Hey, remember this?
I post this not because I am dreaming of quitting my job (or, more accurately, one of my million or so jobs), but because guess who got a subscription to Nielsen SoundScan? That’s right- this nerd. Basically, what this means is that I can look up physical and digital sales by week for just about any audio track, single, or album. You know you’re jealous, and suggestions for how I can do research with this are most certainly welcome. (I do have a current project that I am using the data for, so I didn’t entirely just get it for shits and giggles.) At the time this video was posted on YouTube, there were a number of news reports that pointed out that the song used- “Gone” by Kanye West- hit the Billboard Hot 100 for the first time shortly after the release of the video, despite having been released all the way back in 2005. That’s interesting and all, but we can take it a step further and see what that means for the trajectory of (digital) sales:
Well then…that’s not exactly what I would call a subtle effect. It’s interesting to me that the effect is not only so immediate but also so ephemeral. (For context, the video was posted on September 28, 2013.) Looking at the graph at a different scale, however, gives a bit of a different interpretation:
Looking at this closer view, it does seem like there was a bit of a sustained impact, and, looking at the raw numbers (which I can’t show you), sales in 2013 had averaged about 75 per week prior to the release of the video, and sales for the last year (i.e. April 2014-April 2015) averaged a little over 100 per week. This makes intuitive sense, both because people continue to view the video even though it’s not as popular as it was in October 2013 and also because the initial spike in sales likely led to people playing the song in front of other people, setting off a bit of a chain reaction in terms of recognition and, as a follow on, sales. (Never underestimate the power of the mere-exposure effect!)
This analysis is important not only for the purposes of interesting cocktail party conversation (though such benefits shouldn’t be dismissed entirely) but also because it leads to a conversation about the economics of copyright law and raises some questions as to whether current legislation is actually maximizing value for the creators and consumers of recorded music. At present, copyright law is basically of the “use my stuff, get sued” form unless said use falls under what is known as fair use doctrine. The “correct” way to use a recording in a video is to get a sync license, which, if you follow that link you will see is not the least complicated or expensive of processes. (Fun fact: songs have two copyrights, one for the recording and one for the composition, so someone who wants to use the recording has two potentially distinct creators to deal with.) In all likelihood, therefore, it’s unlikely that the creator of the video would have gone through this channel and done things “right,”* so the video creator has the option of either acting illegally and risking punishment or not using the song. In a world where Kanye West (and his label) benefits from the use of the song, his incentives are not aligned with those of copyright law in this situation.
Copyright law, as written, applies well to situations where, let’s say, I make and sell photocopies of Greg Mankiw’s favorite textbook. In this case, my activity almost definitely takes away from Greg’s profit, and, more importantly from an efficiency perspective, decreases Greg’s incentive to come out with new editions of his book. In contrast, copyright law doesn’t work so well from an efficiency standpoint when the use of a copyrighted work benefits both the user and the creator. In these cases, it’s not clear in which direction, if any, payment should go, since, for example, it would both make sense for the video creator to pay Kanye West but also for Kanye West to pay the video creator in order to get his music placed and reap the sales benefits. If the goal is to create incentives to produce, the direction of payment should be towards the party who needs more of a kick in the pants to keep creating output (i.e. has higher elasticity of output), and more research would need to be done in order to figure out which party this is.
The problem gets even more complicated when you consider that it’s not clear when use of a copyrighted work is beneficial to the copyright holder and when it’s detrimental. Given the number of counterintuitive outcomes we see, it’s important to conduct the proper research in order to gain objective insight into the matter.
tl;dr: The music business is a lot harder than you’d think.
* If you look at the YouTube page for the video, you will notice that the YouTube system provides a place for the content creator to give credit for the music used. In addition, YouTube enables the music copyright holders to scan videos for their content and collect ad revenue from such content. This is a step in the right direction but is still suboptimal if it turns all revenue from the video over to the creator of the background music.