# Oral argument — George Johnson v. CRB

> Briefs, arguments, decisions, and more.

URL: https://www.frixlaw.com/law-library/documents/argument%3A69147

## Record

- **Collection:** Oral argument
- **Document type:** Oral argument
- **Published:** March 10, 2020

## Text

L. Mr. Kington for Appellate and Copyright Owners. Mr. Johnson for Appellate and Support Services. Mr. Schuchat for Appellate and Copyright Owners. Mr. Shadman for Interpreter and Copyright Owners. And Mr. Kington for Interpreter and Streaming Services. Thank you. All right, everybody's ordered. Mr. Kington's the point. Good morning. Thank you, Your Honor. And may it please the Court, I'll address the rate structure and rate level issues, and Mr. Unstretch will discuss the bundling and retroactivity issues. To begin with, the majority's two key decisions regarding rates, setting particular rate levels and adopting the uncapped TCC structure, both are infected by the same error, reliance upon unsupported assertions that an increase in royalty payments for the publishers would decrease royalty demands by the record labels, what's been termed in this case the seesaw effect. With respect to rate levels, the model put together by the majority to justify its 44% rate increase concludes that the labels will accept an approximately one-third cut in what they currently receive, a reduction of more than $500 million per year, notwithstanding the board's recognition that the labels exercise super competitive oligopolistic power in their market. The majority relied entirely on that predictive drug and label demand to determine that rate levels satisfied two key elements of the statutory Section 801B factors, that rates afford a fair income to the copyright user as well as to the copyright owner, and that they reflect the relative roles of the owner and user, and these are the rate factors relevant to satisfying the Constitution's reasonable rate of return requirement. It was plainly irrational, the opposite of reasoned decision-making, for the majority to rely on the model's highly counterintuitive conclusion that the labels would give up more than $500 million a year, especially because the majority itself acknowledged at JA 1061 that its model did not reflect reality because it didn't account for the label's oligopolistic market power. In addition to the lack of reasoned decision-making, no substantial evidence supports the conclusion that the labels would reduce their demand. The dissent said that at Joint Appendix 1241-42, and the proof is the meager authority cited by the government and NMPA. Not a single one of the cited experts supports the prediction of the majority's model. Most just say that the increase in the publisher rates will be taken into account by the labels, but they don't say that there'll be a reduction for how much there will be, and in fact, the copyright owner's expert, Professor Gann, said the burden would fall on the services and that the label would keep their demands at the same level. To the extent the majority relied on Professor Watts, even he didn't predict the dramatic model-level reductions of $500 million a year by the labels, just said that the labels would reduce their demands by most of what the amount of increased rates. But even those say things. The board's response to the failure of the models to address the oligopoly, you said it better than me, oligopolistic nature of the sound recording holders' interests, that what they were going to do is, so what mattered then was the level at which the rates were set. So it was too high, there wouldn't be enough counterweight on, I guess, their steep shot. And the debate just seems to be about, at some level, predictions of economic consequences. And so I'm wondering why, when it comes to predicting what an economic consequence will be of a change, that's not the type of thing, I have two questions. One is why wouldn't it defer to the board? And secondly, if more evidence was needed, the concern is there wasn't evidence for their prediction, what more would you have than the expert testimony you had here, saying that there wasn't a seesaw and that there would be this effect? It wouldn't be perfect, but there'd be this response. Well, just to take the last question first, Your Honor, the evidence is not just not perfect. The evidence here doesn't come, even Mr. Watts, Professor Watts' statement, which is flawed for a number of reasons, but even if it were taken at face value, he did not say that the $500 million drop per year would occur. All he said is, for the amount of increase, which is a much lesser amount, the labels would decrease their demand. The critical thing here is, in order to satisfy the 801B factors regarding fairness to the copyright user as well as the copyright owner and appropriate shares of the product, the majority has to look at what the impact of the rates would be. And if the section of this decision that addresses that issue, which is that A, zone appendix 1076, relies entirely on this model and the conclusion that the services will actually make more than they were making today, and that's why, based on the model's predictions, and that's why things would be okay in terms of the balance between the services and the copyright owners. So the problem here is not a dispute about prediction. It's that their prediction has zero support in the record. Nobody believes that the labels are going to reduce their demands by $500 million. Maybe the majority could have justified its rate of increase under those factors by relying on something else, but the only thing it relied on was the predictions of the model, and the predictions of the model are just totally irrational and not supported by the record. To be clear, your point is when they said, look, we can adjust for the model's failure to deal with these oligopolistic tendencies by reducing the amount of the rate increase, that that was just not a rational way of addressing it? It's mostly like the model, and I thought the seesaw was generally talked about, not just by one witness, and so if there is a seesaw, then it's just a question of how much pressure you need to change it. Two points on that. The reduction in the rate that the majority talked about only applied to the PCC prong. It didn't apply to the revenue prong. So just forgetting about the PCC prong entirely, which could result, as we explained, in dramatically unreasonable rates, just looking at the revenue prong and the predictions that the majority's model makes about the revenue, not adjusted because there was no adjustment, the prediction of that model is the services will keep $500 million more a year because the labels will reduce their revenue, and that was the basis for the majority saying this rate level will fairly treat the copyright users. So their adjustment doesn't matter. It's just a question of what they relied on to make that determination just is incredible and was not supported by Professor Watt. In addition, Professor Watt, not only did he not support that dramatic drop, he did not take account, as the dissent says, of the fact that rather than sitting by and accepting a dramatic reduction in their share, the labels might decide to reorient the market. In fits on getting the money from the services, see what happens. Maybe they'll pay us. Maybe the services will go broke and the labels will take over. Professor Watt didn't consider that possibility, and the dissenting judge, who is the designated economist on the board, said the failure to consider that means that it is implausible to rely on Professor Watt's evidence. Can I ask you about the fair notice question? I was wondering why you don't lead with that. Well, we do lead with interim relief. Yes, I noticed that. And we believe strongly in it, Your Honor. But I think that the issue that I was talking about goes to both the right structure, because it justified the uncapped nature of the TCC, and it goes to the right levels because of the discussion on the 801B determination. So it seemed like something that put the whole thing together. I'm happy to talk about the lack of fair notice. As you know, the dissenting judge, this isn't just an argument we're making, the dissenting judge said there was no fair notice so that we would have an opportunity to address the impact of the uncapped TCC. And that is a significant failure. It's a failure that this Court has relied on before in setting aside agency determination, as we explained in our brief. Was there not enough evidence in the material that was already available to consider the uncapped TCC? Well, we didn't have a chance to put in affirmative evidence about it in the affirmative part of our case because the issue wasn't discussed. As the case was litigated, the copyright owners wanted a per-play rate. The services wanted the continuation of the phone-02 rates, the prior rates. Those prior rates had a cap on the critical premium $9.99 a month part of the service, which supplies the overwhelming majority of revenues. So nobody had any notice that the uncapped, the capped nature of that rate was going to be an issue until the majority adopted that proposal. What about the Google proposal? Well, the Google proposal came in after the evidence had closed. I understand, but then there were filings, and the services responded to Google's filing and didn't object to the uncapped. Well, they didn't object to the uncapped nature because Google very carefully couched its filing as tied to the particular level. The board had made clear early on that this is de novo review. Everything's on the table. Whatever you want, you have to prove it. Don't assume status quo. Whatever you want, you have to prove it. So maybe that put the burden on you to show the need for caps even at the initial hearing, but certainly when someone threw on the table, okay, you're right, this is rather messy what we have now. Uncap the rates, but we would like these to be the rates if you're going to uncap. I don't understand why you then, in filing a response to that very submission, wouldn't have addressed the argument, saying, look, we don't think they should be uncapped at all, here's why, and then at least echoed Google and said it. I forbid you to uncap them, set them at the low level like Google says, but you didn't do anything. You didn't object to the uncapping at all. I guess two responses to that, Your Honor. The Google proposal came after the record closed. It was not possible to induce new evidence to address the Google proposal. Really, is it not true that in your response to the Google proposal, you couldn't have made your arguments and said, and if the board disagrees, we think this is such a dramatic, the same thing you've argued about, perhaps it's a dramatic change we didn't have notice of, and you would need to reopen the record. Perhaps we could have said that, Your Honor. But you didn't. We didn't, but the fact is that Google very carefully said, if you raise the rates from what we're suggesting, then this has the possibility of moving away rates. I thought your concern here was we had no notice that the thought of uncapping rates would even be passed for a fleeting moment through the minds of the royalty judges, and, in fact, it was right there on the table. And, I mean, maybe I'll wish you'd written it differently now, but the notice argument, I guess I'm having a little trouble. And these post-hearing filings like this happen. No one says it's not a legitimate thing to do. So I don't understand the notice argument. It's certainly legitimate to have that filing. That does not answer the question whether we had an opportunity during the hearing, which is the critical time, to address that issue. I think even if Google's proposal had been adopted, it would have been possible to say there's no fair notice here of an uncapped rate. But the critical thing is Google specifically said in its proposal our rate is tied to the level that we specified, and that level was 15% compared to the majority's 26.2%, so a very, very dramatic difference. So I think it was reasonable for the services to conclude that since Google had framed it that way, the general issue of uncapped rates was not put on the table. It wasn't put on the table by anybody. Can I ask one more question? The response to your argument on the other side, or maybe it's only in my mind, on the question of whether it's really reasonable to believe that the record labels won't raise their rates. It doesn't apply all the time. If they have all this power, couldn't they drive you out of business today anyway? And if that's really in their economic benefit, as you seem to think it is, why haven't they just done that? I think, Your Honor, as the dissenting judge said, the record labels are happy with the status quo. They're getting a large amount of money, and they're happy with the current situation. But if your point is they could get another $500 million then, why don't they just take the other? The issue is how much different is this compared to the bad economic position you're in anyway? Well, I think it's – I guess there are two points to that, Your Honor. With respect to the uncapped rate structure, we don't know how bad it's going to be because we don't know what will happen. And because the consequence of the – with respect to the rate levels, because the reasonableness, the determination of rates are reasonable depends on the reduction of $500 million a year for the labels. This issue doesn't really come into play because it's quite a different world for the labels, where instead of making what they're making now, they're being told $500 million less. That isn't going to happen, but that's the entire basis for the majority of the determination that the 801B factors were satisfied. Thank you. Thank you, Your Honor. May it please the Court. The majority made two additional legal errors. First, it changed the initial determination for bundled services. Second, it made the new rates in terms of effective retroactive. And this Court should reverse both of those decisions. Looking at the bundled services question, the judges recognized that among the copyright owner's requests to change the initial determination were some that were for, and I'm quoting from 1251, some that were for substantive relief. And if you look at page 1267, where they actually analyze the copyright owner's proposal, it's clear that they are engaging in substantive review. They say we are going to pick the copyright owner's proposal based on our evaluation of the Section 801B1 factors. There is not a word in their actual analysis and justification that speaks of fixing a clerical error or making a technical change, where those kinds of language appear elsewhere in the order where that's what they're doing. So what they did here is rehearing, and Congress said, Ford, if you're going to do a rehearing and change an initial determination, you can only do it in exceptional cases, and that's 803C2. And we know what the judges thought about that. The copyright owners, and I'm quoting from 1251, did not meet or even make a prima facie case of meeting that exceptional cases standard, which long-standing precedent, repeated here in footnote 3 on page 451 and on page 138 of the appendix in the denial of Mr. Johnson's motion, makes clear that exceptional cases is 59-8. You need a change of controlling law. You need new evidence unavailable at the time. You need manifest injustice. That's when you can make substantive changes to a new standard. No, that standard doesn't just apply to parties. And the court itself can, if it looks upon review of its initial determination and repeats from filing and finds that there's a glaring error, it can fix it. Sure, because in 803C2, Congress said that copyright royalty judges may, in exceptional cases, upon motion of a participant, order a rehearing on such matters as deemed appropriate. So it's tied to, you know, there's no sua sante authority to make changes in the emotion. It's limited to exceptional cases. This is very different language from what we see for the FCC or FTC or FERPA. In B-5, they have sua sante authority to make changes based on paper submissions. They have the authority to make changes based on sua sante to make decisions based on paper submissions. In B-5, B-5. Just to be clear, we're not complaining that they did this without having a live hearing. We're not saying that rehearing requires a live hearing. Sua sante authority to make, I'm talking about the ability to actually make a change. Right, but that's the question of where the exceptional cases standard is met. Does the board have to have a live hearing if you do it on the paper? B-5 doesn't have an exceptional cases limitation. Your Honor, I understand B-5 to be about the 803C2 hearing. It's not, as I understand it. I guess I'm getting the question of whether. It just seems a little odd for Congress to set up a scheme where you have an initial determination, then you take these filings, and then you have a final determination, and to think that the board itself, after issuing its initial determination, is helped by the submissions of the parties, and in doing so, discovers there's a significant problem that's just missed. And then I have to go, I'm sorry. I guess because we just missed it, there's nothing we can do about it. Before our final determination, we just have to lock it in. And that seems to me, it's one thing again to limit. You don't want parties relitigating everything at this stage, and so we try to confine the parties. But if they discover that something's really haywire, that they can't fix it themselves, assuming again that they allow people to make paper submissions, they just all say they're to everybody and everyone's allowed to. So, Your Honor, again, if Congress wanted to create that scheme, it could have, and it has in all sorts of other administrative proceedings. Here, Congress carefully articulated down to the month. They create deadlines for this phase of the proceeding, deadlines for that phase of the proceeding. Everything is very short. The deadlines don't have to do with what they can do as long as they're within that time frame, whether they can really fix something. They just discover. Why would Congress want to lock in the Board itself from fixing a mistake that it recognizes, and to do it in a timely manner consistent with Congress's time frame between the initial and final determination? Why would it want to lock that in? Your Honor, I'm speculating. This was built on another older regime, and there is some mention in the legislative history of concern about people evading that actual hearing process. Was that the Board or the Congress? That's in Congress's legislative history. No, no, I'm sorry, but the people that were evading it were those— I believe the parties. Exactly. But so, given that we do have, and this sort of leads into the second argument, there are instances where retroactivity is authorized in the law, and there you can understand Congress thinking that the initial determination ought to look a lot like the final determination, except in exceptional cases so that parties can plan accordingly. Now, our situation, and I'll get to that in a moment, is not one of those where retroactivity is authorized, but you could understand— You're talking about the bundling here. That's not a retroactivity issue. No, it's not, but you're asking, Your Honor, why would Congress want to limit the Board in this way? Because I mean, it's retroactive between the initial and the final determination. Well, where the final determination gets to apply retroactively, right? So the parties see the initial determination. Okay, so this is the regime I'm going to live under once this takes effect. The retroactivity argument is a bundling issue, right? Exactly. I have the argument, if it's retroactive, that you are making on the date of the rate. I guess all I'm trying to do, Your Honor, is to suggest why Congress might have wanted to give this particular set of Article I judges less power to do re-hearings than it gives the FCC or the FTC or FERC, all of which have much broader authority. But turning to retroactivity, this is real retroactivity. We spent 13 months buying licenses at a price, and then the Board increased it. And it was a surprise. In May 2017, before the Board acted, we pointed to 803D2B and told the Board that this act pretty clearly says it's not published until it's published in the Federal Register and it doesn't take effect. And then the decision comes out, and they agree with us. On page 818 of the record, this is from the penultimate paragraph of the decision, The rates and terms established in this initial determination shall supplant the existing ones at such time as the Register of Copyrights completes her statutory review and the Librarian accepts and publishes the determination. That's the same thing the Board told us on their website at page 75 of the record. The initial determination comes out, and we think the Board has agreed with us. Come November of 2018, we learn nobody even asked the Board to make this change. Now they're going to make the new rates and terms retroactive to January 1, 2018. Nothing in Section 803D2B authorizes that. Yes, there's an accept-as clause, which the government relies on, but the majority never cite it. But under the Bowen principle, that accept-as clause isn't the kind of express authorization needed for retroactive rate setting. And we know that because we can see in the immediately prior sentence, Congress actually used the phrase, shall be retroactive, to refer to when they set rates initially. And in the immediately prior paragraph, 803D2A, Congress said that where other kinds of rates have a specific end date, the new rates can go into effect on the next day, even if the decision is published later. So given the background here, the right way to read 803D2B is that accept-as clause lets the judges pick a different post-publication date than the default, not that they get to pick a pre-publication date. Can you explain to me how the retroactivity even worked here? Because I take it it went into effect. There's no stay, and so rates had already been paid, and then they had to leave. Transactions had already occurred under the formal treaty. So were they renegotiated, or what happened? As I understand it, everybody had to recalculate 13 months of transactions that had occurred and completed under the Formal 2 rates and terms, and recalculate and figure out what the delta was and make additional payments or seek refunds based on the Formal 3 rates and terms. Did that just to the music publishers, or did you go negotiate with the sound recording? This is specific as to the mechanical error. As to that one? Okay. This decision doesn't change the negotiation. So during that time period, you had no offset from the gravity-cutting shock of the negotiation? Not that I'm aware of, Your Honor. I see my time has expired if there are no questions. All right. Mr. Shanmugam? Thank you, Judge Henderson and may it please the Court. I'll be quite brief because I'll be back up here in my capacity as intervener. While the services broadly challenged the Board's final determination, my clients, the copyright owners, challenged that determination only in one very narrow and discreet respect, and that is with regard to the Board's decision to reduce the number of subscribers, and therefore the royalties due under the mechanical floor prong of the rate structure, and that's the fallback, the third prong of the rate structure, where the services offer discounts to students or family members. On that prong, our submission is a quite simple one. The Board reasoned in its final determination, and this is at pages 1079 to 1080, that the services were offering those discounts to build market share, consistent with the broader efforts on the part of the services to engage in revenue deferral. But the Board also reasoned that the services were offering those discounts to reach consumers with a low willingness to pay. Our submission on our affirmative appeal is simply that there was no evidence in the record to support the proposition that students or family members in fact have a low willingness to pay, and in light of that, that narrow aspect of the Board's final determination should be vacated. Now, on this issue, the government and the services point to three categories of evidence in their briefs to support the Board's determination, but none of those categories in fact provide factual support for this key proposition that students or family members have a low willingness to pay. The first testimony... If they had students, they're short on funds. College students in particular don't have funds. And because of their generation, if they have to pay more, they'll go to YouTube or perhaps some alternative means of acquiring the music. There was evidence about that. Maybe you don't think it was enough, but it doesn't seem right to me to say there was nothing in the record. Well, I think what there was, Judge Millett, was that there were statements made by representatives of the services to that effect, but they were very conclusory statements. They were not backed by any evidence. In fact, with regard to students... Why isn't testimony as to why we are getting less money than we otherwise could pretty relevant? Well, they were... It would be pretty much across the industry. They were really statements, Judge Millett, about the services' motivation in offering the discount in the first place. When it came to the question of whether students or family members in fact had a low willingness to pay, there was nothing in the record. In fact, the only study that was in the record, and we point to this, I believe, at page 31 of our brief, was a study showing that students are, if anything, more willing to pay because they are so eager to obtain access to music. And so beyond that, beyond the bare and, frankly, somewhat equivocal statements of the services' own representatives, all you had in addition to that was analysis by the experts on price discrimination more generally, and that is essentially generic expert testimony that price discrimination can increase revenue, really without reference to students or family members other than a very passing reference by Professor Marks at page 440 where she included students and family members in a laundry list of parties as to whom she essentially assumed the price discrimination would be beneficial. And the reason why I think this is so critical is because there's a very fine line here between low willingness to pay on the one hand and revenue deferral on the other. And, of course, where you have the services competing for the market and not just in the market, as the board said in its final determination, the services have an incentive to attract customers of any variety through a variety of attractive pricing mechanisms. And, for instance, there was testimony from the… These are universally chosen, and they explain why. They said these folks are difficult to monetize otherwise, and it just makes honestly common sense that college students, for the most part, won't have much money, and they have technical abilities to get music another way. And so if we get them in, get them in the practice of paying, get them a price point that they're willing to do it, and get them in the practice of buying this music legitimately, that's better for everybody. Across the board, it helps everybody. And then as they mature, hopefully they'll be in the practice of making this payment. To the extent you have that intuition, Judge Millett, it presumably would have been easy for the services to come forward with studies or other evidence to support… They come back and say, this is why we do it, and it's across the board, and it's consistent with common sense. I guess it's common sense, is it not, that the vast majority of college students like music and are not flush with money. I mean, you might have that intuition, Judge Millett. So, again, the evidence that was in the record suggests that if anything… But it's not my intuition. It was clearly the intuition of the people who are losing money by offering cheaper packages. My intuition has nothing to do with it. But that is also consistent with the notion that what the services were doing was simply competing, trying to get in customers who might be customers for life. And the problem, if anything, is magnified with regard to family plans, particularly because there is obviously no one accepted definition of family. And, again, our submission is simply that if the services could, in fact, point to some factual evidence that these groups had a lower willingness to pay, that might be one thing. Again, there is testimony. It may not be the fancy expert witnesses at your studies that you want, but there is testimony, and it seems perfectly rational for the board to credit that if we have to pay for our 3-year-old and our 8-year-old to have their own accounts, we have to pay full freight. You know what? We're going to get one account, and we're going to share. And there was testimony that that would be more convenient. And cheaper. And families, again, most people aren't looking to maximize the amount of money they give to families. In this colloquy, we're, by definition, talking about discounts. But I think the reason why the court should be perhaps a little more searching than it might be otherwise is precisely because all of that is also consistent with the broader enterprise of revenue deferral and competing for the market through any means. And the last thing I would say with regard to this. It's not revenue deferral if you're trying to bring in, again, for the college students, people who otherwise wouldn't pay at all. But you would also be competing for them regardless. And to the extent that discounts for students or family plans are convenient or otherwise attractive, that is consistent with that broader motive, which, of course, explains the broader rate structure more generally. And the last thing I would say on the subject of the testimony of the actual witnesses is I would encourage you to look at the testimony of the services on witnesses. I would point the court to Gary McCarthy's testimony, which is under sealed, but which is at pages 1327 to 1328 of the appendix. That testimony, I would respectfully submit, was ambivalent and unsupported. And I would also point to the testimony of Rishi Merchandani from Amazon at pages 1365 to 1368. So we've already looked at that. And if you characterize it as ambivalent, but the board found it relevant. Well, I think that the board in its analysis really did not say very much at pages 1079 to 1080. And it, too, relies on the substantial evidence argument. This is a substantial evidence. They don't have to repeat it all themselves for it to count. That is certainly true. But, again, my submission is that the testimony is ambivalent or, in the case of Rishi Merchandani at 1365 to 1367, unsupported. And while there was references to the fact that the services had conducted research on this issue, none of that was presented in the record. And, indeed, the witnesses were not involved in that research. Thank you very much. Mr. Johnson. Good morning, Your Honors. And Judge Henderson. And may it please the Court, my name is George Johnson. And I'm a pro se songwriter. And I thank Your Honors for letting me speak today, participate. Since I'm not an attorney, I intentionally didn't argue any constitutional questions in this field because I'd probably lose. So I've kept my issues to a few simple practical requests inside the bounds of the compulsory license and inside the rules of 37 CFR Section 385. First, asking for a long overdue subpart A inflation adjustment. And second, eliminating free limited download. The basic argument in this case is that songwriters used to make 9.1 cents per download, but now they make zero cents per download. And that is not reasonable nor reasonable rate under the Copyright Act. This lost sale created by the limited download has decimated the American songwriting industry. Then, under subpart B, the streaming rate is essentially zero. And that is also not reasonable nor reasonable rate. For decades, the U.S. government guaranteed American songwriters and independent music publishers 9.1 cents per sale. But now this sale is literally being given away for free as a limited download. The Copyright Royalty Board is required by law to set reasonable rates. Zero is not a reasonable rate because it's not a rate at all. Nor is it an equitable division of music industry profits between copyright owners and users as required, since there is no division of profits if one side is getting zero. By eliminating the free limited download, a term created out of thin air in 2008, the court would only be restoring what was taken away from property owners in Formal Records I. This decimation caused by the combination of lost sales and essentially zero streaming rates is evidenced by the loss of music or songwriters and publishers from 4,000 to less than 400. Add to that the rate should really be 50 cents instead of 91 cents. Your Honors, one important point I would like to clarify here is that while 50 cents is a break-even point where 110 years of inflation has been properly weighed in Subpart A, it is not an all-or-nothing request. And while we pray that Your Honors will determine the full amount of inflation that's approximately 50 cents per mechanical for vinyl, CD, or download under 3853, we welcome any determination of 20, 30, 40 cents per mechanical or one that gradually increases the rate over a several-year period if a full 50 cents is not possible at this time. With that said, the full 50 cents per song correction is supported in this appeal with an amicus brief by Formal General Counsel of the U.S. Copyright Office, and I hope this carries weight in addition to the public comments supporting the full 50 cents made by Pelham and NPA found in my final opening brief. Mr. Johnson, can I ask a question? Yes, sir. Why is it that the principal songwriter organizations are not appealing these costs? I don't know. It's a good question. I think that they should. They brought them up before. If it is as dire as you represent, do you think that they would, from an economic point of view, be making the same money? Well, it is absolutely counterintuitive, and I think I put that in one of my briefs, that they wouldn't want the extra 9.1 cents, but my guess is that the three major labels that fund NPA and NSAI, they represent the three major labels. That for some reason they have said, we just want these streaming rates, we get them in a full billions and billions of performance, add up to enough money for them, but still way below what the rate was for selling CDs and albums. So I think that they've decided that the three major record labels don't want to... I think the three major labels, when there were still records and CDs, you had to manufacture those things, printing cost, plastic. So they didn't want to do that. But on the download, the costs are already paid for. So I don't understand why they wouldn't want to just have a download that the customer could download if they wanted to and pay the 9.1 cents. Your Honors, as you're aware, on page 539 of the Joint Appendix, or Exhibit A of my addendum, is my subpart A inflation chart based upon the public record, and is one of the most important pieces of evidence, because it helps you visualize the 2 cents to 50 cents in the green line. And the current price fixing and central planning is what we're doing is the red line. And we've basically gone from 2 cents to zero over 110 years. And while we're referencing the Joint Appendix, there's a 2016 Billboard magazine article starting at A553, but mainly graphic at A554, which is part of my evidentiary record. The reports were the artist Adele's last album sold 3.8 million copies and where 25 songwriters split $13 million. And that's very significant. And NMPA said publicly in 2017 that downloads are irrelevant. And while I would argue that these 25 songwriters would disagree, the album sale evidence here proves that subpart A download and the sale of 9.1 cents are very relevant. And Adele herself made $3.8 million. And so to recap, you know, the subpart A compulsory rate has failed to keep pace with the actual value of music orders. Combined with the compulsory subpart B streaming rates have been inadequate to support professional songwriters. And third, the compulsory limited download loophole is a lost sale of a permanent download and must be eliminated. In preparing for this hearing, I noticed a few sentences. In number four of section 385.14, where in 2008 even DEMA and NMPA and RIA contemplated the data with a free non-interactive stream as well as the limited download might need to be eliminated. And they actually used the term eliminated. So that is why I use it as well. And, you know, they also said the law is based upon the industry practices of the time. So it may be time for elimination of the limited download. And, of course, this 385.14 also says that the rate should be set to no vote. And, you know, the rate's been 9.1 cents since 2006. So during the former records one, two, and three, the rates weren't set to no vote. And they weren't set in this case. So the limited download is essentially a permanent download. And even if it's not permanent, it's letting the customer take possession of the musical work for months at a time. And this 385.14 says, you know, this requires payment of the applicable royalties. So just my other two issues real quick were a voluntary buy button, which I hope the services would consider to replace the limited download, and also, you know, an equitable rate structure. And I don't think the rate structure is what it should be. It's just, you know, what they set up in 2008. So I think my time's up. And I appreciate you taking a look at this case and listening to it. Mr. Johnson, you may be in the wrong business. You did a wonderful job presenting. Well, I certainly appreciate that. Thank you so much. Ms. Utrecht. Good morning, Your Honor. Jennifer Utrecht on behalf of the Copyright World Court. May it please the Court. It's important to remember here that the Copyright Act has vested the judges with an intensely practical task. They have to establish reasonable rates that should be paid based on various policy factors, some of which have countervailing concerns. And it's the judge's task to weigh all of the evidence and to determine which way these policy factors best cut. And here I think we've been presented with parties on all sides saying that the judges have drawn incorrect inferences or that they perhaps should have weighed certain policy concerns more heavily than they did. But ultimately, these concerns are the sorts of things that the Copyright Act asks the judges of doing and that this Court owes them deference. Now, I'd like to begin, if I may, by addressing the services arguments regarding the rate structures and the rates. And I think here the fundamental point of this agreement is extremely narrow, and I'd like to explain why. Everyone agrees here that what the judges are doing is essentially figuring out how much this mechanical royalty is worth, so that we can divide the profits from streaming equally. And if the services proposed in this proceeding, one way of figuring out what a song is worth, and again, everyone here agrees, is to look at how much profit the streaming services are making, to look at their revenue. But, of course, there are problems with a sheer percentage of revenue rate, as was recognized extensively in the judges' decisions and as the services experts acknowledged. The services are engaged in a competition for the market, which means that they have been consistently driving their own prices down in order to get as many customers as possible. And while there are certainly competitive advantages and reasonable reasons why they might do this, the judges acknowledge that the effect of this is that copyright owners, songwriters, will walk away, if there's a sheer percentage of revenue rate, will walk away with less money. And so the way that the services proposed balancing this was to have an alternative to the revenue rate, which has been referred to as the total content cost rate. So the services have proposed here and before the judges below that the total content cost rate should be subject to a series of caps. And I believe the services have contested and have characterized the judges' decision as rejecting those caps based on the fact that they believe that record labels will accept less money. And I don't think that's actually properly the true basis of the judges' decision. I think what the judges said here is we're trying to set a fair rate for copyright owners. We're trying to give songwriters their reasonable share. And once we do that, the remaining money that's left over is going to be negotiated between record labels and the services as fair market actors who are going to be acting in their own rational self-interest. And there are a number of things that could result from this. The record labels might insist on the same amount of money. They might accept less because they have to. There are multiple possibilities, but that's the market acting as it would ordinarily act. And I would point your honors to there are several places where the judges explain that what they're doing is setting a fair rate for the copyright owners, even if this means that the services might come away with slightly less money. So on footnote 107, which is JA 1045, or the Federal Register version, if you prefer, 1944 to 45, the judges discuss the fact that the services proposed a continuation of the present rates, even though they all state that they are losing money. And the judges say the reason for this is their loss in a bottle for market share and less the losses they are incurring are not a serious competitive detriment. On footnote 75, which is Federal Register 1934, the judges say any increase in the mechanical royalty rates, whether or not they are computed with reference to regular company royalties, has the potential of leading to a bad outcome for the services. Even maintaining the status quo could be a bad outcome for the services, as it surely would be for songwriters. The judges have to trust that reasonable market actors will reach a meeting of the minds regarding the remainder of the profits. And again, JA 1077, lowering mechanical royalty rates to provide services profitability in the face of the acknowledged problem that a lack of scale, of the lack of scale of this competition for the market that I was referring to earlier, would constitute an unwarranted subsidy to these services at the expense of copyright owners. And that, this principle was what was motivating the judges here, which is that we have to ensure songwriters receive a fair share. But didn't they also have to ensure that the services would retain a fair return? And so if, and they had already recognized that the services were overpaying royalties to the songwriters, or the sound recording. And then there's a pretty massive increase, 40%, 44%, 40 or 44, whatever, large percents in the payments to the mechanical royalty. And the question is, on what possible basis could they think that if the sound recording rights holders don't budge, because businesses don't like to surrender hundreds of millions of dollars voluntarily, and they now owe a massive increase to the songwriters for reasons before it. Explain that instead of a seesaw, you're going to capture the services in the bikes, that they will not be able to survive. And how would that ensure a fair return? Well, I think that argument really goes to, I think, both the amount of services received as well as sort of the disruptive effect, which is the fourth policy factor. And I think there are really two things that need to be said here. One, of course, as this court has recognized, the policy factors sometimes are countervailing, right? Sometimes ensuring a fair rate of return to the copyright owners is going to mean that services received less money. So B requires you to look at fairness on both directions. It does. You have to look at both sides of the coin. And so I guess it doesn't work to say we were taking care of one party in there really well and paying no attention or not addressing the consequences or possible consequences. And I don't think that – and so the second thing that's important to say is I don't think it's fair to characterize the judge's decision as not accounting for the services here. And the judge's at length explains that the services are competing in this race to the bottom, this competitive market where they drive their prices down, and that they could, in fact, decide to raise prices or have a different model. And the level of profits they have is not fixed. But I surely did not read the board decision, and tell me if I'm wrong, to say that this will all be fine even if the sound recording holders do not lower their rates at all. We think there's enough fat in the services operation that they can absorb this and still be viable. You're not suggesting that. So I think the judge's decision – I'm not suggesting that. But I think the judge's decision is fairly right in saying, oh, there is some legal room in the services profits and that they can accept some losses. They say this is not the maximum amount of losses because they're engaged in this race to the bottom. And they also say, and there is evidence to suggest, that record companies, if necessary, in these negotiations that are going to happen, the record companies will lower their prices because they will have to. Well, they have to. They have to if they want to continue making well over a billion dollars from the services. There are options. If there is really no money left, if we give the songwriters their fair share, and the amount of money left after that is insufficient to pay both the record companies at their current level and the services to maintain a business, the record companies in their negotiations have two options. They either say, okay, we're going to lower our prices because we know you can't pay us more. Or they're going to say, well, I guess we're walking away from the table and we're forfeiting well over a billion dollars. Or we're bringing it in-house. That is – I mean, I suppose that's another option. But, of course, there are transaction costs and startup costs for that as well. And what the judge has said here is that there's no evidence in this proceeding that the record companies, that this level of loss is going to lead the record companies to leave the table entirely. Where is the evidence that it won't have that kind of success? So, I mean, I think the strongest evidence is that they won't. Yeah, the strongest evidence that they won't is remember that the copyright owners here proposed a rate increase that was even more substantial than the ones the judges, in fact, adopted. And if it were the case that a substantial increase of that kind were going – were going to lead the record companies to walk away, surely it would be the case that an increase of the kind the copyright owners proposed would lead the record labels to walk away. Right? This is a concern about the degree of increase. Because if we're splitting a pot of money, if we're giving a certain amount to the copyright owners and there's less money to negotiate over, if we give more to them, there's even less to negotiate over. And so if that evidence – if that evidence was not presented with respect to the copyright owners' proposal, which was even more significant, then of course, you know, the lack of evidence here is – sorry. The lack of evidence here, you know, it should be here. If it were – if there were evidence, it should be here because it would also apply to the copyright owners' proposal. I guess I just want to – because their whole rationale was – the board's, as I understood it, was we can't completely do it just to model safe if they're not taking account of the oligopolistic nature of the record companies. Yes. And so we think that the magic bullet here is to assume the seesaw, take that that's been proposed by these folks, but just not use their numbers and use lower rates, somewhat lower rates. And that's how we will deal with making sure that the industry players stay arranged as they currently are. But I just don't – I don't – and then somebody would pick a number. And I just don't know where the evidence is that by lowering rates the amount they did, they have somehow found a magic bullet point, a magic point where they can ensure the seesaw will – well, they didn't say how much it was going to balance, but will adjust somewhat – just enough, I guess, to keep the services breathing. Well, I think what's important to note here is when the judges set the rates, they first figured out what the appropriate percent of revenue rates was. And they did so based on predictions that the experts had used. One expert's prediction about how much the services should be paying to both, and another expert's prediction about how that money should be split between record companies and songwriters. And from that, they derived their reasonable revenue rate. And then they took – they had to find a total content cost rate, and they took – they picked a total content cost rate that as applied to current market conditions, as applied to the amount of money that one of the services is currently paying to record labels, it would equal the same amount of money as that reasonable revenue rate. Those numbers are designed to produce the same general amount of money. And so the question is, how have they determined that? The answer is the two rates are effectively supposed to, at least in current market conditions, produce the same amount of royalties. Could you explain the meaning of footnote 137, which is where the judges discussed the sense of concern about the consequence to the freedom services? Yeah, I absolutely can. And I think footnote 137 is best understood in light of the discussion on that case, in which the judges say there's no evidence in this proceeding that the record companies are going to walk away from the table. And in footnote 137, they're explaining, in response to Judge Strickler's concern, that a change in rates is going to change the way the market adjusts, right? It's going to change the negotiations. They say, well, of course, you know, whatever we do in this proceeding, it is a static, it's a fixed point that will influence other negotiations in the market. If we lower the amount, if we raise the amount, it's going to affect how much money is left for the services and the record companies to fight over. And that might lead maybe someone to decide that this isn't worth it. Maybe one service or one record company will decide that this isn't enough money anymore. I agree. That's the way I read this. But what about the consequence that the entire industry will go out? I don't take – I mean, I think that the judges absolutely did not say that there was a reasonable possibility that the entire industry would go out. Did they address that question? At least on that same page in the text, they say there is no evidence in this proceeding that the record companies will entirely walk away, which will lead to the entire industry going out. But don't they need evidence the other way to make sure they're accounting for – I mean, you're asking questions, but you say there's no evidence that they will, but don't they need no evidence they won't to make sure that we're protecting the services' fair return? Well, so I think the evidence that they won't, the judges relied on, is that these are all reasonable market actors who have the ability after these proceedings to negotiate on their behalf in their own self-interest. These are major tech conglomerates who will be negotiating with record labels in order to figure out how much money they're going to have to pay in the future. And they can renegotiate these contracts on a continuing basis. The one party here that does not have the ability to renegotiate is the songwriters. The songwriters are fixed in this proceeding. And I think that's why the judges determined that it was more important here – It's assumed the continued existence of both the services and the record companies. I'm assuming that they're making a lot of money, but my assumption is that they're rational economic actors. And when asked to take a hit to that big profit, wouldn't they rationally sit back and go, this is a better decision for us just to take this in-house, or say no, drive them to bankruptcy and buy up the services ourselves, rather than just giving up money, which is going to make our shareholders unhappy? I suppose that is – I mean, that's a speculative possibility of things that could happen. But this is all speculative, predicting what was going to happen economically. So I don't think that label is going to get out of vain. They needed evidence that they were protecting services. And to be very clear, Your Honor, these are all things that could happen regardless of what happens in this proceeding. And so the nature of – No, because my assumption is that the current – or the previous program status quo was that not just that the record companies are making lots of money hand over fist, but that money – they make more money that way, having their services pay these high rates to them than they would if they took the product in-house or bought up some of the service companies. But that all changes if they are suddenly – instead of a seesaw, have their services in between high rates to the record companies and high rates to – well, justify the higher rates to the songwriters, they squeeze them in there. And one of the options then is they go bankrupt. They're already living on the edge. They go bankrupt and they'll be cheap to buy, or we just take it in-house on our own. And that reduction – you have to address whether the reduction, the assumed reduction in the amount of return to the record companies would at that point make it more economically beneficial to them to do one of these other courses of action rather than just following the law. And again, Your Honor, I would repeat the point that I made earlier, which is that the copyright owners proposed a very substantial rate increase, and that these concerns that Your Honor is identifying, that the services have identified here in their briefing and in this argument, would all apply with respect to any rate increase of any kind. And if they did not produce evidence of that type of disruption – We're at this point where we're totally squeezed. Or they proposed different ways altogether to confuse – Well, Your Honor, I think that is certainly what they say, but I think the judges explained that the cap is sort of irrelevant to these concerns, right? But the actual concern here is that we are splitting a pot of money up, and if we give too much – if we give so much money to one side, it means that one of the other two participants is going to have to walk away with less, right? Because there is this fixed pot of money. And the judges say that these concerns about the record companies walking away or the services losing money and maybe having to go bankrupt or services leaving entirely by going in-house, all of these are concerns with respect to any rate increase. And it's a question of degree and not of kind. And the judges determined, as the Copyright Act tasks them to do, that the rate increase they picked, the percentage they picked, and the total content cost percentage they picked, which both amount to the same, was not sufficient. It would not lead to these kinds of results, because there was no evidence in this proceeding that even the much higher proposals that copyright owners had proposed would lead to these sorts of results. That whatever they do here, there is going to be some effect on the market, and that that fear of possible speculation without any concrete evidence to support it, the speculation should not allow paralysis to make the status quo the status quo. Are both sides speculating what the consequences are going to be? The speculation label, I'm trying to understand why the board is speculating, because it wasn't totally accepting the experts. It took them so far and then just made its own adjustment speculating, predicting in the future what the consequence would be. And it seems to me the dissent in the argument in the services is not here to call it any more or less speculative than what the board did. They say, no, here's what we predict could happen. It's just as likely, if not more likely, to happen. And so you need to address those arguments. Well, I think it's important to note that the rates the judges actually picked were within the range of the rates that the experts proposed. They did not go outside of this range at all. And so to that extent, that is important to note, because these concerns about the services not having an opportunity to talk about these particular rates is not founded. These rates were well within the range of what the experts predicted. And the judges also said that the two experts whose assumptions they relied on, Professor Marks and Professor Gantz, they said that Professor Gantz's prediction was fully consistent with Professor Marks' analysis. And it was very well supported by the evidence and directs your honors to the end discussion in the final determination of the section addressing Professor Gantz specifically, because the very last sentence of that section states that Professor Gantz's analysis is confirmed by Professor Marks' analysis. Can I ask you to address the fair notice question? Yeah. Which specifically? The fair notice with respect to the rate structure? With respect to the uncapped PCC prog. Right. Which did not occur until after the record closed, which you rely on the fact that Google recommended one, suggested one, but only in a context where they said we would only do this if the rates were a certain amount. Why didn't you have to open up the record for more evidence and argument on this question? Well, a few points, your honor. I mean, as a general matter, it is always the case. In my knowledge, in every single rate-making proceeding, the judges have made various adjustments to the party's proposals and have never adopted them for sale. And the reason for this— That's not really the question. My question is maybe there's a distinction between this and our normal logical outcomes analysis with respect to a new rule that is proposed after rulemaking, but it's different than the original rule. And the question is whether it's a logical outcome of the rulemaking and whether people would have been on fair notice. And the argument here is that this is not just different than what the parties proposed, but this is quantitatively and qualitatively different because it's the difference between kept and unkept. That is the argument, and I just don't think it's founded. And I will explain why. So a few points. And separate and apart from Google's proposal, what the services proposed here was a rate structure with multiple different— that was segregated for multiple different— But the mainstreaming issue, their proposal was unkept. I don't know what you mean by mainstreaming because the service has had 10 different types of services and unkept services. Which one was unkept? So non-subscription bundled services like Amazon Prime, who is one of the petitioners here, were unkept. Ad-supported services, Spotify— Yeah. Sorry, ad-supported services. What is ad-supported? Ad-supported, so advertising-supported services, so services where a consumer does not sign up for a subscription, but the revenue that they accumulate is— What was the kind of service that they asked to be kept? I'm fairly certain all eight of the other 10 services. Right, so that seems like a lot. And that seems like there's a difference between the ones that they were willing to use as unkept and the ones that they wanted kept. So maybe you're right. Maybe they're really the same thing. Well, notably, Your Honor, I think it's important to note that one of the primary concerns here was that the previous rates had been set because of the past rate structure. I'm sorry, one of the primary concerns of the copyright owners, so something that was heavily litigated during the administrative hearing, was whether the previous rate structure, in fact, was fair. And the copyright owners identified a number of problems with the revenue and said that the total contact cost aspect was not sufficient to balance out these problems, in part because the caps were set at such an artificially low rate. So that goes to how high you set the caps, not the unkept ones. Well, not necessarily, Your Honor. There are many things that could come from that. You could pick a rate like the copyright owners had proposed, which scraps the entire system entirely. You could set higher caps, which no one had proposed, and there was really no evidence as to what those caps should be. And I think more fundamentally, this goes to the reason why the judge's decision here is supportive, is that the services had proposed these caps as part of their rate structure in a proceeding that was supposed to be conducted de novo, and they produced no evidence as to why these caps were reasonable. They didn't explain why certain services were capped and others weren't. You're saying they removed the caps totally because the specific caps weren't reasonable? Well, no. They removed the caps totally because the judges determined that the full weight of the evidence did not support a cap at all, and that is a determination that they're entitled to make. Entitled to make, but only if the other side has a chance to put the arguments opposed to it. I mean, all of the arguments that the dissent makes here are ones that could have been supported, or maybe they couldn't have been supported, but it seems like there should be an opportunity to support them. Well, and I guess that goes to, and I've been setting this aside for a moment, but this goes to the fact that Google proposed an alternative at the close of the hearing that they were entitled to do under the judge's regulations. And I believe the reason that Google did this is because it saw throughout the course of the proceeding that they did, and they were very clear, only under the circumstances of particular rates. They didn't say we're in favor of uncapped regardless of the rates. That's true, they did not. But the concerns that the services have raised here are not that they are upset. Their concern about the cap is not that they're concerned about a no cap of the rate. They have said that a cap is per se unreasonable. And this court should look extremely skeptically on those arguments, given the fact that they themselves proposed an uncapped rate for several services, and also given the fact that when Google proposed this and they had an opportunity to reply, not a single service identified in any of these replies any of these concerns that they are now raising here. They had that full opportunity to do so, and they did not. Again, maybe they didn't have the concerns with respect to the rate that Google suggested. Let me just switch, because we're out of time now, to two other procedural questions. I also have some difficulty on the bundling question about the rehearing. What statutory section are you relying on for that? I think that the judge's decision could probably be understood under either the rehearing or the C-4 provision. So the C-4 says technical or clerical errors? Yes. And it also says changes to the terms but not the rates in response to unforeseen things that would frustrate the proper implementation of the state. It says measure and amendment to the written determination to correct any technical or clerical errors in the determination and modify the terms but not the rates in response to unforeseen circumstances that would frustrate the proper implementation. What were the unforeseen circumstances? I mean, the unforeseen circumstances would be that they carry forward a term that was not supported by the record and that was, in fact, substantively unreasonable and would frustrate the proper implementation. And that was not foreseen? It was not foreseen until someone had pointed out to them that they had carried forward this term that was inconsistent with the rest of the judge's decisions. And if I may, I'd like to explain why. The judges, at length in their final determination, explained the problems of revenue deferral and displacement, this race to the bottom that the services were engaged in. And part of this was done through the calculation of bundled service revenues. And there are field examples that are cited in our brief that I would direct your honors to. And judges discussed this at length. At no point in their final determination did they discuss this term other than to just include it in the attachment of the regulations and when the copyright owners pointed out to the judges that, in fact, this term is inconsistent with their decision and is not supported by the record, the judges acknowledged this. And the services procedural arguments here really boil down to an argument that if the judges notice a fundamental error in their decision, that they are incapable, really, of correcting it unless someone files a motion for rehearing or a petition in this court. And that can't possibly be the way that we read the Copyright Act. The judges have recognized in this and other proceedings that they are capable of correcting errors in their initial determination prior to the federal register publication, even if they don't formally grant rehearing. Is that an inherent power, or is that what you're putting under three? I think it's both, right? Sometimes it will fall under the C-4, and sometimes it will fall under the C-2, the rehearing power. And I don't think it's necessary for this court to address which one it is, because I think it could properly be understood as both. And it's also clear that the judges said, directly to page 2 of the order on rehearing, the judges said to the extent that this is viewed as a decision on rehearing, then we grant it without argument. And that really, I think, addresses this concern. Let me just be my ignorance. How is this about a term and not a rate? Because this is how we sort of count heads for the floor. It seems to me computing the floor would say what that rate is. No, Your Honor, this is about what the base of revenue is that the rate applies to. So remember that the rates are the percentages that have been set by the judge that are set forth on the first page of their determination. And it's really just a question of, well, percentage of it. Doesn't this directly impact the rate? It might directly impact the amount of money that's paid, but the rate itself, the percentage, does not. The rate here just means don't change the percentage. It doesn't mean that changing terms allows you to change things that directly impact that percentage. I think that many of the terms might impact, just as a general matter, many terms might impact the amount of money that's actually paid. So, for example, one of the terms that got changed here that the service suggested was the definition of a fraudulent stream. And, of course, if something is a fraudulent stream, it would not fall within the money that they're paying, and it would then lead to other consequences. And so, of course, changing the term there will impact the amount of money that's paid, but that doesn't mean that fundamentally the rate itself has changed. It's very narrow here. This argument about unforeseen circumstances, I didn't see it mentioned either in your brief or in the judge's determination. Am I right about that? So our brief did, in fact, say that you could rely on either C2 or C4 in supporting the judge's rehearing here. Yeah, but the C4 was about the technical or clerical. Was it about the unforeseen circumstances? We did not specifically focus on unforeseen circumstances because I think our broader point was that the judges have the ability to correct errors in their determination and that where you find the source of that authority is a little less important than the overall point that they certainly have the ability. Let me ask you one more procedural question. On the retroactivity question, I didn't see any discussion at all in the judge's determination of the registrar's statement about not being able to engage in retroactivity. And I think that's because the judges properly determined that this was not retroactive rate setting. So the registrar's decision in the determination that has been cited by the services was addressing a very different circumstance that we have here. So here we had the judges set in 2016, say that the effective date would be 2018. In their initial determination in January 2018, on the very first page of the initial determination, and I believe the services cited – Are you answering by relying on your image or some other argument? No, Your Honor. The judges – I'm saying that the judges had always set the effective date prospectively to be January 1, 2018. They did so before the final determination. They did so at the very beginning of this proceeding. They did so in the initial determination that was announced in January 2018. I would direct you to JA-725, which is the very, very first page of that decision, which says that this proceeding was to set the rates during the rate period beginning January 1, 2018. But the statutory – to the extent we can derive from the statute what they mean by retroactive, there's to be before the final certification of the result. So if there is a circumstance that would be considered retroactive with respect to this statute – I mean, so looking at 115, it says that the effective date can be no earlier than January 1 of the second year following the year in which the petition requesting the proceeding is filed. And so to that extent, if there were any question of retroactivity, it would have to be because the judges made an effective date before that very first initial date, which is what happened in the registrar's decision that the services relied on. The judges set an effective rate that went beyond what the statute would otherwise allow. And that was the problem the registrar identified. That did not happen here because, of course, January 1, 2018 was the first of January, the second year after the proceedings initiative in 2016. And so that is fundamentally the difference between these two things. The discussion about the agreement I think is properly understood as the judges saying, under 115, the effective date begins after the successor rates end. We are allowed to set the effective date. We did. And we know under 115 that the parties might agree to a different period. It says so at the very end of this section. It says, or such other period as the parties agree. So the parties had the full ability to sort of supersede the default date that we had set, to set an additional or even longer period or perhaps a shorter period, different end date. They did not do so. Throughout the course of this proceeding, they operated as if the dates that we set would apply, and that is what that discussion about agreement means. If there are no further questions about any of the other issues, I would ask that your honors adjourn. Thank you. Thank you. Thank you, Judge Henderson. I intended my brief time to focus primarily on two issues. I'm going to start with the issue of fair notice that Judge Harlan raised and then turn to the issue of the seesaw and the disruption to the services, which Judge Millett raised. And I might have a brief word in my remaining seconds about retroactivity. On the issue of fair notice, it is certainly true, as the government points out, that the possibility of an uncapped TCC prong was in place throughout the proceedings. It was in play because at least three of the services had proposed an uncapped prong with regard to at least certain categories, and it was certainly in play because Google proposed an uncapped prong across the board. And when Google did so, it was precisely in order to defeat my client's alternative proposed model, the proposed per-play, per-subscriber model that the copyright owners had advanced. And notably, when Google proposed an uncapped TCC prong, none of the other services objected. Now, to be sure, as Judge Harlan rightly points out, Google made it quite clear that its proposal was tied to its proposed rate. But it is a fundamental principle that the board was not bound to adopt any particular participant's proposal in total, and Google could have been under no illusion that the parties were vigorously disputing what the appropriate rates under the services proposed model should be, and in particular, what the appropriate TCC rate would be. And so, again, I think from a fair notice perspective, while Google may have been disappointed that the final rate was higher, Google was certainly aware of the fact that the parties had presented experts who had proposed a range of TCC rates, and, of course, in the end, the court did adopt a somewhat higher rate. So I think from the point of view of fair notice, although it is certainly true that the ultimate proposal that the board adopted was not identical in every material respect to the proposal as any of the parties, the parties were on notice of the issues and had ample opportunity to present evidence. And, indeed, Google's own proposal obviously relied on the evidence that had been presented at the hearing. And so to the extent that the services point to the purported lack of an opportunity to present evidence, no one was coming in and saying that additional evidence was required. Now, turning to the substance of all this and to Judge Millett's questions about the appropriate rates, to step back here and look at what the board actually did. The board appropriately calculated the rates for its final determination, and, in particular, the board, I think, appropriately took into account the prospect of disruption even as it was setting the rates. As the court is well aware, the court ultimately adopted the TCC rate that was proposed by Professor Gantz. But having done that, precisely in order to account for the perceived market power and the complementary oligopoly on the part of the record companies, the board applied a haircut. And as Ms. Utrach rightly pointed out, the way that the board did that was to take into account the existing revenue streams of the services in an attempt to establish a TCC rate going forward. There would be a rough proxy for the revenue rate going forward based on existing data. And so to the extent that the board did that, it was precisely in order to account for the prospect of disruption. Now, there was a lot of discussion about this issue of the seesaw, and I think it's the better reading of the board's decision that the board did not dispositively rely on the existence of the seesaw. It certainly noted the fact that there was evidence, and, indeed, there was evidence from the experts on both sides that there might be some degree of a seesaw effect, which is to say that once more surplus was allocated to my clients, to the copyright owners, that the record companies in the negotiations could take that into account. But ultimately, the relevant question for purposes of the 801B1 factors was, would there be disruption? And on that question, the board correctly concluded that the services had failed to present evidence of disruption and evidence of the level of disruption that you were talking about there, Vladimir, colloquially, with the G-TRAC, namely disruption that would drive the services out of business. The board correctly recognized that, of course, any rate increase on some level could have a disruptive effect, and a disruptive effect on particular players in the industry. But I would respectfully submit that it was incumbent on the services to present evidence of disruption. The services were well aware of the fact that there were proposals on the table, both under our proposed rate structure and under their proposed rate structure, that would lead to even greater rate increases. And yet, they did not present evidence that that sort of disruption would, in fact, take place. Isn't part of the problem that they wanted to require under D an immediate or imminent impact? That's how they read impact, is immediate impact. And maybe that's the problem, is they have too narrow a view of how they ought to assess the consequences of the changes they're making. Well, I think that the board took a sort of common sense view of all of this. And that view was simply that it was not their role to protect particular market actors. And, of course, to sort of step back here, that the board has only a certain degree of control here. Obviously, it is regulating the rates for the mechanical licenses. It's not regulating the recording companies themselves. All of the experts acknowledged on the issue of the so-called TCC rate or percentage that under existing market conditions, the percentage of the royalties that my clients were getting was too low. And I think that the board appropriately accommodated for that. And it recognized that, of course, any rate increase would have a potentially disruptive effect and took steps to mitigate that. The only disruptive effect they care about is one that's immediate and irreversible in the short run, a substantial immediate and irreversible in the short run. Then they go, okay, so we'll just phase them in over five years. And that takes care of immediate irreversible in the short run. But it doesn't address the reality of the impact over the five-year period. Well, I do think that to the extent that the board phased in the rate, obviously that had the effect of essentially lowering the average rate of the increase over the five-year period. And starting next year, the board will be initiating proceedings to consider the rates for the next five-year period. And at that point, if, in fact, there are long-term disruptive effects, the board will take that into account. My position is that the – But they also need to – to the extent they thought by increasing these rates, we will put pressure on the record companies to lower their demands. But isn't the difficulty, then, with phasing in the rates is that, in fact, you're not going to have that pressure on the record companies because it's just – it's going to be too incremental for them to really be able to insist upon it. And each year, they'll be able to say that's your own. Take it out of your own profits. Take it out of your own profits. I am happy to acknowledge that the board didn't analyze it at that level of macular detail. But I do think that the board plainly, simply acknowledged that there was good reason to believe that the recording labels had every incentive to continue to deal with the services, not to drag them out of business. In fact, I think – Why is there good reason to believe that? Sorry? Why is there – where's the record evidence that there's good reason to believe that? Well, I think you actually need to go no further than to look at Google's own proposed findings of fact. I think notably, as we pointed out in our brief, if you take a look at Google's own proposed findings, Google stated at page 696, quote, An uncapped TCC prong protects copyright owners, as it is beyond doubt that the labels will not price their product below what they believe to be reasonable, given the nature of a services product offering. But they will also not price so high as to disrupt the industry. And I think the board, in its analysis, as Ms. Utrecht rightly pointed out, appropriately took that into account, appropriately recognized that market participants – Just to be clear, so that reading from Google was not part of the evidence in the record, right? Because that was post – That was their own proposed findings of fact. I'm just not correct, but that's – Right, but I – After all the evidence was supposed to be in. Right, that is true. But my point is that there simply was not record evidence to support the proposition that this rate structure would have a disruptive effect. And above and beyond that, I think the board appropriately recognized the reality that the one thing it could not control was these free market transactions between the recording companies and the services. And yet, the board, I think, appropriately said these market participants are self-interested. And, of course, to step back a little bit, Judge Millett, from the details of all of this, the very reason why the experts on both sides were using these Shapley analyses was to attempt to model what would be an appropriate allocation of surplus as between all of the relevant market participants – the copyright owners, the services, and the recording labels. Both sides, of course, presented experts using those models. And so to the extent that the 801B factors weighed the interests of all of the participants, I think that the very way in which the board went about its enterprise appropriately took that into account. Can I ask you a fact question? Do you know if under the prior regime, the Sonar Record Settlement, if the caps that were in there, if they were getting hit, were those caps serving an actual function? I do believe that they were getting hit. I don't know exactly to what extent they were getting hit. But I think what's notable about those caps, Judge Millett, is the very word cap is a little bit misleading. What they really were was sort of an alternative per-subscriber metric so that if you hit the cap, the number, it would essentially switch to a per-subscriber metric. And that's precisely why I would submit the board ended up deciding to adopt the uncapped TCC model was to ensure that the TCC prong served its intended purpose of addressing the risk of revenue deferral. And so, you know, as a substantive matter, I don't think that the services really can have a valid objection to that. And then it really does become just a question about the rates. And, again, I think the way in which the board went about setting the rates reflected reasoned decision-making and certainly was supported by substantial evidence. I think the last thing I would say on the issue of retroactivity, I'm happy, of course, to answer any questions on the service revenue definition. But on the issue of retroactivity, I think the last thing I would say is that I really do think that the parties were consistently operating on the understanding that these rates would take effect on January 1st, 2018. And I would actually point you to the same document that my friend, Mr. Angstreich, pointed you to, the initial determination itself. That initial determination came out, I believe, on January the 26th, 2018. And now, for the first time in oral argument, Mr. Angstreich points to the somewhat equivocal statement at the end of that initial determination. But if you look at the first page of the initial determination, it makes clear that the rates were being set for the period 2018 to 2022, as the final determination did. And you would think that if the services, in fact, objected to that, that you would have heard from the services during this time period or supported retroactivity that they were unhappy with that. And yet, you didn't. The only thing that the services point to is a single sentence in a brief that was filed in May of 2017 that essentially misread the applicable regulations to suggest that the new rates could become effective only two months after the publication. And I would submit that that is insufficient to essentially preserve this objection. And, of course, we agree with the government's submission that there really isn't retroactivity here at all in light of the fact that everyone was on notice before the January 1, 2018 date that that would be the effective period. All right. Thank you. Thank you very much. Mr. Fink. Thank you, Judge Henderson. In this part of the argument, I want to return to my friend Mr. Shanmugam's argument about the student and family discounts, which, as the court noted, is a very, very focused factual argument whether there was substantial evidence to support the factual premise that students and families have a lower willingness to pay. I think, as Judge Gillette pointed out, the services all do that, and they do that because they've determined, and they're experts in the market, that there is a lower willingness to pay. But the additional highly relevant evidence in the record are the agreements with the record labels, who, notwithstanding their great market power, accept agreements that have lower quality payments for student and family plans because they, too, realize that this is a way to maximize revenue that otherwise would not be captured. So we think those agreements are a very, very powerful factor in support of the board's determination. My friends on the other side say, well, there are a lot of other elements of those deals, but certainly it was within the board's authority to conclude that they provide substantial evidence for this determination. And one last factor is Mr. McCarthy testified with respect to a study that was conducted that found that students have a lower willingness to pay. It's true, as my friend says, that the study wasn't introduced into the record, but there was no objection to that testimony about what the study found. So we think all of that provides very significant support. All right. I don't know if the court – I know I'm out of time. I would ask for a rebuttal, but I don't know if the court wants to hear all the interveners and then have us come back for a rebuttal. The message I got from the clerk is that both you and Mr. Shanmugam have given up your rebuttal time to respond to each other. No, no, no. I don't think that's right. I think the government ceded time. These two minutes saw us, and then secondly, I've reserved two minutes for rebuttal, but we'll get you some. So why don't you use two minutes now to rebut? So let me turn to the rate level question. The critical issue here is 80B, but it's not, as my friend says, the disruption factor, although that's relevant. It's the fair income factor. And it's not the TCC problem of the rate. It's the revenue problem of the rate that doesn't work, the problem that didn't have any adjustment. When the board analyzed the fair income factor – and, again, that's on page 81076 of the joint appendix – it didn't say there's fat in the services, we have to give an increase to the copyright owners, but we conclude that sharing the burden between the services and the copyright owners, recognizing we can't do anything about the record labels, the real-world additional burden that we're imposing on the copyright users is fair, and we'll leave them with a fair income. What it said is we have this model that we've constructed. It shows that the copyright owners will retain a dramatically greater portion of the revenue than they're retaining today, and it's based on the model determination, which would, if it were happening in the real world, result in the $500 million reduction in what the labels paid them. It's based on that determination that they concluded there was a fair income. So it was a determination that has absolutely nothing to do with the reality of what was going to happen once these rates were put into place. As I say, they could have done it another way, but they didn't. And I think the fair reading of footnote 137 is them saying we don't really care what the impact is on the services. If the market gets totally reorganized, as Judge Strickler said, with a reasonable possibility, that's sort of tough luck for the services. And we think, therefore, it's critical clear that that fair income determination can't possibly be upheld because it rests on a model that the majority itself, on page 1061, said does not comport with the real world because it does not take account of the labels of monopolistic power. Maybe the difficulty with a cap is that they explained that, look, our goal here is there's a huge distortion between what the record companies are getting and what the songwriters, who really are the genesis of all this business, are getting. And we don't have authority to regulate directly the record holders in this proceeding. But we do have an obligation to protect the songwriters. And so we've done what we can to protect the songwriters. We believe, based on this economic analysis, that it will still leave enough for the services to survive. We're making predictions here, but we've looked at all this testimony. We're making that reasonable prediction. And after that, there's really not a whole lot more we can do to help you with the record companies. And capping the songwriters seems to just punish them for the overreach of the record companies. Why isn't that reasonable within the range of reasons? Well, I want to distinguish between the uncapped TCC structure and the rate levels. What I was just talking about are the rate levels. The uncapping of the TCC structure, just to start from the beginning, the theory was this will get more money. It will prevent revenue deferral. There certainly would be a way to do that based on percentages without leaving it wide open if that was a goal. But I think the critical question here is when it came to the rate levels, the majority didn't say, what are we going to do here? We have to get more money to the copyright owners. But looking at the real-world implications of what we're doing, it isn't going to be, it isn't going to drive the services out of business. It is going to give them a fair income because, after all, driving them out of business is not the test. Fair income is the test for the factors B and C. And what they didn't do is look at the reality and say, yes, there will be a fair income for the copyright users. That's what the statute required. They said we're going to base our fair income decision on a model that we know isn't going to produce, isn't going to happen in the real world. So they could have made the assessment that Your Honor hypothesized if they provide an explanation and if there was evidence in the record to say that wouldn't, that would still provide a fair income to the services, but they didn't do that. They basically said we're just going to rely on this model, and we're not going to look at, or they didn't at least look at, what the implications of their decision was going to be in the real world for the services. That's dispositive, we think, under those factors. And we think under the disruption factor, as your colleague, Mr. Shanmugam, indicated. Submit 137 basically weeds out of the statute the disruption factor by saying, well, if the disruption is that the industry is totally reorganized, that doesn't matter. That's not something we have to consider. And there was no evidence that they cited for their conclusion that this wouldn't happen. And, in fact, just to give a real-world example, it's not in the record, but Disney, a content creator, has created its own streaming service in competition with streaming-only services. It concluded that that was actually, given the economics of that industry, a logical thing to do. So the idea that this is impossible to happen in this industry, notwithstanding, even if the record labels will face the choice of lose $500 million or reorganize this industry, as your honor said, the logical thing for them to do would be to say, we still want to be paid what we're paid. Maybe the current system will survive and we'll keep making our money. But if it doesn't, we have alternatives. Maybe the services will go bankrupt. Maybe we'll be able to start our own competitions because they'll be led dry. Those are hypotheses that Judge Strickler talked about significantly, and there was no evidence to support the majority's decision that they wouldn't happen, and no evidence to support Professors Wright's decision that they wouldn't happen. And I wasn't clear from your brief. I thought at one point you were objecting to the board's definition of disruptive impact, that it was short-term, immediate, and irreversible, but that it didn't seem to be fleshed out as a statutory construction-type issue. And I haven't heard anything about that today, so I take it then that that's just not really what you're pushing. No, that's the point that I was just making. And I think we did try to make that point maybe more explicitly in our brief, that the board's definition, its determination in this 137, that changing the structure of the industry is not disruption that it has to consider, it has to be wrong, just as a matter of the plain language of the statute. If I could start with retroactivity. The first page of the initial determination, page 725, just repeats what the board's intention was at the outset of the proceeding. And had they published the final determination in a timely manner, the rates could have taken effect on January 1st, 2018, or at some point in 2018. But it's on the last page where they actually address the effective date, and there they get it correct. 803-D2B, interpreted in light of Bowman and the rest of the statute, doesn't allow retroactivity. 115, which Ms. Utrecht pointed out, has nothing to do with retroactivity. What that says is, board, even if you had finished this in mid-2017, even if you had published this in the Federal Register on June 1st, 2017, 1003 couldn't have taken effect until January 1st, 2018. That has nothing to do with retroactivity. It's simply the earliest allowable effective date. As to bundles, I think it's telling that the government can't identify which provision the judges relied on here. This is not technical. We see technical changes. We asked for some. The copyright owners asked for some. We cited one in our briefing footnote on page 60. The copyright owners mentioned the fraudulent stream definition. Those are ones where the definitions, they used an uncapitalized word, they pointed in confusing directions. This was a substantive change, and the government can't identify a provision that supports, gives authorization to do that. They can't identify a prior example where the judges ever made that kind of change without finding the exceptional cases standard to be satisfied. And they expressly found here that it wasn't. They can't say it wasn't satisfied, you didn't make the pre-reflected case, but if we need to rely on our hearing authority, I guess that's what we're doing. That's not the way Congress set up the system, and so they didn't have authority to make this substantive change. If they don't like the constraints on them, and I recognize this is a complicated undertaking, and the judges may well not get everything right the way they want to in the administration, but that's a dispute with Congress. What about the unforeseen circumstances language? Sure, as we read the unforeseen circumstances, first I agree with Judge Millett that this feels like a rate change, not a term change. But the only unforeseen circumstance here is, as Judge Strickler pointed out in the dissent. You described this as a term question. I would describe this as a rate question. But I think more importantly, there wasn't anything unforeseen here. The copyright owners made the conscious decision not to propose an alternative. As Judge Strickler noted, they made the decision to try to throw the baby out with the bathwater, his words, and say, we don't need a revenue-based system at all. We're going to go with a per-play system. Had they said, but in the alternative, you should change the bundled services rule we could have had that dispute on the record in the initial determination. But that was their litigation posture and their litigation decision. To come back afterwards and say, oh, now that we've lost on our per-play regime, we'd like actually a chance to litigate this issue. I don't think that's allowable, especially when the board finds that there was no evidence adequate to support their proposals, and yet adopts it. Judge Blatt, this is very much like what you found in prompter and settling devotional claimants, where they're making a decision based on a lack of evidence in the record. It's made worse here by the fact that the copyright owners gave up on their chance to make this proposal during the hearing. I think you would have said that it was ironic that they thought you all were on notice of everything to no vote, but they weren't on notice, so that would affect no vote as well. It may be ironic indeed. Thank you, Erickson. Thank you. Thank you, Your Honors. On our affirmative appeal, I just want to address one point that my friend, Mr. Pincus, said. He alluded to the fact that there was testimony from Barry McCarthy of Spotify, alluding to the fact that Spotify had conducted research on this question of willingness to pay. But not only was that research itself not produced in the record, Mr. McCarthy in fact himself acknowledged that he had no part in the preparation of that research, and indeed, even beyond that, he acknowledged that he had not seen the data and was not even aware of what form the data was produced in, and that's pages 1336 to 1337 of the appendix. So our fundamental submission is that there was simply no evidence in the record beyond mere speculation to support that proposition. And finally, I just want to say at the court, receiving a letter from me, I want to provide just one citation for the court for something that I said earlier. It says that the court had taken into account the interest of all the parties by adopting the Shapley model, and the citation for that in the record is appendix page 1075. Thank you very much. All right, thank you. Madam Clerk, I have that Mr. Johnson has rebuttal time. Is that correct? Yes. All right. I don't have much to rebut, but if I can just make a couple of points I just wanted to make, Your Honors. I would say in defense of my friends at NMTA and NSAI, one relevant point is that the 44% increase in streaming rates is clearly not going to make or break these streaming services financially. And the reason why is Apple's not here. And if it was going to make or break these companies, I think Apple would definitely be here to appeal. I've made the point before again that American copyright creators, painters, photographers, illustrators, they're not subject to a compulsory license. And it's still a copyright registration. It's an individual copyright, an individual expression of art. And in these hearings, you know, it's important to have market share. I have no market share. It's important to have benchmarks. I tried to get a benchmark. I couldn't get one. You know, I have no expert witnesses. I have no evidence because the judges didn't use my evidence. And I have no Nobel Prize winning promise, you know, to come in and be an expert witness for me. So it makes it tough for us. Just, I would say that last thing. Sorry. That if you look on, sorry, page 9 of my May 12, 2017 conclusions of the law, I cited the famous Supreme Court case precedent from 1917, Herbert v. Shanley. And I hope it's still valid precedent. But in that, Justice Oliver Wendell Holmes says famously, if music did not pay, it would be given up. And whether it pays or not, the purpose of employing it is profit, and that is enough. And as you can see from Music Row, you know, all these people have given up music. And that's been the effect of this compulsory license and the 2008 rate structure created in 2008. And if you look at, you know, there's a lot of cases, Supreme Court cases that use the word specific profit. They're not using income or payment or make money. And in American Geophysical Union v. Texaco, they say, quote, accordingly, copyright law celebrates the profit motive, recognizing that the incentive to profit from the exploitation of copyrights will redound to the public benefit by resulting in the proliferation of knowledge. So the profit motive is the engine that assures the progress of science. And when you look at the 801B standards, and I made this argument, I think, in my motion for rehearing, it seems like that those four criteria, it's 100% for the services and zero for the songwriters as far as a fair return. And it's really about the profit. And, you know, in 2015, the Copyright Office did a copyright reform study called Copyright in the Music Marketplace. And it says there is no policy justification or standard that requires music creators to subsidize those who seek to profit from their works. And I think that's what's going on here. And it goes back to, you know, the equitable division of music industry profits between copyright owners and users. And we don't profit whatsoever. And I think there needs to be a balance of 50-50 instead of just 100% services and zero for songwriters. And I thank you. Thank you. Thank you all for the exhaustive – I didn't say exhausting, I said exhaustive. Thank you so much.

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A69147. Public record. Not legal advice.