The 13th World Conference on Tobacco OR Health

Objective:

A number of studies have used cross-country datasets to attempt to quantify the impact of advertising restrictions on tobacco consumption. All have been restricted to developed countries and mostly to OECD countries. The conclusions drawn in the literature are diverse with some papers estimating that advertising bans and restrictions have lead to a negative impact on consumption, while others suggest no significant impact while another even suggests a positive impact.

Methods:

This paper revisits this literature and based on the well-known criticisms designs a new methodology that makes use of panel models and draws on the partial addiction hypothesis to account for the addictiveness of tobacco in estimating the demand for tobacco. It uses instrumental variable techniques in the panel models and also investigates the time series properties of the data.

Results:

This paper validates the results of Laugesen and Meads (1991) and finds that advertising bans have had a negative and significant impact on tobacco consumption in OECD countries between the 1960s and the 1980s and that comprehensive bans have had a greater impact than limited bans in reducing tobacco consumption.