"To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries." U.S. Constitution


A number of economists have argued against the retroactive extension of copyright to existing works. Since both economics and the U.S. Constitution encourage copyright only to the extent it promotes the production of literary and other copyrightable works, the argument is pretty self-explanatory. Extending the length of copyright of works that are already produced can scarcely make them more likely to be produced.

There are also quite straightforward computations showing that extending the copyright period the length of time currently mandated by law can have little effect on encouraging the production of more works. Simply said, the extending the copyright only increases the income of the creator at a fairly distant time in the future, and income in the far future isn't worth very much.

But why stop at the theory? Beginning in 1919, the length of copyright has been continually extended. At the turn of the century it was 28 years and could be extended for another 14. Prior to the Sonny Bono/Mickey Mouse Act of 1998 it was 75 years for works for hire, and the life of the author plus 50 years otherwise. So the length of copyright term roughly doubled during the course of the century. (Today it is 95 years for works for hire, and the life of the author plus 70 years otherwise.) If this approximate doubling of the length of copyright encouraged the production of additional literary works, we would expect that the per capital number of literary works registered would have gone up. Below is a graph of the number of literary copyrights per capita registered in the United States in the last century. Apparently the theory works. As predicted, the various copyright extensions have not led to an increase in the output of literary work.

The Data

The data is from http://www.copyright.gov/reports/annual/2000/appendices.pdf, which reports annual copyright registrations. However, beginning in 1909, non-literary works were covered by copyright for the first time. However, the next appendix in the same source gives a breakdown of copyrighted works by category for the year 2000 at which time literary works are 46.3% of the total. Assuming that the number of non-literary works increased linearly at a fraction of the total from 1909 to 2000 should provide a high degree of accuracy in the early and late parts of the century, and a decent estimate in the intervening mid-century. The strange decrease in registrations in 1976 is due to switching the starting date of the fiscal year; the downward spike represents a "year" that is only 4 months long. The population data is from http://www.census.gov/population/estimates/nation/popclockest.txt.
Eldred versus Ashcroft

Statement of Economists

Affidavit of Hal Varian