Pricing policy and some other predictors of smoking behaviours: an analysis of French retrospective data

https://doi.org/10.1016/j.drugpo.2004.06.003Get rights and content

Abstract

In France as in other developed countries, cigarette smoking has been acknowledged as the most preventable cause of death and disability. Tobacco control has become a priority for the French government, which decided a 20% rise of the cigarette price at the end of 2003 and plans further rise in 2004. This study adopted a life-course perspective with retrospective data extracted from a national telephone survey for assessing the impact of pricing policy and some other predictors on smoking behaviours in France during the period 1965–1999. Our findings suggest that during a period marked by cultural trends favouring more frequent and earlier initiation of cigarette smoking, pricing policy has been a poor deterrent to youth smoking. Such findings highlight the necessity to implement simultaneously other preventive strategies, including restrictions on tobacco advertising, youth access and smoking in public places, as well as mass media campaigns and school-based educational interventions.

Introduction

In France as in other developed countries, cigarette smoking has been acknowledged as the most preventable cause of death and disability. Every year, smoking causes at least 40,000 premature deaths (Nizard, 2000), and its social cost is estimated at 1.1% of the Gross Domestic Product (Kopp & Fenoglio, 2000). Tobacco control has become a priority for the French government, which decided a 20% rise of the cigarette price at the end of 2003 and plans further rise in 2004. But what is the impact of pricing policy on smoking behaviours?

This issue is rather easy to investigate in countries such as the United States or Canada: indeed, as cigarette price varies across states or provinces, a national cross-sectional survey can capture its impact on smoking (Chaloupka, 1991; Chaloupka & Weschler, 1997; Mullahy, 1997; Stephens, Pederson, Koval, & Macnab, 2001). On the contrary, in most countries this price is identical across the whole territory but time-dependent, so another strategy consists in merging data collected over time to introduce the variability of price (Anguis & Dubeaux, 1997; Holly, Gardiol, & Zurn, 1999; Townsend, Roderick, & Cooper, 1994). At the most aggregate level, short-term changes in annual sales of cigarettes illustrate the power of pricing policy: in France, from 1965 to 1999, the curve of sales perfectly mirrored the curve of cigarette price (Fig. 1).

Nevertheless, this relationship may reflect a number of possible behavioural changes: as cigarette price increases, some smokers may decrease their consumption, others may quit smoking, and some non-smokers may be deterred from taking up smoking. The surveys conducted since 1974 by the French Board for Health Education (CFES) provide further information (Fig. 2): among people aged 18 and over, the male smoking prevalence decreased since the 1970s, while the female prevalence remained stable. Comparing Fig. 1, Fig. 2, it is evident that cigarette price has a strong impact on the quantities smoked, but a weaker one on the decision to smoke or not (as in the United Kingdom, Townsend et al., 1994).

Fig. 1 mingles smokers and non-smokers together and Fig. 2 does not distinguish never-smokers from former-smokers, thus both figures draw an incomplete picture of smoking. Repeated cross-sectional data provide a set of smoking rates to compare one with another, but each smoking prevalence mixes lifetime consumption paths observed at various stages for successive birth cohorts, and these paths may change across cohorts (with delayed or more precocious initiations and cessations).

In order to improve our understanding of the determinants of both smoking initiation and cessation, it is therefore useful to adopt a life-course perspective (Kenkel, Lillard, & Mathios, 2002a). Moreover, as a recent study established the usefulness of retrospectively reported information about smoking (Kenkel, Lillard, & Mathios, 2003), we opted for the life-course perspective to analyse French retrospective data. Thanks to this perspective, it was possible to assess jointly the impact on smoking cessation of several well-known predictors of quitting: cigarette price, but also gender, age, age at initiation, academic achievement and parenthood.

Section snippets

Sampling and questionnaire

We used data abstracted from the French Health Barometer (FHB), a national survey conducted by the CFES in December 1999 among a representative sample of French aged 12–75. The CFES drew a random sample of telephone numbers from the database furnished by the French National Telephone Company (France Telecom), which contains almost all phone numbers in France. A letter that introduced the survey and promised anonymity was sent to corresponding addresses. The telephone survey began two weeks

Data collected

Overall, 13,685 individuals participated in the survey (response rate: 70.1%). In this representative sample adjusted by sampling weight, 47.8% were non-smokers, 19.2% were former smokers, and 33.0% were current smokers (28.2% of daily smokers and 4.8% of occasional smokers).

Life-course smoking patterns across cohorts and price

For each cohort, Fig. 3, Fig. 4 give regular smoking prevalence curves and cessation cumulative rates (computed with the cohort size at the denominator). The rate of initiation can be easily deduced from the two other

Methodological shortcomings

The life-course perspective adopted in the present study enables an analysis of the natural history of smoking within and across cohorts, and to separate initiation and cessation effects on smoking prevalence. However, we have to acknowledge that such perspective cannot separate a ‘pure’ price effect from the impact of other factors that have changed across cohorts, including other tobacco control policies as well as cultural trends toward women and youth emancipation, in France as in other

Conclusion

From a practical perspective, we do not claim that increasing the cigarette price is not a relevant policy. Our findings rather highlight the necessity to implement simultaneously other preventive strategies, including restrictions on tobacco advertising, youth access and smoking in public places, as well as mass media campaigns and school-based educational interventions. Increasing the cigarette price could be more effective if coordinated with other interventions (Lantz et al., 2000).

The

References (32)

  • Anguis, M., & Dubeaux, D. (1997). Les fumeurs face aux récentes hausses du prix du tabac. INSEE première...
  • Beck, F., Legleye, S., & Peretti-Watel, P. (2002). Alcool, tabac, cannabis et autres drogues illicites parmi les élèves...
  • G.S. Becker et al.

    A theory of rational addiction

    Journal of Political Economy

    (1988)
  • F. Chaloupka

    Rational addictive behavior and cigarette smoking

    Journal of Political Economy

    (1991)
  • Chaloupka, F. J., & Warner, K. E. (2000). The economics of smoking. In: A. J. Culyer & J. P. Newhouse (Eds.), Handbook...
  • DeCicca, P., Kenkel, D., & Mathios, A. (1998). Putting out the fires: will higher cigarette taxes reduce youth smoking?...
  • Cited by (11)

    View all citing articles on Scopus
    View full text