EclectEcon

Economics and the mid-life crisis have much in common: Both dwell on foregone opportunities

C'est la vie; c'est la guerre; c'est la pomme de terre . . . . . . . . . . . . . email: jpalmer at uwo dot ca


. . . . . . . . . . .Richard Posner should be awarded the next Nobel Prize in Economics . . . . . . . . . . . .

Friday, November 12, 2004

Return policies


One of the really neat things about most of the major retailers is that if I buy something and don't like it, I can return it for a full refund, no questions asked, no problem. As an example, I sometimes find that when I'm working on something around the house, I buy too many materials; the store lets me return the extras for a full refund. As another example, I recently bought something at a Walmart in London, Ontario, but when I got it home, found it was defective. Rather than make the 1-hour drive to London to return it, I simply went to the Walmart in Goderich. They zapped my receipt, pulled up a record of the transaction, and provided an exchange.

In most instances, in fact, even if you don't have a receipt, the major retailers will provide a refund or, at the very least, a store credit.

I've often wondered not if people abuse this system, but how many abuse the system and to what extent. I guess in some cases the answer is that the abuse is serious, judging from this article in the Washington Post. Some major retailers are beginning to track refund histories and are denying refunds to people who seem to return a large portion of their purchases.

I don't blame them, but they have to be careful about type I and type II errors. If they have a strict policy, they'll cut down on the number of instances of people who "buy" something, use it once, and then return it (a wardrobe for a special event, a tool for a special project, a generator and propane stove prior to the Y2K scare, etc.); these folks are essentially using the store to provide them with the use of something for a short term without compensating the store. This practice can be quite costly for the stores.

At the same time, if the store is too strict, it will anger some customers who, due to nothing more than randomness, have a high number of returns. Trading off the costs and benefits arising from the policy and the type I and type II errors requires careful use of binomial distributions combined with an understanding of shoppers' tendencies. This is not a simple problem, but computer tracking will almost certainly allow/encourage other major retailers to move in this direction.
 
Who Links Here