Accept no substitutes? Vol. II

by Andreas Baumann

The city I grew up in – Aalborg in Northern Jutland – recently passed a resolution by the city council, which meant that e-cigarettes are subject to the same regulations in terms of smoking on public property as regular cigarettes. I believe this is a bad idea – and I’d like to explain why, as this underscores how one should think about substitution in terms of public policy.

The e-cigarette is technically a vaporiser, in which a fluid containing nicotine and aromatic components are vaporised with the purpose of inhalation. The base of this fluid is propylene glycerol and vegetable glycerin – common food additives. 

Over the last ten years, more and more restriction on tobacco smoking has come into place in Denmark, including municipalities regulating the smoking of tobacco on municipal grounds – even outside. Recently, the Aalborg City Council stipulated that this ban extends to the e-cigaretes mentioned above (nb: link in Danish).


One of the main arguments presented is the fact that second-hand vapor might have adverse effects, i.e. that there might be carcinogenic or otherwise adverse effects of second-hand inhalation, akin to the well-documented risk from second-hand smoke. This is a serious risk that needs to be assessed, since nicotine itself might be carcinogenic (Toxicol. Sci. 2004). As such, there might be a valid argument from externalities that such products should be regulated. However, very little research exists on this topic. One might then prefer to err on the side of caution.

Substitution effect
What the city council and others, who argue for the regulation of e-cigarettes haven’t considered properly, is the effect of regulation on consumption substitution. People who smoke e-cigarettes tend to use them primarily as substitutes for regular cigarettes, and this substitution was prompted, in part, by the smoking regulations enacted. When it was illegal to smoke a cigarette, but legal to smoke an e-cigarette at a bar or restaurant, the “cost” of the e-cigarette went down compared to the ordinary cigarette, while utility was held constant – leading people to switch consumption to e-cigarettes (i). Increasing the cost of e-cigarettes by subjecting it to the same legal constraints as ordinary cigarettes then decreases the marginal benefit of using an e-cigarette instead of an ordinary cigarette – which should inhibit the substitution rate and might even lead to people switching back to the ordinary cigarette!

While both the short-term and long-term health effects of e-cigarette smoking is relatively unknown, I would venture the guess that they’re milder than those of ordinary cigarettes, since there is no tar, benzene etc. in the vapour inhaled. This means that people substituting e-cigarettes for cigarette consumption is reducing their own risk of disease, and indirectly following from this, e-cigarette substitution has positive externalities under a regimen of socialised medicine.

The point
The point of this post is not to argue in favor of e-cigarettes or whether or not one should ban them or their use in public facilities. Rather, the problem is that when you’re raising the cost of them without raising the cost on ordinary cigarettes at the same time, you’re making it less attractive to take up e-cigarettes instead of ordinary smoking. The analysis of the City Council and the sources they cite is not technically wrong, but they consider the wrong comparison: one should not consider e-cigarettes against tobacco abstention, but against smoking ordinary cigarettes, because that’s the usage frame indicated by actual human behaviour. 

This is exactly what ties this discussion in with the one about cannabis legalisation: people are comparing the use of cannabis to not using euphoriants, when they should actually compare it to using alcohol and evaluate those two scenarios against each other.


(i) One could argue that substitution should be much more rapid and wide-spread, considering the adverse long-term health effects of tobacco smoking. However, there is some empirical research indicating that smoking behaviour might be the result of greater time discounting, which would diminish the influence of long-term effects on substitution behaviour (see for example Kamijo et al. 2011)