What is the welfare cost of Amazon supply restrictions on books?

Maybe that welfare cost is not very high at all.   After all, if Amazon does not carry a book you can sign up at the Barnes & Noble website and that takes a few minutes at most.

There is a tension in most criticisms of Amazon.  On one hand, the critic wishes to argue that a “not carry” decision by Amazon has a big impact on how a book does.  On the other hand, the critic wishes to argue that the loss of access to particular titles is a big deal.  You cannot easily have it both ways.  If readers won’t switch to B&N.com, they must not care very much about particular titles, in which case the Amazon refusal to carry (or delay in shipping) is small even relative to the size of the (small) trade in books.

Krugman’s column today, which covers Amazon vs. Hachette, appears terrible at first glance, but in fact he presents a new and original argument.  Get past the mood affiliation and you come to this:

…what Amazon possesses is the power to kill the buzz. It’s definitely possible, with some extra effort, to buy a book you’ve heard about even if Amazon doesn’t carry it — but if Amazon doesn’t carry that book, you’re much less likely to hear about it in the first place.

If I may fill in some blanks, one possible version of the hypothesis — to pull an idea from Gary Becker and Steve Erfle — is that readers consume both “books” and “buzz around books” as complements.  The marginal gains from books can be low but the marginal gains from the bundled package may be much higher and those higher gains will not be measured by the (high) price elasticity of book purchases.

In the early stages of this war, Amazon boycotts have often increased the buzz for a book, such as with Beth Macy’s Factory Man.  But if these practices continue, they will cease to be news stories and an Amazon refusal to carry or promote plausibly will damage how books will do, without much potential for upside.

How much of the value in a book/buzz package is due to the buzz?  65 percent?  That would explain the concentration of reading interest among bestsellers and books your peers are reading.  But if Amazon won’t carry or promote a book, does the total supply of buzz fall?  Or does the buzz simply transfer to other titles?  In the latter case we are again back to small welfare costs from an Amazon refusal to carry.  Krugman’s idea is fun, but I am still inclined to think the welfare cost of Amazon supply restrictions on individual books likely is small, again even relative to the size of the book sector, much less relative to gdp.

It is fine to argue that Amazon is being unfair to some authors and to object on ethical grounds.  The economist also should add that readers don’t seem to mind very much.  Most of the objections I am seeing are coming from authors and publishers, who of course in this sector are much less diversified in their interests than are readers.


Comments for this post are closed