Spotify Wrapped 2022 year end listener statistics just became available. I very much like this summary, finding it a fun way to look at the last year of music. I sometimes find my most played songs surprising, and my most played artists aren't always who I'd have expected going into the year. But every time I start thinking about Spotify and listening time, I end up tilting back to the subject of Spotify revenues and artist payments. I love music, listening much of the day to something, when possible. My own Spotify stats showed I listened longer than 94% of Americans. This doesn't even include my still lengthy Pandora usage nor my (decreasingly relevant) personal music collection listening.
As so much of my daily enjoyment is music, I want Artists paid fairly for their creations so they keep creating. But as part of this, we should make sure payments for listening reflect actual listening...this is where I start finding Spotify and other Streaming revenue sharing...odd.
Spotify, like most streaming services, use a stream-sharing model where they add up all the streams for all the rights-holders, and then distribute to them based upon their share of total streams. There's other articles that cover this, easy to google for, so I encourage you to do this if you are unfamiliar with this model. But this means that if you stream very little, yet pay the same $10 month as everyone else, most of your money goes to artists you may never have listened to.
My top five songs this year were all niche songs with comparatively few global listeners. Some groups I currently enjoy have fewer than 1000 monthly listeners. I had a total of around 58k listening minutes. If songs average 3.2 minutes, this is over 18125 songs streamed. Using an oft-cited average of $0.004 per stream, this means my $10 a month fee payed about $72.50 to rights-holders. Yet this is only 60% of what I paid to spotify. Supposedly they pay nearly 70% of total revenue to rights-holders. But I'm in the top 6% of U.S. listeners. Assuming Spotify makes more from paying users than free-accounts with advertising, shouldn't I expect my streams' estimated payments to be far higher than the Spotify average, perhaps even exceeding my total payments to Spotify? But it isn't. Now take another individual, who streamed around 23k minutes, more than 64% of other Americans. Using the same song duration and per stream estimates we have around $29 of their revenue going to the artists they streamed. Potentially $120/year of fees and only 25% going to who they listened to. One thing this doesn't take into account is that, being well above the majority of users in average listening time, we both likely skew the statistics in favor of our selected artists. But since our top artists or top songs are not themselves necessarily(or in my case, even likely) in the top 50% of songs streamed, this skewing may only help get our favored artists back to par. The situation is far worse for those who stream in the bottom 50% of users *and* listen to obscure music. Under the stream-sharing model most of their money *must* be going to artists they would never listen to. The argument in favor of this weird system seems to be that the more listeners you bring in the more the total pot grows, and so everyone gets paid more in the end, and on average artists are paid similarly for their streams. On Average.
We would never allow this model for any industry where products shipped have actual cost. It would be ludicrous! You're going to build something, consumers will pay a subscription to get it, but the more products your competitor ships, the less you make? To earn more from that common revenue pool you have to ship more product, making even less per-product than before? If your opponent builds a cheaper product and ships far more of them to inexperienced consumers who are ok with mediocrity so long as it comes in a flashy box, you make less still? Again, ludicrous. But that's where our music industry is at.
Unknown niche artists enjoyed by those users who don't listen to the "Today's Top 40 because we're promoting it endlessly" playlists, are only going to see their "fair share" of the revenue if their listeners also keep the music going longer than the average listener. Of course there's no reason to believe that your average niche listener listens to less(or more) music than your average top-40 listener, streaming companies don't actually pay per stream, and we're only aware of the averages, but I don't know how anyone can argue this isn't sketchy. Unfortunately the numbers aren't public to see exactly
what happens here but it all feels very "off". I do know of at least one user who
has many times my total listening time, and was stated as in the top 0.1% of U.S. listeners by time. If my top 6% of listening time can't even get my average pay per stream to reach that average 70% of revenue, does it mean that the top 50%
of streams are actually condensed in the top 5% of users? If so, the bottom 95% of users are potentially all subsidizing artists they may never listen to, and to a huge extent. I'm not arguing anything
underhanded is going on, though it isn't hard to see how this system can be abused by bot-driven fake listeners. But in any case, this is far too opaque of a system for a
very centrally-controlled industry to operate. It may actually be true that the vast majority of rights-holders and their listeners would be better served by excluding these top users from their financial universe. If those heaviest users don't correlate to the listening habits of the 95% of streamers, why should those streamers send their money to artists they don't actually enjoy? I certainly would like to know that when I pay for music, I'm supporting the artists that I want to hear more from.
We have no idea who we're actually paying to support. Sure, users pay for access to the library of songs, and get access to the library of songs. Perhaps you could argue that any other considerations are beyond the users' care as a consumer. But clearly, they're not. We're told not to pirate because we need to support the creators we enjoy. We're told we need long copyright terms to support the creators we enjoy. But then when we actually pay, we don't know if we're actually supporting the creators we enjoy? Please. This is not right. A strict per-stream model would be opposed by users as they expect unlimited play at this point. Perhaps just exposing the private details of these contracts would be enough to keep users informed on whether they're paying the right amount for what they're getting. That would be a nice start in any case.
Aside from that, what could we support instead? If there is an optimally fair and obviously just model, I can't think of it. Part of this is due to what I mentioned earlier: no industry where products shipped have actual cost would have tried what we have with streaming. That's because music, like all copyrighted material, is not actual product. We had to grant Congress an express right to create an artificial "property right". And for the sake of encouraging creativity, it is surely needed(even if it is too long a duration). But there's little sense in saying recorded music should make more money if it's played more. If I wear boots twice a year during heavy rains, that doesn't mean they are only 0.5% as valuable as the street shoes I wear everyday, their makers shouldn't get paid $2.75 while street shoes get $47.25. But that's how we value recorded media. It is nonsensical.
For those that want to argue that "more popular" correlates to higher quality, more skilled musicians, etc, and we certainly do pay higher amounts for higher quality products built by more skilled workers, the lack of per-play "unit cost" and natural time limits users have to search and listen skews any real impact that quality may have on listening decisions. The negligible cost per copy of recorded music means it doesn't play by the same rules as real merchandise, governed by supply, demand, and competition between price and value. It's true that bad songs are not very likely to become popular; but that does not mean unpopular songs are likely to be bad. If 100 great song exist, and everyone has time for 10. Those first few to be hyped as "great"(either by listeners or advertisers), will end up with more listens than the rest of the great songs. If we're in a world of abundant, quality music, listen counts are governed by discoverability and familiarity, not quality. If our goal with copyright is to encourage the creation of more good content, it is counterproductive to use a payment model that doesn't take creation cost, effort, and quality into account. It especially doesn't make sense to detach an individual listeners' payment from their payments to artists.
Since we're clearly going to keep the unlimited models, and I don't know of a way to objectively define quality and value to an individual song without referencing skewed global popularity, it would be nice to know we're using a payment scheme where revenue from an individual customer equates to payment for the product they consume. If Spotify is going to pay 70% of revenue to rights-holders, then 70% of what you as an individual pay should go to the artists you as an individual listen to. If a short-duration listener pays the same $10 as a long duration listener, then they have already declared that these fewer listens have a greater value *to them*. Lacking normal market forces of supply, demand, and quality vs cost considerations, individual user stream-share pricing is the most logical way to gauge value per song. If artists want a piece of the pie you're paying for, they need to make something you individually want and make sure you individually find it.
Due to the tenuous relationship that copyright has with cost and value, and the artificial nature of these copyrights to begin with, we may need Congress to re-balance the playing field so all creators get a sustainable portion of the pie. Keeping in mind we want to do this without turning streaming into a "make work" effort any untalented person can claim a piece of, we should start by dictating that subscription streaming revenues must be transparent to both users and rights-holders. At least then we can know what content our fees are actually funding. Beyond that we need to be more careful with stepping on potentially creative business models while ensuring new or unknown creators are not dissuaded from creating or forced to accept non-competitive pay structures due to the weight of industry monoliths.
Complex and private payment structures, with numerous side contracts between the largest players, with revenue sharing hidden from the paying users who have limited real competition to choose from, are ripe for abuses that harm both the consumers and the creators.A reasonable regulation might be that any "payment pool" scheme be governed by exactly one, public contract with accountability access open to all signatories, and that any side-deals between the platform and other content owners cannot include terms that share aspects in common with that pool. For example, you cannot use 50% of your revenue for a shared pool divided by stream count, but then pay some content providers even more money per stream from another pool of 20%. You would, however, be able to create two pools that have overlapping content but are common to all signatories. For example, 50% of revenue pooled by plays for all content, 20% of revenue pooled by play for content with original release dates newer than 5 years(notice all content owners can fit into both pools equally, but it still can reward "freshness" of content and new creativity). Those "popularity suggests quality" types might object that rights-holders representing the largest names may object to their content being paid the same rate as the more amateur creators. In such a case, it would not be illegal for them to have a side deal where their group gets a fixed portion of revenue(say from 5% of total revenue while 50% goes to the "pool"), only that they cannot base that share on an aspect used to divide the 50% pool(like stream counts, or total minutes). And such a contract would have to be disclosed publicly. In an industry governed by giants, individual content creators should know the rules that impact their value, and users should know exactly who they're funding.
Most performers see the bulk of their recorded-media profits in a short window if they reach mass-market awareness, and the rest of their revenue comes from live performances that are more valuable to them the more their recorded music is shared and their fan-base enlarged. The public is arguably harmed by long copyright terms diverting money for recorded music to the biggest, well established and long profitable names while spreading far too little to the lesser known or new artists who would most benefit from higher "per play" fees. Copyright is about encouraging creation, not blindly creating a few ultra-rich celebrities.
Aside from what we should do(limit copyright terms to a narrow 20 year window, more in line with the original purpose of encouraging creation of new works, rather than a method of locking our shared popular culture behind 120 year long paywalls), the best option would probably be to mandate that while rights-holders maintain a full monopoly over the right to profit from their works, they may only license them on either a strict ownership or universally equivalent per stream basis. They can set whatever price they wish. If they think their songs are worth 10x that of other songs, they can charge it. But it must be knowable to the user how listening to that music is impacting them. Users would choose the streaming platform which they found best suited to their functional needs, and it could access the entirety of the world of music. At this point, it would be up to the platform to help the user choose what makes most sense for them. A good platform helps the user extend their listening budget while finding the best songs for their taste. Users might hate to see unlimited streaming go away, but as they clamor for more content at a fair price, the market system would finally be allowed to work properly on recorded music, something it hasn't been allowed to do pretty much ever in history. Switching to such rules would ensure complex and opaque acquisition agreements cannot hamper the ability of users to know who they're funding or harm smaller creators who lack the clout for fair contracts. Obviously, I don't hold out hope for such a rule to ever come into being, but it's where we should try to head.
No comments:
Post a Comment