade and Kahneman, 1998) – and projection bias – i.e., the tendency to exaggerate the degree to which their future preferences resemble their current preferences (Loewenstein et al., 2003), people are prone to mispredicting the future hedonic consequences of their decisions (Gilbert and Wilson, 2000; Wilson and Gilbert, 2005).

    View all citing articles on Scopus
    View full text