Rethinking Music: What The Industry Could Learn From Netflix

It seems fair to say that no one in the music business right now is happy with how it’s being run. As streaming, including both paid and ad-supported, has replaced CD sales as the industry’s main economic engine, the record companies have seen gross revenue decline sharply, artists and songwriters have seen their royalty income diminished, and the companies doing the streaming are losing so much money they’re losing the ability to raise more of it.

In an interesting thought experiment at the Future of Music Policy Summit in Washington this week, musician and CEO of touring van rental service Bandago Sharky Laguana, considered how one component of the industry’s current business model — how subscription revenue Music_Festivalfrom paid streaming services is ultimately allocated to individual artists — might be made more fair, if not necessarily more lucrative.

In very broad strokes, of the $10 a month most subscription services charge consumers, the streaming service keeps $3 (30 percent) and $7 (70 percent) is paid out in royalties (theoretically to artists and songwriters but in practical terms to labels and publishers who are supposed to then distribute them). The portion of that $7 accruing to any one label is calculated based on how many times songs recorded by any of the artists under contract to the label are streamed by subscribers, typically resulting in a per-stream value of a fraction of a penny.

Now suppose you have two bands, and there are 10 people subscribing to a service. Band A has a few tracks out that are broadly popular and are streamed by seven of the 10 subscribers during the month. If each of those seven streams an average of 100 tracks a month, that’s 700 total streams, of which, say, 100 were of songs by Band A. So Band A is credited with 100 plays when the pool of money available for paying royalties is divided up (I’m paraphrasing Laguana’s presentation).

Band B has a smaller fan base, and its songs are streamed by only 3 subscribers in the month. But Laguana then asks us to assume that one of those subscribers is a super user, and streams 500 tracks in the month, of which 200 are by Band B. The other two stream the industry-average 100 tracks each, for 200 total streams, of which 14 are by Band B. When it comes time to divide up the royalty pool, Band B is credited with 214 plays.

On its face, that seems fair enough. The more your music is played the more you get paid.

The problem with that scenario, however, Laguana argues, is that Band A, with its broad popularity and larger fan base, is responsible for generating more of the subscription revenue from which royalties are ultimately paid than is Band B with its smaller fan base. In effect, fans of Band A are subsidizing Band B, and Band B is in part free-riding on the revenue generated by Band A.

Worse, Laguana argues, the message the system sends artists is: don’t get fans, get clicks, because royalty payments are based on clicks, not fans. That creates an environment ripe for click fraud, in which tracks are clicked on, often by non-human bots, despite no one actually listening to them.

A fairer way to do it, Laguana proposes, would be to base royalty calculations on the number of listeners an artist has, rather than on the number of times their tracks are streamed. Under that scenario, Band A would be paid more than Band B because it had more, albeit less avid, listeners.

A per-user system, Laguana maintains, would more fairly reflect artists’ respective contributions to subscription revenue — and thus the royalty pool — would eliminate the incentive for fraud and incentivize bands to focus on building their fan base rather than driving clicks.

Others speakers at the Future of Music conference pointed to possible flaws in Laguana’s proposal. A per-user allocation of royalties could end up offering artists even less transparency than they have now under the per-click system, for instance, because they would have no way of knowing how many other artists were being listened to and, thus, no objective standard for how much they should be getting paid.

A per-user system might also leave indie and local labels and artists with small, but avid followings worse off than under the current system because their small numbers would be lost amid Taylor Swift’s legions. There may be no perfect system.

But Laguana’s thought experiment does throw light on a deeper disconnect at the heart of the industry’s business model. For better or worse, the music business is becoming an all-you-can-eat affair, whether the buffet is being monetized through advertising, paid subscription or is bundled with some other product or service.

In any all-you-can-eat business, the total amount of money in the system is a function of the size of the total user base and average revenue per-user (ARPU), not of the frequency of use. There’s a reason Netflix has long refused to disclose “ratings” data or a tally of how many people watch “House of Cards.” Those data are ultimately not relevant to Netflix’s revenue model but could influence its dealings with vendors and partners in ways that could misalign the incentives in its cost base (i.e. its licensing deals) with the strategic goal of growing the user base and increasing ARPU.

When Netflix first began streaming movies, in fact, the studios that had been selling it DVDs (either outright or by revenue sharing) sought licensing deals pegged to the number of times their movies were being streamed — that is, to unit sales. The studios eventually figured out, however, that Netflix’s all-you-can-eat subscription model meant Netflix was getting paid whether or not anyone watched a particular studio’s movies. As Netflix’s user base grew, moreover, its revenue was increasing even as the studios’ per-stream licensing fees did not. That is, Netflix was capturing 100 percent of the upside to which the studios’ content contributed.

Eventually, the studios began to structure their licensing deals such that their fees increased as Netflix’s user-base increased. That is, the studios aligned their own interests with the incentives of Netflix’s all-you-can eat model.

The music industry, however, remains mired in business practices, licensing models and a royalty system premised on maximizing and rewarding unit sales — or streams — rather than on maximizing the user base and ARPU. The industry’s intense, at times obsessive, strategic and legal focus on accounting for every unit sale and the precise apportioning of the spoils is at best and exercise in wheel-spinning. At worst, it’s robbing the industry of strategic focus, and increasingly now the capital, it needs to invest in increasing the user base and driving up ARPU.

Changing those practices in the music business won’t be as easy as it was for the studios, unfortunately. Much of the licensing and royalty system in the music business is governed by statute or by court orders, and would require an act of Congress or new legal rulings to change. But coming to grips with where incentives lie and where the real upside is in an all-you-can-eat world requires no more than a thought experiment.