By Jeremy Watson , Megan MacGarvie, and John McKeon
Technological change has led to a rapid and dramatic evolution of business models in copyright-intensive industries, especially in the music industry. This paper examines the nuanced effects of recording copyright on the availability of music, using a data set of musical releases, live performances, and digital availability of musicians popular in the UK in the 1960s.
We find that copyright has different, and even opposing, effects on the availability of popular music dependent upon the nature of the distribution channel. In contrast to most prior research, we are able to identify the impact of a change in copyright status during an artist’s lifetime (and can therefore observe changes in the supply of performances). We also exploit the extension of copyright terms in 2013 to empirically disentangle copyright status from year and age effects, correcting for potential bias due to 50th anniversary effects.
Examining data on re-releases of songs distributed in retail channels, our results suggest a substantial increase, of approximately 160-340%, in the number of UK re-releases of songs once recording copyright expires, relative to songs of the same age and approximate vintage remaining under copyright protection. This result is consistent with previous findings for books. However, we show that public domain songs by a given artist are less likely to be performed in concert than copyrighted songs by that artist. This finding is consistent with a model in which artists choose to perform (and thus promote) songs that bring the highest sales royalties.
Our findings on digital platform availability further depart from prior results on copyright and availability. We show that the distribution model of the typical streaming platform, with its comprehensive licensing of large catalogs of rights and nearly complete availability of songs for most artists, negates the effects of copyright expiry for any individual song. Popular songs with recording copyrights in the public domain are no more likely to be available on Spotify than songs whose copyrights have yet to expire. However, it is the predominant streaming platform strategy rather than digital distribution per se that explains this difference, because estimates based on the availability of permanent digital downloads from Amazon (an à la carte platform similar to the distribution model for physical releases) show similar effects around copyright expiry to what is observed for physical releases.
Taken together, these results show that copyright has nuanced effects on availability, and can lead to different effects across different distribution channels. They also demonstrate that the impact of copyright on availability differs based on the business model of a digital platform. As the revenue shares, and relative importance, of dissimilar distribution channels shift in an industry in transition, it is important to understand how the impact of intellectual property rights on product availability may change as new channels rise in prominence.