Brad DeLong discusses Larry Lessig's views on the harmfulness of copyright laws -- that is, the harm of extending the term of copyrights at the behest of big corporations (like Disney) so as to prevent valuable cultural creations of exactly the sort that Walt Disney himself invented when he put out modified versions of Grimm's folk tales, etc.

Something to consider: One of the all-time great law-and-economics duos -- William Landes and Richard Posner -- have a brand new article on SSRN, in which they defend the idea of indefinitely renewable copyrights. They rely on two possible bad effects that might happen when a work passes into the public domain: First, the destruction of the incentive to maintain and exploit the work, as well as too-great an incentive to make otherwise-unnecessary changes so as to retain something copyrightable.

The second problem they call "congestion externalities." What they mean by that term is that if too many people use a work or image, it might be so devalued that there is a general loss in social welfare. On p. 13, for example, they say this:

If anyone could use Humphrey Bogart's name or likeness in advertising, . . . the total utility might fall if the lack of excludability and resulting proliferation of the Bogart image led to confusion, the tarnishing of the image, or sheer boredom on the part of the consuming public. Eventually the image might become worthless.

They go on to examine empirical evidence on the small number of copyrights that are renewed, and on the effects of a renewal fee (evidently, the demand for copyright renewals is fairly elastic, which means that raising the fee decreases the number of renewals). They then suggest that if we shorten the copyright term, and raise the fee, hardly anyone will bother to renew copyrights anyway, which means most works will be in the public domain rather quickly. But very valuable works, the kind that the inventors should benefit from, should be able to be renewed indefinitely. Indefinite renewal would prevent the rent-seeking efforts that currently proliferate whenever a major player's copyright term is about to expire.

Despite my hesitation to go against the august authority of Landes and Posner in this realm, I'm not sure I'm convinced. Their argument on congestion externalities seems rather weak and is not fleshed out with much detail or evidence. Still, their paper should be considered in evaluating the arguments about copyright terms.