Academic journal article Journal of Economics and Finance

Smoking Prevalence in the United States: Differences across Socioeconomic Groups

Academic journal article Journal of Economics and Finance

Smoking Prevalence in the United States: Differences across Socioeconomic Groups

Article excerpt


This paper uses recent US state-level data to examine smoking behavior across four population groups divided by age, literacy, income and ethnicity. Another contribution lies in the consideration of various smoking control policies and the full costs of smoking. Overall, the results show that non-price smoking policies are relatively more effective than price or tax policies in reducing smoking prevalence and that these policies gain effectiveness as the population begins to age.

Keywords Smoking prevalence * Smoking restrictions * Cigarettes * Socioeconomic groups

JEL Classifications D12 * H70

1 Introduction

The prevalence of smoking varies significantly across various population groups and these groups have responded differently over time to various smoking control measures. For instance, Jha et al. (2002) estimate global smoking prevalence across various groups and find significant differences (also see European Commission 2006). This paper examines the determinants of smoking prevalence across various socioeconomic groups in the USA. Whereas our understanding of the smoking behavior of the adult population is quite good, especially for the USA and for a few other developed nations, attention to population subgroups has been limited (see Cameron 1998; CDC 2000; Chaloupka and Warner 2000 and Goel and Nelson 2006 for surveys of the literature).

From a public policy perspective, an understanding of different smoking patterns and determinants of behavior across different groups would enable formulation of specific policies to be effective in each case. On the other hand, an absence of significant differences across groups would imply that uniform smoking-control policies may be implemented. Due to the habit-forming nature of tobacco products, a handle on prevalence is necessary for the design of early intervention programs. Currently, the effectiveness of smoking-control policies across nations and groups varies considerably (see Chaloupka and Saffer 1988 and Wakefield et al. 1992; also Goel and Nelson 2006 for a review).

Four categories of socioeconomic groups considered in this paper deal with age, literacy, income and ethnicity. These groups can affect smoking participation in unique ways. For instance, in 2002 the average smoking prevalence in the US population without a high school education was 32.95%, while prevalence went down to nearly half that (17.49%) in population with more than high school education (Table 1). Age can have a bearing upon smoking due to the habit-forming nature of cigarettes. Thus, different smoking control policies may be suitable for youth and adults. A more educated population might be better informed about the potential risks of smoking and therefore be less likely to smoke. On the other hand, greater literacy could give a false sense of confidence to some in terms of their ability to deal with adverse health effects of smoking. In other words, some educated smokers might be more belligerent in their attitudes towards smoking. Further higher incomes negate the effects of high cigarette taxes and render price-based smoking control policies less effective. ' Finally, the race or ethnicity of a person might play a significant role in smoking participation. Some cultures frown upon smoking, while others view smoking less unfavorably. Consideration of all these aspects would educate us as to which influences are more powerful across different socioeconomic groups. Smoking-control policies may then be better tailored to be more effective.

2 The model, data and results

The estimated smoking participation equation follows the literature in controlling for smoking behavior by including the price of cigarettes, income and educational attainment of the population (see Gallet and List 2003). In addition, we control for the effects of anti-smoking policies, cross-state cigarette smuggling, and medical costs. Thus, besides the focus on the various socioeconomic groups, the consideration of these influences may be seen as an additional contribution of this work. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.