Georgia law professor Paul Heald’s new working paper suggests so:

By examining what is for sale “on the shelf,” the analysis of this data reveals a striking finding that directly contradicts the under-exploitation theory of copyright: Copyright correlates significantly with the disappearance of works rather than with their availability. Shortly after works are created and proprietized, they tend to disappear from public view only to reappear in significantly increased numbers when they fall into the public domain and lose their owners. For example, more than three times as many new books originally published in the 1850’s are for sale by Amazon than books from the 1950’s, despite the fact that many fewer books were published in the 1850’s.

Copyright law provides artists with a limited monopoly over their work to incentivize the creation of such works. Many copyright owners want the copyright length extended so that they can have a monopoly over both their past works and future works for a longer period (See the Mickey Mouse Curve). But extending the copyright term for past works has no effect on incentives – after all those works have already been created. Congress should have no reason to grandfather in those artistic creations.

In return, lobbyists have argued that increasing the duration of copyright protection will allow owners to profit from their work for a longer period and give them greater incentive to make their works widely accessible. They maintain that extending the copyright term for past works is in the public’s best interest even if it has no effect on incentives.

Heald’s paper pushes back on this argument, finding that works in the public domain are more widely available than those still under copyright protection. He took a random sample of 7000 books on Amazon and matched them up with their year of publication in the Library of Congress. For a couple of technical reasons, he was only able to match 2317 of the 7000. Here’s the distribution of when those books were published:

All books published before 1923 are currently in the public domain. All those afterward are not. The graph shows that when a book enters the public domain, its availability spikes.

The usual caveats apply here. There’s nothing that suggests with certainty that copyright law causes this distribution. Correlation does not equal causation. The sample size is not huge and Amazon does not represent all retailers. Amazon probably finds it most profitable to license and sell recently published books as well as those that are in the public domain (no license necessary). Books in the middle – not recently published but still copyrighted – are least profitable. There certainly could be many other factors that caused this distribution as well.

Heald continues on to look the availability of songs in the public domain versus songs that are not. That analysis is less convincing thanks to some strange proxies. Nevertheless, Heald’s analysis of Amazon’s book offerings stands. It’s not decisive empirical proof by any means, but at the very least, the next time someone says that a longer copyright term will increase the availability of a work, take it with a grain of salt.