Tuesday, January 8, 2013

Are Loss Leaders an Anticompetitive Free Lunch?

All experienced shoppers understand the concept of a "loss leader." A store offers an exceptionally low price on a particular item that is likely to be popular--indeed, the price is so low that on that item, the seller may lose money. But by advertising this "loss leader" item, the store hopes to bring in consumers who will then purchase other items as well that are not marked down. Of course, from the consumer point of view, the challenge is whether, in buying the loss leader item and making other purchases at that store, you actually end up with a better deal than if you bought the loss leader and then went to another store--or perhaps even did all your shopping in one stop at a different store. 

The terminology of loss-leaders is apparently nearly a century old. The earliest usage given in the Oxford English Dictionary is from a 1922 book called Chain Stores, by W.S. Hayward and P. White, who wrote: "Many chains have a fixed policy of featuring each week a so-called ‘loss leader’. That is, some well known article, the price of which is usually standard and known to the majority of purchasers, is put on sale at actual cost to the chain or even at a slight loss..on the theory..that people will be attracted to this bargain and buy other goods as well. Loss leaders are often termed ‘weekly specials’."

But every economist knows at least one example of the classic loss leader from the late 19th century, the "free lunch." At that time, a number of bars and saloons would advertise a "free lunch," but customers were effectively required to purchase beer or some other drink. If you tried to eat the free lunch without purchasing a drink, you would likely be thrown out. Thus, the origins of the TANSTAAFL abbreviation: "There ain't no such thing as a free lunch."

What I had not known is that there is a serious argument in the industrial organization literature over whether loss leaders should be treated by antitrust authorities as an anticompetitive practice. In 2002, for example, Germany's highest court upheld a decision of the Germany's Federal Cartel Office that Wal-Mart was required to stop selling basic food items like milk and sugar below cost as a way of attracting customers. Ireland and France have also been known for their fairly strict laws prohibiting resale below cost.

The theory of "Loss Leading as an Exploitative Practice" is laid out by Zhijun Chen and Patrick Rey in the December 2012 issue of the American Economic Review. (The AER is not freely available online, but many academics will have access to this somewhat technical article through library subscriptions.) Their approach has two kinds of sellers: large firms that sell a wide range of product, and smaller firms that sell a more limited range of products. It also has some buyers who have a high time cost of shopping--and thus prefer to shop at one or a few locations--along with other buyers who have a lower time cost of shopping and thus are more willing to shop at many locations. They then build up a mathematical model in which large firms use loss leaders as a way of sorting consumers and attracting those who are likely to do all their shopping in one place. Once they have those consumers inside the store, they can then charge higher prices for other items. Thus, the result of allowing "loss leaders" in this model is that a number of consumers end up paying more, and large stores with a wide product range may tend to drive smaller stores out of the market.

Of course, the fact that it is possible put together a certain theoretical model with this outcome doesn't prove that it is the only possible outcome, or that it's the outcome that should be of greatest practical concern. Back in 2007, the OECD Journal of Competition Law and Policy hosted a symposium on "Resale Below Cost Laws and Regulations." The articles can be read freely on-line with a slightly clunky browser here, or again, many academics will have on-line access through library subscriptions.  The general tone of these articles is that loss leaders should not be viewed as an anticompetitive practice. In no particular order, and in my own words, here are some of the points that are made:

  • In general, business practices that reduce prices of at least some items to consumers should be  presumptively supported by anticompetition authorities, unless there is a very strong case against them. In general, regulators should spend little time second-guessing prices that are too low, and more time looking at prices that are too high or practices that are clearly unfair to consumers. 
  • There are a number of reasons why loss leaders might tend to encourage competition. Offering a loss leader can encourage consumers to overcome their inertia of buying the same things at the same prices and try out a new product or a new store. Sometimes producers may want to reward customers who are especially loyal or buy in especially large volumes. It may be far more effective for a store to advertise low prices on a few items than to advertise that "everything in the store is on average 2% cheaper than the competition." Loss leaders may be linked to other things the seller desires, like become a provider of credit to the buyer or raising the chance of getting detailed feedback from the buyer. Loss leaders may especially useful to new entrants in markets, seeking to gain a foothold.
  • Evidence from Ireland suggests that the grocery products where loss leaders are prohibited tend to have higher or rising prices, compared with other products. Since the products where loss leaders are prohibited tend to be more essential products, the prohibition on selling such items below cost tends to weigh most heavily on those with lower income levels.
  • In general, there has been a trend in retailing toward big-volume, low price retailers. But this trend doesn't seem to have been any slower in places where limitations on resale-below-cost were in place. And if the policy goal is to help small retailers, there are likely to be better targeted and less costly approaches than preventing loss leaders. Indeed, small firms may in some cases wish to entice buyers by offering loss leaders themselves.
  • It's not clear how to apply prohibitions against loss leaders to vertically integrated firms, since they have considerable ability to use accounting rules to reduce the "cost" of production--and then to resell at whatever price they wish. Even in a firm that is not vertically integrated, invoices for what is purchased can often include various discounts and allowances, and it is an administrative challenge for rules preventing resale-below-cost to take these into effect. If a firm buys products at a high wholesale price, and then the market price drops, presumably a resale-below-cost rule would prevent the firm from selling its products at all.
  • It's worth noting that laws preventing loss leaders are not the same as laws that block "predatory pricing," where the idea is to drive a competitor out of business with with very low prices and then to charge higher prices on everything. This intertemporal scenario is quite different from what happens in a prohibition of loss leaders.
  • Allowing loss leaders doesn't mean allowing deceptive claims, where for example there is a very low price advertised but the item is immediately out of stock, or of unexpectedly low quality.
Market competition is a multidimensional affair, happening along many dimensions at once. I lack confidence that government regulators with a mandate to block overly low prices will end up acting in a way that will benefit consumers.