Should we ban loss leaders?

Loss leaders are one of those retail marketing tricks that are often complained about, for many different reasons. For example, back in 2009 the NBR ran a story Lack of loss-leader sales good news for brand conscious wine industry. In 2011 the Otago Daily Times ran a story claiming,

Supermarkets were also driving binge-drinking by using alcohol as a "loss leader", sold below cost to attract shoppers.

Others see loss leaders as anti-competitive or predatory.

Big chains have the power to buy in bulk, reducing the individual price of products. They use tactics like ‘loss-leaders’ – purposely selling a product below cost to entice customers into their store to buy even more with their ‘savings’. The smaller retailers must match their prices, for fear of losing customers, which reduces their margins. When their rents rise markedly, they are increasingly being forced to move or close.

But are loss leaders really so bad? Are they are anti-competitive or predatory?

A forthcoming paper in the American Economic Review - Unplanned Purchases and Retail Competition by Justin P. Johnson - suggests that loss leaders are non-predatory and possibly even pro-competitive.

Abstract

I propose a framework in which asymmetric multi-product retailers compete for one-stop shoppers who have biased beliefs about their future purchase probabilities (and so make unplanned purchases). One firm carries a full portfolio of products while the other carries an incomplete but endogenous one. Using this framework, I examine the phenomenon of loss leading, the optimal product portfolio of the smaller firm, and the effects of banning loss leading. Among other results, I show that there is a non-predatory (and possibly pro-competitive) justification for the observation that such larger firms may charge below cost on the core product lines of their smaller rivals.

So, sorry, but banning loss leaders may not be such a great idea.