Intellectual Property Roundtable

Once the nucleus of patent litigation, the Eastern District of Texas is giving way to districts in Delaware, California, and other regions, following the U.S. Supreme Court and Federal Circuit's game-changing rulings this year on venue selection. Meanwhile, damages remain a central and contested issue as practitioners continue to wrestle with the contours of apportionment. Our panel of experts discussed these issues as well as the current state of patent eligibility, and the latest trends in inter partes review.

California Lawyer met for an update with panelists Christa Anderson of Keker, Van Nest & Peters; G. Hopkins Guy, III of Baker Botts; Fabio E. Marino of McDermott Will & Emery; and Matthew D. Powers of Tensegrity Law Group. The roundtable was co-moderated by independent mediator Jeff Kichaven with California Lawyer.

JEFF KICHAVEN: Let's begin with the topic of damages in patent litigation. What have been the leading developments, and what do you anticipate in the next several months?

MATTHEW D. POWERS: The biggest development is Mentor Graphics Corp. v. EVE-USA, No. 15-1470 (Fed. Cir. 2017). There's been a spirited debate on secondary apportionment tests for lost profits claims. Loose language in a couple of Federal Circuit cases could be interpreted as suggesting such a requirement might be appropriate.

In Mentor Graphics, the panel decided no, that once you have satisfied the Panduit factors, you have by definition apportioned because satisfying Panduit means you have lost that sale, in significant part, because of the patented component. If you lost a sale, your lost profits are whatever you would have made on that sale. En banc review was denied, although there were spirited, divided opinions about that question in the briefings.

HOPKINS GUY, III: There appeared to be a moment in time in which it was pretty clear that apportionment was required for all damages analyses, and that you were only entitled to damages for the incremental economic value for the claimed invention.

What I'm seeing now is the plaintiffs' bar attempting to avoid real apportionment, whether it's using a lost profits analysis and/or using comparable licenses, where apportionment is presumed to be part of that. So I think that there will need to be some fine-tuning of what is required for apportionment. That's what I think we'll be seeing.

CHRISTA ANDERSON: Yes, the apportionment issue has been front and center for years now. The nuances change, but it's always been a point that plaintiffs' counsel struggle with to try to apportion the incremental value for their invention.

I think we'll continue to see attempts by counsel and judges to get a handle on how to make the test for apportionment firm. Intuitively, apportionment makes a lot of sense. How you define it and how you avoid folks being able to sidestep around it is the tougher question.

POWERS: There is no question that you should apportion as best you can. The challenge is that apportionment is inherently imprecise. There's never going to be a perfect basis to apportion. The proof will never be perfect. It cannot be perfect. So you can't require perfect. The question is what's sufficient and on whom does the uncertainty of that fall? That's the issue that we're all going to be wrestling with.

FABIO E. MARINO: To me, the whole concept of talking about apportionment in the context of a lost profits damages model makes no sense at all because lost profits require but-for causation. Either you did cause the loss of the sale or you didn't. Apportionment really has no place in that debate.

When you go to the reasonable royalty environment, then I agree with Matt Powers 100 percent. We all agree there has to be apportionment; the question is what form should it take? If you're trying to find comparable licenses, there's this great irony with respect to apportionment: most real-life licenses that I've ever seen are based on total revenue. So, on the one hand you say, yes, we need to have apportionment. On the other hand, we throw out most of the licenses because there wasn't enough apportionment. That's the debate.

What the experts have been doing recently, and some of them have been validated, is they try to go around it by saying, okay, I'm going to assign a per-unit royalty as opposed to a percentage of all sales. But that just moves the debate to a different arena. So I think that's really going to be the analysis we are going to see in the development of how you define these tests.

POWERS: Your point on comparable licenses is an important one. In real life people often do licenses on the total value of the product and pay what they think is a fair amount, because it's easier. If it's truly a bare license on the same patent, you better believe that licensee negotiated as much as they could for apportionment, and yet everybody says there's no apportionment. Real life ought to intrude a little bit more into the legal analysis.

KICHAVEN: How are you expecting this to play out in litigation strategy?

ANDERSON: It is continually an issue of trying to require more proof than we're seeing in many of these damages reports.

From the defense perspective, the damages reports tend to contain lots of broad-brush proxies that do not meet a level of proof that's acceptable, especially if there's a more scientific approach available. For example, while it's not available in all cases, sometimes multifactor regression is a useful tool from a scientific perspective. There are some cases in which it could be used but isn't used. You have to wonder, why use a really crude proxy when there is a better analysis available?

So it is a continual struggle between the plaintiffs' bar and the defense bar over these issues. How much do we require folks to satisfy a more scientific standard? And when is that not practical?

POWERS: That's a great example of the law pushing us to a sense of false certainty, false precision by requiring a conjoint survey or a multifactor regression analysis. Licensors and licensees are not doing conjoint surveys or multifactor regression analyses to decide the royalty rate. They negotiate it based on their perception of the value being contributed--based on a lot of qualitative information typically. We've all been in those discussions. To require that sort of level of scientific discipline, (a) increases the cost of every case, and (b) divorces us from real life.

KICHAVEN: How are you seeing judges respond to the kinds of arguments that Matt [Powers] is articulating?

GUY: It depends on the district court you're in. There are some courts that are going with a model that is more aligned with what Christa [Anderson] is saying which is at least look at it. If you have a large company who signed multiple licenses for patents over the last 20 years, draw a line through all of them and figure out roughly what the value is.

I think what Matt [Powers] is not taking into account is that there are judges that will allow you to cherry-pick. As long as there's one license that someone signs somewhere, regardless of what the total economic value was, if there's a per-unit royalty mention, you could get stuck with that royalty. So that's where there is a big divide. Finding some consistent law on a nationwide basis would be nice, but I don't think we're there.

POWERS: That's an example of the judge intruding into the role of the jury by dictating to a plaintiff or a defendant which license agreements they should be relying upon out of the group. You should be able to disagree about whether those are the most comparable and have your experts argue that point in front of the jury. It's a fact-finding question.

MARINO: The underlying assumption in this debate has been that what the judge is going to allow into evidence at the Daubert stage is the whole question. If the defendant, in fact, has a more credible, scientifically sound approach if the judge lets the plaintiff's report in, why wouldn't the defendant just put in the sounder report and then let the jury decide?

We've been focusing too much on just the admissibility standard, as opposed to the whole trial strategy. I wonder if, over time, that's actually going to lead to a more balanced approach by the courts because just throwing the plaintiff's damages report out at Daubert may not be the sine qua non of every single case. And then you'll have better evidence come in, better reports being presented to the court of appeals to define the standard in a more reasoned way.

One of the things I see is that there's a big difference between various district courts in the level of comfort they have actually calling balls and strikes on what are fairly complex economic models. Some judges are very comfortable with it; many others are not. So to focus everything on just the judge's Daubert decision creates a lot of forum shopping. There are other tools that the defense bar can use to ameliorate the situation.

ANDERSON: That raises an interesting point. I notice that there are some judges at least that are requiring earlier damages reports, earlier fronting of Daubert challenges. On balance, whether the evidence stays in or out, whether the report gets tailored back in some way or not, is such a valuable assistance to the proper management of litigation. I like that trend. I hope it continues.

POWERS: I agree with that. Too often damages is the afterthought of litigation. In an evolving and still nascent area of the law, it makes sense to have an orderly process. Doing it as part of a Daubert motion at the pretrial conference stage, which some are doing, is crazy at that point.

The fundamental problem with damages law right now is the uncertainty it gives plaintiffs and defendants. You have no idea whether this model or that model is going to even get to the jury until very, very late. So neither side can price the case in a reliable way. The law should give parties some greater sense of certainty on what the case is worth.

GUY: Let me push back on that. You're saying on the one hand that knowledgeable people, licensees and licensors, can sit across the table from each other and in a few months determine what the price of something should be. And yet for the first ten months of litigation, it's impossible to figure out what the price per case should be?

POWERS: That's because you don't know the evidence that's going to be considered for that process. If you knew that and you knew that Daubert was going to be done in a more restrained way, you would know what each party is going to be asking; then you could price it just the same way licensees can. But when it's all left until the end, you can't.

MARINO: So in other words, the process can be effective if there is enough early discovery of damages to allow that discussion and decision by the court to be made with actual knowledge of what the evidence is going to look like. It's not going to be complete, but it's something.

KICHAVEN: Are courts willing to give you that guidance early enough?

ANDERSON: Some are.

POWERS: The Northern District of California has led the way in adopting early damages contentions. There are very few districts that are doing anything like this at all in terms of having serious early damages discovery and contentions.

There isn't a track record yet in terms of how effective it is--how much actual data you're getting and whether the parties are able to settle better because of that. But it's a start. The downside of that, obviously, is the case is more expensive earlier. For both sides.

ANDERSON: Of all the things that a judge could do to facilitate settlement, frontloading the resolution of challenges to damages reports and Dauberts is most helpful because that really puts some brackets around what the dollar range of the case is.

KICHAVEN: Do you encourage judges to do that? Do they do it on their own?

ANDERSON: Some judges do. Some districts do it. But it's not universal by any stretch. In many courts, it's been pointed out, it's pushed to the last minute as part of the pretrial motions and resolved as you're getting ready to pick a jury.

GUY: And that encourages sandbagging and bad behavior.

KICHAVEN: Can you and other counsel manage this without judicial supervision?

ANDERSON: You can certainly try. You can try to have informal discussions. Let's exchange some broad-brush information. Let's see if we can get to a mediation. But until you know the judge's view of the reports and where the judge is going to come out on some big-picture issues, it's not going to be as helpful. It would be more helpful for the judge to make plaintiffs frontload the damages opinion and get that resolved up front.

GUY: I think so, too. The more variability there is in the high-low values at the very beginning of the case, the more parties spend on their counsel and the more contentious it becomes and the longer the case takes. When you can establish a long-term baseline of what a case is actually worth, the cost of litigation goes way down. Judges can be proactive about this by frontloading damages and creating a high-low boundary early in the case. Then it's a matter of counsel to argue whether it's high or low but at least it is somewhat constrained, so it can't change too much. That would significantly change the practice more than anything.

MARINO: There has to be an enforcement mechanism in place. If you just make it a voluntary disclosure and there are no consequences, guess what, the guy that has something to lose from the exchange is not going to be very forthcoming.

We can all agree intellectually that early disclosure should be done. But how do you make enforcement effective? The problem with damages disclosures is that it's based on information in the hands of your opponent. So you can't really just apply that rule because it's just going to incentivize your opponent to not give you the information you actually need. Finding an effective way of enforcing the early disclosure rule is the difficult part.

KICHAVEN: The issue of venue has been front and center after the T.C. Heartland v. Kraft Food Group Brands, 581 U.S. ____ (2017); In re Cray, No. 2017-129 (Fed. Cir. Sept. 21, 2017); and In re Micron Tech., No. 2017-138 (Fed. Cir. Nov. 15, 2017), decisions. Are there many open issues left?

POWERS: Cray and Micron put the biggest stakes in the ground that we need, and the remaining issues are going to be fairly easy to mop up at this point. It's relatively clear what the rules of the road are. The effects are clear. Filings in Delaware are up by about 35 percent; in Texas, they are down by about 35 percent or more. So Delaware is now number one in new filings. Many cases are coming into the Northern District of California, including most of the transferred cases.

KICHAVEN: Do you think Delaware will become the new Eastern Texas? Do you think it will become overwhelmed?

POWERS: Well, Delaware is about to get their two vacant judgeships filled. Those two new judges don't have a track record yet. Delaware is not a defendant-unfriendly jurisdiction. I think you're going to see Delaware as a very, very dominant place for patent litigation going forward.

MARINO: Delaware will become overwhelmed just because it's the default now. I tend to agree with Matt Powers that it's not going to have the same draw to plaintiffs that the Eastern District of Texas has.

KICHAVEN: So is it a good thing or a bad thing for the progress of science and technology that this concentration of cases in Eastern Texas is going to be dispersed to Delaware, California, and other places?

ANDERSON: Whether or not this is a good or bad thing is driven by who the client is, what kind of business they're running, whether they have offices in the particular venue, and whether or not they are contributing to that part of the country in some way that they think will give them an advantage for having their case litigated there. It really varies from party to party whether or not that will be viewed as a positive or negative development.

GUY: On the defense side, it depends on client size and longevity to some extent.

Any company that's doing business nationwide can be expected to be sued anywhere in the country, particularly if they have brick-and-mortar facilities scattered everywhere. For example, Microsoft and Apple have brick-and-mortar stores in Plano--those are going to be established places of business. In recent cases, they haven't even challenged venue.

I think T.C. Heartland will matter more to small start-ups in a particular locale, whether it's Southern California, Silicon Valley, or Seattle. They'll be given the option of incorporating in the state in which they exist at least during their initial stage and then pay, perhaps, a tax penalty for transferring to another state. That's a consideration now that will be brought up. That's something that we counsel clients on as a choice.

Most of the VCs will say they like Delaware for its reliability, predictability, and speed when it comes to corporate issues. On the other hand, getting hauled to another district--even Delaware--to defend a patent lawsuit can be expensive at a certain stage. So I think it depends on the size.

MARINO: The class of companies that benefits the most from T.C. Heartland are Silicon Valley start-ups because they're not going to be everywhere. They're going to be right here and you'll have to sue them here.

ANDERSON: It will be interesting to see what practical effects will happen in the districts that may end up with a lot more litigation than they had before. Will the court system there manage it well? Will they have the right judges in place that can handle the new high load of cases caused by the change in the law? It will be interesting to see how the different jurisdictions react.

POWERS: For example, in California, the average time to trial is pushing two and a half years before this. Does it go to three and a half? I've heard from several Northern District California judges that their schedules are in chaos because of all the cases that are being transferred to them from the Eastern District of Texas. Trial dates are getting canceled. I just had a trial date in the spring canceled and maybe it will be nine months later--maybe. Delays increase the cost of litigation. At some point that's no way to run a railroad.

MARINO: One practical effect that you're already seeing in the Northern District of California is the likelihood of the motion to stay pending IPR being granted is very, very high. As the dockets become more crowded, you'll see that trend expand.

KICHAVEN: But the total number of patent cases is going down, isn't it?

POWERS: It's coming down a bit. Economically, you have cases that aren't being brought because you would have to spend X hundred thousand dollars per IPR; it would take two years longer to make money because of the stay pending IPR. So the math doesn't work for 25 percent of them. Some of the patents have dropped out as well.

What's interesting, of course, is the world evolves. This is why litigation funding is now a thing. It's now a thing precisely because of all of these influences.

GUY: That's the trend we've seen: more of the cases appear to be partially or fully financed by a funding organization. It can't be that suddenly the quality of patents just doubled overnight. It has to do with the need and their ability to cut deals and stay the course, something that the law firms can no longer do.

MARINO: In my experience, the amount of due diligence that goes into a litigation finance case is much higher than the kind of due diligence that went into law firms taking contingency cases maybe five, ten years ago. And that's probably because you have a much more rigorous financial analysis being forced by people that are actually in the business of lending money to finance litigation, as opposed to lawyers who are in the business of litigating cases instead of the financing parts of it. My gut feeling is that the awards are actually going up for the cases that are being filed because there is more rigorous analysis at the due diligence stage.

GUY: Like any group of funds, the early pioneers will find better deals. A huge factor here, though, is in all of the markets, yields are incredibly low. They're looking for some venture component to every fund--it's represented as a high yield payout, or at least that's what they're selling--and it's putting a lot of pressure and a lot of money in the patent litigation space. The problem is, like many assets, this one is actually quite finite. The number of good patent cases out there, I would say, is pretty thin.

All of the funding in the world isn't going to change the next quartile into the top cases. I think you'll see some fallout probably not next year, but the year after, in terms of funding organizations.

KICHAVEN: What are the latest developments concerning Alice Corp. v. CLS Bank Int'l, 573 US _ (2014) and patent-eligible subject matter? Has this area settled down?

POWERS: I'd say that we are moving in the right direction but have a long way to go. A few years ago, there was only one Federal Circuit decision finding patent eligibility. Now there are many. And "many" makes it easier to plot a line and figure out data points. The problem is the cases are absolutely impossible to harmonize. It varies by judge, it varies by panel.

KICHAVEN: Where's the biggest lack of harmony?

POWERS: How people are treating what is an abstract idea or not. Everybody is trying to fit within DDR Holdings or this case or that case. You're not seeing a coherent analysis of where the line ought to be. That's leading to increased calls for legislative reform. IPO is pushing an amendment that would effectively obliterate Alice and say that the only exception to patent eligibility is whether it's something wholly within someone's mind.

GUY: A little background is helpful here. There was a time when software patents were largely written in a formulaic way by patent prosecutors who would claim inventions in the vaguest possible terms to make it broad. Basically, the Supreme Court in Alice called them out on that. We're seeing a process now of those being weeded out. I do think that the law is becoming clearer, however, by definition, when you're dealing with something concerning an abstract idea, what is or is not an abstract idea will defy precise definition. So the question is really whether you want to go back to a world where you can claim that sort of thing or not. That's the gist of the legislative intent. From the defense bar perspective, large companies in the software space are getting claims allowed over Section 101 rejections, so we've probably still got a ways to go before we clear out that path. It's going to take some more time.

POWERS: I agree with Hop [Guy] about the problem, just not the solution. I agree there are many really bad patents written in exactly the way you're describing. But 101 is an incredibly blunt instrument that has been used to kill all sorts of great technological inventions and for that matter medical diagnostic inventions that are hugely valuable and important.

The right solution is some combination of 102, 103, and 112, all of which could be used to attack a vague, non-enabling, broadly claimed invention without any real spec support, which is really the problem.

GUY: To the extent that the patent specification and claim are equally vague, then I think those should die. But I think the debate is where the specification itself includes a significant amount of detail and yet the claim drafter chooses to claim the invention so broadly. The analysis currently limits you to the claim; you'll see lots of cases now which say, "I'm not asking what is in the specification, what is in that claim?" You end up penalizing the patent-holder for the approach that their prosecutor took in trying to claim it.

ANDERSON: I agree that the case law is still developing in this area and it may take some time to establish clear guidelines.

MARINO: I have a radically different perspective from everybody else. This is an area that has been at the center of my career since the very beginning as I'm a software guy. I used to build computer systems before I went to law school. Here's the interesting thing: I tend to agree with Matt Powers: section 101 was not really the right tool to address the problem. The problem was that when software became patentable, the Patent Office wasn't ready for it and the volume of literature that we use to research the prior art wasn't ready for it, because patent examiners mostly look at patents, more so than other publications. Since there were no software patents at that point in time, guess what?

KICHAVEN: Everything seemed novel?

MARINO: Everything seemed novel. So they issued a whole bunch of patents that were clearly invalid. I would posit that sections 102, 103, and 112 with a better class of examiners would have been a very good tool to deal with that more so than section 101.

However, I do not think the test that currently exists for section 101 is as loosely defined as Matt Powers characterized it. I do agree that for a long time the Federal Circuit was saying look at our prior cases and try to draw a line through them as best you can. And there's always a lot of variability to that.

But I think that if you do try to find the best fit of that line, there is a pretty clear demarcation between improving the computer and just using the computer to do something known in a better way. The courts are starting to define a better test and be more predictable in the outcome, so we can advise our clients accordingly.

POWERS: The problem is not defining what the line should be. I totally agree they've settled on are you improving the computer or merely using the computer to do something old. The problem isn't false positives, it's false negatives. You see courts reject serious patents, saying, "You're just using conventional components." That's a 103, 102 issue. Almost every patent uses conventional components arranged in a way to do something new, but the courts reject them as invalid under 101. It's a broken system.

KICHAVEN: Do you think legislation is needed to solve the problem?

ANDERSON: At least right now, especially in this legislative climate, I think that it will be the courts that parties will have to turn to in order to address and clear up any uncertainties. I don't think the Alice approach is going away, but the case law will continue to develop and clarify nuances around that decision.

MARINO: I agree. The legislative solution is not feasible because it would require a very bright line. The way the law has developed in this field over the last ten years, it's anything but a bright line. It's a very jagged line. Trying to replace a jagged line with a bright line will just make things a lot more complicated.

POWERS: We're still pretending whether something is conventional is a pure question of law. It's not. It is a question of fact. How in the world on 12(b)(6) can a judge say, "I think that was conventional," without expert testimony that you would have under 103? That's happening every day of the week. And it's wrong.

GUY: I think in order to save a good patent, you need to go back to section 101 itself which does not even use the word abstract in it. It refers to new and useful inventions, improvements or new and useful processes, machines, or compositions of matter.

That does bring in a factual issue. For example, if you can say prior to this invention all these machines were out there and none of them used this particular formula in terms of the way in which they operate. There shouldn't be a litmus test that just because you claimed in a formulaic way, it necessarily invalidates it. I think you just need to go back to the statutory language, which is pretty basic.

POWERS: The problem with going back to the statutory language is you're going to have to change nine people's point of view about that. I don't see that happening anytime soon, which is why people are going for a legislative solution.

GUY: But it's a matter of putting together the factual record. It may very well take another Supreme Court case. But I have yet to see what I would call a broadly claimed allegedly abstract idea that actually went to something that was truly fundamental; you could make the factual case that it clearly meets the criteria of invention. That case has yet to be pled to force the pendulum back the other way.

MARINO: Further development in the case law would be helpful. I'm encouraged by at least some of the Federal Circuit judges who really care about this issue and have a background that allows them to understand it. I think we will get better decisions as the law develops.

KICHAVEN: Any insights on how to advise clients better or differently now?

GUY: Make sure it comes up in prosecution and you get over the rejection.

POWERS: Hop Guy's point is a good one on patents coming out that have gone through 101 prosecution under Alice guidelines. That might reduce the number of good patents going down under Alice that shouldn't go down.

KICHAVEN: Let's talk about the interesting and complex interplay between IPR proceedings, federal courts, and the U.S. International Trade Commission. What are the latest developments?

POWERS: The fight between Cisco and Arista produced some interesting decisions both out of the ITC and the Federal Circuit. The ITC found infringement, and issued a limited exclusion order; then the PTAB invalidated the patents. The ITC was asked to stay or modify the enforcement of the LEO pending appeal of the PTAB decision, but said no. The Federal Circuit was asked to stay the ITC's enforcement of the LEO pending the Federal Circuit's review of the PTAB decision. The Federal Circuit said no, but let the revised products come through.

That is the most interesting recent situation where you have two administrative bodies--one who's in charge of deciding whether patents are valid, one who has its own jurisdiction--and the question whether one defers to the other. The ITC said no. But the Federal Circuit drew a half compromise.

ANDERSON: This is a really complex area. The interplay in terms of timing and approach when you have overlapping ITC proceedings, PTAB proceedings, or appellate court proceedings is difficult to navigate and the right course of action is very case specific. The Cisco case example is a good one to demonstrate the difficult timing and procedural issues faced by clients in those scenarios.

MARINO: Oil States Energy Services v. Greene's Energy Group, 639 F.App'x 639 (Fed. Cir. 2016), cert. granted (June 12, 2017) (No. 16-712), might resolve the whole issue if you get rid of IPRs.

POWERS: From your lips to God's ears. Or at least nine people's ears.

MARINO: That said, I think Oil States is a closer call than most people make it out to be. It's really a case of separation of powers, in my view, whether once a patent issues, is it the proper province of the judiciary to review its validity or potential invalidity or can it still be part of the executive branch? I haven't seen a good response to that. They all go on about whether Congress had the power to legislate this particular issue a priori. And usually, as a litigator, when I don't see a good response that's a pretty tough argument. There are many intangibles that go against the appellant in that case, because obviously it would create a huge mess by de facto reinstating patents that have been invalidated over the last few years. So I don't know what's going to happen, but it's an interesting case.

ANDERSON: Another trend we are seeing is that IPRs are not invalidating as many patents as in the past. That trend certainly affects litigation strategy, and what arguments parties can make on a stay motion. Many of the motions for stay pending IPR used to discuss the statistics of the likelihood of invalidation of patents. Those statistics are less favorable to stay motions at this point.

GUY: I file quite a few IPRs. A lot of the weaker cases are seeing those institutions on the petition occurring and then going all the way through and then they stand up on appeal. So I've seen more appeals on IPR PTAB work in the last two years than I have on anything else. It's worked out fairly well. It definitely is a way of taking the weak cases out.

At the same time I think juries have historically done much as what the ITC does. Juries occasionally take patents away, but rarely. If they feel a case is weak, particularly on validity, they'll read the claims a little more narrowly. They may limit damages, or split the baby on infringement--for example, find one or two patents not infringed and one infringed.

If you're going to treat validity as a real defense, I've been very pleased with the IPR process. You can point out the technical detail and defeat the petitioner early on. But it can be a scary process for a patent owner if you don't know what you're doing. It is a very different animal, and you better do everything you can to defeat the petition. If you have any argument whatsoever, that's the place to make it.