The Value of Music

Know Your Enemy

I’ve been thinking about the concept of “devaluation” of music. It’s such an emotional topic, so difficult to grasp. On the one hand, the vast majority of music on platforms has essentially zero economic value, such as the large percentage of tracks on Spotify that are never consumed. On the other hand, individual artist catalogues sell to private equity firms for hundreds of millions or even billions of dollars.

There’s a fashionable quasi-economic argument among tech world people that music has no economic value whatsoever. Zero. I come across these sorts of essays now and again, this one is typical. The TLDR is that - because the writer doesn’t particularly care what music he’s listening to and the music itself is infinitely reproducible in the digital realm - therefore it’s in infinite abundance and, like oxygen, is without economic value. Music to him is just another easily accessible domain for generative AI models to conquer, to produce fungible music more efficiently and effectively than human competitors. This, by the way, is the greatest threat to human creativity in all of history, full stop.

This sort of “no value” argument has been kicking around awhile. It’s an evolution of that old tech bro rallying cry, “information wants to be free.” I don’t entirely disagree, but my own takeaway from the so-called Free Culture Movement is nuanced. I don’t appreciate that news media is so often paywalled and inaccessible as a matter of policy, but I don’t think it’s a good outcome for all media on the Internet to be free. Art and entertainment products, in contrast to factual information, should be free only if the creator or rights-owner decides it should be free. There is a market for creative works because we have intellectual property laws, which are intended to create artificial scarcity. If we weaken the laws to the point art loses its marginal value, it’s a failure for our culture. I believe we should renew and strengthen our commitment to acknowledging value in the creative works that bring joy and meaning to our lives. Renewing and strengthening our laws, that is, specifically to address the threat of generative AI.

As Joe Shithead says in the protest anthem posted above, you got to know who your enemy is. The enemy of creative art is the self-serving, self-fulfilling tech world association that all art and entertainment is fungible and has no value. If information technology is the pipes and arts and entertainment products are the content, devaluing the content increases the value of the pipes. If the content is free, and people want the content, then the pipe-makers can overcharge due to the demand….for the content. Devaluing content benefits the pipe-makers, aka tech bros, aka the enemy of art.

Incentives

Our copyright laws are established in the U.S. Constitution, but the ideas have been around even longer. Article I Section 8 of the Constitution provides as follows: “[Congress has the power] To promote the progress [of the arts] by securing for limited times [to creators] the exclusive right [to their works].” The way I read this clause, our laws grant exclusive rights to creators to promote innovation and incentivize creativity. They permit artificial scarcity so that the works will have economic value. If our laws are effective and consistent with our Constitution, the value should never go to zero like oxygen. If the modes of distribution allow the value to go to zero, then the modes are inconsistent with our values. Our founding document, the source of our shared values, believes we need to create incentives. Therefore incentives for creativity are central to our values so long as our laws are derived from our founding document. In an era of abundance, the challenge is creating some form of scarcity. That’s the point of the law.

Devaluation vs. Revaluation

I feel like I say this a lot in these pages, but I’ll say it again: I’m not anti-tech or anti-innovation. I am pro-innovation so long as we stick to our values. I’m not anti-AI, though I recognize an existential risk. I’ve always been tech-forward in my thinking; however, when the self-serving, money-grubbing, anti-art tech bro narrative conflicts with the values that are codified in our laws, I call bullshit.

Ever since Napster made file sharing technology accessible, thus disrupting our industry’s longstanding business model, there has been a lot of hand wringing about devaluation. By the time the on-demand streaming model emerged, paying fractions of pennies per stream, the music community has cried foul. On its face, this model represented a profound devaluation from, say, $16.99 for a compact disc or $1.29 for a track purchased via the iTunes Store. The move from a buck a track to $0.003 per stream made a lot of people’s heads explode.

In my formative days as a music lawyer, I developed my world view with a broader perspective than simply a lawyer representing copyright owning clients. Before I had a music practice, I was an adjunct law professor. Academics are dispassionate in how they think about these issues, because they typically don’t have clients whose interests are impacted. Academics have the luxury of thinking about issues in a vacuum.

In my teaching days, back in those very different times of the late ‘naughts, I had a go-to class discussion that I found fascinating. I asked my students if they would give up their iPod if they could stream all the music they wanted on their phone. I asked them to assume they could stream anywhere, any time, at CD quality, versus having a collection of MP3s. Almost every student said they preferred the iPod over access. Why? Because they owned the music, it was “theirs.”

But what did they own, really? Some bits n’ bytes, probably stolen, probably generating no revenue or income for the artist or rights holder. The point is they perceived value in their “collection,” which fit in an object smaller than a deck of cards. Now that every single one of those then-20-somethings, now 40-somethings has moved on to the access model and pays a streaming service, I doubt they could even locate their precious device they hoped to have forever.

I was advocating for the access model, because I believed the access model represented the only sane path forward for our industry. Music was already more or less free on the Internet, I thought it would make sense to paywall access to music so that it would generate revenue. In general, the music business resisted this evolution and it took the tech bros - in Sweden of all places, home of IP theft aggregator The Pirate Bay - to provide the solution. Just like when Apple showed us the way to create a market for digital files with the iTunes Store and the iPod. Were they the enemy?

No, not exactly. But they weren’t innovating for the purpose of establishing value in music. In the case of Apple, it was to sell devices. In the case of Spotify, they have demonstrated in a variety of ways that they don’t really care that much about music or musicians - note the pivot to podcasts and audiobooks (potentially more lucrative content with a lesser royalty burden) and establishing through their business model that a significant percentage of the music on platform is non-royalty bearing. Like in the case of generative AI, music provided an opportunity for innovation and ROI.

The Value of Music Today and Tomorrow

The access model has resulted in a literal devaluation of music, of course. It’s also resulted in a revaluation of music. Streaming has brought stability to our industry, freedom from the disruption that resulted from innovation over more than a century - from piano rolls to wax cylinders to records to radio to television, cassettes, CDs, CDRs, MP3s, and social media. The access model has brought stability so that we can predict the value of evergreen copyrights over decades. Music is an attractive asset class because those elite copyrights that have evergreen value can be bought and sold, not as copies but as entire bundles of rights under copyright. If an artist is able to grow a significant and sustainable streaming audience for a music copyright - recording or song - they likely will have the option to sell their copyright for a price that is close to the actual market value of the song based on the projected streaming market. That’s new, and it’s good for artists.

The downside of the access model is the perception, the general availability of free music that leads to the tech bro conclusion that music is devoid of value. The abundance of free music on the Internet has created the opportunity for the worst actors - the Sunos and Udios of the AI world - to simply take what they need to aggregate training datasets an charge customers to create original music that is entirely derived from existing music copyrights. This is the real devaluation, the most visceral statement that music has no value.

We’re at a point where music not only has value, but it is more easily measured than ever before, due in part to the stability of the streaming model. If you talk to a lifelong music fan and musician such as myself, we’ll tell you that original music is an incalculable value-add in our lives. But I want to be very clear about this point because we are at a point of existential crisis. Anyone who argues that it is fair use under copyright law to use protected music without permission in an AI training set to create derivative works is making the statement that music has no value. This is the action that follows the shitty opinion of that tech bro. For anyone who values music, who perceives value in music, and would be harmed by the outcome if the incentives to create went away - know your enemy.