Studies on demand for cigarettes
Studies on demand for cigarettes have applied different economic models to two different types of data, aggregated and. individual level. Analysing each of two data-types has some advantages and disadvantages. The aggregate data are either time-series data or pooled cross-sectional and time-series data. High correlation among many of the key independent variables and prices can be a problem with time-series data. Consequently, estimates of the impact which prices and other factors have on demand can be sensitive to the inclusion and exclusion of the other variables.
The problem with using the pooled data is the measurement of cigarette consumption. Using these data, smoking is normally measured by annual state-level tax-paid cigarette sales. Both cross-border shopping between the neighbour states and the long-distance smuggling from low-tax to high-tax states can occur due to differences in taxes on cigarettes. Failure to account for this will produce upward-biased estimates of the impact of price on cigarette demand.
Finally, with aggregate data the demand and supply of cigarettes need to be modelled simultaneously since cigarette price, sale and consumption are simultaneously determined. In contrast, the use of individual-level data can ease some of the problems associated with aggregate data such as simultaneous biases resulting from the price and consumption, and multicollinearity between cigarette prices and other factors affecting the demand. In addition, using individual-level data can allow researchers to study the price responsiveness of different subpopulation groups such as those based on income, education, and age. The problem with individuallevel data is the accuracy with which consumption of cigarettes is measured. Self-reported consumption is typically under reported.
The problem with using the pooled data is the measurement of cigarette consumption. Using these data, smoking is normally measured by annual state-level tax-paid cigarette sales. Both cross-border shopping between the neighbour states and the long-distance smuggling from low-tax to high-tax states can occur due to differences in taxes on cigarettes. Failure to account for this will produce upward-biased estimates of the impact of price on cigarette demand.
Finally, with aggregate data the demand and supply of cigarettes need to be modelled simultaneously since cigarette price, sale and consumption are simultaneously determined. In contrast, the use of individual-level data can ease some of the problems associated with aggregate data such as simultaneous biases resulting from the price and consumption, and multicollinearity between cigarette prices and other factors affecting the demand. In addition, using individual-level data can allow researchers to study the price responsiveness of different subpopulation groups such as those based on income, education, and age. The problem with individuallevel data is the accuracy with which consumption of cigarettes is measured. Self-reported consumption is typically under reported.
