Engelberg Center Live!

Proving IP 2019: Proving Markets

Episode Summary

Proving Markets: What are the best ways to approach market definitions, demands, and harms? Featuring Dale Cendali (Kirkland & Ellis), Rahul Guha (Cornerstone Research), and Mark McKenna (Notre Dame Law School), moderated by Scott Hemphill (NYU Law)

Episode Notes

The full video for this panel is available on YouTube.

 

Episode Transcription

- This next panel is about proving markets and we all know how important it is when you are, when you're litigating a case. Similarly, when we were talking about proving how consumers think about the mark or the product or the similarity between two photographs, what markets you're talking about because, if the markets for the products are unrelated, then it may make it much easier to show that there's not a problem but if they're closely related it can be a significant issue and so controlling that conversation, understanding how it plays out, is critically important. So please join me in welcoming the next panel for proving markets.

 

- Terrific. Hi, everybody. I'm Scott Hemphill. I'm on the faculty here. My work is in both intellectual property and also anti-trust and as Michael mentioned a major proof issue in a lot of IP cases is to assess the nature and extent of competition facing the rights' holder. So just to frame this with a minute or two of reaction, one example would be the extent to which a defendant's product or service competes with the right holder. So this comes up in copyright, in fair use in the fourth factor. It comes up in trademark in the context of likelihood of confusion where the closeness of competition between the senior mark and the junior mark can be relevant for determining the likelihood of confusion. A second place where this shows up would be to what extent has the plaintiff been successful in competing for customers? So, in patent, this arises in the context of obviousness where secondary factors under 103 includes commercial success and some of these that we'll be talking about are closely connected to anti-trust determinations of market power. I'm an anti-trust person. Everything looks like anti-trust to me but when I teach my students about whether sleek craft boats and slick craft boats, there's a prospect of confusion, I'm doing quietly, I may not even mention to them, but I'm doing a kind of quiet market power, market definition discussion in the background and then finally, of course, this could affect not only the question of liability but also damages and so we wanna spend some time today thinking about these different contexts where the issue arises and also how do you prove this up? Is it documents? Is it testimony? Is it economic experts? Is it, is it industry experts? And we have just a fantastic set of panelists to help us think about this and let me just take a minute to introduce them. To my right, Dale Cendali is a partner at Kirkland & Ellis. As most of you all know, she heads up Kirkland's copyright, trademark, internet and advertising practice group. Her practice spans the areas of law I just mentioned and more. She litigated the TVEyes case that we'll be talking about in a little while on behalf of Fox, Harry Potter, Lexicon, Oracle v. Google, Daystar v. Fox, Moseley versus V. Secret Catalog among others. Let me catch my breath for a minute. To her right, Rahul Guha, is a senior vice present at Cornerstone Research, a leading economic consulting firm where he co-heads the anti-trust and competition practice. His work spans intellectual property, anti-trust, and product misrepresentation among others, has a particular expertise in pharmaceutical and healthcare matters. Recent work also includes AT&T Time Warner, a merger that's been quite in the press recently and the recent combination of Essilor and Luxottica. Finally, to his right, Mark McKenna, is the John P. Murphy Foundation Professor of Law at the University of Notre Dame. His research and teaching focus on IP and privacy law. He has a particular expertise in trademark law though his work ranges more broadly and he's written, among other topics, on the role of market definition in anti-trust intellectual property cases. Some of his most recent work focuses on questions of design protection. So, as you sense here, we have some domain expertise and copyright and patent and trademark and also some different approaches. We've got a litigator, an academic, an expert economist. So I think what we'll do is take some time on each of a few different topics. Someone's gonna offer, one of us will offer, one of them will offer some initial remarks and then we'll try to have a little bit of a discussion. So why don't we start with Dale and some of the copyright questions that arise?

 

- Okay, thank you, and I'm really honored to be here with so many amazing professors and others and thank you NYU professors and others for training some of our terrific associates, some of which are here today. One other thing, I'm a lecturer at Harvard Law School and one of the things that I always say to my students is, "The facts are not a given." And that was one of the things I was attracted to about this panel. Now, at first they got nervous and thought that I was telling them to make up evidence. That was not the point but if you have a red light and your client went through a red light, being able to use discovery or experts or otherwise to show the context of that could be vital to what happens to your client but yet, too often when we read these cases, the judicial decisions just, it's as if Moses delivered the facts and we all know as litigators that actually they may only reflect the tip of the iceberg, they may reflect some of the facts and the judges won't know the facts unless you can find them out and present them in a persuasive way. So I was asked to talk about the, our TVEye's case that my team and I litigated as a way of explaining how you would use and develop evidence to help present facts in a copyright fair use case and for those who may be unfamiliar with it, TVEyes was a service that copied everything on television 365 days a year, 24 hours a day, and they made it available in seriatim, high quality, 10 minute clips to paying subscribers and they gave their subscribers a lot of freedom for what they would do with the clips. They could, as their own marketing materials say, watch live TV 24/7, download unlimited clips, play unlimited clips, email unlimited clips, store unlimited clips, edit unlimited clips, and post unlimited clips to social media and I mention this because we use, as part of proving IP, right, the topic of this, we use their own words because very often in a case sometimes the lawyers characterize things a little bit differently from how the clients actually said things. Oh, well, we don't really use our service for any of these things. Our service is really, you know, for high level research. It has nothing to do with social media and marketing campaigns but if their documents say we're pitching this to social media and marketing campaigns, that's less of a compelling argument and you see that too in the Salinger case, the one about the six years after case where at one point their original marketing documents characterize that work as a sequel and then later they tried to say, oh, it's really not a sequel, it's a commentary. The district court judge said, "But you called it a sequel!" And it was hard for them to back off that. So then when you're looking at proof of market harm in a copyright case, you have to start with the statute which says, in determining whether the use is a fair use, the factors to be considered, shall include for the effect of the use upon the potential market for or value of the copyrighted work. And as a litigator, that starts by informing you that this isn't supposed to be a damages analysis. In other words, if we needed Cornerstone to figure out how many exact widgets were made or lost, well then we could do that in the damages part of the case but in this part of the case, it's supposed to be, is there harm to not just a market but a potential market? And the Supreme Court gave us even more thinking on that and they said look at the harm caused by the particular defendant but also look at the harm not just to the product in question but to derivative markets, markets that you may not be in now but you might reasonably go into and, by the way, also look to see what would happen if everyone and his brother did it, if it became widespread. So once you know the law, it's easier to figure out what evidence you should present to help make those points. If you don't know the law, it's kind of like garbage in, garbage out, and it's not useful for anyone and it's wasteful for the clients. So in TVEyes, what we did is that we needed to explain to a somewhat skeptical district court judge why we cared and we needed to explain how the television industry was changing and I would often say to my class, "How many people get their news by every night "watching the six o'clock nightly news?" No one, one person, okay. Most people don't. Everyone that ultimately agrees that people like mobile delivery, they like short form clips, they like small videos, things like that and what we needed to present and explain to the court was that that was how the industry was going. That was the value and interfering with the ability to monetize their clip business and to take that for itself was harmful and so what we did is we had both the client testimony, we had an expert, Dr. Beth Knobel, who was a, is a professor at Fordham and a former TV producer on journalism and we had documents that helped illustrate our points and the points that we made, we had lots of evidence of market harm. This relates to the punchline of this in that we talked about direct harm to Fox's TV Everywhere program that let's you watch TV live, Fox news programming. We had lots of harms from TVEyes, Fox New's website where it, just shortly after things are on the air, you will have all the clips of basically everything that happens on Fox News is available for free for you to look at and it's all archived for free. You just have to watch a pre-roll ad which is how it's monetized. So we talked about that. We talked about agreements with syndication partners, with our clip sales business, marketing for media monitoring services that are out there and also diversion since they advertise this as watch live TV from the Fox News Channel itself because we explained that the Fox News Channel is not available over the air, that the only reason Fox News and CNN are able to exist is because they have paid subscription services and that was the thing. We also came up with trying to tie these facts together into an overall theme of the case and the theme that we developed was finding versus delivering. The TVEyes aired by going similar to what Judge Leval was saying. It actually delivered content as opposed to helping people find content and as you can see from this excerpt from my opening argument to the Second Circuit, that's what we argued, that there's a fundamental differences between services that find copyrighted works and those that deliver copies of such works. At the end of the day, the Second Circuit held that it was not fair use and in particular with regard to factor four, the court said, look, there's clearly a market here. TVEyes own success shows that it's a million dollar market. We got it. There's harm. Now if you saw and compared this though to that list of harms that I told you we put in evidence on, you'll see this is just a small subset. It's like we had the court at hello. We presented so much more evidence than that. We were expecting a 20 page session or part of the opinion possibly on all this different market harm. The court didn't need to go there. The Second Circuit can do what it wants. It found enough. It didn't need to but in litigating the case, we didn't know what was gonna be persuasive and so we had to put in everything and if the other side saw it cert, cert was denied but who knows if the Supreme Court might've wanted to know more information. So that is ultimately our point is you have to know the law, find the facts, find them in a persuasive way, and be prepared for the fact that you don't know what the court is really gonna find impressive so you may have to go in more than one direction.

 

- Mark, let me turn to you. Thinking about whether it's a fourth factor test or related areas in trademark, how does this proving markets question play out?

 

- Yeah, I mean, so one of the things that I think is especially a challenge always in the copyright context is that when you're thinking about that fourth factor and you're explicitly contemplating derivative markets, you have this sort of well known circularity problem, right, which is the question of there's an inescapably normative conclusion that's involved in there which, basically, which kinds of derivative markets are you willing to credit and are you just essentially drawing the conclusion that those are derivative markets to which the copyright owner is entitled, right? Because, obviously, any market that you give people legal rights over then you're gonna be able to find market harm in those markets. So I guess, I'm curious, Dale, from your perspective, how you think about that question when you're trying to prove the existence of the market? How do you differentiate between derivative markets that you think you can prove to a court are ones that are clearly within the established right and those that seem like we would really like to have this as a licensing market and if you extend our rights to that market then we will.

 

- I mean, the flexibility you have, it's not just about you and your client. Maybe your client isn't big enough or hasn't done everything but if other people in the industry have, it could still be reasonably expected that they may go into it. If you're expecting, has anyone used their copyright on the moon or something like that, something that's hard to reasonably anticipate, then I think you have a more challenging situation but if you had marketing materials where the clients were brainstorming and maybe we want to do this and maybe we want to do that, that could help prove that that was a market that was reasonably contemplated.

 

- I mean, implausibly exploitable gives you, under that language, a fair , breathtaking might be too strong but a fair amount of running room.

 

- That's right and that means the more you can, similar to what we did with Rat-meister, you have to explain to the court so they see.

 

- Mm hmm.

 

- And that's part of what you need to do on markets often with experts and with your own client.

 

- Rahul, let me turn to you for a minute. I mean, putting on your economist hat, you're thinking about this not just on damages but on liability, what role do you see expert economic testimony playing on a question like this?

 

- Yeah, so first of all, I find this framework of finding versus delivering content to be very persuasive. I mean, to me, there is a direct analogy between that framing and the notion of complements versus substitutes that is sort of the natural framing for economists and if you're delivering content that is arguably a substitute and if you are, sorry, if you're finding content that's arguably a complement and if you're delivering content then you're a substitute. So the framing, to me, makes a lot of sense and there's a tremendous amount of intuitive appeal between that framing and how economists naturally tend to think of things. So that was my first observation. The second thing, of course, is, you know, being economists, we tend to get excited about things that, that the rest of the world finds absolutely ludicrous and this case strikes me as one in which there is variation in outcomes, right, and anything that has variation in outcomes is useful for economists to prove the liability point. So let me explain what I mean by that. So in this instance, at least the way I understand the facts is there is a world before TVEyes and there is a world after TVEyes and there are measurable outcomes. What are the measurable outcomes? They could be GRPs for Fox News, so eyeballs, number of eyeballs or as you said, the clips themselves are loaded after the fact on the website so clicks are also measurable outcomes. So this then starts to look very much like an event-sten, so there's a before and after analysis. If these things are truly substitutes then you can do the analysis. You know, and again, there are some econometrics involved. You have to do this thing right but if there is a difference in either the number of clicks or the number of eyeballs before and after TVEyes starts delivering content, then the case is clear. Then you are substitutes. You're competitors and so I think those are the, if I were to ask, if I were asked to kind of address this question empirically, I think that's what I would like to do. You know, I think if you were creative enough, I think that this sort of question can be addressed with surveys as well but I think in this particular instance I would first turn to the kind of data on either clicks or eyeballs to get at this question of whether, you know, are you competing or not?

 

- Great. Mark, let me turn to you and I'm tempted just to say, trademark, go, 'cause I know you've thought about a lot of these related issues. One obviously piece of connective tissue here is, you know, this plausibly exploitable market idea as one particular clause on the fourth factor, reminds me a little bit of merchandising cases and I know you've been critical in academic work of the extension of, or the application of trademark in that context. Sort of putting that normative piece to one side and thinking about the positive question in proving that, I wonder if you have reactions about how we might go about that in a merchandising case or more generally in some of these related trademark issues.

 

- Yeah, I'm gonna treat that like you just said trademark, go, right? So, no, so I do think that, as I've written before, there's a number of contexts I think in a trademark context where at least implicitly what's happening is something akin to market definition, right, but I think as I was sort of suggesting about copyright, there's two, I think, important things to highlight about it. One is not withstanding the frequency of the relevance of that, there's really not a lot of considered methodology for how courts are supposed to do it. It tends to be proven just through argument by the parties and I don't mean to say that just through argument in the sense that there was no evidence marshaled but there's no systematic doctrinal structure through which the court is supposed to filter that evidence and look at it and I think there's a bunch of reasons why that's problematic but one reason for it that I was trying to imply before is that a lot of these market definition questions are not merely empirical questions about the way the world looks. They're sort of inescapably normative questions about what kinds of markets and what kinds of competition we want, right, and so it's really, it's not just do we go out and find facts about the world, it's more like, is this, are these facts about the world ones that we want to encourage or discourage and we might define markets that way. So I'll maybe just use a couple of examples. So likelihood of confusion, I think, is the most obvious one. It's the one where people, I think, probably think the most in the sense that competitive proximity of the products is explicitly a factor that's relevant to likelihood of confusion and in some cases that's true because the parties aren't just arguing about whether the goods are similar in some abstract sense but about whether they're similar in view of the context in which they circulate in the market but I don't think in most of the, in many trademark cases at least, the question is really about whether the goods compete, it's about whether they're similar enough that consumers might think that they come from the same source, right? And in that sense it could be true that they're both goods you might expect to come from the same company even though they don't compete in any meaningful way. So it could be that laundry detergent and dishwasher detergent are not substitutes for each other but it might be reasonable to assume that the same kind of company would make both of them. So in those cases, it's really not a market definition question even though market proximity is relevant but I think there are two other contexts in which market definition plays a much more explicit role and then one thing that I want to highlight at the end is the way market definition implicitly informs a lot of the evidence that gets, so I think the two places are in functionality, especially in the context where what courts are evaluating is whether the features put the, exclusive use of the features would put a competitor at a significant non-reputation related disadvantage. That's relevant especially in the aesthetic functionality cases. It's not relevant to the question of utilitarian function but somebody needs to tell the federal circuit that because it keeps hanging onto it, but in any event, I think the point is that the legal determination in those cases implicitly depends on some kind of market definition in the sense that you can't determine competitive disadvantage without knowing the competitive context of the good. So this is one of my favorite examples. It's an easy one. You might know this project that's called Dippin' Dots. Dippin' Dots is the ice cream of the future. Interestingly, it's been the ice cream of the future since the 80's and apparently there's no future because it's still the ice cream of the future. So Dippin' Dots sells this flash frozen ice cream that comes in these little beads and they sued another company, notably after the expiration of a process patent on the process for making this ice cream, and they claimed that the shape, color, and size of these ice cream beads was its trade dress and Frosty Bites, the defendant in that case, argues that all those features are functional and so, one of the things that was most notable about the case is that the plaintiff says, you know, listen, the defendant doesn't need to make it's ice cream look like this. Look at all these other ice cream companies that exist in the world, Haagen Dazs, Hershey's, they all make ice cream and they compete just fine and the defendant says and the court accepts that that's just not the right question. The question is, the defendant doesn't wanna make ice cream, it wants to make flash frozen ice cream and the court accepts that that's a meaningfully different market and, of course, having defined the market that way, the competitive effects look really different than all of a sudden if you accept that the right framing of the market is just in terms of ice cream. There's a sort of analogous thing that happens in genericness cases where the question is whether the primary significance of the term is to refer to a particular producer as opposed to a type of goods. So in lots of cases, what that means is, the question is really a fight about how we understand a product category, right, what the right way to name that category is. So you might be familiar with the sort of running dispute about the registrability of the mark Pretzel Crisps, right, where the question about whether Pretzel Crisps is generic is really about whether we're supposed to think that Pretzel Crisps is a relevant category of product or instead the product category is just salty snacks or even pretzel snacks, right, and so you, the question about whether it's generic depends on whether you think consumers are sort of drawing market categories in the sort of finer or in a more gross way and that is essentially a market definition question and there's no, I think, really sort of founded nature about that. It's about a question about how you imagine that competitive context. So as I said, what's kind of notable about that is that in none of those contexts is there really developed law that is supposed to tell a court how to systematically think about defining the market. So the parties argue about it. Good lawyers find good evidence about how parties understand who their competitors are and they surface that, but there's nothing kind of in the doctrine that walks you through that. So I think the place that actually you see the most systematic thinking about market definition is actually, in some respects, invisible to courts but it's really, in the context of when you're trying to design the survey. So you get your expert who comes in and if you just take, one easy example, likelihood of confusion survey, the first step is defining the relevant universe. Who are the people you're asking the questions about whether they're likely to be confused? So the people who are supposed to be relevant there are consumers or potential consumers of the defendant's goods but you could, of course, describe those goods at lots of different levels of generality and then increase or decrease the size of your universe with sometimes significant effects on the way you find confusion, right? So if the defendant's using a mark for potato chips, you could define the respondents as consumers, or potential consumers, of potato chips, or salty snacks or snack foods or food, right? And all of a sudden, you're sort of growing or shrinking the size of the universe and what you're doing is defining the market in a different way so as to figure out who are the, the universe of relevant respondents. All of those are things are sort of subject of argument. Do you hire another expert and they dispute the definition of the universe? But again, there's nothing, often that's, that sort of discussion is sort of invisible to the court and it's, or at least it's not done with any sort of clear doctrine that's supposed to guide what the right situation is. Let me just say one more thing without running out which is that one thing I think I see happen a fair amount is also a sort of game that's played in market definition and surveys is that, I just framed it a second ago and I think the right question in these questions is asking about the consumers or potential consumers of the defendant's goods. What plaintiffs sometimes try to do in those cases is to say well, it's supposed to be the universe's, the people who are both plaintiffs and defendants, consumers, right, and I see David shaking his head, because I agree that's got to be wrong because the question of the overlap of the consumer markets is an empirical question that's supposed to be illuminated by the survey. It's not supposed to be baked into the structure. So if you take the sort of assumption, imagine a case where the defendant's consumers, let's say only 5% of them overlap with the plaintiff's goods but if you define the universe to consist of the overlap, you've just defined 5% of the consumers to be 100% of the universe, you might then find 20% confusion but it's really 20% confusion among 5% of the people but you've masked that by the way you've defined the universe in the survey and so I think it would be really useful as a matter of trademark doctrine to get a little better guidance from courts so that it was easier to imagine how you would criticize these games that I think people play with universes.

 

- Okay, so Mark's at the table, and it sounds like we're basically in the wild west here, right? Law sounds like it's not constraining us too much in terms of what proof we ultimately put on so I wanna sort of, I don't know, lay out a challenge for Rahul and then for Dale how we might fill in these blanks whether it's shaping the proof that gets put on or thinking about it from an economist's standpoint. What kind of tools could be brought to bear to answer some of these questions, whether it's genericness or functionality or what have you?

 

- So, I think, you know, let me give you a two part answer. The first, when the products are competitors, I mean this seems like an easier question to address. I think, because why does confusion matter? I think the answer to that is better to be straightforward. So I think confusion matters because it can affect like, as a simple matter, it can affect my sales or profits and so in that case the measurement and like, you know, how that is related to harm seems clearer. The part that I really struggle with and I'm not the lawyer here, I mean to me, it seems that even if the products are completely unrelated, there is a possibility of harm, and to me, it's, and regardless of whether that manufacturer has current plans to make a different product, you know, just make up an example, think of Mercedes Benz. Mercedes Benz may or may not ever have plans to make consumer products, however, the brand has a value. I mean, you see people put numbers on brands. You see these interbrand rankings. So that brand is worth x billion dollars and if you have a producer of a poor quality product in a completely unrelated market, so it's really sort of not sort of competitive markets in any traditional sense of cross elasticities or being competitors, completely unrelated market but something, nevertheless, that consumers would see. It's arguable that the value of the brand goes down and that's the part that I, I mean, to me that seems, again it's an empirical question but the notion that a brand value could be damaged because of confusion seems like the possibility exists. How that fits into trademark law is the part that I don't understand and like, you know, how courts look at this question of like, you know, and I know that I've heard people talk about that, in assessing sort of the likelihood of confusion, I am gonna look at whether the products are close or not but I'm not exactly sure how it plays out because to me it seems like confusion could have a hugely detrimental effect even if there aren't, the products aren't close.

 

- Right, right. Dale, you wanna jump in?

 

- Yeah, I think it's, we're kind of in a new world with trademark law. A lot of trademark law was based on, you can have someone in New York use a trademark for widgets and someone in California use it for red widgets and no one would be confused but now there's so, geography's broken down because of the internet but also you have so much more brand extensions and expansion. So it used to be Hershey's to me was Hershey's candy bars but now Hershey's are sweatshirts and t-shirts and backpacks and things like that. So building up what you were saying, Rahul, if I saw a Hershey's backpack, I could easily think that that must be licensed from the Hershey people and if it isn't, that could be confusion and could be infringement. The other thing I wanted to build up is something that Mark said and that is in how you present evidence about it, is another way that survey evidence comes up is do you do a Squirt form survey or do you do an Ever-ready form survey depending on how the products appear in the marketplace. An Ever-Ready is, you just kind of hold something up, what, who puts this out? Why do you think so? Are they affiliated with anybody else? Do they have a business or did somebody give permission for this or not? That's the kind of thing. A Squirt survey, there's different versions of it, but essentially it can be an array, you look at one thing, you look at another thing, kind of like a line up or some sort of juxtaposition. As you can appreciate, when you have a juxtaposition, there's been a lot of criticism that those kinds of surveys could lead to higher rates of confusion especially if the control isn't good and so there's a lot of debate about and courts have held, and we've gotten surveys excluded in cases we've done for Disney and others that say you can't put together in a Squirt survey two goods that would not be encountered in proximity by consumers at the same time and that's an important concept to know in designing survey evidence because you might think I want to do that but it's not gonna help you if it's gonna get excluded.

 

- Mark, do you wanna chime back in on this one briefly?

 

- Oh, and just to respond quickly to say, I mean, so I think the important question is whether the confusion predicate is present in what you were just talking about. So yes, it might be the case that for some brands, that even relatively distant products, there would be confusion, especially for relatively well known brands. I suspect if you were to test Mercedes Benz products in almost any market, you'd find some measurable amount of confusion. So, but the doctrinal question is just, is not whether there's some abstract harm to the brand but whether the brand, whether the harm derives from confusion. So if it does, then what these surveys are measuring is not harm, they're measuring confusion and then we just, we make the doctrinal assumption that harm follows from the confusion but there's also, of course, there's potentially, there's a dilution action but that, I mean, it's limited to a small number of brands of which Mercedes Benz might be one but famous brands that are kind of household names but that, I think, better captures the harm intuition which you had which is basically that even if products are really quite far apart, there might be some sort of negative effect. Now I and others in this room have some negative things to say about dilution but I won't bore them.

 

- [Dale] And some of us have good things to say about it!

 

- I'm gonna move us on just in the interest of time to patent. Rahul, go.

 

- Yeah, so let me start by saying that, like Scott, I spend a great deal of my time thinking about anti-trust issues but within anti-trust, aside from, outside of other things, a great deal of what I do is in the life sciences industry and in life sciences you cannot, you know, think of anti-trust analysis without fundamentally thinking about intellectual property and patents in particular. So the, you know, so intellectual property issues and anti-trust issues in pharma life sciences are very, very closely aligned and so this idea of market definition is something that I've thought about a lot even outside the context of this panel and to me the difference is interesting. In anti-trust, that is, the idea that you have to define markets is explicit. That is the starting point of the analysis. In IP cases, in patent cases, especially in a field like life sciences, it is implicit. I think virtually all of the important questions that you need to answer either on the liability side or on the damages side fundamentally turn on certain assumptions about market definition. So let me just explain and elaborate on what some of those issues are. So let's start with validity. The validity questions in patent cases, in large part have to do with just issues of prior art and those are issues that scientists deal with and economists have very little if anything to say about that but courts also, in addition, to sort of those first order questions ask questions about secondary considerations and obviousness, right? So the particular question that the court asks is, was the invention, im-boarded by the patent, was that obvious? And some of the things that courts look at in that context are, was there a long-felt need for that product and was the product commercially successful? And if you think about questions like that, you cannot answer those questions without implicitly having some notion of the market in mind, right? So if you think about a long-felt need, you are asking a question of what the market is. If the answer to the question of was there a long-felt need is yes, you cannot escape the conclusion that it has to be a narrow, indeed a new market, right? So it is built in to that idea. Commercial success, so why do you look at commercial success? I think, you know, at the risk of sort of oversimplifying a little bit, the idea is that if I have an invention that is commercially successful, surely that couldn't have been obviously because if it was, somebody else would've come and discovered it first, right? It's the same idea that the reason you don't find dollar bills laying around in the street is somebody else would've, has picked it up already. What do you do in this analysis? As a practical matter, what do you do in this analysis? One of the first things you do as economists is you look at price cost margins and that's not different at all from the way you look at anti-trust analysis. So you look at how different are prices from marginal costs. And if you have big margins, that suggests that you are operating in a market that is narrow. In other words, you don't have lots and lots of competitors. Specifically, in the context of pharma, one of the things that we have found to be persuasive is we look at, when we have a new product and the validity of the patent covering the product is being questioned, we actually look at health plans and ask the question, what did health plans do with that product? Because if the product is undifferentiated and operates in a large market, you would expect health plans to put that product in a place on their formulary or ask the consumers or patients to share more of the coverage than otherwise. So looking at what health plans do, tells you something about the markets in which the product operates. So those are some of the issues that, in the analysis that you look at on the, on the validity side, on the liability side. The damages question to me, it seems like, you know, are very, very related to anti-trust analysis. So if you're in a world where you're seeking lost profits, the bandwidth factors. The first step is you have to show that the four Panduit factors are satisfied and the second Panduit factor explicitly says that you have, you know, asks you to look at whether or not there are acceptable substitutes. That is a market definition question. In my mind, that particular question is not at all different or not substantively different from the question that the agencies would look at in evaluating the merger between two entities. So that's one, so when you're seeking lost profits, that's the question that you're looking at. It's a little more complicated in the case of reasonable royalties because if you are in the world where you are not entitled lost profits but you're seeking reasonable royalties, then there's a first order question as to whether the relationship between the patent holder and the infringer is a horizontal one. By that, I mean, are they competitors or is it a vertical one? For example, a licensing relationship between a university and a pharmaceutical company, like a Pfizer or a Merck. Now, but in either case, I think in determining what the, okay, let me back up for a moment. The easy and kind of, and reasonable royalty analysis is to say that, oh, I found a few comparable licenses, here's the average, that's what applies here. There's not a whole lot of economics in that. I understand that gets done but that gets, you know, there's a different kind of expert that kind of deals with that, licensing experts, but as an economist, I think when you ask the question of what the right royalty is, you do have to ask the question of like what are the markets? So if it's a vertical relationship, I think you have to ask the question, in determining the rate, is that technology available only to that university or are there multiple universities? So, in other words, you have to look at the competition at the upstream level and conversely you have to ask the question like if you are a university, is Pfizer the only buyer of your technology or if Pfizer doesn't buy, can you go and shop it to Merck? Right, so again, in determining the royalties, at least the economic part of the analysis really has to think about a market definition at both the upstream as well as the downstream level. So those are some of the comments I had in terms of thinking about the role of market definition in liability and damages and patents.

 

- Great, so I'm gonna turn to Dale in just a minute but just to come back on one of the things that you said about non-obviousness and the idea that price cost margin might teach us something helpful about non-obviousness. Part of why I like teaching and doing work both in anti-trust and IP is that they're always in a certain amount of tension and price cost margin is a really nice example for me because, for a pharma company, pointing to a high price cost margin, it's kind of a mixed bounty. On the patent side, if you're in a patent case, it helps as you say to establish non-obviousness under 103 but the flip side is that in an anti-trust case, it makes it, there's a lot more to be said but to oversimplify, it makes your life difficult when you have this very high price cost margin 'cause that tends to be evidence of market power and it's difficult typically for an anti-trust defendant to resist the conclusion that whatever the market is, it's relatively narrow and that they have market power, in that market partly as shown by their high price cost margins. Dale, I wonder if you have some reflections and, you know, within 103 and we're generally here. One question I have for the panel if anybody has a reaction is, one puzzle for me when I teach some of this material is, okay, commercial success is basically a proxy for long-felt need or failure of others. Isn't it easier to prove long-felt need or failure of others or wouldn't that be more directly probative of the question of non-obviousness than commercial success which of course could have a lot of different, a lot of different causes, that you'd have to tease out.

 

- Yeah, now, I'm gonna defer to the many patent lawyers in the room on that issue but in talking about something slightly different related to commercial success related to copyright law but as you heard in patent law, if your product is commercially successful, it's more likely you can defeat a claim of obviousness so, yay, for you to have a commercially successful product. In copyright, normally a commercially successful product is a good thing because if someone infringes you, you can have a good damages claim, et cetera, but I believe just like other speakers today have said I believe the law is wrong and needs to have something done about it, I believe that the law, that the recent trend that some people have been saying which is that if your copyrighted work is so popular, well, that means you have lesser rights and that other people should have freedom to infringe it and take from it and use it too just because it's so popular. I believe that totally flies in the face with the constitutional incentives that, if someone who's created something new and wonderful should, and has done a wonderful thing for the world and they should get the benefit of that. I don't think others should be able to use that against you. I think it directly flies in the face of things and I know that this was argued in our Harry Potter Lexicon trial. The other side argued, well, J.K. Rowling, you know, the Harry Potter books are so successful, you're not gonna be harmed. It doesn't matter, we can do whatever we want with them, you'll be fine and she said on the stand in emotionally gripping testimony of a real creator that if when she was deciding between money for typewriter ribbon and feeding her baby, she knew if she came up with something and it was popular, that she would have less rights, fewer rights, because it was popular then someone who came up with something that people didn't like as much, that would not, that would be a very hard concept for her to appreciate and understand and hard for creators to understand. You should have more rights not fewer and I believe that Judge Leval was endorsing that concept on the last question that was asked to him about factor four, which goes back to TVEyes when he said, "Well, you know, I don't know, "I don't think you can just take something "that's somebody else's and use it "because you have some other purpose in mind to use it. "I think that's wrong," and there may be some future cases that will talk more about it.

 

- Mark, I wonder if you have any reactions to that?

 

- Not specifically that but I was just, as I was listening, one thing I just sort of thing is interesting to highlight is that the one place it seems in IP where there actually seems to be some doctrine about how to think through market definition questions is in patent remedies and so if you just sort of juxtapose that with, in a copyright case, when there's an infringement, there's virtually, almost never any effort to say, well, what's the alternative, non-infringing thing you would use and let's try to figure out the difference between what the value is in this and that. We just sort of assume that this is not replaceable by something else, right, which is to say it's a market unto itself and so we don't actually have substitutes and it's sort of interesting to think about why it is that in patent law we get, because parties litigate so hard over the remedies question, why we get some doctrine that is formed there where maybe through the trademark cases, the questions are almost always implicit, on the validity side in patent, they're implicit. They're no less important but they're just sort of, there's not doctrine that's developed there about how to do it but there is in patent remedies and I'm just sort of curious about why.

 

- Mm hmm, mm hmm, so we're now at question time. Please come to the mic, ask us, ask them questions. I mean, one place just thinking about your patent versus other areas for just a minute, I mean I think apportionment in remedies in copyright might, to some degree, implicitly bake in a kind of market, power market definition idea to the extent that you're saying, okay, look, a lot of other things were successful or were factors in our success and by the way implicitly we could've swapped out some of the details of the story that, okay, admittedly we infringed with some other similar story, perhaps, to the extent that we say look at all the other things that contributed to the success, to the extent that we're downgrading what the plaintiff did. There might be a kind of implicit market power.

 

- It doesn't strike me that we do that very often.

 

- No, that's fair. That's fair. Yes?

 

- [Male] Actually, that gentleman was first, please.

 

- [Male 2] Could you talk a little bit more about the aspects of TVEyes and what the use was, what it was used for in contrast to the Google case where they digitized all the books and they, in effect, got away with it and the concept we were talking about before of de minimis use being okay and also new uses? Google digitizes all these books and says I'm not gonna use it the way you ordinarily do, I'm gonna have 14 other different uses. Shouldn't that be appropriate-able by the copyright owner rather than by some new --

 

- Yeah, I, I --

 

- Dale, this is to you, I think, in the first instance.

 

- I think that the main difference with Google Books is this finding versus delivering thesis. Both sides tried to say Google Books is our case but I think we won that argument because in Google Books, Google went through great pains to not show people the whole book. They blacklisted 10% of each page of hits, 10% of any work was blacklisted. You could opt out of it. If it was a short form thing like a recipe or a dictionary definition, that wouldn't even be included in snippet view, a far cry from as many high definition 10 minute snippets, they called snippets, we said they were clips. The court said, yeah, you just can't call something a snippet and make it a snippet. It's a clip. So that was the difference. In fact, my opponent at the Second Circuit started her argument off saying, "This is the audio visual version of Google Books," and judge, I think it was judge, I'm not sure which judge, maybe Judge Newman, leaned forward and said, "Are you saying we wouldn't have come down on Google Books "the same way if Google was making all the books available?" Well, I don't know, that would have been a closer, "Of course, we wouldn't," and you heard Judge Leval talk about that too. So I think that there is a difference between just doing enough to help people find things and giving people what they're looking for so that they don't need to go back to the copyright owner.

 

- And just to make sure I understand the distinction that you're raising: I mean, TVEyes, on your distinction, has some finding in it, it's just that they were also doing delivering? It was, in part, a finding tool judging from the list.

 

- That's right. I mean, it is in part, you can do sort of searches to kind of find content but unlike say, Hathitrust which was earlier than Google Books which would list on this page, this happened, on this page, this term, and on Google Books went a step further than Hathitrust and gave people the small snippets of content and information and TVEyes, they tried to take it a step further and give people everything. Now our opponents tried to argue, well, we tried to limit it contractually for kind of internal research purposes and so that was somehow okay and our response to that was Texaco, Texaco, Texaco and our second response to that was that's not how you marketed it.

 

- So there might be some alternative video finding tool that was differently marketed and more limited in its functionality that would pass muster under Google, it sounds like?

 

- Well, I don't know. I don't think that the court opined that all video but if you're delivering, if it looks like TVEyes, I think it's pretty clear that's an infringement.

 

- [Male 2] This is a great panel. That was a great answer.

 

- Thank you.

 

- [Male] Thank you. So I really appreciate the insights with respect to markets, proving markets and the existence and scope of markets for purposes of the harm suffered both by creators and potentially by users. I wonder if you all could comment on what insights these market understandings could tell us about the scope of rights creation in the first instance? What I have in mind is the patent context but I imagine the same thing would work in trademark and copyright and what I mean is, when you're trying to figure out how to define the right, patent right say, in such a way that it isn't so broad, so abstract as to pre-empt the field, right? That's something that courts care a lot about, certainly the Supreme Court, but what field are we talking about? Is it the, all computer software? Is it computer software that deals with this problem in this way, et cetera? So for problems like preemption, for problems like enablement and scope, and then corresponding problems in copyright and trademark, what can we use market definition for to figure out how broadly a right should sweep so as not to be invalidated later on?

 

- So we're already stepping on lots of other panels in our scope and you want us to take on like all of 101, too, right? No, no, I welcome the, I welcome the challenge. Mark, do you want to start this?

 

- Well, I guess I want to say in the first instance, it's a sort of, it's an important question to ask whether that's supposed to be a market definition question or not, right? Is it supposed to be a proxy for a competitive harm or is it supposed to be a proxy for something else? And I do think that there's some impoverishment that's happened by virtue of trying to take all these doctrines and define them in terms of competitive harm, right? So I think if you were to, I'll just take the trademark examples. If you were to take, there's one thread of functionality law that thinks about these questions entirely in terms of competitive harm. There's another that thinks about it in terms of preemption of patent policy. If you define it in that way, the market questions go away. You don't have to actually define the markets in that same, same thing with genericness, like you could think about it just in terms of consumer perception. It doesn't make them go away entirely but if you think about this in terms of competitive need to use the term, then you, that serves all of the four, and so there's a lot of pressure that's brought to bear on these doctrines by conceiving of them as means of operating sort of competitive balance, that situation. So that's one answer is that we could take some pressure off some of these doctrines by not necessarily conceiving of them in that way and then I think, if we are gonna conceive of them all in competitive terms, then we do need to actually start thinking harder about how to answer some of those questions and I think you just gave a couple of examples but I actually think they're pretty pervasive throughout all of IP.

 

- Let me get David's question in.

 

- [David] Sure, so I talked this morning about how survey controls are allowing us to measure survey harm or the confusion much more specifically. We see the same thing when it comes to measuring markets. With help from people like Rahul, we're very good today at measuring markets and actually measuring the harm and I was thinking, Dale, about the sort of extensive evidence you gave the court, of all the ways in which this was gonna harm us, but I also sometimes worry when I do that that I'm undermining my ability to get a permanent injunction because you're laying out for the court that there's a very specific way to actually show what the reparable harm is, that the harm's not irreparable but actually, we can pay you damages later at the end of the case so let's just let TVEyes continue to do this for a few years while we litigate it and we'll deal with what the payment should be later. How do you, and I see Rahul's shaking his head, how do you think about the tension between the desire to show irreparable harm but also in the world now of eBay against Mercexchange, this need to show that there's some actual harm that's occurring?

 

- I'll let Rahul deal with the economics aspect of it but from a practical perspective, one, this is not harm that would be very hard to quantify because we're talking about likely markets, we're talking about some, a lot of actual markets but it's not easy to figure it out especially when you're dealing with one actor. The actor by themselves may not be as big, right? They only have so many subscribers but if the use became widespread, that would really be the biggest harm. So that's the difference and the other difference is in irreparable cases, as we know, one of the factors is also, is the defendant able to pay the judgment? And if your harm is big --

 

- So just make the number big enough, if the number is large enough to lead to judgment proofness then the problem goes away?

 

- Well, that is a, yes, it does go away but I'm not saying to create --

 

- No, no, I didn't mean to suggest that.

 

- I'm just saying that it often is the case that the harm to you is a lot more than the person would ever be able to pay.

 

- Rahul, you wanna --

 

- If I could just add, so this is actually a challenge that comes up a lot in these Hatch-Waxman pharma cases where you have these big drugs that a drug company has, there's a generic that's threatening to enter and you go with outside counsel to the drug company and you say, you know, I need your perspective on what's gonna happen and the first thing they say, oh, if the generics enter, our sales are gonna drop off a cliff, which is absolutely true but there is nothing unmeasurable about that. There's nothing irreparable about that, right? So what does happen though is, is if you, let's say if you're a drug company that has one drug that has $2 billion in sales, you lose virtually all of it, you might have to lay off all your people. You have to lay off all your scientists. Now, then if the generic drug company, the infringer is subsequently enjoined, you get the money back, the losses of the, the losses in sales and profits back but can you hire the people back? Can you hire your scientists back?

 

- [David] Are you seeing that run in cases, that argument?

 

- So we, when we file declarations in favor, you know, in support of irreparable harm, one of the things that I think we are very careful, like and this is sometimes a struggle with clients because the first thing they say is, oh my God, sales are gonna drop off a cliff, and you have to say that, yes, that's true, but that's nothing unmeasurable about, at least not to an economist. So when we, when we make arguments about irreparable harm, we try to be very careful about describing things that will happen and making a clear distinction between what is actually measurable or at least reasonably measurable by economists and things that are not, right? The idea that if you lose your best scientists, you may not be able to hire them back, that, you know, I mean, you will find economists that'll say they can quantify anything but that's a tough one. That you could sort of say there are categories of things that fall, and that you could make a reasonable case that I can't measure this thing, right? I can't make you whole after the fact.

 

- [David] Excellent, thanks.

 

- There might also be a point that even if you could make the firm whole, that there would be a social harm that couldn't be undone. So to the extent that that was a factor. It wasn't your transfer but there's a public interest. Chris, you want me to give you the last question?

 

- [Chris] Yeah, quick question about the fourth factor in fair use and market definition within the fourth factor, and I want to use as a lens something that Supreme Court Justice Souter said in Campbell versus Acuff-Rose. So he quotes from Somerset Maugham's On Human Bondage and he says, "People ask for criticism but they only want praise," right? So the district court in TVEyes said that that was true of Fox News. So they won't license to people who don't praise them. They won't license to people who criticize them. That would suggest a couple things. Rahul, you said, well, market definition is a question of substitutability. That's an economic question but there's also a legal question. So in Campbell versus Acuff-Rose, the Supreme Court said there is no licensing market for rap parody because people only want praise. They're not gonna license to rap parody. So that's a separate market. There's no licensing within it. The defendant's work is fair use with respect to that market. So that would suggest it's a legal question as well as an economic one. On TVEyes, if the district court's right that, I don't know if it's right that Fox News won't license to critics, does that suggest that if TVEyes had a different business model, which is instead of making people sign a contract that said you're only gonna use this internally, to making them sign a contract saying you're only gonna use this in Elizabeth Warren's campaign or you're only gonna use this for The Daily Show or you're only gonna use this to say mean things about Fox News, that that would in fact be fair use?

 

- Three points. One, as Judge Leval was talking, answered the question earlier, taking someone's thing and using it, if it was, even if it's for a campaign, I don't think that is enough to get you there. Second, on the district court's comment, one of the challenges we had in the case was making sure that the court didn't think that this was in any way about stifling criticism, rather it was about copyright. So we presented extensive evidence that Fox News, in fact, did license people for, who used the license for purposes of criticism and, in fact, a lot of people use the clips from Fox News' own website and we had our expert detail this with lots of examples for purposes of criticism. They also had, CNN saying this is bad for the industry. There are lots of industry too but, the judge seemed to be less concerned about, is this really about censorship when he realized that what many people think of as a big competitor, had the same concerns about monetization of news.

 

- So I think we're at the end of our time slot. Please join me in praising our panelists.