Academic journal article Contemporary Economic Policy

Does Advertising Matter? Estimating the Impact of Cigarette Advertising on Smoking among Youth in Developing Countries

Academic journal article Contemporary Economic Policy

Does Advertising Matter? Estimating the Impact of Cigarette Advertising on Smoking among Youth in Developing Countries

Article excerpt

I. INTRODUCTION

The worldwide pattern of tobacco consumption has been shifting away from developed countries toward less developed nations (WHO 2011). Smoking in developing countries can begin very early in life, and it is estimated that, worldwide, a quarter of the teenagers who smoke have tried their first cigarette before the age of 10 (Shafey et al. 2006). Because smoking habits are established primarily in youth, policies such as advertising restrictions that have the potential to alter youth smoking behavior attract special interest. Advertising restrictions are an established part of the policy toolkit for tobacco control in developed countries, and are becoming increasingly so in developing countries as well (WHO 2011). However, the vast majority of the current economic evidence on the impact of cigarette advertising on smoking is based on data from the United States and other advanced economies. It is unclear whether empirical findings from this literature can be applied to developing countries whose populations may have different patterns of responsiveness to tobacco control policies.

The existing evidence on the impact of advertising on smoking in the United States and other developed countries yields conflicting results. A literature review in Blecher (2008) lists 17 studies which find a positive and significant impact of advertising on smoking and 18 studies which find the opposite. Similarly, Saffer and Chaloupka (2000) identify nine studies with no advertising effect and nine studies with a positive advertising effect. A meta-analysis of 49 studies on cigarette advertising regulation in the United States and other developed countries in Nelson (2006) finds insignificant advertising regulation effects. Saffer and Chaloupka (2000) distinguish between different intensity levels of advertising regulation in 22 Organization for Economic Cooperation and Development (OECD) countries and find that weak advertising bans are ineffective while highly comprehensive bans can impact aggregate consumption. Blecher (2008) evaluates aggregate data from 30 developed and developing countries and finds similar evidence that the impact of bans is largest when they are most comprehensive, however, both studies may be limited by nonstationarity of cigarette consumption data. Using U.S. data, Iwasaki, Tremblay, and Tremblay (2006) conclude that even if advertising restrictions do not decrease market demand, they are still effective in decreasing the equilibrium cigarette consumption in the United States due to supply side effects from reduced price competition. In contrast, Nelson (2003a) evaluates data from 20 OECD countries and concludes that the causal link between advertising bans and cigarette consumption flows not from bans to consumption but, rather, from consumption to bans--i.e., the change in political climate that was concurrent with reduced consumption led to the introduction of stricter bans.

While the aforementioned studies support a variety of opinions, virtually all are based on macro-level data, i.e., the dependent variable is aggregate cigarette consumption at the state or country level. However, the direction of the causality between bans and consumption is harder to identify when consumption is an aggregate measure than when it is measured on an individual basis. For instance, identification of a causal advertising effect may be complicated if higher overall consumption in certain locations prevents advertising bans from being passed, or if tobacco producers respond to higher consumption in certain areas by boosting advertising spending. The reverse causality problem can be addressed by modeling individual smoking behavior rather than aggregate consumption since this permits the assumption that no single person's response to advertising can influence advertising policy. However, micro-level data that contain information on both individual consumption and advertising exposure are not readily available even in developed countries, and such studies are rare. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.