Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Royalties from Writing a Hit Song with Justin Bieber (digitalmusicnews.com)
156 points by microtherion on Oct 6, 2017 | hide | past | favorite | 156 comments


If each radio spin at 347,820 spins hit 10,000 radio listeners average (a Denver hit station has about ~250,000 weekly listeners, a popular rush hour/drive time show with mixed talk/music format has ~80,000 listeners) , then the $53,000 was at a rate of 0.00001524 dollars per listener.

If Pandora paid that, his royalty would have been $582.55...

So it looks like the industry from a streaming point of view assumes about 20,000 radio listeners per play on the radio.

This might even average correctly across the US market.


Another difference is that most of those 20k people did not ask to hear this song while every (well, more or less) streamer has.


Pandora has to pay more if it was requested.


but we all know that the whole pandora business is exactly to get around that.

you get to hear the song you want, and a few similar ones, in a guise that it worked like radio and you didnt select any song, so they can keep the royalties down.

...that said, i'd be very curious what's YouTube excuse. if they need any and are not just shoving their weight around. for tv series and cartoons they are using the "but thats our users streaming content live! we will punish them, after suggesting their live stream to everyone and collect ads revenue" excuse


The per-stream rates between radio & interactive has narrowed significantly in the last 10 years so it's only a 10-15% difference.


Yeah, radio is one->many, YT and Pandora are (usually) 1-1. Totally different schemes...


Aren't royalties paid per play (i.e. spin)? The rate for the radio is about $0.14 per play (53/347), in that case. Stations have no way of knowing how many listeners actually heard the song. They can estimate with ratings services, but not get exact numbers like say Spotify.


Right. But the point is that it's sort of weird to directly compare album sales, radio plays, and online streaming view count. Those are fundamentally different both in terms of audience size and income potential.


Right - the point though is that playing a song on the radio and playing a song as a user of a streaming device have wildly different audience sizes (other than maybe for my 2am college radio show) and should pay very different rates.


I think you meant $0.0014 per play...


> How will I be able to support my gift?

In the end what we are seeing is that that golden era of creators making a living (and much much more) was an anomaly. Control of new types of distribution created scarcity and new forms of expression and shifting norms jacked up demand. But now distribution is free, instant and global. Demand has flatlined and production has exploded.

The hard question a creator needs to ask themselves now is not "how do I get paid for what my art is worth?", but instead "was my art ever worth much to begin with?"


If even people reaching a truly stupidly huge number of people don't get even close to making a living (I get more than that off a few hundred people on Patreon for the work that I do), in a situation where those operating the streaming services are renting out entire floors of office buildings in Manhattan, your question is disingenuous.

Just because some of us are eking out a sort of existence in the new era doesn't make it a functional system.

I would be perfectly happy to see it become 'this is the Star Trek future, your entire compensation is the satisfaction of a job well done and being personally liked for what you've created', if our world operated on some sort of UBI and we didn't have to have our performance linked to survival at all.

Under those circumstances, things naturally shift down the Maslow hierarchy of needs, and people take for granted that they'll live, and get desperate for validation and popularity, that being what they lack. I have NO problem with a world that completely unlinks money from performance, because I'm not a free-market capitalist type demanding that financial reward behave like a 'meritocracy'.

Self-evidently it doesn't anyway, so literally nothing of value is lost.

BUT, we have that completely decoupled world and yet demand that people pay for even the most frugal living with money they got somebody to give them, in a world where people simply can't and don't do that anymore.

Silicon Valley clever-boffins: come up with some disruptive alternative really fast. We don't have time to mess around, and what these songwriters face is the same fate waiting for all of us, in turn, including you. How about abolishing money and replacing it entirely with 'likes'?


The problem is the middlemen and the way the revenue is distributed through them.

Someone who produces and uploads their own video that reaches a stupidly huge number of people will make bank guaranteed, just like someone who is reaching their audience directly through Patreon.

On average you should be able to get $5-$10 per 1000 YouTube video views if it is on your own monetized channel. So that 35 million views probably generated around $245,000 dollars for someone.

Probably most of it ended up in the hands of the record label that owns the YouTube account, and by the time they finished taking their cut the final $200 made it back to the song writer.

My point is that new methods of distribution flattens the distribution and gives you the potential to make much more money if you spend the time to build your own thing rather than going with a record label.


>On average you should be able to get $5-$10 per 1000 YouTube video views if it is on your own monetized channel. So that 35 million views probably generated around $245,000 dollars for someone.

Not disputing your numbers but just putting it out there that Hank Green estimated $2-per-1000.[1] That calculates 35m views to be ~$70k.

Google also recently changed its engagement algorithms and everybody's views and monetization went way down. I'm not sure what the latest 2017 cpm estimates would be. That Hank Green estimate was from 2015.

[1] https://medium.com/@hankgreen/the-1-000-cpm-f92717506a4b


From what I heard it is/was $5-$10 per 1000 if your video is eligible for the high value video pre-roll ads. If you only get text AdSense ads then it is about $1 per 1000.

Either way, my point is that a 35 million view YouTube video generates a lot more than $200 dollars for someone. Even if you take the minimum of $1 per 1000 you are still looking at $35,000 of ad revenue. That ad money must be going somewhere, even if creators don't get it. So YouTube isn't to blame, its the people who are skimming off so much money that only $200 is left for the writer


One data point, my wife's channel has averaged $1.14 per 1000 views in 2017.


No, absolutely not. The power is with record labels and streaming services. You seem to be assuming that there's such a thing as a 'hit song' quality resident within the song, or a 'hit game' quality resident within the game, and that these are merely facilitated by the distribution channels: that's not how it works at all.

The hit things are a mechanism OF the distribution channel, which is how the 'hit' scales to those levels. If you go 'new methods' it only guarantees that you will never, ever have a hit thing. You might well have a culturally important or creatively valuable thing: they say the Velvet Underground spawned a thousand bands. But you simply don't reach the stupidly huge number of people unless you're part of the mechanism that's taking all the percentage out of that profit.

There's no such thing as that flattening of distribution, truly there isn't. There's no such potential. It's just tighter control of the channels by the record labels, media companies etc. and it's all middlemen, running the show. New media didn't make that better, it's substantially worse than it was.


Probably most of it ended up in the hands of the record label that owns the YouTube account, and by the time they finished taking their cut the final $200 made it back to the song writer.

But the video will have more 'creators' than a song. I.e. song writer should not be getting 20% royalties of the video spin.


Yes, but YouTube is itself a middleman at this point. Possibly a generous one or not but still a middleman who can vary the size of their cut.

Only competion can alter that very much.


Simply reaching huge numbers of people in and of itself isn't necessarily valuable.

They may listen to what you've produced and enjoy it. But are they paying to listen? Did they buy the track or sign up for a subscription? Are advertisers paying to be featured alongside the song, or to use it in their commercials?

If not, maybe your song isn't so valuable. If those listeners could easily listen to some other song for free, then maybe your song isn't so valuable.

And this is talking about the whole song. If you're just the writer, it's even more stark. How good of a writer do you need to be to write a hit Justin Bieber song? If you didn't exist, how easy would it be to find another writer who could write a hit for Justin or some other star?

I would guess that what's really bringing money to the table is Justin's star power, not the uniquely amazing abilities of a single writer. So 20% seems generous.


To your last point, it's not quite like that. What happens is the Andre Lindalls of the world re cord demos. The artists listen To the demos and more or less "bid" on them. Top tier artists get top tier songs. Or rather they get songs they think will be hits.

So there is a hierarchy among songwriters even for pop stars, it's definitely not a vast ZMP pool.


> abolishing money

Communes have been tried, voluntary ones and forcible ones, over and over. They never work out. People just aren't wired up that way.

> Star Trek

is an imaginary utopia. You can always make things work when writing a script.


I know of several Communes that worked just fine for 20+ years. What's interesting is not that they die off, but that the form so naturally. To survive they need to be like collage fraternity's and focus on new membership regularly, but naturally occurring Communes don't have that drive out of thin air.

Monastic orders for example can be stable over hundreds of years, but to pull that off they need a focus on new membership. They fail not when the last member dies, but when nobody signed on 20 years before then.

PS: There are ideological reasons that people oppose communes, and many had significant issues. However, the jump from 6 young people sharing rent, to 20 young people living together is not that difficult. And at 20 people economy of scale kicks in where buying in bulk saves money but meticulous record keeping is not worth the effort.


Another thread suggested I check out a long lived commune in the US, so I did. It turns out that it was indeed long lived, but the average member only stayed a year and a half before moving on.

The larger point is, in America, anyone can form a commune or choose to live in one. There's no law against it. What you don't get to do is force people to join and prevent them from leaving.

There seem to be orders of magnitude more people advocating communes than are willing to live in them.


Interesting, I would simply be cautious that publicity is not sought out by many communities. Also, people can have significant financial incentives for these places to break apart so communal property may be sold off.

PS: I am not personally interested in living on one. However, I do know people who have for ~30 years and see a host of upsides and downsides.


> people can have significant financial incentives

Of course. People who believe they will be better off leave the commune. Those tend to be the highest producers. What's left are the low producers, and that can't sustain the commune, so it collapses.

The next obvious step is to prevent the high producers from leaving, and history shows how that plays out.


I think you are assuming people make their money inside the community. That's far less efficient than having an outside job and enjoying significantly lower cost of living and a low effort social life. Rent, food, house supplies, utilities, cable etc for ~500-600$/month/person + regular parties etc.

It's often the low producers without significant retirement savings who have larger incentives. ~100k can be a lot of money or almost meaningless very easily.


You obviously are not convinced, so I suggest you join a commune and let us know how that works.


I am speaking from personal experience, so you can not believe me, but that's about it.


Respectfully I disagree with your assessments.

Abolishing money does not need to exist only in a commune format; currency is a mode of exchange that exists because there is a natural scarcity of resources. Currency allows us all a medium to acquire what we want; usually as a result of personal investment of some source.

Star Trek, while imaginary, is a technological utopia and I would argue that we are not that far off from the beginning of that type of society. Since we're on HN i'll assume you're familiar with Kurzweil's Law of Accelerating Returns and the consistent and predictable exponential tech growth. As we make advances in energy, nanotech, robotics, and genetics the problems that plague society today won't be relevant to the people of tomorrow. If we aren't working towards Utopian society as a collective then the fault is ours, because working towards anything else is just a failure of imagination.

Of course this assumes the political will and foresight to address the challenges and growing pains that come along with the transition.


People will always want more. It's just the way we are.

Consider the state of the poor in the US today. It is far, far better by any measure than the state of the poor in medieval times, Roman times, etc. Yet the conventional opinion is it is not remotely good enough.

By today's standards, the Star Trek universe is a utopia. But by the standards of the inhabitants of that universe, they may regard the state of their poor in crisis terms.


>Communes have been tried, voluntary ones and forcible ones, over and over. They never work out. People just aren't wired up that way.

Communes from the 60's still exist.

https://en.wikipedia.org/wiki/Hog_Farm


"Members have their own jobs and bank accounts,"

https://www.voanews.com/a/a-13-a-2001-08-07-25-hog/395182.ht...


> If even people reaching a truly stupidly huge number of people don't get even close to making a living (I get more than that off a few hundred people on Patreon for the work that I do), in a situation where those operating the streaming services are renting out entire floors of office buildings in Manhattan, your question is disingenuous.

Seems like a weird detail to fixate on. Plenty more money has been spent on office space (whole buildings, even) by music industry and music industry adjacent companies that acted as gatekeepers and ensured only a hand-selected few ever made a living off their songs.

> I have NO problem with a world that completely unlinks money from performance, because I'm not a free-market capitalist type demanding that financial reward behave like a 'meritocracy'.

I also have no problem with this, but I think creative people won't be sharing our opinion anytime soon.


I think the point is the middlemen are clearly rolling in revenue, while the content creator is barely scraping by. That suggests the content really is valuable, and the profits just aren't making it to the artist for some reason.

(unless you believe marketing & distribution is all of the value)


In the end what we are seeing is that that golden era of creators making a living (and much much more) was an anomaly.

Historically, musicians and actors were very poorly paid. For a brief period in the 20th century, some of them made a lot of money. That's over.

There were, at peak, several million Myspace bands. Some of which didn't suck. Supply of musicians far exceeds demand.


Myspace was great for smaller bands. It made it very easy to connect with like-minded artists and set up tours and shows on your own.


Clearly, it did not work for myspace.


The question of worth changes as time goes by and technology changes. When your only option for consuming a new song was to either wait for it to play on the radio or go buy the album, the value was much higher because supply was controlled by a few and consequently in short supply, so pricing power was much higher.

Now most of the power lies with the consumer who will typically consume whatever artists choose to offer their creations for close to free. There are some artists like Garth Brooks and formerly Taylor Swift, who wield enough power through enormous demand for their songs to limit supply and charge higher amounts, but these are the exceptions.

What's ironic about this article to me is that Justin Bieber would likely not have ever become famous in the first place if not for YouTube because that's where he was discovered. The same goes for a large percentage of today's most successful artists. So previously the record labels wielded all the power and only the chosen few who they deemed worthy made large sums of money. Now pretty much anybody can put their stuff out there but the consequence is that they can't control the supply.


Turns out writers weren't being paid for their songs, but were being paid for a portion of the massive markup on a plastic disk. Now that the disks are gone, the value of the song is laid bare. The same has largely become true for software. How many new Lotus's or Corels are being funded today?


Yup. The value of software has been driven literally to $0. Yet I still keep writing software :-)


Me too. But I write it for the web, so I get paid for the massive markup on web servers running the code.


I'm not sure if I'm missing sarcasm here; if I am I apologize.

But... this is obviously not how the world works. In the records and CDs era revenue was proportional to disks shuffled around. It is absolutely not the case that a typical web business has revenue proportional to the number of servers they employ.

The important point here is what economists call "trapping the value" and "pricing power": a CD full of music has lots of value, but the person selling the physical item was at one point the person controlling scarcity, and thus had pricing power, so they trapped most of the value.

Nowadays web business models allow developers to trap the value of their software more effectively. That doesn't mean they are piggybacking on the markup on web servers.


I meant we markup the computing power of the server (not exactly the purchase price). My host rents me a server that can perform X calculations for $Y/hour. I divide and resell that computation power, plus my software, to customers at Z markup.


You are confusing "distribution" with "supply". Having lots of people making music is a different issue from having music freely available via youtube. Music has been pretty darn cheap to reproduce for decades.

Yes, availability of substitutes can diminish bargaining power of music rights holders, but I'm not convinced that's at all how we should interpret these numbers. I think this is more about market power and network effects of the middlemen (google, pandora, the labels, etc...)

I would be interested in the revenue split between rights holders and YouTube.


> Music has been pretty darn cheap to reproduce for decades.

in a sense yes, but it depends on what you refer to by 'cheap'. other factors like convenience, availability, quality, storage, etc. were often quite different.

it's one thing to wait around for your song and then tape if off an FM radio broadcast, another to just download some 'lossless'-encoded bits off the web at will.


> Music has been pretty darn cheap to reproduce for decades.

"Cheap" is relative. It is relatively "cheap" to copy tape or Vinyl. Now, we have digital files. Copying them is perfect and free. "Cheap" is infinitely more expensive than free.


> But now distribution is free

That is laughable. Middle-men like YouTube (Google), record companies and other channels take most of your money. I wouldn't call that free.

(Discoverability is part of distribution)


Free as in cost-- the cost is nearly free.

The point is--- if the cost is nearly free, why not do it yourself? You're right that youtube/google/spotify are going to take a cut for distribution, but you really don't need the record companies. And if your music is good, and you already have a following, you can just sell it on Bandcamp.


>The hard question a creator needs to ask themselves now is not "how do I get paid for what my art is worth?", but instead "was my art ever worth much to begin with?" I disagree. The hard questions for creators of anything is how to control sales of the product. The digital platforms have failed musicians. The challenge is how to enforce prevention of piracy (ie illegal uploading of material). This will ultimately be legal hardball, and what 'record companies' used to do to protect their label and artist roster. Someone needs ot take on the big digital platforms in a landmark case. Regarding creativity, a tangible asset (like a box and booklet containing the music in some new format for example) would go a long way to making people think they had actually bought 'something' rather than just listening to an fM radio like stream of tunes. Art is rarely worth very much at the time of creation, but can subsequently explode in future popularity. Creators need robust protection mechanisms in place to help them keep creating, and that is legal precidents.


I'm not too worried about people pirating my music. It's already up on Spotify and Apple Music legally and I make no money from that. Most people I know in smaller bands are not in it for the money and support themselves through other jobs. I'm happy as long as enough people turn out to shows and have a good time and we make enough money to help out a bit with practice space rent or gas and food while on tour. People who are into it are supportive and do buy tangible assets such as records or tapes at shows or order them online. There are enough people out there who'd rather pay for a record they can hold, look at the artwork and listen to as a complete album instead of just skipping through some mp3 files or hearing a random song in a stream. I'd say that people coming out to shows and buying records and merch is what helps us keep creating.


What? Piracy isn’t the evil that it used to be 15-20 years ago where people would rip cds and tapes.

Nowadays everyones music is hosted and streamed legally on youtube/spotify/play music/any generic music service. The days of people paying $20 for an album cd just to listen to the one hit they like are long gone.

If artists trust their popularity then they can pull their music from streaming services, indirectly forcing fans to purchase their albums (like taylor swift did a while ago). But even this isnt an option for smaller artists, who would just disappear into obscurity because of how over saturated the industry is.


> The days of people paying $20 for an album

anecdata, but I see a lot of people (myself included) buying $10 CD's and $20 vinyl with a download code at shows in small venues.


What parent said holds up if you include the other half of that sentence.


"But now distribution is free, instant and global" The only word in this phrase that's new is global. You could always go out on the corner and "instantly distribute" your music for free. What we are seeing in music is the same as we are seeing in the retail industry with Amazon. Those who own the distribution channel create the winners and the losers. With certain distributers having so much leverage they take power away from the creators. It becomes a race to the bottom.


You are making a semantic argument. I apologize for not being more precise with my language.

Going out on the street to sell your music is hardly meaningful distribution at all compared to putting up a Bandcamp with your music for sale. One has the upside of maybe a few hundred people in a day, the other has billions of Internet-connected consumers.

In the past, the distribution channels were owned by radio conglomerates and retail conglomerates and distributors/publishers/marketing/etc. Basically the music industrial complex. And it was very hard to penetrate that complex with your art and have anyone consume it, much less pay for it. And they picked winners and losers capriciously.

Now you can put your music on YouTube, Spotify, Bandcamp, iTunes, etc. If one of those won't take you, there are others that will. And you can promote a song on Reddit and it can go to the front page and you can have millions of people listening to your music tomorrow. That's what I mean by distribution.


Is this when your being held upside down out of the window over Denmark street (londons tin pan alley)?

BTW the father of one reality show host is alledged to have done this to a famous British musician back in the day.


I thought that was Suge Knight's move with Vanilla Ice.

Music business can be rough: https://www.youtube.com/watch?v=f3zesMAuJrM


Apocryphally it was Don Arden (Sharon Osbourne's Father) did this to Tommy Steel


Get with the times! The way people consume music has radically changed. The music gold rush is over. The simple fact is people listen to music on YouTube and Spotify now. Music is also absurdly high in supply because everyone wants to be a musician. Guess what that does to the value? That's right, music is cheap because it isn't worth much.

Whenever I see people complain about this, I just feel sorry for them. The world changes, and there are winners and losers because of the change.

It's so strange to me how people think the world owes them something. The value of objectively looking at the market is so underrated.


Completely agreed. I get that it's difficult to look at things objectively when it's your paycheck on the chopping block, but that's really not the point.

I'm glad to be in a growing field that pays really well. But I'm also working to save a ton and invest wisely so that if my skills are suddenly not in demand anymore, I won't have to worry about being out on the street. I'm definitely lucky that a confluence of factors came into play that allow me to do this, but anyone and everyone should be thinking about how to hedge against employment issues down the line. That's just basic financial responsibility.

If you want to be a musician, you are already putting yourself into a difficult spot financially. If you're ok with that, more power to you... but you'd better figure out how to take care of your basic needs, at least, and you'd better have a plan for downturns since you're already living on thin margins.


On the other hand there is money in music, and it could and should increase. When you have 1 billion people paying $10 / month you have $12 billion per year to split between the artists, songwriters, record companies and distributors. That might still be lower than before but it is some money.


$120 billion per year, you missed a factor of 10.


Fortunately for Hacker News, most of that economic windfall goes to middlemen.


Unfortunately, it goes to the labels. The streaming services aren't rolling in the dough.


It also goes to the engineers working on the services. $200,000-$300,000/year to a senior, even if the service itself is losing money (And when it goes bust, you can always find another job.)


I feel like this irony is horribly lost on about half of the commenters here. Creator's job is seen as worthless, but the deliveryman's job value is indisputable.

And yet if Spotify were to pay its engineers scraps, I doubt the voices here would be as supportive.

The only reason the engineers are paid as much is because the companies can't get away with paying less. And just because they can get away with paying musicians less, doesn't mean the musicians should "get with the times". I am not saying every musician should be a millionaire, but there are plenty who collectively carry the multi-billion dollar industry on their backs, but get scraps in return.

On a more general note, when a vulnerable part of the population gets shafted, they shouldn't "get with the times" either. It's a sign of our failure as a society.

I guess it is not surprising that people in a capitalistic society think that the value of an activity lies in how much profit it can generate to the performer of that activity. I guess it's beyond reasonable to expect people to think of externalities and understand them. I guess it's not at all surprising that a bunch of people who claim to be in STEM have a collective amnesia when it comes to understanding that M in STEM is an art akin to music, but with laughable ROI and a far more limited audience (and the number of corporations that hire mathematicians to do mathematics in the entire world is probably... 5? Go on ArXiV and see how many math papers there are by people outside academia, i.e. are the product of the "free" market system).

None of that is surprising. But that doesn't make victim-blaming any less tiring to see.


Actually, the musicians don't carry the industry. Even if Taylor Swift died tomorrow, and all of her music was erased from history, the music industry wouldn't be threatened even slightly. There is a virtually endless supply of musicians.

Society hasn't failed either. Society is made up of individuals. If the individual wants to stream music instead of buying a CD, and if that means less money for the musician, then the only choice is to adapt as a musician or "get a job".

I say all of this as a musician myself. I realized all of this the hard way. Just about everything in the world that can be bought or sold is valued based on perception, which means anything can be worth millions tomorrow or squat. Nothing is permanent. We're all gamblers in a huge game of perceived value, and technology only makes it more crazy.


Victim blaming? These people aren't victims. They're people who picked a profession where it is (and always has been) difficult to find success, both in terms of popularity and material wealth.

I don't see the creator's job as worthless, but I do think I see it as more of a commodity than many creators would like. The biggest hits in music today aren't there because they're somehow objectively the best at their craft. Don't get me wrong, they do have talent, but their success is largely driven by promotion and marketing. There are a ton of largely-unknown artists who could see just as much success if only they'd been able to get their material in front of someone with influence and reach in a better way than the current incumbent successes have done.

> On a more general note, when a vulnerable part of the population gets shafted, they shouldn't "get with the times" either. It's a sign of our failure as a society.

I'm very sympathetic toward people who are actually in vulnerable places. People who can't kick drug addictions, people who are homeless, people who just made one bad decision or just had enough bad luck to end up in a bad place. Creative types are not that. They've chosen to follow their passions to the place they are now. Even older artists simply can't remember a time when their craft was a "gold mine", because that's never been the case.

> I guess it is not surprising that people in a capitalistic society think that the value of an activity lies in how much profit it can generate to the performer of that activity.

I think you give people too little credit. I think people understand that an activity can be very valuable to society but not be lucrative at all. Grade-school teachers are the prototype example of this that pretty much everyone has heard before.

I don't think many people would argue that music and art have low value to society. They're certainly not essentials like housing, clothing, and food. They're not going to push the sum of human understanding further like scientific research. But they're certainly more valuable than some bullshit apps that we talk about here, things like "Yo" that attracted an obscene amount of investment for what I'd consider very little value.

But I feel like you contradict yourself with this point you've made. If you do manage to separate the value of an activity from the profit generated for the performers, you can easily make the argument that no one "deserves" to get paid for creative output. Just because something has non-monetary value, it doesn't mean people will want to pay a lot for it.


Everyone knows that songwriters do not make their money off royalties. They make their money off charging pop stars hourly to write music with them.

The same way performing artists don't make money off their albums, they make money by touring and increasing their brand reach as an artist.

Music industry is not about creating IP, the same way the tech industry is not about creating IP. Making things and making money are two separate, yet intertwined tasks.


> is not about creating IP

I wish more people would come to terms with that. Protecting "IP" should not be the focus, but for many it is, so we have Copyright, DRM, etc.


My grandfather had a single hit song over 50 years ago and he still lives comfortably off accumulated royalties and other proceeds in 2017. Those days are long gone. In this current system artists can still become millionaires via concert and show money. It's the writers and producers who are getting shafted since they depend mostly on royalities and points.

What I find intriguing are the genres where producers are arguably MORE important than the artist (think trap music). Some producers in that space are circumventing this problem by starting their own labels and signing artists under them. This is essentially what Mike Will Made It has done with Rae Sremmurd. As the music industry evolves, people participating in it have to continually think outside the box and innovate to maintain revenue streams.


I expect that your grandfather is still an outlier, even back then. It's not like that sort of thing was common 50 years ago. It's still possible today, just unlikely. Maybe more unlikely than it was 50 years ago, but I hardly see that as representative of a huge shift.


"Santa's Super Sleigh"?


No but my user name is a big hint. Artist who grew up in South Philly 50-60 years ago and had a hit song so big he still lives off the royalties today.....


Hmm, my best guess is a member of the Silhouettes (Philly group with one huge hit - "Get a Job" - and nothing else very notable...)


Chubby Checker?!


I don't know that Chubby Checker did so well. I saw him at a county fair 25 years ago. He was performing in a tent the size of a large restroom, with only two rows of benches, at about 2pm. There may have been only 3-4 people there altogether.


Is your grandfather Bobby Rydell?

Or maybe Fabian or Frankie Avalon?

I love this kind of challenge.


Boyz II Men is my guess


Bobby?


Yea... I think the system is rubbish, but I don't really feel very sorry for someone earning $149k from lyrics they wrote 3 years ago.

Edit: helped write - per wgj there were 5 authors on that song - https://news.ycombinator.com/item?id=15419567


I believe that was a total? I didn't read it as just for this year.


That could be, but it seems like he's being well compensated for his effort either way.


Is he? Is he being compensated well when looked at relative to what others have made off of it? It would be interesting to see those numbers.


Just consider it flat out: Pick something you did three years ago. Let's say it took you a day, or a week, or a a good chunk of a month, off and on. With no followup work after that, what's a "reasonable" rate of return for that effort?

I make a good hourly rate, but at the end of each hour, my earning from that hour is done, and none of my hours have every billed at $149k.


You can make that argument about any kind of job. Is the CEO who makes $10M a year really worth 100x the marketer who makes $100k a year?

It's all in what you value, and who has the most leverage to get more money for whatever value they may provide. You may not think it's fair that the record company made some large multiplier of what the artist made, but that's just how the business works. If you want to change it by cutting out the record company, it's a lot easier to do that than it was 20 years ago, so I'd say the situation for artists in that regard is better, not worse.

In the end it absolutely does not matter what anyone thinks is "fair". Absent some form of regulation, people are going to extract as much wealth as they can out of a series of transactions, and some people are going to be more advantaged at doing so than others, whether because of talent or for structural reasons.


It's a good amount for the time spent. But you guys need to remember that with creations like this it's probably 1/100 at best when it becomes a hit with that many spins. If you take out prob of already working with somebody really famous that's more like 1 in thousands.

So it's like comparing what founder have earned per hour in some unicorn startup. Yes it's not bad, but when you think about EV of the work it doesn't look all that great.


Sure, but even with those (or worse) odds, I don't think it's that bad. I personally don't think writing a 3-4 minute melody with lyrics to go over it is worth all that much, at least in general. In that 1/100 case where it blows up and you make $150k for that time, I think that should be an exceptional case.

But of course that's just my value system at work: other people might value creative work of that sort much more highly; if a creator can get that person to pay them, great. There's no intrinsic worth or value in a particular instance of creative output; it's just up to how much demand you can create based on how well you can promote/market it.


Everyone is very focused on the contrast between the Radio royalties and the Streaming royalties - but we all skipped right past that first item: $149,000 in royalties for writing a song (just the first 3 years).

That's a lot of money - more than most people make in 2-3 years - for what probably amounted to a few weeks of work. And we're supposed to feel sorry for him because he couldn't retire on it? Seems to me if he got 1 of these per year he'd be pretty set.


Writing a top hit song is a 'few weeks of work' as much as writing a program is 'hitting the 'submit' button in your source control system.'

You don't get paid for all the non-hit songs you wrote.


I get paid the same for all the code I write, but none of it ever has any chance at all of still paying me three years later. None of it. Ever.


I don't think his problem was with the $149,000, but rather, with the trend toward $200 for tens of millions of streaming plays.

I.e., it might not be long before a comparable song with comparable popularity earns him $1000 instead of $150,000.


I have to accept that the value of the work I do may change in time, either up or down. Why do people in the music industry not have to accept that as well? Why are they entitled to the same economics decade after decade, while the rest of the world changes?


I personally think that music is a bad career choice. I started out majoring in music in college 20 years ago before switching to computer science, and while I still enjoy playing and writing and recording music, I enjoy whatever money I may get from it like a bonus.

I don't think that music industry people are entitled to anything in particular. I do think there's probably something suboptimal going on with streaming payments. If I were watching payment for my work drop 100x, I would probably be both looking for ways to "correct" that if possible, and looking for other work to do...


Wait for GANs and Style Transfer. Your "blessing" will evaporate as soon as the first Machine Learned Song hits Billboard 100.



The title point out that it's "overfitted". Basically the AI just reproduced almost the exact song that it was fitted with. It's basically a failed experiment where the machine didn't learn how to make music, but insted just repeated the note that it was taught. Not the best example to show case what RNN can do for music.


I can't say I've done a lot of research on this dispute, but so far I side with the streaming services since they aren't constantly conflating completely different numbers to try to trick readers. "Spin" is a radio term that isn't used in streaming, for good reason; it's just not the same thing at all.


In the absence of significant income from album sales, artists can make a respectable living through touring and merchandising. This path isn't going to work for songwriters. They need to find a different funding model. Perhaps bill artists by the hour? Probably a flat-rate contract for x dollars per song would be more feasible.


Famous artists can yes. If you are a semi-known band with four people playing a twenty date tour playing 400 capacity venues with two local support acts you're probably not making so much.

That's 8,000 people who care enough to come out and see you. So let's say at least 100,000 people need to like you enough that they would at least consider buying the ticket.

$4000 dollars gross in ticket sales per night, 30% to the promoter. $2800 left. If you get 50% of that with 25% to each support act then after commissions to booking agent and manager you're left with maybe $950. You need to pay travel and accommodation costs out of that, and possibly a tour manager. For a four piece band plus tour manager you probably need $30 a head per diem for incidentals of living in the road. Add in food - another $30 per head and you've got $650 left. Hotel rooms aren't cheap. travel isn't cheap.

Sure you can sell 25 t-shirts a night at $20 each and make $15 a shirt. And throw in some CDs - maybe 30 at $10 a pop... you maybe net another $600 a night.

$1250 a day for your hotels and travel for 5 people - maybe you have $600 left after these costs. Oh and you have to pay your tour manager...

Each of the people in the band makes... what - $100 per show?

Can artists really make that respectable a living from touring and merch?

Sure, scale this up, play bigger venues - 2,000 capacity. But then your costs increase - people want a more impressive show, not just spit and sawdust. You need a set, lighting design, all the rest of it.


I've got a relative in the music/movie business. He was pretty successful at both, most people in the US over 40 would probably recognize him. He and his family live well, but he still tours; usually small venues, travels by tour bus instead of fancy planes. Small crew, just a manager, sound guy and his wife. I figure he might pull $5K a stop, maybe 2x that depending on the size of the venue. Not bad, but not FU money. He still acts in movies, even some that have not held up well over time. I doubt he makes a ton with those roles either. But he still does them. My mom asked why he still does it, at his age; he replied, "I've got kids to put through school."

I saw him perform for the first time in 40 years, and realized how much he lied to my mom. On stage, he was (as the kids these days say) lit. He clearly loved the music (he still writes and performs new material), but the light in his eyes showed that what he really loved was performing and sharing his music with the audience.

Most musicians can't count on a career lasting 50+ years. The era of singles is gone, and performances are where the next generation will have to make their living.


This is a story about middlemen.

30 years ago, the supply chain for recorded music went something like this:

creator -> music label & radio -> retailers -> audience

Now it's something like this:

creator -> (label) -> digital platform -> audience

Huge power is concentrated in the platforms, which essentially combine curation, distribution, sales, and marketing. They control pricing. They control the playlists. They control the relationship with audiences and understand them far more than record retailers or radio ever did.

Labels now provide questionable value -- I would expect that some platforms may start to move into that area, in order to guarantee exclusivity, which is one of the few ways for labels to establish competitive differentiation. This is already happening in other media, such as books (ex: Amazon-owned book labels signing top authors or strong up-and-comers) as well as Amazon video and Netflix funding promising scripts and creative teams.

What can artists do? They have a few cards to play. One of them is exclusivity. Another is the ability to build their own connections with audiences (often using free or low-cost communications platforms). And dedicated audiences will do what is asked of them, within reason.

Some also perform, which is one of the few areas to make money these days. But platforms are also horning in on that pot. Ticketmaster is ruthless when it comes to manipulating pricing and forcing artists to play by its rules.

If I were an up-and-coming artist, I would concentrate not on becoming a "star" but rather making music that both you and your audiences can love, and building indestructible bonds with audiences that they will follow you no matter where you go, whether it's jumping platforms, going on tour, buying new march, or taking part in other activities that respect their love for the music while helping you make a living.


I'm worried about platforms moving into label territory too much because as soon as label start seeing them as competitors, they'll start rolling their own platforms and then we get that fragmentation which we know from video.


Yes and no.

30 years ago the supply chain for recorded music went something like this:

Songwriter - publisher - recording artist - record label - wholesale distribution - retailer - audience.

Someone would write a song. Ideally they would have a publisher, whose job it is to help them maximise the commercial exploitation of that song. That song would be recorded by an artist, funded by a record label. The writer and performing artist may or may not be the same person or people.

A record label would sign a recording artist, and often pay them an advance against future recording royalty revenues. That advance - say $100,000 - was exactly that - an upfront payment, and until the artist's royalty payout hit $100,000 they did not make any more money and did not share in any profits the label might be making from the recording. At the point that they have paid back the $100,000 advance the artist has "recouped" and starts to get a share of profits.

The label would pay for the recording of the song, and then pay for mixing, master, and production of the finished product. They would then ship this product to a distributor who would get it into shops. The distributor would pay the label a unit price considerably below the retail price. The publisher representing the songwriter would get a small fee per unit shipped based on the wholesale price. The artist would get a small recording royalty per unit against the wholesale distribution rate - maybe 10% or 20% or whatever deal they had struck with the label.

Retailers would mark up from what they paid. Sales would be driven by a marketing campaign paid for by the record label - sometimes the publisher would contribute to this marketing spend. Marketing would comprise interviews/features/reviews in a limited universe of music publications and newspapers, some advertising spend, a radio campaign (convincing radio stations to play the music) and perhaps some TV slots. It was fairly manageable, and labels had good connections into major media so were quite easily able to generate good hype for artists.

If the record didn't sell, then the label had sunk a lot of money into marketing and promotion, and the advance to the artist. And if the record didn't sell, the artist would not begin to recoup against their advance.

Now the model is something like -

Songwriter - publisher - recording artist - record label - music platform

Someone still needs to write a song. Someone still needs to pay for that song to be recorded. Recording is easier and cheaper than ever though, so labels don't have to invest so much here. However, because it's easier and cheaper to record than ever before, it also means that more people than ever are doing it. Distributing physical product was HARD. And expensive. And now that you can quite literally with a couple of clicks send a recording to Spotify or Apple Music, that barrier to entry has also dropped. So it's easier and cheaper to record, easier and cheaper to distribute - and the audience has far more choice than ever before. Marketing music effectively is harder than ever - there are more channels to reach consumers than ever before, and labels have far less influence over these channels than they did before. So marketing music effectively requires a real understanding of the potential fan base for an artist, and real understanding of how to structure and run a campaign, and reach the curators who power the platforms.

The role of a label has changed considerably in the last 30 years. Do labels provide questionable value? Against the old model, yes - you can record and distribute your music without a label. To reach an audience and really cut through against all the thousands of other people recording and distributing music - you need real expertise, and considerable experience of what works and what doesn't work.

Running a record label in the old days was very different to running a record label now. But labels make money by being experts.

If I were an up-and-coming artist I would make music that I and my audience could love - and I would hire the BEST POSSIBLE PEOPLE to ensure that I'm maximising my chances in every possible area of my career, and cut them in for a percentage of whatever increased success they could bring me.

Are those experts working for a "record label"? Maybe, or maybe not. But they will definitely be doing this day in, day out, for similar artists, and they will be learning every day what does and doesn't work.

A lot of people in technology seem to think that music is a "build it and they will come" sort of thing. Just because your song is on Apple Music or Spotify does not mean that people are going to discover you. Once they discover you, you have to hope that they will stick with you on your journey, but actually because they way a lot of platforms work the music comes first and the "artist brand" comes second. It's a pretty widely acknowledged fact in the music industry that if you play a bunch of relatively well known songs to a bunch of kids they will probably know the songs, and be able to sing along. They probably WON'T be able to tell you the artist behind those songs though. Great songs deserve great audiences - but all too often great songs suffer because they can't cut through and find those audiences, and even when they do get listened to, converting someone who hears your music into someone who remains a fan and follows along on your journey is something else entirely.


I can't speak as to the other platforms, but spotify's revenue sharing model really sucks for small artists.

For example, if someone signs up for Spotify to listen to exclusively songs by one artist, that artist is not getting that user's monthly fee... they're probably getting a couple pennies tops, if that.


I've heard this before. I think it's definitely possible that Spotify's revshare works out in favor of big artists, but I haven't heard an argument for that that didn't involve making broad, unsupported assumptions about users' listening habits.


I really do wish that the part of the money I pay Spotify that goes to royalties ONLY went to the artists I listened to...


But why should they get most of the monthly fee? Spotify offers the ability for that user to listen to that artist plus a million others and thus carries the overhead required to create a platform for that artist to publish in the first place. This seems a bit like saying that I as a taxpayer should only have to pay for the public services that I, at this point in time, take advantage of.


That's why Spotify takes its cut, no?

Comparing listening to different artists to funding multiple agencies with tax dollars isn't a valid comparison.

It's like saying that when I buy my favourite brand of ice cream at the store, that the other ice cream brands will get a cut simply for gracing the shelves.


No, it's like saying that when you buy your favorite brand of ice cream at the store the grocery store keeps most of the profit because without the store the ice cream manufacturer wouldn't have a place to sell to the masses.

If said artist is in such high demand, then he or she can go make their own streaming platform, charge $10/month, and keep 100%. Obviously that isn't the case, so they must get some utility out of using the Spotify platform.


No... you're missing the point entirely The store is Spotify. The brands are artists.

If I buy brand A, the store gets a cut, and brand A gets a cut. Brand B gets nothing.

With Spotify, if I listen to Repartee, Justin Bieber is still getting some of my money.


Saying Justin Bieber gets some of your money isn't really the whole story. Justin Bieber's songs are streamed billions of times a year, so his music accounts for a very significant percentage of total streams, and consequently use of the Spotify platform. Thus, he is bringing an outsized amount of value to Spotify, so he rightly is deserving of more money.

Again, Repartee is free to pull their songs and sell them directly to you if they think that will bring them more fans and more money. It's kind of funny for two consenting parties who both use a platform to complain that the platform is treating them unfairly.


Are you sure you are understanding the argument being made here?

I think the argument, which is fair, is that it is very weird that if I pay Spotify $10 a month and listen exclusively to a single artist, that artist may get a penny from me and Justin Bieber may get $3 from me, despite me never listening to a single track from him.

Nobody is arguing Spotify can take it's 20% or whatever, people are just arguing that the calculation Spotify uses to distribute the monthly royalties severely penalizes small artists. (because it looks in aggregate across all listeners instead of divvying per user)


I think it's pretty well known that artist's make really less from streaming or even album sales.

That's why i usually buy t-shirts or something like that of the artist to support them. Fortunately for the type of music i prefer, soundcloud is enough for as of now.


The point of this article is that spotify's revenue sharing model really sucks for BIG artists. Justin Bieber definitely qualifies as a "big" artist.


To be fair, Spotify isn't mentioned at all


Deezer is apparently beginning to address this.


There are some interesting points in the article's comment thread.

>>> Amir Epstein - Money for big hits is in streaming on DSP’s. Spotify pays $7,000 per million streams. That may sound like nothing, but when artists like JB get 100,000,000 streams (that’s a conservative number) that’s $700,000. And then there’s apple itunes, which pays around the same, and google play which pays a little more. And if you are lucky and getting decent numbers on Tidal, it pays $23,000USD per million. Not too shabby

>>> Anonymous (response) - That Spotify money you just referenced $7,000 per 1 million streams goes to whoever owns the MASTER. Typically the record company, NOT the songwriters. Do your homework

There is a pot of money out there, but you're unlikely to see it unless you're the one writing the record contracts.


I pay GBP10 a month for Spotify. I've never listened to Justin Bieber on Spotify, but lots of other people (who don't pay) do.

As a percentage of my cash, how much goes to Justin Bieber vs how much to the bands I've actually listened to who have many less total plays?

Or - do neither of them get anything?


The other comment was correct that most of the pooled money will go to Bieber, however I feel there's a better way to frame it.

Economically speaking, you can imagine if Spotify had 100 users paying 10 GBP each; 99 listened exclusively to Justin Bieber, and 1 to 'Band.' All users listen to music in equal frequency. Though 99% of the pooled money — and if divvied up equally, 99% of 'Band' fan's money — goes to Justin Bieber, 'Band' fan's listening weight results in ~10 GBP of the total pool being pulled towards 'Band.' So, in effect, all of 'Band' fan's money could be classified as going towards 'Band.'

Also, as a couple side notes in Spotify's revenue structure:

1. 70% goes to artists, 30% goes to Spotify.

2. Over 70% of Spotify users become paying subscribers iirc, so there's not much of a problem with listening weight being pulled by free users.


>Economically speaking, you can imagine if Spotify had 100 users paying 10 GBP each; 99 listened exclusively to Justin Bieber, and 1 to 'Band.' _All users listen to music in equal frequency._ [emphasis mine]

Well obviously not all users listen to music at the same frequency (which i know you are likely well aware). What is the reason that they don't just take each users $10 then dole that out only according to the songs that person listened to. So if the user listened to one artist the entire month, then that one artist get's the full portion of the $10 that is given to artists.

That will help with the situation where: 1. Someone else listens to bieber on repeat 24/7 (why should that affect where another's $10 goes) skewing the sample massively towards that artist.

2. In terms of free users, segment them off and pool the ad revenue again boxed on a per user basis.


I completely agree with this solution being ideal, but I think it's a matter of engineering effort and memory. It's not worth the time investment to build a system more complex, and it's a lot easier memory-wise to just pool it all together for allocation.


Very minor correction - 70% goes to rights holders, 30% goes to Spotify.

A rights holder may or may not be an artist - quite often, the rights holder for the recording royalty is a record label, and the artist is getting a percentage of what the label takes. There is also the publishing (songwriter) royalty - and the songwriter may or may not be the recording artist.


Isn't the whole problem that users don't listen to music with equal frequency? If I use Spotify more than you do, then in effect some of your subscription money goes to the bands I like instead of the bands you like, which is weird and kind of destructive.


Pulling numbers out of thin air:

Let's assume that Justin Bieber gets 100m streams this month, and that's 5% of all streams on Spotify. That means there are 2bn streams across all artists. If your 10 favourite bands between them get 2,000,000 streams, then they represent 0.1% of streaming activity. Out of your £10, spotify gets £8.33 (taking off the VAT). Spotify pays out 70% of revenues in royalties - so there's £5.83 going out the door. Bieber gets 5% of this - £0.29 - and your bands get £0.0058 - so about 0.6 pence. If there are ten bands, then they are each getting 0.06 pence of your £10. Assuming those bands are self releasing, and there are four people in a band, each band member is making 0.0015 pence out of your initial £10.

Whether that money actually goes to Bieber personally, and whether the 0.06p actually goes to the members of the band is another thing entirely, and is dictated the deal that they have (or don't have) with an intermediary such as a record label.


A fair chunk would go to Bieber, almost nothing would go to the band you listened to exclusively.

It's because Spotify pools all the subscription money earmarked for artists into one great pool then divvies it up based on number of plays or play-time or some weird proxy measurement.


Wow. The amount of hate in the comments is amazing. Eveyone knows best! I know many artists who complain about their low streaming royalties turn around and listen to music on Youtube, Pandora, Etc. HYPOCRITES. Do you really care about directly supporting your favorite artists and encouraging the work of upcoming acts? Then buy their music on BandCamp, go to shows and buy t-shirts/merch.


In case anyone wants to glorify the "good old days": http://www.negativland.com/news/?page_id=17


On this theme, now is a good time to re-read the incredible letter Steve Albini wrote to Nirvana about potentially producing In Utero, which includes his perspective on royalties, etc. https://consequenceofsound.net/2013/09/read-steve-albinis-pr...


The guy wrote a song. It took him, what, a month? For that he earned $150,000? Annualized, that's almost $2 million. I don't see the problem here.


The problem is that writing a hit song is not something he can do every month. It's like finding a nugget of gold. It still has to fund the operation.


His pay probably lasts as long as that song was popular.


It would be fair to count in years of education and training. Just like with any other profession.


As with any article about music finances, it misses one crucial detail: elaborating how the money is split between every single entity in the value chain. There's plenty of YouTubers who make orders of magnitude more money on the same amount of views. I feel it's probably largely due to the fact that youtubers usually only split the cash two ways between the star and the editor


I don't understand, so Bieber made $596,000 (assuming he owned the other 80%) from this song? No way, it would seem like?


There are actually a completely different set of royalties payed to the performer vs. the songwriter.


Rodney Jerkins is only one of five writers on the song: https://en.wikipedia.org/wiki/As_Long_as_You_Love_Me_(Justin...

Bieber also has songwriting credit. If the royalty was divided evenly, Bieber would also get 20%. However, it isn't required to divide evenly, and songwriter royalties can be negotiated myriad different ways.


I have similar feelings. My gift generates millions for someone else. I take home above median income, but I'm not getting rich. I'm working toward a plan. I wonder what his plan is?

It's interesting watching this comment section fill with bickering over what commenters think he deserves.


Seconded. I think the software industry is an example where the "creators" just get a salary.If you are lucky or an early stage employee, you may get some equity. Software that we write creates billions for corporations but we never get a share. Question: Why do musicians & "creative" people get royalties for what they do and not others? If value creation is the argument, I would argue we are also creating value. I am not asking to be mean, but I am curious.


It's the risk tolerance intersecting with demand-supply. If you join a startup where you are willing to be underpaid in return for equity which could make you rich at a small probability, you're in a entertainment industry model.

However if you want a salary and comfortable living, you take a salary and don't get a shot at wealth. You could refuse to work on such terms but your company could persumably hire someone else who won't refuse since the supply of risk averse individuals significantly exceeds those who take risk..


Because they have a union that has negotiated terms on their behalf.


They are talking about songwriting royalties which are easily less then 5% of the royalties paid out by streaming companies. The vast majority of the royalty goes to the recording artist and their label in this case Justin Bieber.


Right, this is an important point which appears to be have been missed elsewhere in this thread (and was probably assumed knowledge in the original link).


Obviously the streaming numbers are minuscule, but if people bought CDs, they can play it unlimited times for a fixed rate. Streaming does substitute with radio and purchased music, but more the latter.

Expecting radio rates for something that primarily substitutes for purchased CDs is ludicrous. It would have been more interesting (to me) if the article had compared the youtube/pandora payments with the equivalent CD purchases.

Also, this is what google and apple want to do to all of us. Make your complements cheap to enhance your value. Hence the plummeting value of software on their platforms. Apple is busying crying crocodile tears about the difficulty of making a living on the ios platform. They're not gonna care unless and until it damages their platform.


Your first point regarding how streaming cannibalizes CD I think is pretty spot on. Radio play for record companies and artists was a way to entice listeners to buy the record/cassette/CD/etc. and depending on where you sat in the payout chain, you really didn't expect money from radio... often times you paid to have your record played on radio (https://en.wikipedia.org/wiki/Payola ... and there were creative ways around the laws after they were created). But radio didn't play what you wanted when you wanted (that took buying records), radio played what they wanted when they wanted.

Streaming upends that. I don't buy records anymore, I pay a streaming subscription. I get pretty much anything I want whenever I want on demand. I have access to a much larger catalog than I would if I bought everything individually and I'm probably spending on-par with what I did in the old days. (I use to get a lot a free CDs and such because I worked for a record store chain in the 90's, but even so, I have access to a larger catalog today). The fact that I can get the vast majority of what I want means that those few hold out artists that I would like to listen to, but can't because they refuse to play the game (with good reason), simply don't get listened to... I have other choices.

I can't comment on the other issues you raise.


Well, I thought this was going to be about JB screwing over a writer... so at least it wasn't that. Pretty low numbers though. I wonder what Spotify would have looked like if it were on the list...


To some degree that is what happened, though I'm sure he was screwed by a lawyer rather than the artist himself.

When they offered this songwriter some % of song royalties, they knew ahead of time that it wasn't going to be a whole lot of money.

Solution? Songwriters need a new metric to charge by.

One problem, though, is that perhaps songwriting isn't actually worth that much. When you look at a performer like Justin Beiber, how much of his income depends on specific songs that he's putting out? I'm not sure, but I don't necessarily think it's a large amount.


If writers are so opinionated about their works value, they just need to charge a fixed price. The allure of the possibility of writing a song once and having potentially massive perpetual income is why they like these royalty deals (and why said lawyer knew it was a good deal). But that's more of a lottery ticket in todays music industry. However, if you assign a price to your work you will quickly know what it's worth.


The rates paid to songwriters by Spotify & Pandora are similar: about $0.00007 per play.

Source: royalty statements


You want to know how many spins a day some of my code does?


Actually yes please...

I would love to see that math worked out with someone who has written a popular program/website.


Judging by YouTube alone he wasnt paid 20%, but rather 0.01-0.5% of real royalties.


The revenues from live when you write or co-write for a globally recognised artist who tours will always be more lucrative for songwriters than the revenues from recordings. This is because (generally speaking) the revenues for live for songwriters come as a split of box office - and live performance is generally going to gross a lot more than recording for an artist of that stature.

For example, in the UK, PRS takes 3% of box office, and that is then paid out to songwriters based on duration of performance. So for example, if someone sells 1000 tickets at 10 each, there is £300 to be paid to PRS. If there are three acts playing and they each play 30 minutes, songwriters get £3.34 per minute. So if two songwriters co-write a song and agree a 60/40 split and that song is 4 minutes 30 seconds one songwriter would get £9.02 from this show and the other would get £6.01.

Bieber played six dates at the O2 in London - capacity 20,000 - and tickets seem to have been around the £45 mark. Let's assume he pretty much sold out - so those six dates grossed £5.4m

Assuming he did a 20 song set (so maybe 75 minutes duration for songs, with two support acts each playing 30 minutes - total duration 135 minutes) then the songwriters (for both Bieber and support acts) are getting £1200 per minute of performance across that segment of the tour. So actually, if you are the support act and you are a singer-songwriter playing 30 minutes support to Bieber, you're going to walk away with your live appearance fees plus £36k in songwriter royalties.

Streaming revenue splits between labels and songwriters (or sound recording copyright royalties, and publishing mechanical royalties) are based on old record label models, where the label invested a large amount of money to record, manufacture and market product; because of this, they took the lion's share of revenues from record music sales. It's easy to argue that things have changed, but equally, generally speaking a songwriter needs an artist to perform and record the song - and the artist probably needs a label (or someone with some money) to market that song and help it generate the maximum revenues. Songwriters still benefit, because they get MORE revenues than they would have done otherwise. A great song is nothing unless someone records it and performs it.

So the problem is not necessarily with Pandora, Spotify, YouTube and all of these other companies - they are following models which by-and-large have been dictated to them by labels - you can't operate a music streaming platform without songs to play, and you can't play songs without obtaining permission to use the recording, and the recording copyright - and thus that permission - is controlled by record labels rather than songwriters.

The scale of the market is completely different. Bieber's 2016 world tour apparently grossed $250m in ticket sales - if we stick (for ease) with the UK model for compensating songwriters for live performance, this tour generated $7.5m in revenues for songwriters. Let's say that across the whole tour - including Bieber songs, numerous co-writes, and support acts - there were 250 songwriters involved. Assuming everyone's song was performed for roughly the same duration then each songwriter should have walked away with about $37,500. Not bad.

The problem is not streaming - it's people crafting flawed narrative that is based around "my song got streamed a billion times and all I got was this lousy t-shirt" but at the same time ignoring the completely different scales of revenue for live performance vs recording. Bieber has 32 million monthly listeners on Spotify. Assuming that they each listen to an average of 8 tracks a month (for a total of roughly 250m streams) then his label is generating maybe around $1.25m in recording revenues. The songwriters will be getting a fraction of that - and each individual songwriter will get a proportion of that fraction, based on how much their song is listened to. If his most popular song gets 50% of streaming activity, second most popular 25%, third most popular 10%... and his tenth most popular gets 1% then that track is only generating the label $12,500 a month - and maybe $2,500 for the song writer - but it's still getting 2.5m streams a month. If the writer has a 20% share of the song, then they are getting $500 - over a year (all things remaining equal) then maybe they get $6,000 for 30m streams.

If that song was played on every date of a Bieber tour as part of a 20 song, 75 minute set, on a tour where headliner is performing 60% of the total duration, then based on the UK model for songwriter performance royalties the writer of that song would be getting $225,000.

If someone has a 20% share of that song they are getting $45,000.


At least we know why Google earns so much. They don't pay any royalties, rather keep everything to themselves. Interesting business concept.


I don't think they're keeping everything for themselves. I think they're actually not taking in that much money from Youtube ads / youtube red.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: