Copyright protection and cumulative creation: Evidence from early twentieth-century music

This paper uses data from an online database of music sampling to estimate the effect of copyright protection on the cumulative use of music. Using unique panel data that link upstream and downstream music, I use regression analysis to examine the rates at which early 20th-century musical works were used throughout the 20th and 21st centuries. The results suggest that copyright protection causes an upstream work to be used less than half as often as it would be if it were in the public domain after conditioning on upstream-song and downstream-year fixed effects. Placebo regressions in which the copyright expiration date is artificially shifted forward and backward in time by 2, 5, and 10 years suggest an immediate effect of copyright expiration on downstream use.