United States Patent and Trademark Office – Madison Auditorium
Live webcast available at: https://new.livestream.com/uspto/copyright
The Appropriate Calibration of Statutory Damages: Individual File Sharers and Secondary Liability
Moderator: Darren Pogoda, Attorney‐Advisor for Copyright, Office of Policy and International Affairs, USPTO
David Sohn, Center for Democracy & Technology
[sorry, missed this because I went to the wrong building!]
Steven Tepp, Sentinel Worldwide
Statutory damages are old as copyright itself, contemplated by TRIPS, necessary because actual damages are often conjectural/difficult to prove; deters for-profit businesses from encouraging widespread infringement (note how that last rationale is slipped in despite being very new)! Digital age makes statutory damages more important than ever: drive thriving marketplace by giving copyright owners a measure of security a measure of security against misappropriation. The more rampant piracy becomes, harder it is for legitimate actors to compete. (Michael Carrier’s empirical work tells a different story.)
Sandra Aistars, Copyright Alliance
Challenges exist for effective enforcement for all types of creators, and in ensuring that the public understands and respects the law. Overly politicized enforcement/unscrupulous attorneys have created PR problem. Need to find common ground. Tendency is to look at big corporations, but copyright exists for all sizes of creators—need to understand effects on small businesses and individual authors. Statutory damages are often the only legal recourse for an individual or small business to address infringement—their availability is threshold for individual deciding whether or not to pursue a claim, given the costs of bringing action in federal court. Members can’t obtain legal assistance where statutory damages are not an option. Deterrence and compensation are important, as well as difficult nature of proving value of copyright/loss caused by infringement, where work is uploaded and available to the entire internet. Where only direct loss provable is license fee, that is an invitation to infringe without consequence. (Interesting that this is the general tort regime.) Profits can also be inadequate because profits may be too small, or too hard to calculate in terms of attribution to the infringement. Broad range of damages is justified, flexibly applied.
Beyond these, any statutory damages scheme needs to preserve creators’ right to say no. Merely compensating lost license revenue is little more than compulsory license. Need to resort to statutory damages isn’t because they’ve suffered no actual damages but is because harm to creator/community is broader than what can be proven and may also include noneconomic damages where the infringement is unusual in some way.
Prof. Peter Menell, University of California at Berkeley School of Law
Simplistic views of history: to say that statutory damages are well established misses a lot of context. Current system derives from ASCAP/BMI’s problems decades ago, and we now live in a completely different era. Congress wasn’t thinking in 1999 about the enforcement problems that were going to arise a year later. Panel is focused too narrowly: this issue is nested within a larger section about keeping rights meaningful. Any solution must be multifaceted. The issue we’re trying to solve is enforcement, and that needs to be viewed holistically. Statutory damages are only a means.
In the internet age, we want a copyright system that garners public approval. That’s been lost, and statutory damages played a significant role in their disproportion. We ought to be concerned. Judges are seeing very peculiar cases—selected based on incentives created, and statutory damages thus bring bizarre and unfortunate cases, inundating the courts. Porn litigation. We ought to care about the harm being done.
We ought to channel consumers into authorized markets. That’s the longterm goal for most players in the system. Statutory damages was thought to be successful, but last decade showed it didn’t work for recording industry, which backed away.
To what extent is this system promoting tech and creative advances? Statutory damages aren’t helping. Distinguish between noncommercial and smaller players and big commercial players; think about orphan works which has solvable problems; then the large-scale enforcement problems—even there, the system is out of whack. Aggregating $150,000 across hundreds and thousands produces obscene numbers; we can think about how to scale statutory damages.
Markham Erickson, Internet Association
Statutory damages need to be thought of in the context of primary and secondary liability. While statutory damages are old, secondary liability is completely judge-made. In other parts of the law, the auto industry isn’t held liable for consumers speeding, even though they know those cars will be so used. Increasingly we’re seeing more litigation around primary infringement—Cablevision, DishHopper, Aereo—that would traditionally have been secondary cases. Because of the scale, nascent tech can’t come to market because the threat of damages is so out of proportion. It’s hard for counsel to talk publicly about clients who’ve been suppressed. Key question: deterrence of what? Legitimate noninfringing uses; good faith, objectively reasonable belief in noninfringing use. Reasonable minds will always differ at the margins, and the statute should encourage litigation to clarify matters, as produced Grokster and Betamax.
Q: How do we conduct cost benefit analysis, if statutory damages are indeed chilling innovation? How do we measure this?
Tepp: Doesn’t accept the premise. We have a multitude of very successful online services (… that various entities have tried to sue out of existence). Other side: to what extent do statutory damages deter piracy and allow licensed services to move forward with confidence they won’t be undercut. Legit services are most vulnerable to piracy because they pay for their content. The system is working well.
Erickson: certainly we want to encourage licensed services and take down clearly infringing services. But there are gray areas: services that are operating in good faith are exposed to statutory damage regime out of whack. Look at cloud locker services: there is no possible way that every piece of content can be licensed. As long as users are allowed to upload, Amazon’s not in a position to determine whether they are infringing. Our companies want to allow users to store their content and space-shift it. Google and Amazon have taken the risk, but they’re big companies that can tolerate litigation. Tepp’s view is not practical.
Aistars: Legitimate cloud businesses like Amazon can be compared to business models employing functions more clearly intended to drive infringing content, like Megaupload. That’s where you see cases being brought, not against staple articles of commerce.
Sohn: takes the issue to be more calibration than existence of statutory damages. Can we minimize costs by focusing statutory damages more appropriately on real bad actors while imposing less risk on entities navigating uncertain copyright regime? Hard area to quantify, because deterrence is a hard thing to prove (though Michael Carrier did try!). But it’s also very hard to prove what infringement has been deterred. Can’t have it both ways: assume that statutory damages deter infringement, then say deterrence of legit activity has to be proved. When asking what behavior has been deterred, apply same standard.
Menell: most investments are best thought of ex ante. Many entrepreneurs don’t want to run these risks, and we don’t want them to have to—we want them to be able to make better guesses. Cloud services: a decade ago, Michael Robertson tried to introduce a cloud service! Resulted in one of the poster child statutory damage awards, in which the record company took over. And now we accept the same result! In an ideal system, we don’t get this damage because people can make informed judgments. We can’t make informed judgments with long drawn out cases and unpredictable juries. Make it easier to assess risks before we get into bringing in lawyers.
Erickson: uncertainty is part of law—if we lurch too far to delineating what’s ok, you do tend to lock in innovation in unhelpful ways. One appropriate measure to allow courts to work as they should: scale down insane awards. Company that thinks it has a lawful service but knows it might well be sued can test that without destroying the company.
Menell: doesn’t disagree with that premise. We’ve found ourself here just by the peculiarities of our constitution: SCt decided that juries decide, and that creates uncertainty. Moving towards a system where beyond a certain range you have to prove some measure of damages would be good—we have sentencing guidelines in other areas—other ways to better correlate to actual damages. Statutory damages is in part about combating underenforcement, but in these big scenarios we don’t have underenforcement—someone is going to go after Aereo, which is not a small nightclub/bar which was the target of statutory damages. Think about risk settings distinctly.
Tepp: we’re being told that there’s so much uncertainty that we can’t have statutory damages, but also being told to embrace uncertainty, which I took as a reference to fair use. That’s your prerogative, but let’s not import policy debates over scope of exclusive rights into discussion of statutory damages, only available to adjudicated infringers. (This is the worst argument I’ve heard today, though the day is young. Does he counsel his clients not to worry about the potential damage awards when considering a litigation or pre-litigation approach? No, let’s not make this personal: this is a disingenuous argument because legal analysis does not work this way. Of course damages and substance interact, and among other things they interact on willingness to litigate out what the actual boundaries of the exclusive rights are.) Range of statutory damages is intentionally wide. Nothing but anecdotal evidence of outsized statutory damages awards. There are checks on awards: timely registration requirement.
Q: requiring statutory damages to more closely track actual harm in some cases. Natural counter is: how could Congress/guidelines reconcile that with fact that a lot of aboveboard copyright owners face significant obstacles identifying infringers and providing evidence of harm for something like P2P filesharing. Would identifying actual harm be possible/fair/strain on judicial resources? Wouldn’t it be hard to prove ownership, registration, etc. of thousands of works in a sharer’s library, instead of a sampling as we saw in Jammie Thomas?
Sohn: There are a variety of ways to do it. One could imagine a regime where higher damages require a showing that there are substantial damages, even without proving them specifically. Distinguish probably harmless infringement from infringement that is probably causing a bunch of harm. Could be a presumption of minimum range without a threshold showing of harm. Point would be to have a middle ground, without full on proof of damages we believe too difficult to show.
Aistars: courts are already serving that function. Only cases where there truly is some greater societal harm see the larger damages awards; even default judgments against filesharing users tend to be on the lower end. Need to look at wide variety of creators relying on statutory damages and their deterrent effect, not just the business models of music and movies; need to look at newspapers, photographers. New proof is completely unmanageable for small business and overlooks noneconomic damages that individuals and small businesses often pursue infringement claims for. They have no track record of licensing; there may be no directly provable profits; photographer whose work was used without permission by clothing designer in large department store.
Menell: on the music side, if we started afresh we wouldn’t build a system built around massive statutory damages. We have experience that doesn’t work. Small claims, parking ticket style system would be much better. Recalcitrants could get ramped up. Using fed courts to resolve disputes: already much more than most of these works are worth. We need some other system. But not for the P2P network itself, which can scale.
Samuelson: fewer than 14% of WIPO countries have statutory damages, most being post-Soviet states; the statutory damage states have many limits, such as Canada’s cap on noncommercial infringement damages and judicial discretion to reduce statutory damages to meet a just award; many countries don’t allow per infringed work which is particularly important in the secondary liability context (Google Books, statutory damages exposure in the billions). Other countries have 2x/3x guidelines. There are a number of things to look at for limitations that make them more just. Not arguing for repeal, but more limits.
Tom Sinder (sp?): Korea didn’t have statutory damages, and that didn’t deter infringement. 4 jury trials of individual filesharers, in which plaintiffs introduced reasonable royalty evidence: what would the license fee have been to do what D did, which was the whole economic value of the copryight—that means the award was compensatory. (No it doesn’t! No one thinks that Thomas’s making available worked like American Idol’s official website’s making available. There was no chance she’d distribute on that scale.) Do you think these awards were excessive?
Sinder: but what if it’s compensatory—what you’d have had to pay?
Sohn: for individual behavior, you want the amount to reflect the damages. The real focus we should have is on tricky questions of copryight law—it’s a problem to have a regime that suggests that if they make a wrong interpretation the consequences are $100,000.
Tepp: we are naturally more sympathetic to a single mother, but she can impose just as much harm on the copyright owner for millions of people to download—the harm may be just that great. (Hm, I hadn’t noticed these songs losing their economic value entirely.)
Q: Themes like lessening risk—maybe have discussions on secondary liability and orphan works then revisit statutory damages. If we fixed those, what would be left?
Erickson: that would be helpful, but primary infringement is a growing issue too.
Menell: enforcement can be thought of up front.Aistars: along with enforcement, there’s room for public enforcement to reduce harm to individuals and small businesses—small claims process. Voluntary stakeholder process including necessary players making enforcement less burdensome for all players (all!) whether we represent small creators or internet innovators likewise burdened by enforcement challenges. (I can’t think who’s left out of that either/or.)