More advertising = More consumption?

A debate about alcohol advertising is going on in Finland currently. Those who want to reduce overall alcohol consumption argue for banning  image promoting advertising (eg. The suggestion that a fox will be sexually more successful if it drinks a certain brand of cider – yes this particular ad was already forbidden). Will this reduce overall consumption? 

The easy answer might seem to be yes. The argument used by many who promote this view (lets call them the “pro ban” people) is what an economist would call a revealed preference argument. The fact that firms advertise suggests that advertising increases profits. From this those in favor of a ban conclude that advertising must increase demand, thereby increasing the total quantity consumed. If this was not the case, then the firms would not advertise.

This argument has a superficial appeal of logic (and economics). At least until we study it more intensely. The fact that individual firms choose to advertise does not mean that this increases the total quantity that eventually is consumed.

One easy, albeit perhaps not entirely realistic hypothetical case that disproves the pro ban argument is one whereby producers compete for existing consumers. By advertising, producer A seeks to get those who used to buy from B to buy from A instead. In this situation the only effect of advertising is to shift demand from one producer to the other producer. Total quantity consumed stays constant.

Lets make the picture a bit more nuanced. Assume that  firms seek to differentiate themselves from each other through advertising (eg. the sexually successful fox case), but at the same time advertising expands the total market. By differentiating himself from A, B can charge a higher price, as consumers now perceive B to be something different from A. Simultaneously, A can also increase advertising, in order to increase demand facing him. With A and B differentiated from each other thanks to advertising, each can charge a higher price. For a given demand, this means lower realized consumption. Call this setup the advertising induced differentiation setup.

On the other hand, if we ban advertising, then the producers will need to compete on price, instead of on image. This results in a lower price, and thus for a given demand increased realized consumption.

Thus we see that the effect of ban on advertising is not unambiguous. To sort out the effect, we need more information.

Lets assume that advertising increases the total potential market size. If this is the case, while consumers’ price elasticity of demand doesn’t change too much, then realized consumption is indeed higher under advertising induced differentiation. 

However, if advertising affects consumers’ price elasticity of demand enough (making consumers less sensitive to changes in price), while potential market size stays the same (or doesn't increase too much), then the realized consumption will be lower under advertising induced differentiation.

Thus, it all depends. The reason that advertising can decrease total consumption is that advertising can differentiate products, giving their producers market power, and thus an incentive to limit production so as to increase prices, in order to increase profits.