Kernochan Symposium 2024: Panel 3

Kernochan Symposium 2024: Panel 3

INTRO SPEAKER: So we're here for our third panel of the day on the music industry. This is another reminder that if you did not sign back in from lunch for CLE, please do so now, before we get started. And it's my pleasure to introduce Professor Shyam Balganesh, who's going to be moderating this panel.
MODERATOR: Great, well, welcome back. So we move from publishing and audiovisual content and photographs to the music industry, an area that is actually very well acquainted with licensing. The problem is perhaps too much licensing, that it's so complicated that it all needs to be taken apart. So we have a wonderful group of experts to talk to us about licensing in the music industry. But given the complexity of the content, our structure for this panel is going to be a little bit different.
Each of our speakers is going to speak for about 15 to 16 minutes, laying out in great detail the licensing structure within the industry. And then I will ask a series of questions and begin a conversation and opening it up. So to kick things off will be our first speaker, Elliott Peters.
[APPLAUSE]

ELLIOTT: Thank you. Very kind. And this works. Let's see, clicker, all right.
VIDEO PLAYBACK: Actually teamed up with YouTube.
ELLIOTT: I didn't mean to do it that fast. And now, how do I go back? There we go. You got a spoiler. So I want to talk about the music industry.
I joined in 2000. I was doing corporate law for five years before that. I did not take any copyright classes. I took a class with Professor Ginsburg on legal methods. And I got to take her final exam twice. So, yeah, thank you.
So Napster was quite a disruptive technology, so the music business that I stepped into might have been a little bit different than the music business that preceded it. We spent a lot of time trying to figure out how to address that disruptive technology, how to pivot, grow, transform our business rather quickly. We were a bit surprised.
And I noticed when I was going through my presentation, a lot of it-- there was a lot build up explaining where the music business was. And a lot of what I'm going to talk about were the new digital models and basically the death of the CD, and what we did very quickly to try to pivot and adapt and deal with uncertainties in copyright. But I think there are good lessons learned.
But because all the sizzle on the steak, and even in the steak itself is toward the end of the presentation, I thought I would share with you something recent and something that we've done in the past year with Google. We partnered with them around gen AI experiments called Dream Track.
Our company is really built on disruption. Our CEO is really about disruptive technologies and leveraging them, leaning into technology, not running away, but doing it in a thoughtful, judicious way to return more value to creators. And so, the first clip is the sizzle. And then I'll bore you with some detail. And then we'll get back into a little bit about what we did in 2000.
This is a clip from one of our artists, T-Pain. And he was part of the 2023 Dream Track experiments with gen AI. And he created a YouTube short that I think speaks for itself. It's kind of fun. But you'll get a sense of what the tools allow a musician, an artist-- recording artist to do. Will it play?
[VIDEO PLAYBACK]
T PAIN: - I actually teamed up with YouTube to do a project called Dream Tracks. So, look, here's what it looks like. Create a track. And these are the voices you can use. These are all the voices.
It's going to be crazy. You're going to love this. Here's what you do. You pick a voice at the bottom. Use a artist. There I am, right?
Now, Chat, what is this song going to be about? Describe what it's about. Buying new rims on Temu and loving them. So look. Look how quick this is. It wrote the lyrics out. Let's see what it sounds like.
[POP MUSIC] My rims on Temu
She said that she want to come and chill
I said, shawty, you can come and hang
If you make sure that my wheels is clean
And you wipe off that fingerprint
Cause I be looking at myself
When I be driving down the street
That's why my
[END PLAYBACK]
ELLIOTT: Oh, he has a really funny thing at the end. But just in the interest of time, we'll go on. But that's what we're doing today. Let's talk about 2000 and where the industry was.
I say, what's old is what's new. In 2000, I'd say that was probably one of the most disruptive technologies that the industry had faced. There are lots of disruptive technologies, but that was, up until that time, the mother of disruptive technologies, gen AI, I think is also disruptive. We don't know how disruptive.
We want to embrace the technology. I know a lot of what I'm going to talk about in the next couple of slides, talks about replacing formats and that phenomena. And that gen AI is sort of a brand new thing. It's a new paradigm in that all of a sudden there's technology that really relies on creation and questions of attribution and ownership. But I think there are lessons learned. So let's go to the retrospective.
In the music industry, it really enjoyed this phenomenon for over 100 years. And, I mean, it's great. You think about Thomas Edison coming up with a phonograph cylinder hundred-some years ago. I don't think a lot of people owned it. I don't think the commercialization was wide.
But it was the beginning. And with all of these formats, every epic period of time, they were replaced by a new format. And what you'll find is there's the advent of another format, the prior format declines.
So following the phonograph cylinder-- I always call them phonorolls-- you had the shellac record. After the shellac record, you had the vinyl record, you had the 45, and then you had the LP. The LP actually created the album market instead of moving away from the singles market. Interestingly, when we get back to the digital download and iTunes, we go back to the singles marketplace.
But the shellac record had limited life because it was a limited format. If you lived in a cold weather-- nobody here remembers the shellac record, I think. You may have them in your museum pieces. But if you lived in a cold weather state, shellac freezes and gets really brittle. So if you take it and you drop it, it's going to crack like a glass. And so vinyl was better.
But vinyl, if you live in warm weather, if you don't store it correctly, warps. And the sound is distorted. Then you move on to cassettes, the 8-track, the reel-to-reel, and you have to wait or rewind if you want to hear something again. Not terribly convenient.
Then you have the 8-track, which was a little niftier. I'm sorry, the compact cassette, which is a little niftier. But there's degradation of sound. But with these replacements, which leveraged new technologies, there is this expectation this nice, built-in feature in the recorded music industry in that.
And before I get into that nice feature, I just thought it would be interesting-- and I used ChatGPT, because I've been telling this story for two decades, and my belief in the music industry, where it's going, how it will always overcome adversity, how it's a good idea to leverage new technologies. But I didn't ever have to put it in a deck or a slide show. So ChatGPT told me and shared all the errors of my ways and my inconsistencies and inaccuracies.
This quote, when I put in, I said, tell me about the format replacement boom in music, spit out this wonderful blurb, which is crazy. I would have never come up with this. But one of the things that landed and they talk about in technological innovation, consumer behavior, nostalgia stuck with me. And I'll come back to that.
And it goes on to say that with this replacement, the changes reflected broader changes in music industry and society as a whole. Never would have come up with that. So if I think about the effect of buying new, replacing your collection every 10, 20, 30 years, whatever the period is, I think it's a windfall.
And the record industry really had become accustomed to just minting money every period of time. And it also masked a lot of issues in the music industry. Historically, I'd say beginning of time, you have IP, you have copyright, and you have the different uses.
And the two didn't really bear any relation to one another. There was so much money in the ecosystem, things didn't have to work. You had labels that would write big, big advances. They really didn't report. They didn't, regardless of recoupment. And nobody cared because there was just so much money in the ecosystem, you didn't have people complaining.
An example of this is with the advent of the '76 Copyright Act. You have the 115 Statutory License to automatically get a license for the musical composition in records. And you're supposed to get the permission before you actually distribute the record. But in modern day, and even in the past, Adele puts out a new record, and there are multiple songwriters, and you have producers on it, and everybody has a piece of the musical composition.
Splits, the ownership interest in that composition, are often not sorted out for years until after which the record's been distributed. That's an infringement. That is an unlicensed work. Nobody complained about it because, again, there's so much money in the ecosystem that changes.
And I think I touched on this. So I think this is the only other graphic I'm going to share. But I think this is a really fun graphic-- the next one-- because it shows the rise-- let me just go to it.
It goes from '73 to 2021. And it shows the rise, on the basis of revenues, of new formats with the similar concurrent decline in the prior format. And I think it's interesting to pay attention to the rise of the CD, its decline, when these things happened, and look at the concurrent increase in digital downloads, bearing a few things in mind.
An album, in the good old days, when people were really paying full boat for an album on a CD, would sell for $18.98. A full album on iTunes, a download iTunes launched in 2003, was 9.99. But people were really buying $0.99 tracks. So I will just play this, if I can. And I'll try to do a voiceover.
So we start in '73. Vinyl's pretty dominant. Please work.
There's another Play button right below it.
Oh.
Down below it when you move the cursor.
Oh, there we go. Thank you.
So vinyl's doing well. Here comes the cassette. We're at '79. Cassette overtakes vinyl in '83.
And oh, there comes the CD. The CD landed in '82. And it looks like it's got a future. And there it goes. There's no stopping it. What could take its place? What could happen?
1999, Napster invented. Oh, it's kind of slowing down. And we see downloads on the rise. That's 2003. iTunes launched their store in 2003. This does not reflect all the file sharing and downloads that were not paid for or monetized.
So it looks like it's great, but it really isn't. And I'll go to it in the next slide. I just think that's kind of fun.
Let's see. Back to here. And so the CD was the golden child of the industry. Folks were drinking the Kool-Aid. It was great.
There were CD clubs. You pay a penny, you get 12 CDs. You just have to keep subscribing. My company owned half of one of those things. And I got to see what was under the hood.
I can't share any of it with you. It's mind blowing how significant and how much money there was with that replacement. And I think folks just forgot that they needed to create a new format, that the CD would not last forever. There are lots of tests with digital downloads.
There's one called the Manhattan Project. Was it? No, that was not the Manhattan Project. It was a project. It was not-- that would be something.
But there was a project in 1998. So there was limited testing done. But we were just caught flat-footed. And we were caught flat-footed by a few things. See if I forgot any of that.
Yeah, why did the CD crash and burn? Mainly because it was not copy protected. There is a consortium that came up with standards for the DVD. They decided to copy protect it.
The CD wasn't copy protected. It was shiny. It was fun. They're shipping it. It was fast. People just saw dollar signs and forgot about the next phase.
They also didn't see technologies coming-- rip, mix, burn. Apple, '97. '99, you have now small file sizes because you can rip, you can compress these small files relative to the DVD. And you have Napster that makes it easy to share.
Everything is free. Gosh, OK. I'm going to really pick it up. So the effect, 2000, sales were $30 billion. 2010, sales were $15 billion.
Record companies tried to enforce. They sued citizens, P2P. We worked hard to compete with free, create new services.
Here's a little bit of a primer on-- I don't know if people need it. I'm going to go through this really quickly because I have a lot of examples I hope I might get to. Musical compositions are the music notes.
And they're the words written by writers, administered by publishers. Sound recordings are the embodiment of musical compositions, administered by record labels, sometimes owned by the record labels, often. DSPs are Digital Service Providers, like Apple Music, Spotify, YouTube, TikTok.
Here's an example. Dolly Parton wrote, I Will Always Love You. She also recorded it, so she has it on the record side and the publishing. Whitney, a singer, an artist, recorded it and put that I Will Always Love You on the Bodyguard soundtrack in '92.
There was a lot of loss in terms of revenues for the record companies. This was the Great Reset for the creators. Publishers historically have not gotten their due. I will be the first to admit-- I work at a music company, not just a record company. We have music publishing, as well. And they took this opportunity and the uncertainty around rights to hold their ground and really reset the table in the 20 years, 20-some years, their revenue earnings have gone up two- to five-fold.
And I have to point out, this is different, as I did before. This shift is different. This isn't just a format shift. This is a paradigm shift in terms of creation and ownership. I'm not going anywhere near anything specific on gen AI other than to share what I just did around the testing with Google. And there's a version 2 of that recently announced. We're in other conversations. And we've got a long way to go.
Perfect storm, we needed to develop new models fast. Lots of questions were raised as we tried to do this. We tried to create new formats. Now I'm just going to jump around.
One was the ringback tone and one was a multi-session disk. Multi-session disk leveraged the DVD platform. There was more storage space, you could have multiple instances of the same sound recording on it, all these virtues, but the problem was, all of a sudden, you have five mechanicals that are due on a disk.
And so with the ringback tone, it wasn't clear how to license the publishing rights associated with that. And so, in this instance, industries got together. Really, just the majors, the majors that had recorded music in-house, music publishing, and they cross-licensed each other to fix these problems.
So this is industry coming together. Not entirely accessible to everybody else. It helped the majors, created precedence, but not 100% efficient.
There's questions around rights implicated in-- rights associated with a stream. End of the day, we turned out there is a reproduction right. That was through industry negotiation settlement and the interactive streaming settlement, which became part of the regs associated with 115 licensing. I will not go through the case study of trying to send notices of intention to obtain compulsory licenses to clear content for the first subscription service, MusicNet. But it was challenging.
There are other instances where Congress got involved. Barbara Boxer and Orrin Hatch dragged a bunch of us on both sides of the Digital Media Association and the RIAA to figure out interactivity as it related to internet radio services. And internet radio service, the intention was that it's listening to radio, but in a non interactive way, like riding in your car, you don't like a song, you turn the channel. But it was much more than that.
We couldn't agree. It was fun seeing DC for four or five days. That ended up getting settled in a lawsuit. Wrong outcome. Litigation isn't always great. And I'm going to be wrapping it up soon.
So if I had to summarize everything, we tried litigation, we tried to re -- voluntary licensing, enforcement. We tried to renegotiate with YouTube and take all of our content down and go dark. Impossible. We ended up going gray. The safe harbors that protect YouTube really don't balance the rights of the owner and the service. But in short, we were caught behind the eight ball. We did everything we could to catch up very, very quickly.
And again, I showed you the T-Pain clip at the top. That's an example of collaboration. YouTube has gotten smart and learned it's better to try to agree up front rather than foot fault or do more than that as a transgression and beg forgiveness.
Here's an example of actors that are taking a different approach. And this is Suno. And this is some back and forth and some love language in the industry. In the industry, the language of love is litigation, unfortunately. And I don't think that's always optimal.
I got nothing-- oh, it's not loading, so I think that's a sign that I should wrap it up. But I do think, fundamentally, that things will work out. Music is messy. The intersection of technology and IP, where it doesn't work, isn't the sky isn't falling, but it's an opportunity.
We have to be patient. This is just part of the iterative process. Fundamentally, I always believe people like music.
If technologies take away creators' ability to make a living, music goes away. It's self-correcting. I prefer not to crash and burn and hit the ground, but to do it voluntarily and stay ahead of things. But these slides are available. There's more in there, fluff or not. But they're available. Thanks for your time.
[APPLAUSE]

MODERATOR: Great, our next speaker is Lidia Kim. I'm just going to transition her slides.
LIDIA: Thank you. Is it this one?
MODERATOR: Yeah.
All right, [INAUDIBLE] short, OK.
LIDIA: Hi, everyone. My name is Lidia Kim. And I work at Concord Music Publishing, which is the Music Publishing Business Division at Concord. In my role at Concord, I oversee the digital affairs for the company.
So this typically involves negotiating blanket license deals with the digital services, some rights enforcement matters when we're dealing with services that aren't cooperative, or are massive infringers of copyrights. And so, my perspective is from that of a rights holder or music publisher and concerned with revenue generating and value protection for our songwriters and our own copyrights. This role requires an understanding of how global licensing networks work generally. And we work closely with the various societies and management organizations in order to ensure that we are capturing the revenue that is being generated for our copyright.
So I'm just going to go over a brief overview. I assume everyone knows copyright law. But in music publishing, we do operate in the full bundle of rights, so reproduction, making copies of sound recordings, for example, preparation of derivative works, so the sound recording itself is a creation of a derivative work of the musical composition, and then, distribution of said copies. Public performance, which may come in the form of live concerts or radio or in streaming, which implicates both the mechanical and performance rights. And then, public display, if you see lyrics in a karaoke setting that you are seeing the implication of music publishing rights.
But in the music publishing industry, we operate a lot on business standards. So I think of this in terms of major streams of income. So if we're looking at mechanical income, we're looking again at reproduction and distribution.
For performance income, we're looking at public performance, grant rights, or the performance of music in a dramatic context. For synchronization, we're talking about commercials, film, TV, video games, and nowadays, maybe some social media-created videos. And then, the other source of income is, for example, print, sheet music.
And then, the music publisher's role in this music industry ecosystem, while it includes A&R and creative services, at its core, music publishing is a licensing and administration service business. So publishers license and monitor the use of works and will enforce our rights either through cease and desist letters or takedown notices via the DMCA. We resort to litigation where it is necessary. We publishers also register the copyrights in the works and manage the registration of the represented catalog in various societies.
So the MLC, the Mechanical Licensing Collective, is one that manages mechanical rights. And so we need to produce or deliver metadata that shows these are the compositions we represent, these are all the details related to it to be able to track, claim, and collect revenue on it. So we then claim and collect the income from there, which can be very data driven, resource heavy. We have internal teams that have to go through and work with the services or the admin service providers to ensure that the metadata is clean in order to actually capture that revenue. And then, once we collect the income, we then process and pay out to our songwriter clients and other rights holders.
Typically, for musical works, there are multiple rights holders, there are multiple writers and producers in a session. They may all control a songwriter interest. And then, they may all have different publishers that they work with to help administer their rights. So we then end up with a number of stakeholders on a per song basis. So that can get complicated very quickly.
So when we think about licensing, what we consider going into these negotiations is to ensure that a service is providing a certain level of what I call the four tenets of licensing. So we want to be able to have consent over whether or not our music is used or a songwriter's music is used. Credit, which can come in the form of attribution of who wrote the song and then when it's being used in a particular work. So in maybe film and TV, if you actually stay in the movie theater and get to the end credits, you'll see a section that's on music licensing and who wrote what song and who the recorded artists are. And then, proper compensation, which we ideally negotiate at a fair market value. And then, transparency of how and when that music is used.
So in the music publishing industry, there is a mechanism in place to negotiate license fees directly and partner with trusted tech and admin service providers to manage those rights properly. So I want to speak a bit about the blanket licensing regimes. So this is accomplished in a few ways, either by direct licensing or through societies or collective management organizations. So we have compulsory mechanical licenses and performing rights organization licenses, which are handled on a collective basis.
In direct licensing, we, of course, have the ability to license on an individual basis, talking about a specific TV program or a specific film or a specific commercial. But I'm going to focus on blanket licensing. So for blanket licensing with digital services, we need to look at the nature of usage, which can include user-generated content or programmed music use.
The idea here is that a service may be using a ton of music. And it would be too onerous to issue individual licenses for each music use. And so we look at a sort of blanket agreement that can cover all of the contemplated uses and properly monetize it.
So the trick here is to be able to identify the use and then provide the appropriate value for said use. These deals often have a section of the contract. And this usually requires a level of discussion on how rights are managed on those services.
For example, if we see something like Peloton, they're curating programs so they know what music they're using exactly. So they just report on that usage. But if we're talking about a YouTube or a TikTok, there isn't a as clear mechanism to understand exactly what music is being used. And so we need to make sure that there's a rights management policy in place and a mechanism, which can include engaging audio identification tools through a tech vendor, or if you're a big enough service, you build that yourself, the tools.
And then, a rights administration mechanism, which, again, they can build themselves. If you're a YouTube, you have-- I believe someone brought this up earlier, but there's a content manager system. And that's sort of an in-house, internal built system that rights holders can then access.
But if you're not a YouTube and you need to manage your rights, you would then engage as a third-party administration service, such as Music Reports or Harry Fox Agency or otherwise, in order to get that administration done correctly. And that lends to how accounting and reporting is achieved.
So then, when we think about direct licensing, we also, as I alluded to, have to deal with these online service providers that may not necessarily have a clear, direct music use, but their end users use a lot of music. They may provide a library of music for content creators to attach to the social video or what have you that they create. But they may also-- sorry, I got distracted by that timing thing.
But where was I? Music library-- oh, or they may provide some music-based services. Some of these services are just pure UGC platforms. And so, then the question becomes whether or not there is a reasonable basis to license the activity on the platform. And then pure UGC user-generated content may also provide some reasonable protections under the DMCA.
So a lot of times these services hide behind the safe harbor provision of the Digital Millennium Copyright Act, saying that they are not held liable for the activities of their users. And as long as they have a reasonable notice and takedown policy in place and they comply with it, they should not be held liable. This has been a bit confusing and litigated to various degrees of success.
So then, I also wanted to focus on the performing rights organizations, which when we issue direct licenses, it usually excludes the public performance licenses. Those are handled by the PROs. There are four PROs in the United States. ASCAP and BMI are subject to antitrust consent decrees overseen by the US Department of Justice.
And SESAC and GMR are invitation-only. And by comparison, ASCAP and BMI cover the PRO licenses for the majority of the performances in the US. And so SESAC and GMR cover a much smaller market share.
PROs issue non-exclusive licenses for public performances. They are commonly on a whole repertoire basis. The licensees include radio stations, television, streaming services, et cetera. This is for the public performance right only. And as a reminder, for digital streaming, we have a mechanical component and a public performance component.
The way that PROs work, songwriters and publishers will enter affiliation agreements with a PRO. Songwriters typically choose one society to affiliate with, whereas publishers can affiliate with multiple. Usually the publishers want to have a complement society to be able to collect the rights on behalf of their songwriters-- or the royalties on behalf of their songwriters.
The way that royalties work, so PROs will collect income. That income is split between a writer share and publisher share, which is usually 50/50 split. And then, the publisher's share is paid out to the publisher, writer's share directly to the songwriter.
Because PROs grant performance rights on a non-exclusive basis, copyright owners also retain the right to engage in direct licensing with users. But there has been some question about withdrawal of full scopes of rights, for example, digital rights. So this is more of a limited scope basis.
The ASCAP and BMI are, as I mentioned before, subject to consent decrees. These were originally entered in 1941 to protect licensees from anti-competitive practices and the arising from the market power each organization acquired through the aggregation of public performance rights held by their members, songwriters, and music publishers. The DOJ Antitrust Division reviews the decrees every five years for the operation and effectiveness.
Consent decrees require ASCAP and BMI to open membership to anyone who wishes to be a member and meet the minimum requirements as a songwriter or publisher. They only issue performance licenses on a blanket and non-exclusive basis. And the PROs have to agree to reasonable rates and terms on a non-discriminatory basis among similarly situated licensees. So it's a set rate, either a set flat fee rate or a set royalty share structure that occurs. If the PRO and a licensee cannot agree to a fee, the licensee may petition the rate court to set a reasonable fee.
The last updates to the decrees were in 1994 for BMI and 2001 for ASCAP. And the last review began in 2019, reviewed by the DOJ, and ended in January 2021 without any proposals for reform. There was some discussion about how it was too complicated and that there was a pandemic, and so there wasn't any advice there provided by the DOJ. And something to note here is that SESAC and GMR are US PROs that are not subject to consent decrees. So they are not regulated. And they can operate outside of the scope of the consent decrees that cover the ASCAP and BMI operations.
In comparison, there are CMOs outside of the United States that also operate in representing music publishing rights. Those societies cover local performances. And they may also include digital mechanical rights in streaming. Rights are often mandated by the writer to the society.
And publishers have the right to collect income via agreement with the writer, which is different from the way that it works in the US, where the publisher is the one granting the performance rights in the affiliation agreements with the PROs. Reciprocal agreements between the societies ensure the flow of royalties to the appropriate rights holders where we're talking about local societies that represent writers and their economic interests globally. They'll collect the local society income from said societies and then pay it out to the writer.
And then, in addition to that, there are multi-territory licensing hubs in place that have agreements with societies and publishers to cover digital rights for an economic territory. So, for example, in Latin America we have LatinAutor, which represents the mechanical and the digital licensing for most of the territories in South America or Latin America, excluding Brazil and Argentina and Mexico. And then they will pay the mechanical rights to the publishers and then the performing rights to the appropriate local societies.
And I just wanted to just say that so the music publishing industry isn't a stranger to these large-scale licensing schemes. And so we have had to address through various technological developments how to manage a high volume of copyrights and still find a way to negotiate a fair market value. And so that is something that we'll have to continue to figure out how to develop and grow as technology advances.
[APPLAUSE]

MODERATOR: OK, we next have Judge Steve Ruwe who is going to--
[SIDE CONVERSATION]

JUDGE RUWE: Steve Ruwe, with the Copyright Royalty Board, I didn't mean to advance. So I have to keep a light touch on my thumb. I'm going to just give an overview of the statutory licenses that the CRB administers, a history of rate setting in the US, and close on the current structure and the process that the CRB operates under.
First, I'll go back. Other people have addressed this before. Statutory license, license created by operation of law as opposed to by contract or agreement by the parties. It is also distinct from the blanket license that Lidia was addressing, which are by authority of the copyright owner, ultimately.
These are the current licenses that the CRB administers. And there's quite a few of them. I'm not going to go into detail of them now, maybe a little later, depending on how the timing goes here. But we have rate proceedings and distribution proceedings.
One of the important things about that is that's-- well, a rate proceeding is when we're deciding what rate to set. And these are robustly argued by stakeholders. Distribution proceedings was, I believe, alluded to in the previous panel, where the cable and satellite companies are entitled to retransmit certain broadcast signals. And they are obligated to pay royalty fees into a pool.
It is then the CRB's obligation to distribute those monies. And we do that through an allocation phase and then a distribution phase. So in the first phase, the copyright owners would generally get together in groups of similar interests, similar sort of colloquial categories of audiovisual content and musical works, so sports programming and commercial producers, which would be syndicated second run stuff. But the notion is the cable and satellite companies don't know what is going to be on the broadcast signal, so they need a blanket license to be able to retransmit without knowing what exactly they're retransmitting at the time.
The individual copyright owner groups argue for, in essence, a slice of the pie. And then, that's the allocation. Then, a later proceeding, it gets down to who actual individual copyright owners make claims within that slice.
A little bit less intuitive than the rate proceedings, I would gather. So that's why that took a little longer to explain. The other item in here, administrative assessment for the MLC is funding for this newly created collective for the mechanical rights, the 115 license that was recently modified by the Music Modernization Act.
A little bit of history of how we got here. So music compositions protected very early on. I'm going to jump to the end where mechanical reproductions were not considered copies. So there's no right to be infringed at that point. Then Congress creates the right subject to a mechanical license, and at the same time sets a $0.02 rate per use in 1909 all the way to 1976, when it was increased to 2.75 cents.
So leading up to the '76 Act, there was a motivation to consider adjustments of these rates. And that brings forward the concept of, well, you need to have a tribunal to make those adjustments. That was the Copyright Royalty Tribunal.
It was five commissioners appointed by the President, confirmed by the Senate. They were adjusting rates for three licenses. I don't believe that the cable and satellite were under them at that moment in time, but soon would be.
Another license-- actually, it wasn't included on there. No, DART was not. So that's another license. So they're getting more work to do. But there was criticism that there's five people, there's not that many licenses. Maybe we need to reconsider it. It was confined down to three members. But there were still some complaints, as I noted in here. Yes, the 119 was actually by a separate panel, an arbitration panel.
Well, that gives rise to-- as they were concerned about the inefficiencies, too much staff, not enough work for the CRT. And there was another arbitration panel set up for the 119 license. That brings about the Copyright Arbitration Royalty Panel in '93.
And that was an ad hoc system. So there was not a lot of predictability as to what arbitrators you were going to have in a given proceeding. But I want to note the involvement of the librarian and the register of copyrights through the librarian had a lot of input on the CARP decisions, would modify them, reviewed them for modification, and I don't know the percentages, but there are some notable modifications made by the librarian through recommendation of the register.
And that is the register's role in rate setting continues to be a profound one with regard to substantive copyright law, not about the rate settings and the fact finding itself, but the contours of the license are occasionally reviewed by the register. Or routinely reviewed, occasionally, she voices a concern about the direction.
During this time of the Copyright Arbitration Royalty Panel, we did get more licenses. There's the criticisms of the Copyright Arbitration Royalty Panel is, not mine, but what I've read to understand, the cost, the lack of consistency, institutional expertise, because the ad hoc nature of this arbitration panel for each new proceeding.
Additional licenses are taking care of that concern of not having enough work to do that the CRT had. Establishment of the 114 licenses, that is for non interactive use of sound recordings. And prior to '95 the sound recording performance right was very limited, if non-existent, I guess. It was established at that point, along with the establishment of the right was the statutory license.
That's similar to what I said with 115. It's when the right is recognized that it's also encumbered by this license. That's not always the case. The middle column talks about 112.
Well, there was always a reproduction right, but it became a little bit different in how the marketplace needed to deal with it in the digital world. So 112 allows server copies for uses that are either exempted under 114 or have a statutory license under 114. The 2004 Copyright Royalty Distribution Reform Act establishes the Copyright Royalty Judges replacing the CARP system.
So again, the licenses that we administer, I'm going to go through which ones I didn't address already in some sense. And the coin operated phonorecord players, not a huge deal for anyone. Public Broadcasting, this is a pretty confined statutory license. The CARP-- So let's see, the DART I've addressed to some degree.
What's notable is how we separate the proceedings out. So it's not just one 114 proceeding. There are different categories of licensees and licenses under 114. 115 works a little different. It is one rate setting. And the rate settings end up, often by settlement, but also by determination, designating specific uses that are subject to different rate setting structures.
What's notable about 114 and 115, I alluded to the administrative assessment for the central collective that receives and then pays out the 115 royalties. There is also a collective that is set by the CRB-- designated, I should say, by the CRB at each of those separate 114 rate proceedings. It has been Sound Exchange that has been designated consistently. They operate a similar role as the MLC. Analogous, I should say.
The CRB designates Sound Exchange on the 114 end. The Register of Copyrights makes the designation of the MLC who serves this, again, collective role. Another thing that's worth mentioning about the 114 license, the license established at the same time as the right being more broadly established.
And you'll see, the revenues do not go solely to the copyright owners. The artists do basically get a 50% share. Two bundles of those goes to unions. Those are the 2.5% for non-featured musicians and vocalists. That's set by statute is that the 2.5% goes in that direction, as with the 45/50 split.
Again, if I didn't cover-- one notable thing here that I'd say I've addressed some of how we divide up the license proceedings under 114. Business Establishment Services is somewhat unique as they are not obtaining a 114 license. They're exempt from 114, the public performance end of the license. But they do need to pay for the server copies, the 112 portion.
Again, just another look at all the licenses that we are dealing with. Excuse me. I've addressed the non rate proceedings and how they-- whether I got to those up front or in the back end, I guess, I got them on the up front portion.
What is the CRB? They are three judges. I went in order of when they were named. They have three main categories. There's a chief judge, and a judge with economic experience, or significant knowledge of economics. I shouldn't shortchange anyone. And I just said copyright experience. I'll keep it at that and not go with the statutory terms.
How does it work? The CRB announces commencement of a proceeding. Generally, the lists of various proceedings are on a five-year cycle. This is a general matter. That's how we keep up to date. That's the congressional directive.
There have been proposals for additional licenses, which have also brought up concerns about whether you extend or how do you maybe accommodate more license proceedings occurring in a tighter schedule. We do allow-- very open as far as participation. Any interested party may participate. If you need to be an active participant for the most part, there are motions to dismiss some participants that happens where we grant the dismissals.
But more often than not, I think most participants have been able to sustain their interested party status. Participation is important. It can have significant impacts as to the ability for someone to voice their opinion on a settlement. Some of the larger parties may come to a settlement. And there can still be interested parties who retain the opportunity to object and have their voices heard about that. We also do allow pro se participation.
We very strongly-- the statute, I should say, very strongly encourages settlement. There are some obligations with regard to settlement. We do need to consider, as I said, participants, interested parties, established participants.
In the absence of a settlement, we have a robust record that's developed to set a rate. We have extensive hearings. The, for a given proceeding, rate proceeding, four to five weeks is not unusual for in-person hearings, four days a week. It's a very in-depth process. We develop a significant record. We do have a lot of restricted material in the records. But we do take great efforts to try to have a final determination that is understandable without exposing access to some of the sensitive marketplace information.
I mentioned settlement before. The statute definitely encourages it. But there are some obligations and obstacles to a settlement becoming enshrined in regulations and generally applicable. And one of the key ones was objectors. If their participant objects, that is a key requirement for a broader ability to scrutinize a settlement.
We always have the obligation to keep any settlement within the confines of the statute. So occasionally, there have been instances where a settlement might push-- alleged to allow some uses that are not strictly within the confines of the license. And it has happened that those have been corrected. And therefore, through a number of other steps, they don't become enshrined in the regulations for the five-year period.
I mentioned the Register of Copyrights role. She has a role to review for legal error any of our determinations, whether they be a settlement or a determination that is fully written by the three judges. And that correction is something that it doesn't automatically correct what the judges have made in their determination, but there are avenues for us to correct something in keeping with a legal review that the register has issued.
The other aspect of this is the review by appeal. And a lot of our determinations as a matter of percentages are appealed to the DC Circuit. We're doing pretty good on being sustained. But we have had some remands recently. And as people are considering various licensing models, this is the whole purpose of just giving a background of the way things have worked and the way things work.
One of the consequences of these various review procedures, be it from the register or from the DC Circuit, is that we can get work sent back to us, and when you have the five-year schedule, that things can get interrupted as a matter of scheduling. I mentioned there's concerns about the various structures. That's been one that has been raised for us recently. But that's pretty much my introduction to how the licensing system works for us, the statutory licensing system.
Thank you.
[APPLAUSE]

MODERATOR: Right, next, we have Joe Keeley to talk to us about the MMA.
JOE: Thank you.
Good afternoon. My name is Joe Keeley. I recently left Capitol Hill about a year and a half ago, was the former Chief IP Counsel on the House of Representatives for Bob Goodlatte during our big Copyright Review when the Music Modernization Act was adopted. I was also the Chief IP Counsel for the Senate for a rather camera shy, Twitter shy member of Congress by the name of Lindsey Graham. You probably never heard of him.
So anyway, the reason I'm up here today, I want to talk about the MMA, is Congress tends to, quite frankly, repeat itself. They look at what they've done before as a template for what they could do in the future. Sometimes that's a good idea. Quite frankly, sometimes it's not. But that's kind of the go to mode for Congress.
Have we done something in this space? Can we use this as a template for some other future activity down the road? So the biggest, or the most recent, excuse me, compulsory license was the MMA. And there are some reasons, perhaps, to look at that as a template for some activity related to AI licensing. But there's plenty of reasons, perhaps, not to look at it. And I just kind of want to walk through those.
I'll say at the outset that even if your particular interest in a copyrighted work is well represented by a license or you're a member of a PRO or you're a member of MPA or IA or whatever, they will represent you well. And entities that want to license those works know that if they don't license the work, they're going to get sued. So there's an incentive-- it hasn't always been the case, obviously, with many of these AI companies, to reach out in advance to undertake settlements.
But the reality is, when you exclude all of those well-represented folks, there are other creators of content, namely every individual with an iPhone or a phone in general, who maybe they post a blog about, I don't know, living in the Upper West Side of New York or their favorite recipe or whatever. And at the end of the day, licensing, in many cases, is about price. So I assume that, for the most part, the New York Times and the Wall Street Journal, for instance, they probably cover many of the same stories. And I think most people would think of them as fairly equally accurate and up to speed on what they're doing.
My guess is, from a licensing perspective, if one of these AI companies or anyone else was looking at licensing a news source, and let's just say the New York Times charged 10 times as much as the Wall Street Journal, my guess is most companies would go with the cheaper option just because it's 10 times less.
Now, I don't know if that's actually the case or not. But the point is, we're in a market-driven capitalist society. And price is very important, both to those who create copyrighted works and those who use copyrighted works. So if you're an AI company, one, you don't want to have to spend a lot on copyrighted content. That's just the way the market works.
And you might only want to license the Wall Street Journal over the New York Times or vice versa. You might only want to license a certain type of content or a certain source. But at the end of the day, there are folks out there that are creating content for free using their phones to blog about life in New York, their favorite recipe, or whatever.
So there are perhaps areas for compulsory licenses for AI that are not already represented by other parties. So I just want to start with that. And make sure I'm doing this clicker right. And there we go.
So I'm going to make some assumptions today. Congressional sausage making isn't always pretty. I think that's pretty obvious. Direct licensing, very much preferred, but not always possible or feasible. I think most people believe, and I agree, that courts will determine, at some point, that some or perhaps all AI training on works subject to copyright protection will require a license.
And again, many of those works will be represented by entities that can license those works. But again, not all. Litigation as well as licensing negotiation costs become a financial drag for both a subset of copyright owners and AI companies, i.e. lawyers don't work for free.
That, quite frankly, was a huge motivation for folks to enter into negotiations for the MMA. Music companies, beyond being involved in the music business, were in many cases just as involved in the litigation business. They hired a lot of attorneys. They all came to see me and complain one way or the other and complained about each other.
So my view, just having worked up there for a number of years, Congress is going to get asked to step into this at some level. And they'll look at what they've done before. They'll be asked to step in, particularly where they're not forcing a system on parties who don't want it.
So in other words, they're not going to force the New York Times to license their work at a certain price. They're just not going to step into that. But there will be areas where there aren't folks representing copyright owners where there will be effectively a need for someone to represent them.
And Congress has been-- typically, they address narrow policy issues rather than sweeping rights. So I'm not suggesting anything like this is going to go fast. The 1996 TeleCo Act took quite a long time. And they're actually still debating how to rewrite the 1872 Mining Law. So that gives you a sense of how long things can take.
One of the big challenges, though, that the MMA faced and a lot of other industries face, as well as what Congress wants to act on, is the quality of the ownership databases, so who to go to, who to pay, who to find, how do you find-- if Steve writes a blog about Upper West Side life, and I want to intake it, am I going to engage in a direct negotiation with him, never mind find him.
And how much is that worth? Is that worth a few pennies? Might only be worth less than $1. Sorry to insult you there, Steve. But you get the point, which is there's a high negotiating cost. So those are some of the assumptions I'm starting out with here.
So the MMA concept worked. And there are obviously some key differences. So as I mentioned before, it's the most recent creation of a compulsory license. There was a multi-party negotiation effort. And there were some things that were different then. There was $400 million in back royalties that were basically sitting accrued by music services that hadn't been paid out.
So there was an incentive built in to create a system to basically transfer that money. Until then, the money was just sitting effectively in limbo in various music services. Obviously, not the case with AI companies who really haven't collected anything.
There were disagreements on who and how much to pay, but there was a fundamental agreement that someone had to get paid in the music industry. There might have been disputes among which PRO, which publisher or whatever. But at the end of the day, everyone recognized music creators, music copyright owners deserve to get paid. Not obviously, the case here with some of the AI companies that view fair use as their ultimate trump card. So there are some differences there.
There had been a long history of compulsory licensing in the music space, all the way back to the 1909 Act. So the notion of updating a copyright compulsory license was not foreign. People knew some version of a license, had been around for a while. The people around the table that were negotiating the legislation all knew each other.
There was a lot of and forth in the music industry. So for four years, you might be at a label, five years later, you might be at a music service, then you're going to go-- there's just people moving back and forth, so you kind of knew the person across the table. Definitely not the case now with AI companies.
Also, for better or worse, the music industry has been a little bit more tolerant of close enough than others. So the databases have never really been perfect in the music industry. Kind of figure it out as we go along. Got to push the new work out the door. We'll figure out some of the problems on the back end and we'll make it up a couple months later.
That's been tolerated for a while. There's nothing necessarily wrong with that. That was challenging for some of the more tech driven companies-- Apple, Spotify, that kind of thing like that. There were two reasons that the MMA time was right effectively.
There had been a range of court decisions that, at some level, every party disagreed with. And they wanted a solution to resolve them. That's clearly not the case here. Obviously, there's plenty of cases underway.
But there haven't been many decisions, or very few. There's also, as I mentioned, a willingness for parties to talk and negotiate with each other. In the MMA, music parties were, again, I wouldn't say friends, but they all knew each other, they were willing to talk and communicate with each other. Definitely not the case today.
So if you look at the MMA as a solution to AI licensing, as I mentioned, I'm excluding entities already represented or copyright owners already represented. The MLC is a central nonprofit entity that issues licensing and also receives and distributes funds. They're based in Nashville.
Their operations are, I wouldn't say directly overseen by the Copyright Office, but the Copyright Office has some oversight authority. The funds for the operation of the MLC are overseen by the Copyright Royalty Board. So it's a public-private-ish, I guess you'd kind of say. But to the extent you want to look at the MMA, you could see that some new entity being created for a cold cloth kind of thing, that entity could be the ultimate distributor to unrepresented copyright owners, the bloggers blogging about the West Side or whatever.
As an alternative, the entity could just be a pass through of passing funds to-- you've put up photos of your ski trip online. Well, maybe it should go to a visual artist group because they represent other visual artists, that kind of thing. The choice also may depend upon the quality of the relevant database.
And again, databases these days are poor. They're getting better. Metadata is also getting better, but it's still up in the air. Oddly enough, you might be familiar with a service known as Task Rabbit where you can hire folks. There are services like that online.
And literally, there are thousands of people employed in third-world countries for pennies per hour adding metadata to photos because the income demands are so much lower that you have services or bidders, I guess you'd say, providing folks just to go through these photos or other websites, whatever, adding metadata to it. So databases are key.
And also, as part of a potential MMA solution, someone's got to figure out what's the appropriate rate for that blogger and how many words they put up or how many photos they want. I can tell you, Congress isn't going to get involved. They're going to hand it off to some other entity. And I know exactly who that entity is going to be because they don't want to deal with it.
So again, you could see something like the MMA, I'm not going to say it's perfect, but in certain areas you could certainly see that. But there's challenges galore. There's nuances of existing negotiated licenses you heard earlier today. You may not want to be an artist whose work is used in a political advertisement or pornography or whatever the case.
You wouldn't have the ability really to exclude that in a congressionally directed compulsory license. Or maybe they create multiple versions, one that allows political usage and one that doesn't. But there are some nuances that you really can't get in some of these things.
What's worth more, what's worth less, is a nonprofit 600-page textbook, is that worth more than a 600-page Harlequin romance? I don't know. Someone's got to figure that out. I know just the guy. He's right here. What's an appropriate minimum payment?
Let's say I'm the author of that blog about the Upper West Side. I can guarantee you that whatever I come up with is not worth a lot of money. Certainly, in the AI ingestion content space or that world, it might be worth pennies. One of the challenges the MMA has faced, the MLC in particular, is there's a lot of overhead costs. Just the notion of sending a check out the door, that costs money.
Now, granted, these days, you do electronic transfer. But if you think in the old days, you used a postage stamp. You could see a lot of royalties that are due folks be less than the cost of a proverbial postage stamp. So do you create a minimum amount where, yeah, if you've generated more than $10, you're owed money, but less, you're not. I don't know.
There's no back royalties waiting to be distributed like there were with MMA. So there's less pressure to come up with a solution. There's, as you heard, plenty of differences among copyright owners. The size of the table is huge. There's a vast number of interested parties, different views, different opinions, different usage models, different licensing models.
A few members of Congress at the time with MMA cared about section 115 reform. Some members defer to the expertise of a handful of members and staff, which really isn't the case anymore. They all hate each other and scream at each other, so they all have their opinions or whatever. And if enough large parties reach their own license agreement, the motivation for a universal solution decreases.
I mean, if you're only left with crumbs left that are unlicensed, is it worth creating a new MMA-like entity? But again, as I said earlier, I think there's a reason maybe something should be created, because in essence, not that they should be free, but they are effectively free because that blogger about the Upper West Side isn't going to sue an AI company for ingestion of their blog. They're not even going to know about it.
So you're in essence pushing companies to use free in the sense that they're not going to get sued if they use them. That's not always the case. Again, your news creator. But maybe you can't use New York Times, the Wall Street Journal, you're going to use, I don't know, the Tampa Tribune or whatever. Maybe their price is less. Who knows. And that's all I got. But I'm happy to answer any questions with the rest of the panel. Thank you.
[APPLAUSE]

MODERATOR: OK, so we have about 20 minutes left. And I will say those were some very rich presentations. So thank you all for the level of detail and complexity. And I get a sense now why people are scared of licensing.
But this leads me to a question that I was wondering if each of you could reflect on a little bit. And it ties into all of your presentations. Each of you has given us a little bit of the history of how we got to the current regimes that you're seeing in the music industry.
I guess the question that I'll begin by asking is a fairly obvious one, given that we are comparing the different creative industries in this program. How is it or why is it or what was it specifically about the music industry that allowed for it to be regulated through this form of licensing, the various forms of licensing that you're talking about-- blanket, compulsory-- where, in other domains, you're seeing the complete absence of licensing, you're seeing free market, voluntary licensing. But what was it, in terms of the political economy of the music industry, that historically led it to go down the road of licensing, which made it something of an exception within the copyright world?
JOE: I'll take that.
MODERATOR: Please, Joe.
[SIDE CONVERSATION]

JOE: How's that now?
MODERATOR: Yeah, good.
JOE: All right, so I think, to answer to your question, music was different in the sense that when there were the proverbial piano rolls, there was a need to license them. This was not only pre-internet, but it was pre-digital, pre pretty much everything. It was talking late 1800s.
So there was a need to come up with an efficient licensing system that you didn't have the ability to run down to Western Union or the telex company to telex a license because the cost of the telex was too much. It was just easier to come up with a system, the companies themselves, to say, send us a check for X, we'll agree in advance what that amount is, because there was no other effective way to create a licensing system.
There were no telephones, the telex was too expensive, whatever. Whereas, today, or even-- well, I shouldn't say today. But let's say, in the 19-- let's say '50s, you had telephones, you had internet. So I think it's more of a historical anachronism more than anything else that music effectively got into compulsory licenses way more so than other industries have.
I will say that the way out of compulsory license for the music industry, there could have been a way with a high quality system, either one or a universal database, just to figure out who owed what, such that by the '50s and '60s, that if someone wanted to license a particular work, you could easily go find who that was through some database. And the music industry databases have certainly gotten better. But even the MLC reports a match rate of only around 90%. And that's just not good enough, quite frankly.
MODERATOR: Interesting, so you're saying historical anachronism-- that dependence. Anyone else?
ELLIOTT: Yeah, I think a lot of it, there's always a lot going on in the background. I think the Digital Performance [INAUDIBLE] Recording Act. I think that was '95, if memory serves. Getting that, it's always about a trade.
And it's like, OK, finally recognizing the right in a sound recording, digital only. I think the concession there was, OK, well, there's got to be a compulsory license. And so that's how the 114 came about. Yeah, I mean, there are numerous examples in connection with supporting a bill and for artists to support a bill, the exchange, and it was sort of an industry [INAUDIBLE].
I think it's probably [INAUDIBLE]. But in connection with that, record labels agreed that they would not [INAUDIBLE] license 114 services, that that would be paid through Sound Exchange. So there are a lot of things and just disputes. And there's a lot of just collaboration in the industry.
Lots of people have different needs and desires. And sometimes there are personal frictions. And there's a need for legislation or a change. And so it's like an industry consensus with compromises on both sides that sort of gets codified.
MODERATOR: Got it. I guess I'm still struggling to get to the bottom of why it is that the industry was willing to compromise in this particular way. So here's the question, then, if I could just take out the historical anachronism. Is it your sense then, if the industry had to begin on a blank slate today-- I know that's hard to imagine.
But we are trying to mimic that in the AI context. We're beginning on a blank slate and had to organize itself in a way for the different constituents to compromise, would you end up with a system of licensing anything close to similar to what you see right now?
JUDGE RUWE: Generally, it's not something I'd volunteer an answer on. But I do want to note something Elliott said. And I'm sorry if it was in your presentation or at lunch. But the industry will give-- they will provide what the customers need and want.
And whether that's on the licensee's deciding what they need to cover to serve their customers, the public might have some historical background as to why music is seen as broad based and venues, broadcasters don't want to operate on just a limited catalog. And then the public expects that. Whether that's informative of the direction for other licensing schemes, I think, Elliott, that that statement is absolutely correct, is you're going to find a way to serve your customers.
If the demand is for everything-- and some of the things we talking about today, there are indications that the demand for new licensees, they don't need-- there's indication they don't need everything. There's possibly other indications that some of them do want to use-- whether they want to license is another matter-- to use everything. There are different customer bases. And they'll have different needs.
LIDIA: Well, I don't know the answer to your question as to whether we would land in a similar position today as we have in the past. But I think that, echoing Elliott's point, this is a very collaborative music industry by nature or business by nature, the music industry. And it is a high volume business. And especially for music publishing, we operate in fractional interests.
And so we do have to, to some extent, think of things in a more collective manner, because at the end of the day, we are managing a lot of music where the rights are shared by various stakeholders. And so I do think that sort of lends to collaborative thinking in the way that we manage rights.
ELLIOTT: I would just add, I think this was fresh in my mind at one point in time. But like a bad dream, I've gotten past it with lots of therapy. But the music industry was just beset. I mean, there's piracy and revenues going from here to here overnight, flat footed, need to move fast. There's just a desire.
And again, with the technology and the needed uses and the customer wants and trying to build a new business model and quickly, there are very few stakeholders. And you had to get everybody together to cooperate. And everybody had different needs. But folks came together on different things. I mean, interactivity.
I mean, that was a blocker for subscription services. And subscription services were a hope. So there is a lot of back room trading and support of this and doing this. And I pick up the bill at the steakhouse on Friday, and there's a lot of deal making to get things done. But it was collaboration. And there was give and take.
And so it wasn't thoughtful. It was really just being reactionary and addressing a present need. And there were people that behaved opportunistically.
And they were like, this is my chance. Again, publishers reset the table. And so it kind of comes together. It's not pretty, but that's how the music business works.
MODERATOR: Joe.
JOE: I would just add, at a small level, the-- I'm excluding the Napsters, [INAUDIBLE] of the world. But to the extent, your question was about, if you started from scratch, would this be the industry. The one thing you would have to have to be able to do that is-- I see most of you have a water bottle on your desk. Everyone every one of these water bottles has a UPC code, so you know exactly that whatever this number is in this bottle of Saratoga water, corresponds to this bottle.
You would need a musical works soundtrack database, because otherwise, you wouldn't know-- if you had that database, you could say that this work is licensable for this price. But this other work is licensable for a different price. You wouldn't need a compulsory license. That UPC code or version of the UPC code could also say, this is who you need to go see for a license.
And without that central database, it's hard in music. I get it. There's just tons of covers, live versions. It's not easy. I mean, at least, in film world, yeah there's a couple of different-- there might be a popular title of a work of a major motion picture that's been reused a couple of times, compared to music, where there might be thousands of a popular song, recorded live at, I don't know, in Tokyo, and one in London.
Think of all the Taylor Swift versions that she's put out, every concert she puts out. It's very hard in music. But to do anything like what you're asking, to start from scratch, you would need something like a master database that's actually pretty accurate. And that's just hard to do. And it's expensive.
MODERATOR: OK, well, we'll open it up to some questions. I see one in the back over there. Let's start there. If you wouldn't mind starting by introducing yourself.
AUDIENCE: Thank you. Judge Ruwe, I'd like to bring up a scenario exemplified by a 1941 song, Don't Sit Under the Apple Tree, which was a big hit for the Andrews Sisters and for the Glenn Miller Band. The underlying composition was written by an 1833 British composer who died in 1841.
OK, let's suppose today his descendant's assigns or estate showed up to your royalty board and requested their share of the royalties. What I'm just trying to get at is the absurdity, to some extent, with public domain. What would happen if you could establish they were the heirs of the composer?
MODERATOR: OK, do you mind introducing yourself?
AUDIENCE: No, my name is Craig Rosenthal. Thank you.
MODERATOR: Does anyone want to-- or was that directed specifically to the Judge? JUDGE RUWE: I take your question. But I also made note of the years. And 1833, if it was when the song was written, assuming it was published, that would be pretty firmly in the public domain. The sound recording I'd have to consult-- actually, 1941, it would still be protected at this point after the MMA. It would be protected under Title 17.
That person, those heirs, if they owned an interest in that sound recording, and it was the Andrews Sisters who recorded it, not the heirs of the writer, they would be an interested party. They would be participating. They would be eligible to participate.
If someone were to challenge the heirs of the songwriter, if they were to challenge them as being interested parties for our proceeding, I could conceive of them not being because you could probably establish that they were not owners in a work that has fallen into the public domain. I totally get your point about-- well, I don't know if it was a point that you were suggesting.
The term issues are difficult to figure out at times. And there's criticisms across the board about whether they're too long. And some people will say they're too short. Those are policy decisions that I don't have-- we don't really get involved in that.
It's about the rate setting for that which is protected by copyright. But to the extent of whether they would be entitled to participate, as I understood the facts, for a sound recording, possibly, likely, yes, on the musical work, it would have fallen into the public domain, as I understand the facts.
MODERATOR: OK, David.
DAVID: Yeah, I wanted to go back to the question--
MODERATOR: If you wouldn't mind, speak into the mic, sorry.
DAVID: I'm so sorry. David Strickler, also a member of the Copyright Royalty Board. Hello, Judge Ruwe. Good to see you again. This is a question that was posed to the panel with regard to why music is so uniquely subject to regulation. And the answers that were advanced were in part historical in nature, or may I say, institutional, and just sort of serendipity, if I understood it correctly, circumstances.
I would take issue with that and propose another reason. And that is that it's economic in nature in that as was discussed with other panels, there's a transaction cost problem that relates to musical works and to sound recordings, which is that it's obviously not feasible for each licensor and each licensee to negotiate one with the other to be able to hammer out a royalty rate. It would not happen, because the cost of those transactions would be worth more than the individual playing by the licensee of the song. So therefore, you needed collectives.
And that's the historical point that I think is correctly the historical aspect of it, which is that historically, collectives were started, first in Europe, then ASCAP, later BMI. And now we see in the nature of collectives, large record companies, the three major labels, and the music publishers. So what happens is, to solve one economic problem, the transaction cost problem, you have these huge collectives, which are tremendously efficient and good at solving that problem.
But along with solving that problem comes a new problem. They now have this tremendous market power. And I think someone who knows the other industries might say that that's not unique to music. And I would be open to hearing that.
But with regard to music, they now have that power as the Justice Department recognized in the mid 20th century, with ASCAP and BMI, that power had to be controlled because it was abusive. And therefore, the consent decrees were entered into. And then copyright royalty tribunal and the CARPs, and ultimately, the Copyright Royalty Board was set up. And those exist all over the world because this is a fundamental economic problem so that the reason why you see it in the music industry is the total number of copies and the success of minimizing transaction costs necessitates the regulation.
Otherwise, you have this monopoly cost problem. So when the talk comes up as to whether or not you need a Copyright Royalty Board, per se, the answer in one sense is you don't. As I think Joe was pointing out, if you had to start all over again and you had a good database, you'd be able to do this without government regulation, without the government being involved. And I think that's correct.
Private sector has proven that it can handle collections and it can handle minimizing transaction costs and enforcement and do all that. What the private sector can't do is what the Justice Department or the Federal Trade Commission do, which is regulate prices or monopolies or complementary oligopolies that are too powerful. Well, if that's what we're trying to fix, then you need somebody to handle that, whether it's the Copyright Royalty Board or it's the Justice Department or the FTC. And now I've stepped on about 25% of my own presentation. But it seems like the right context to bring this up.
MODERATOR: Do any of my panelists want to respond? Or we can move to the--
JOE: Yeah, I mean, I think that you've seen certainly in the past decade or so, perhaps 15 years, that the balance of power has shifted to some extent away from copyright owners toward services-- Apple, Spotify, Amazon, whatever. I mean, it just, because as the judge points out, there's market power. And it's interesting how the market power has shifted. It's not 100% shifted, but there are some times when-- I mean, I'll throw this out there.
So I was also involved in the section 115 Reform Act of 2005 and 2006. And at the time, the big issue was not streaming, but the iTunes download store, which was the big seller at the time. And there was a fight over, at the time, Apple was offering 30-second previews on its download store.
And some of the PROs said, well, that's a performance. And we want a license for that. And I said, well, that doesn't really make any sense to me. Why do you care? Because you're going to sell more if you're giving away a 30-second preview, you're not downloading the whole thing.
And they had this really complicated argument why it made sense. And I said, I'm a member of Costco. And I go to Costco. And there's inevitably people who hand me a little toothpick with a free sample of something. And I try it and I like it, and I then go buy more of it.
So why is it not applicable here? They said, well, it's just because that's how licensing works. That's not how the world works. Ironically enough, getting back to market power, about a month, no, probably about six months later, Apple changed its policies for rights holders. And they said, if you want to, as a copyright owner, list or make available on our download store, you must give us a 30-second preview for free.
And if you don't, that's fine. Go sell somewhere else. And because at that point, the market power had definitely shifted to them. And lo and behold, I don't think a single song was ever taken off the Apple iTunes store once Apple came and said, you got to give us a 30-second preview for free. We're not paying for it.
ELLIOTT: I think there's equal power. I worked at Apple. And I worked at a major. And when I was at the major, we were terrified of Apple. And when I joined Apple-- this is probably not public information. They are equally terrified of the majors. There is a healthy-- and you don't see it.
There's a lot of saber rattling. But there's respect. And we could get through a lot of garbage if people would just cut through that. And in terms of you're going to go for everything, right. It's the Wild, Wild West.
It's a jump ball. There is uncertainty. People who feel like they've been wronged in the past, it's an opportunity to reset the table. It's publishers are supposed to maximize the benefit of their writers. And they're going to go for it, crazy arguments and substantiated arguments.
I mean, it's just the nature of the beast. And right now, we're in that kind of period where it's a jump ball and everybody's trying to figure it out. And if we're reasonable and we put ourselves around good people and we negotiate two good parties on either side, zealously, we have the opportunity to create law before strangers jump in and try to figure out what it is that we do.
MODERATOR: All right, we have time for one more question. And I saw Reagan's hand up here.
REAGAN: I want to throw out one other potential explanation for why the music market developed a little bit differently on regulation. And that is competition. So when you have, in 1908 and 1909, these two pianola manufacturers, they are putting in, into a physical device, certain songs. And you've also got this underlying decision saying, we're not sure if you have a right and there's a copyright, which maybe gives the will for a solution to overcome that in terms of Congress acting.
But there's this interoperability justification maybe 100 years ago that has justified this sort of a license. And do you see that being any similarities, like, how does that bear into whether this is at all a useful precedent for AI licensing. And maybe related to that also is since you've got the double copyright on music, you have the end of the pass-through license, as well, which is maybe a digital analog to the physicality of where the 115 license was getting used.
MODERATOR: Does anyone want to have a go at that? We have, I guess, 38 more seconds.
JOE: Eh --
All right.
JUDGE RUWE: No.
MODERATOR: All right, we're going to let that comment linger, it looks like.
JOE: I'm sorry.
MODERATOR: But on that note, please join me in thanking our panelists.
[APPLAUSE]

So we will break for 15 minutes and then reconvene at 3:45 for our final panel on antitrust implications of licensing.