Tobacco and Alcohol: Complements or Substitutes? - A Statistical Guinea Pig Approach

Harald Tauchmann, Silja Gohlmann, Till Requate
and Christoph M. Schmidt*

Tobacco and Alcohol: Complements or Substitutes? -

A Statistical Guinea Pig Approach

Abstract

The question of whether two drugs - namely alcohol and tobacco - are used as
complements or substitutes is of crucial interest if side-effects of anti-drug pol-
icies are considered. Numerous papers have empirically addressed this issue
by estimating demand systems for alcohol and tobacco and subsequently cal-
culating cross-price effects. However, this traditional approach often is seri-
ously hampered by insufficient price-variation observed in survey data. We,
therefore, suggest an alternative instrumental variables approach that statisti-
cally mimics an experimental study and does not rely on prices as explanatory
variables. This approach is applied to German survey data. Our estimation
results suggest that a reduction in tobacco consumption results in a reduction
in alcohol consumption, too. It is shown theoretically that this implies that
alcohol and tobacco are complements. Hence, we conclude that successful
antismoking policies will not result in the unintended side-effect of an in-
creased (ab)use of alcohol.

JEL Classification: C31, D12, I12

Keywords: Interdependence in consumption, tobacco and alcohol,
instrumental variables approach

November 2006

*Harald Tauchmann, RWI Essen; Silja Gohlmann, RWI Essen; Till Requate, University Kiel;
Christoph M. Schmidt, RWI Essen and Ruhr University Bochum. All correspondence to Harald
Tauchmann, Rheinisch-Westfalisches Institut für Wirtschaftsforschung (RWI Essen), Hohenzol-
lernstr. 1-3, 45128 Essen, Germany. Fax: +49-201-8149-200. Email: [email protected].
We would like to thank Boris Augurzky and Lars Siemers as well as participants at the Scottish
Economic Society Annual Conference 2006, the European Society for Population Economics
20th Annual Conference, and the Verein für Socialpolitik Annual Conference 2006 for many use-
ful comments. Any errors are our own.

<<   2   3   4   5   6   7   8   >>

More intriguing information