Cigarette smuggling: price vs.nonprice incentives

Abstract

This article uses recent US state level data to estimate the demand for cigarettes. The main contribution of this work is that, unlike previous studies, it takes into account both price and nonprice incentives behind the smuggling of cigarettes. The results show the demand for cigarettes to be elastic and greater than that found in the previous literature. The effects of greater literacy and income on smoking are insignificant. Also, the magnitude of own price elasticity seems affected by whether a correction is made for border prices. Comparing the price and nonprice influences on cigarette smuggling, it seems that price inducements remain the main force behind smuggling. Policy implications are discussed.

Acknowledgements

I am indebted to Mike Nelson for helpful comments and discussions. All errors are mine.

Notes

¹ Cigarettes sold at military bases and on Indian reservations are exempt from certain taxes. The recent availability of cigarettes over the Internet across different tax jurisdictions has further complicated matters.

² Alternately, in recent years, a greater number of educated smokers might be savvier at bypassing the traditional channels and might be buying cigarettes on the Internet. Hence, their true demand might not be measurable.

³ For instance, PRICEbor for Illinois would be the lowest cigarette price of all Illinois neighbours (i.e. Indiana, Iowa, Kentucky, Missouri and Wisconsin).

⁴ The level of detail in our data does not permit us to distinguish between short distance (casual) and long distance (organized) smuggling (Saba et al. 1995; Yurekli and Zhang, 2000).

⁵ The theoretical underpinnings for the incentives behind smuggling can be traced to the seminal work of Becker (1968).

⁶ Since corruption perceptions build over time, a 10-year average (1995–2004) is used for corruption.

⁷ The data reported are in packs sold per-capita and we are taking sales to signify consumption. This assumption is used in many cigarette demand studies (CDC, 2000; Gallet and List, 2003).

⁸ Police are also being increasingly called upon to monitor compliance with geographic smoking restrictions and this might also have a sobering effect on smoking.

¹⁰ In their meta-analysis of about 100 cigarette demand studies, Gallet and List (2003) report the mean income elasticity of cigarette demand to be around 0.4.

⁹ We also ran versions of model A in Table 2 with (i) a dummy variable identifying the six major tobacco producing states (GA, KY, NC, SC, TN and VA) and (ii) without the literacy variable. The results were similar to those for model A and are not reported here. Complete details are available upon request.