Skip Navigation
Back to PolicyBot

Economic Effects of Smoking Bans on Restaurants and Pubs

December 1, 2008

Public health advocates argue that bans are necessary because non-smokers need protection from second-hand smoke.

two tumblers with liquor and a lit cigarrette

Public health advocates argue that bans are necessary because non-smokers need protection from second-hand smoke. The United Kingdom recently enacted smoking bans in public places such as restaurants and pubs (Scotland, 26 March 2006; Wales, 2 April 2007; Northern Ireland, 30 April 2007; and England, 1 July 2007), where a major concern was for the health of employees in such establishments. Advocates also claim that such bans do not harm business owners because of a vast empirical literature showing that restaurants and bars in the United States never suffer harm following bans (Glantz, 2007). In fact, advocates often claim that bans improve sales at restaurants and bars and so, in effect, owners should thank them for promoting bans.

This paper focuses on whether smoking bans harm owners of restaurants, pubs, social and tombola clubs in the hospitality industry. We do not dwell on the issue of protecting employees from possible ill-health effects from passive smoking for two reasons. First, smoking itself is legal and if government was serious about ‘health and safety’ it would act on the primary source of the problem. Second, in competitive labour markets, characteristic of those in the hospitality industry, workers who do not wish to be exposed to passive smoking could choose to work in other establishments. The issue would be different, just as it would be for issues of racial and sexual equality, if the employers were monopsonistic purchasers of labour but this is emphatically not the case here.

We develop our model within the Coasian framework whereby owners of businesses have incentives to deal with smoking disputes between smokers and non-smokers. This model has two important predictions. One, owners attempt to deal with smoking disputes prior to the enactment of bans and therefore it is incorrect to argue that bans are necessary because the private market has no method of attempting to solve smoking problems. Two, smoking bans are expected to exert different effects on different businesses so that it is inappropriate to estimate aggregate effects of bans on businesses. Some businesses will be unaffected while others will experience losses or gains. We argue that studies showing no harm are not as unbiased or scientific as many researchers claim. We also raise the issue of whether claims of harm from second-hand smoke in restaurants and bars are based on sound evidence because this argument is often an important reason given by ban advocates for government intervention.

Michael L. Marlow is professor of economics at Cal Poly, San Luis Obispo.