Without YouTube, I dare say I might not have bought… what… sixteen of the albums in my current collection? Almost all of the music I've bought is from artists I first heard on YouTube.
I think Neil Young said something along the lines of "I look at radio as gone … Piracy is the new radio, that’s how music gets around."
He uses the word piracy which doesn't apply here, but I think what he says is relevant in this context: a lot of people hear music for free, like it, and then buy it. Maybe they hear it for free on the radio, through torrenting, or through youtube. Although the radio and torrenting does generate revenue as well as promotion whereas torrenting only does promotion.
my startup idea that someone else should totally build:
Sell a soundboard recording of the show I went to for $25 on the spot as I leave. I went to the show so I now have an emotional connection to that particular performance. Will this get shared? Sure, but the majority of people downloading bootlegs own the artists' oeuvre anyway.
The fact that larger bands don't up sell a drone video of each show is just leaving cash on the table. I'd happily pay $30-$50 for a video recording of every show I attend.
Touring for your primary income is a brutal lifestyle, especially if you're not in your 20s and unattached to dependents.
I don't think it's reasonable to say "oh, if you want to make money with music, you should just spend 6 months of the year living in a van traveling the country".
I don't see why that's unreasonable. Plenty of jobs in other industries (consulting, sales, construction, trucking...) have a large travel component to them. You don't have to accept work with 50% travel, but if that's where most of the money is for what you want to do, then it should be on the table.
Even an author, typically far more introverted than a musician, has to seriously consider hitting the road to do book signings and interviews for any hope of success.
It's a good article. I think attorneys of Google, etc should repeat every chance they get the part about how music industry destroyed all the people doing what they say they want. And show the greed of the split they ask for with its effects on other business. As in, the music industry is the perp instead of the victim.
Also, people should bring up just how hard and expensive it was (or is) to acquire licenses for things like wedding videos. Distributing your wedding video is a cppyright violation if an unlicensed song happens to be playing. Music industry, if you could get to them, would license you the rights for sometimes $10,000 or something. Absolutely ridiculous. A modern service exists that makes it easier but still doesn't for all songs & still can be expensive. This kind of crap is why they have so much opposition even from those who otherwise would want to pay them.
I see the record labels greed as a feedback loop: People think music is worth less than its price, so they pirate it. Piracy lowers sales. To compensate, the prices go up causing more people to pirate.
When selling a product, what would you prefer? 100 sales at $10 or 20 sales at $20? What the record labels see is that less people buy it less at the higher price, so it must be piracy!
Except radio pays artists way more than Youtube. Just look up what one-hit wonders from the 80s are still earning every year - that's almost 100% radio.
> half of Internet users in the U.S. listen to music on YouTube -- by far the most popular access point -- but YouTube is responsible for only 4 percent of revenue to the industry,
> Song downloads and streaming music account for 70 percent of U.S. music sales. But the industry's total revenue has dropped by 66 percent since 1999.
The profusion of bullshit "statistics" being used the advance a prejudiced position is so grating. I know why the music publishers are doing it, but the journalists don't have to just repeat it. There are actual interesting numbers that could be calculated.
Of course music revenue is down after moving to a massively more efficient platform. Profit is the relevant metric.
50% "listen on youtube", and generate "4% of revenue" ... I don't even know how this could be changed into a meaningful statistic. I guess you could compare TV+radio listener numbers and revenue share 1996 versus 2016 ... for it to really make any sense you'd have to compare listeners and revenue through all mediums. If there is a change of composition away from high-revenue mediums to lower revenue mediums, that would be more relevant.
And why use the year 2000 as the baseline, when recorded music has a 100-year history.
It's bullshit on another axis as well: it assumes that the labels have a claim to a share of all music revenue on YouTube. The slideshare deck they linked for that claim shows the large majority of YouTube songs are "off label content" (slide 46), and it also says that more than half of the revenue is from "user generated content" (slide 43).
The second statement is comparing apples to oranges, but making it seem they are the same. I originally thought the first statement was comparing apples. Until I realized it said "half the Internet listens" and not "half of the music listen to..." It may be the author meant the second statement, but who knows
I don't hear the fashion industry complaining that their creations are being used in photos and videos posted to youtube. So why should the music industry complain?
Music is marketing for a live performance by the performer.
Trent Reznor probably cares more about the thousands of dollars (VIP tickets via his fan club site) I've paid to see him live over the last decade or so than the ~$150 I've spent on his albums (with the label taking most of that).
> Trent Reznor probably cares more about the thousands of dollars (VIP tickets via his fan club site) I've paid to see him live over the last decade or so than the ~$150 I've spent on his albums (with the label taking most of that).
Trent Reznor made millions before the collapse of the music industry in the 2000s THANKS to big labels. It's easy for him not to care about your money, at all. He'd be nothing if no major label invested in him in the 90's .
No, he'd be very famous, and probably just as productive. Certainly not rich, though. He was on TVT records, one of the small labels snorted up by Interscope, later snorted up by Universal Music Group. He was associated with the Wax Trax! people, many of whom were extremely prolific, and extremely successful, and who were snorted up by TVT before they were snorted up by Interscope before they were snorted up by Universal.
The big labels didn't start anything, they just bought it up.
This is the crux of the argument ", the record labels squeezed to death nearly every upstart that wanted to become a Tower Records"
Perhaps if the music industry had embraced digital 15 years ago instead of crushing it, they would be in more control today.
> This is the crux of the argument ", the record labels squeezed to death nearly every upstart that wanted to become a Tower Records" Perhaps if the music industry had embraced digital 15 years ago instead of crushing it, they would be in more control today.
It wouldn't have changed a damn thing. People would still be sharing music for free today, like they did as soon as the medium allowed it (tapes).
Sure. But music sharing didn't kill the industry then; there's little reason to believe it'll be the reason for its death now.
The problem is there is only so much profit to be had. The recording industry is not interested in trying to sell their own product digitally at a price consumers want. So instead, either a company is large enough and entrenched enough (like Google with Youtube) that that company is able to capture the profit, and the music industry complains, or the company is small enough that the music industry captures the profit, and the company goes under or operates at a loss (too many to name, but Spotify and Pandora both fall under this).
There may not be a good split, that sees both the recording company and the distributor seeing a profit, at the prices consumers are willing to pay, which if the case explains why they've fought tooth and nail against every form of digital distribution. But I don't see how the recording companies can stay afloat without it. It may be we'll have to see how music works without them.
Of course it's faster and more densely connected. And? Extrapolating that because people shared before, and now they can share BETTER, to "the music industry is doomed" (or whatever statement you care to make), is what is 'not a valid extrapolation'. I wasn't extrapolating anything, just saying "this is not a new behavior", and given no other data, the fact of that behavior is not in and of itself a death knell.
> Sure. But music sharing didn't kill the industry then; there's little reason to believe it'll be the reason for its death now.
To me that translated to: "Sharing some songs between a handful of friends once every few days didn't harm the music industry; hence being able to share hundreds of songs with millions of people in a few seconds will harm it now."
The second statement sure looks like an extrapolation from the first, and given the mind-boggling difference in scale, an invalid one.
As for lack of data, music industry revenues have been decimated since the turn of the industry. Sure, there are several factors involved, but can you really pretend the terabytes of copyrighted music being torrented every month has nothing to do with it?
I don't where people get this notion that the music business is somehow "greedy" when the profits and margins are small.
The record companies are not mere "middle-men" between artist and fans. They actually fund the costly process of creating and marketing music.
For better or worse, we have an economic system grounded in the concept of intellectual property.
Only with the music business are the producers of a good, who undertake the production of that good at a ridiculously high risk, branded as "greedy" if they attempt to sustain a viable business from their efforts.
As far as I'm concerned, Apple, Google, Spotify, Pandora, Amazon and the like are parasitic "middle-men" who bring a commodity level service to the industry.
Through quirks of the market structure, they have been able to extract the bulk of the value from one of our sublime cultural goods, and leave the actual producers scrambling for crumbs.
If Google thinks it's no big deal, why don't they just undertake the production of the content that they monetize? They won't, because it's a crap business, and only exists because the people who create music are at least in part driven by non-pecuniary rewards.
That's the way it goes, but spare us this attempt to paint this as a morality tale.
"Somehow greedy"? I don't think this is really debatable: Steve Albini put the case succintly in the 1990s, and things haven't really changed from the musician's point of view. Check out the math here:
TL;DR: If a record grosses $3 million, making $710,000 in profit, the band makes $12,000 (1.6%), in this case split between three people at that. Including "salary", since this is their sole payment for working their ass off for a full year.
I'm generalizing, but more or less here is the issue with record companies. Only a small number of albums released come close to making a profit. The vast majority of albums generate very little revenue, and don't even come close to covering the costs of producing the record.
So the business model of the record company is to fund a bunch of albums in the hope that one of them becomes a hit, and generates enough revenue to cover the production and marketing costs of all the others too. Whatever is left after that is the profit.
The contracts are structured to reflect this reality. i.e. very low royalty rates, like maybe 10% of the gross sale price per album. From this small royalty, it doesn't even kick in until the entire costs of the production and marketing have been "recouped" from this royalty!
The artist has an "account" at the label to keep track of how much they have recouped to date. Usually the costs hold over from project to project as well, so the artist can get deeper and deeper in the hole if they release a series of poor selling records. An artist can wind up a situation where they have a massive hit, but receive no royalties on sales as they haven't recouped earlier releases
As an example, let's look at a hypothetical album release. (This model doesn't apply as much anymore as the role of the album is much smaller in music sales, but the principle is the same).
let's say the cost of production and marketing the album is one million dollars, and the artist royalty rate is 50 cents per unit on a wholesale price of $5 per unit.
That means the artist won't see a penny of royalties on record sales until they sell two million units. After that they get their 50 cents per unit. The record label "advances" the funds to make the record to the artist.
While it does look like at first glance like such an agreement is horribly unfair, and can only be explained as exploitive, this is really the only viable business model possible if you want to have a business producing and releasing music.
Why would an artists sign such a contract, with so little hope of making money. The main reason is that if the record fails to recoup, once the contract is over they are not personally responsible for the debt. The record company pools the projects under a business umbrella, which allows the individual artists to pool the risk.
Here's another way to look at it. If you are an artist under a recording contract, you are guaranteed access to the capital you need to make an actual business of being a professional artist under a much less risky structure.
Only a few number of serious musical artist who might be able to fund the project through sources where profit is only possible if the project is commercially successful. If you had to, say take out a personal loan for $500,000 to make and market a record, given that the odds of success are incredibly low, that would be pretty much insanely risky.
The reality is that most musical artist never come close to creating a viable, marketable product, as the talent and capital required are too scarce. Even the few that manage to make a really good product independently do not have access to the capital and organizational resources to market it properly.
Another factor at play, is that there are other ways of making money as an artists outside of record royalties. Having a "record deal" gives you much greater access to those opportunities. For example, there what's called writer's royalties which are connected to the copyright on the composition, not the recording. These are not subject to the same agreement, and are paid on every sales. There are also opportuniti...
The major record companies make an enormous amount of profit, and take nearly no risk. At this point, they're not signing anyone who doesn't already have a fanbase. It's the musicians that are broke.
Mind explaining the costs of producing good music ? Is it cost of studio, salary of technicians, cost of marketing, cost of distribution ?
Specifically, my question is whether the world be better off if even relatively unknown musicians get access to high end equipment through open distribution and marketing afforded by the Internet. Music revenue would be split based on talent and not based on who is popular.
For example, someone puts out a new piece of music on the Internet. A set of music enthusiasts likes it and awareness spreads. The artist now signs up with a studio to produce a high quality version and sells it for money on Spotify.
Those are good questions. I'll take a stab at an answer. It's long, I can't quite figure out how to succinctly sum it up. Sometimes writing a longish post helps organize my thoughts, maybe you will find it interesting...
I'm a music producer, of sorts, but I don't work in the music business per se, so my knowledge of actual costs is limited. I do occasionally work with people who are active in the industry, and so I try to at least keep up with what's going on.
(I usually work either for companies, or on low budget independent projects that don't make money, but are more products of love and passion:)
It's a little hard to get your mind around. The music industry proper is not that big, and what really goes on is not discussed in general forums very often. It's a bizarre little business world unto itself.
One thing to keep in mind is that the music market as an economic structure is an almost entirely "fat head, long tail." The vast amount of revenue is derived from the output of small number of artists (maybe 100 at any given time). The percentages are crazy. The percentage of recordings that even generate any significant revenue, never mind profit, is on the order of .001% (I'm guessing, but it's very low). There is definitely revenue that comes in from the long tail part of the market, but it's relatively small.
This "winner take all structure", which is inherent to a mass media form of entertainment, tends towards record labels being conservative about who they sign, and who they take a chance on promoting. If they do decide to seriously promote an artist, and support the whole business infrastructure needed to create a modern, mass media entertainment "product," they will bet big, and not hedge on production costs.
It's hard to understand in part because the way commercial music is actually produced has changed forms many times over the last 100 years, and the current model is relatively new, and not widely known by the public at all.
Currently the production model for commercial music is that a small team will be put together (or maintained) who work under a name "producer." (Dr. Luke being at the absolute top). These are they craftsmen/women who actually produce the music.
There will be writers, musicians, engineers, sub-producers, and associated creative and business types, who will all work together with the producer, artist, and label to create the actual song product you might hear on the radio.
There are often multiple team members working in parallel, in multiple studios, on the same project, with the best material generated being chosen for continued work. (Sort of the way a bigger technology company would have multiple teams on R&D).
Musical material from outside producers and writers is also be brought in under the production umbrella. For example, sometimes an artist and producer get some traction organically, like you describe. The producer can be brought on board a team to continue the work.
There are lots of independent entities who are producing musical "product" that are used as kind of a farm system for the major labels and producers. A lot of modern music is created from a "beat" which is really just the basic music of the song. There are actually various markets at work for "beats." that are always busy. There are a small number of independent "songwriters" who have a major contribution to the process.
Out of this process, which doesn't necessarily take a long time, weeks, maybe months, a marketable "record" (a single song) is hopefully produced.
I would estimate the costs of this production phase at $70-90 thousand.
From there, a label might be looking at roughly $1 million in marketing costs, videos, and the like. Radio promotion alone can cost hundreds of thousands (getting the song on the radio). Similar funds will be used to get the song visible wherever they can, Spotify, Youtube, anywhere fund...
The graph doesn't show revenue from FM or AM radio.
YouTube is comparable to radio. A place people discover and listen to music, which they later buy as downloads, CDs, vinyl, etc.
Much YouTube music views are of official videos that the artists and labels themselves upload.
If anything, the record industry should be paying YouTube more for exposure and marketing. It's time for the record companies to stop freeloading off YouTube and pay their fair share for their use of this important music marketing channel.
Agreed. I think some of good indie musician on YouTube are much better off alone than with a record company.
People these days have many more means of entertainment. I think it is hard to get super popular by just doing music alone. Pairing with movie, games, or other multi-media seem to be a better strategy.
Although it is probably irrelevant streaming music videos is also the least cost effective way to stream music. So Google has to take a larger cut of the profits in order to offset that cost. (With he exception of physical media possibly)
53 comments
[ 5.0 ms ] story [ 90.9 ms ] threadHe uses the word piracy which doesn't apply here, but I think what he says is relevant in this context: a lot of people hear music for free, like it, and then buy it. Maybe they hear it for free on the radio, through torrenting, or through youtube. Although the radio and torrenting does generate revenue as well as promotion whereas torrenting only does promotion.
Sell a soundboard recording of the show I went to for $25 on the spot as I leave. I went to the show so I now have an emotional connection to that particular performance. Will this get shared? Sure, but the majority of people downloading bootlegs own the artists' oeuvre anyway.
I don't think it's reasonable to say "oh, if you want to make money with music, you should just spend 6 months of the year living in a van traveling the country".
Even an author, typically far more introverted than a musician, has to seriously consider hitting the road to do book signings and interviews for any hope of success.
Also, people should bring up just how hard and expensive it was (or is) to acquire licenses for things like wedding videos. Distributing your wedding video is a cppyright violation if an unlicensed song happens to be playing. Music industry, if you could get to them, would license you the rights for sometimes $10,000 or something. Absolutely ridiculous. A modern service exists that makes it easier but still doesn't for all songs & still can be expensive. This kind of crap is why they have so much opposition even from those who otherwise would want to pay them.
When selling a product, what would you prefer? 100 sales at $10 or 20 sales at $20? What the record labels see is that less people buy it less at the higher price, so it must be piracy!
If a radio play goes out to 50,000 listeners, the total paid is less per listener than YouTube.
> Song downloads and streaming music account for 70 percent of U.S. music sales. But the industry's total revenue has dropped by 66 percent since 1999.
The profusion of bullshit "statistics" being used the advance a prejudiced position is so grating. I know why the music publishers are doing it, but the journalists don't have to just repeat it. There are actual interesting numbers that could be calculated.
Of course music revenue is down after moving to a massively more efficient platform. Profit is the relevant metric.
50% "listen on youtube", and generate "4% of revenue" ... I don't even know how this could be changed into a meaningful statistic. I guess you could compare TV+radio listener numbers and revenue share 1996 versus 2016 ... for it to really make any sense you'd have to compare listeners and revenue through all mediums. If there is a change of composition away from high-revenue mediums to lower revenue mediums, that would be more relevant.
And why use the year 2000 as the baseline, when recorded music has a 100-year history.
The question is: how often?
Trent Reznor probably cares more about the thousands of dollars (VIP tickets via his fan club site) I've paid to see him live over the last decade or so than the ~$150 I've spent on his albums (with the label taking most of that).
Trent Reznor made millions before the collapse of the music industry in the 2000s THANKS to big labels. It's easy for him not to care about your money, at all. He'd be nothing if no major label invested in him in the 90's .
The big labels didn't start anything, they just bought it up.
It wouldn't have changed a damn thing. People would still be sharing music for free today, like they did as soon as the medium allowed it (tapes).
The problem is there is only so much profit to be had. The recording industry is not interested in trying to sell their own product digitally at a price consumers want. So instead, either a company is large enough and entrenched enough (like Google with Youtube) that that company is able to capture the profit, and the music industry complains, or the company is small enough that the music industry captures the profit, and the company goes under or operates at a loss (too many to name, but Spotify and Pandora both fall under this).
There may not be a good split, that sees both the recording company and the distributor seeing a profit, at the prices consumers are willing to pay, which if the case explains why they've fought tooth and nail against every form of digital distribution. But I don't see how the recording companies can stay afloat without it. It may be we'll have to see how music works without them.
Not a valid extrapolation. The Internet is billions of times faster and more densely connected than the sneakernet.
> Sure. But music sharing didn't kill the industry then; there's little reason to believe it'll be the reason for its death now.
To me that translated to: "Sharing some songs between a handful of friends once every few days didn't harm the music industry; hence being able to share hundreds of songs with millions of people in a few seconds will harm it now."
The second statement sure looks like an extrapolation from the first, and given the mind-boggling difference in scale, an invalid one.
As for lack of data, music industry revenues have been decimated since the turn of the industry. Sure, there are several factors involved, but can you really pretend the terabytes of copyrighted music being torrented every month has nothing to do with it?
The record companies are not mere "middle-men" between artist and fans. They actually fund the costly process of creating and marketing music.
For better or worse, we have an economic system grounded in the concept of intellectual property.
Only with the music business are the producers of a good, who undertake the production of that good at a ridiculously high risk, branded as "greedy" if they attempt to sustain a viable business from their efforts.
As far as I'm concerned, Apple, Google, Spotify, Pandora, Amazon and the like are parasitic "middle-men" who bring a commodity level service to the industry.
Through quirks of the market structure, they have been able to extract the bulk of the value from one of our sublime cultural goods, and leave the actual producers scrambling for crumbs.
If Google thinks it's no big deal, why don't they just undertake the production of the content that they monetize? They won't, because it's a crap business, and only exists because the people who create music are at least in part driven by non-pecuniary rewards.
That's the way it goes, but spare us this attempt to paint this as a morality tale.
http://www.negativland.com/news/?page_id=17
TL;DR: If a record grosses $3 million, making $710,000 in profit, the band makes $12,000 (1.6%), in this case split between three people at that. Including "salary", since this is their sole payment for working their ass off for a full year.
http://www.salon.com/2000/06/14/love_7/
The argument is wrong.
I'm generalizing, but more or less here is the issue with record companies. Only a small number of albums released come close to making a profit. The vast majority of albums generate very little revenue, and don't even come close to covering the costs of producing the record.
So the business model of the record company is to fund a bunch of albums in the hope that one of them becomes a hit, and generates enough revenue to cover the production and marketing costs of all the others too. Whatever is left after that is the profit.
The contracts are structured to reflect this reality. i.e. very low royalty rates, like maybe 10% of the gross sale price per album. From this small royalty, it doesn't even kick in until the entire costs of the production and marketing have been "recouped" from this royalty!
The artist has an "account" at the label to keep track of how much they have recouped to date. Usually the costs hold over from project to project as well, so the artist can get deeper and deeper in the hole if they release a series of poor selling records. An artist can wind up a situation where they have a massive hit, but receive no royalties on sales as they haven't recouped earlier releases
As an example, let's look at a hypothetical album release. (This model doesn't apply as much anymore as the role of the album is much smaller in music sales, but the principle is the same).
let's say the cost of production and marketing the album is one million dollars, and the artist royalty rate is 50 cents per unit on a wholesale price of $5 per unit.
That means the artist won't see a penny of royalties on record sales until they sell two million units. After that they get their 50 cents per unit. The record label "advances" the funds to make the record to the artist.
While it does look like at first glance like such an agreement is horribly unfair, and can only be explained as exploitive, this is really the only viable business model possible if you want to have a business producing and releasing music.
Why would an artists sign such a contract, with so little hope of making money. The main reason is that if the record fails to recoup, once the contract is over they are not personally responsible for the debt. The record company pools the projects under a business umbrella, which allows the individual artists to pool the risk.
Here's another way to look at it. If you are an artist under a recording contract, you are guaranteed access to the capital you need to make an actual business of being a professional artist under a much less risky structure.
Only a few number of serious musical artist who might be able to fund the project through sources where profit is only possible if the project is commercially successful. If you had to, say take out a personal loan for $500,000 to make and market a record, given that the odds of success are incredibly low, that would be pretty much insanely risky.
The reality is that most musical artist never come close to creating a viable, marketable product, as the talent and capital required are too scarce. Even the few that manage to make a really good product independently do not have access to the capital and organizational resources to market it properly.
Another factor at play, is that there are other ways of making money as an artists outside of record royalties. Having a "record deal" gives you much greater access to those opportunities. For example, there what's called writer's royalties which are connected to the copyright on the composition, not the recording. These are not subject to the same agreement, and are paid on every sales. There are also opportuniti...
Specifically, my question is whether the world be better off if even relatively unknown musicians get access to high end equipment through open distribution and marketing afforded by the Internet. Music revenue would be split based on talent and not based on who is popular.
For example, someone puts out a new piece of music on the Internet. A set of music enthusiasts likes it and awareness spreads. The artist now signs up with a studio to produce a high quality version and sells it for money on Spotify.
I'm a music producer, of sorts, but I don't work in the music business per se, so my knowledge of actual costs is limited. I do occasionally work with people who are active in the industry, and so I try to at least keep up with what's going on.
(I usually work either for companies, or on low budget independent projects that don't make money, but are more products of love and passion:)
It's a little hard to get your mind around. The music industry proper is not that big, and what really goes on is not discussed in general forums very often. It's a bizarre little business world unto itself.
One thing to keep in mind is that the music market as an economic structure is an almost entirely "fat head, long tail." The vast amount of revenue is derived from the output of small number of artists (maybe 100 at any given time). The percentages are crazy. The percentage of recordings that even generate any significant revenue, never mind profit, is on the order of .001% (I'm guessing, but it's very low). There is definitely revenue that comes in from the long tail part of the market, but it's relatively small.
This "winner take all structure", which is inherent to a mass media form of entertainment, tends towards record labels being conservative about who they sign, and who they take a chance on promoting. If they do decide to seriously promote an artist, and support the whole business infrastructure needed to create a modern, mass media entertainment "product," they will bet big, and not hedge on production costs.
It's hard to understand in part because the way commercial music is actually produced has changed forms many times over the last 100 years, and the current model is relatively new, and not widely known by the public at all.
Currently the production model for commercial music is that a small team will be put together (or maintained) who work under a name "producer." (Dr. Luke being at the absolute top). These are they craftsmen/women who actually produce the music.
There will be writers, musicians, engineers, sub-producers, and associated creative and business types, who will all work together with the producer, artist, and label to create the actual song product you might hear on the radio.
There are often multiple team members working in parallel, in multiple studios, on the same project, with the best material generated being chosen for continued work. (Sort of the way a bigger technology company would have multiple teams on R&D).
Musical material from outside producers and writers is also be brought in under the production umbrella. For example, sometimes an artist and producer get some traction organically, like you describe. The producer can be brought on board a team to continue the work.
There are lots of independent entities who are producing musical "product" that are used as kind of a farm system for the major labels and producers. A lot of modern music is created from a "beat" which is really just the basic music of the song. There are actually various markets at work for "beats." that are always busy. There are a small number of independent "songwriters" who have a major contribution to the process.
Out of this process, which doesn't necessarily take a long time, weeks, maybe months, a marketable "record" (a single song) is hopefully produced.
I would estimate the costs of this production phase at $70-90 thousand.
From there, a label might be looking at roughly $1 million in marketing costs, videos, and the like. Radio promotion alone can cost hundreds of thousands (getting the song on the radio). Similar funds will be used to get the song visible wherever they can, Spotify, Youtube, anywhere fund...
YouTube is comparable to radio. A place people discover and listen to music, which they later buy as downloads, CDs, vinyl, etc.
Much YouTube music views are of official videos that the artists and labels themselves upload.
If anything, the record industry should be paying YouTube more for exposure and marketing. It's time for the record companies to stop freeloading off YouTube and pay their fair share for their use of this important music marketing channel.
People these days have many more means of entertainment. I think it is hard to get super popular by just doing music alone. Pairing with movie, games, or other multi-media seem to be a better strategy.