Prices, Cigarette Consumption and Smoking Intensity

This paper uses novel data sets, which report both the quantity of cigarettes smoked and a measure of nicotine intake obtained from blood or saliva samples for a large and representative population over time for the UK and the US. We show that self-reported quantities, routinely used in the smoking literature, explain only a small fraction of the variance in smoking as measured by nicotine intake. We argue that a large part of the remaining variance reflects heterogeneity in smoking intensity, defined as the amount of nicotine extracted per cigarette. We show that smoking intensity varies with observable characteristics such as gender, age, socio-economic status or race. This paper makes therefore two main contributions. First, we show that although higher prices decrease the number of cigarette smoked, they have no effect on nicotine consumption, as smokers compensate by smoking each cigarette more intensively. Second, our results imply that the economic literature on smoking has missed an important adjustment behavior, and that previous studies on smoking probably suffer from severe estimation biases. * [email protected] and [email protected]. We are grateful to Christian Dustmann, Ken Chay, Martin Jarvis, Steve Machin, Adriana Lleras-Muney and Costas Meghir for helpful discussions and comments and to William Evans for supplying us with data on US tobacco prices and taxes. Prices, Cigarette Consumption, and Smoking Intensity 2