Patent litigation is usually something that is off most people’s radar. The issues are highly complex, both technologically and legally, and usually the affected parties are the litigants themselves. There are exceptions, however, such as the ongoing case of Apple Computer v. Samsung Electronics, which has made the news because billions of dollars matt be at stake. Apple alleges that Samsung’s Galaxy S III infringes on patents it obtained with respect to the iPhone 5. Samsung says it developed the same innovations independently and has filed counterclaims against Apple. InformationWeek’s Charles Babcock provides a summary.
A long-standing issue in patent litigation is the extent to which parties use the courts strategically, either to to improve their negotiating position in licensing negotiations or just to hobble their competitors generally. Individual plaintiffs may litigate as long as the cost of litigation to any individual plaintiff is less than the expected value of actual or strategic benefits. They do not account for costs they impose on defendants or on the public at large. Thus, patent litigation may have substantial externalities.
One way to reduce these externalities is to require losers to compensate winners, such as by having to pay their attorneys’ fees (the so-called English Rule; see Wikipedia and PointofLaw.com). Patent attorney and blogger Gene Quinn highlights a recently introduced bill that would establish a version of the English Rule in certain patent litigation.
Quinn writes that under current law:
whether attorneys fees should be awarded requires a two-step inquiry. First, the district court must decide whether there is clear and convincing evidence that the case is exceptional. Second, the court must decide whether to award attorney fees to the prevailing party even if the case is deemed exceptional.
Yes, even a finding that a case is an “exceptional case” does not always mandate an award of attorney fees in all circumstances. The United States Court of Appeals for the Federal Circuit has acknowledged that ther are many factors which can impact upon a decision regarding whether an award of attorney fees is warranted in a particular case. The types of conduct that could support a showing of exceptional circumstances resulting in the award of attorneys fees include, but are not limited to, willful infringement, inequitable conduct before the Patent and Trademark Office, litigation misconduct, and vexatious or unjustified litigation or frivolous suit.
What this means is that obtaining attorneys fees in any patent litigation is rare.
To be effective, a loser pays rule must be easily interpreted and predictably enforced. That is, prospective plaintiffs must have a clear idea what circumstances would result in having to pay defendants’ attorney fees. Both prospective plaintiffs and defendants also must know how courts will enforce the law.
Quinn raises important an economic question about how the bill would be interpreted if enacted. For example, the bill allows (but does not require) the courts to award attorneys’ fees “upon making a determination that the party alleging the infringement of the patent did not have a reasonable likelihood of succeeding.” What does “reasonable likelihood” mean? Unless Congress is specific about this, the determination will be left to the courts. This is a huge source of ambiguity that undermines the bill. Similarly, by empowering but not requiring courts to award attorney fees, the bill creates a second source of uncertainty that undermines its purpose. Quinn’s view is these uncertainties are quite large:
The federal courts are virtually certain to interpret that very restrictively because they don’t want to punish those who bring a lawsuit and lose. That is not the America Rule of paying for attorneys fees, and historically has been frowned upon in the U.S. because it could have a chilling effect. We prefer to err on the side of allowing cases to be brought and not punishing those who have a legitimate claim but ultimately do not prevail. So you can almost bet your last dollar that “reasonable likelihood of success” will not mean what you think it should mean, and it will be interpreted extremely narrowly.
Quinn does not discuss another major hurdle in devising an effective loser pays scheme. For it to be effective, the underlying law — in this case, patent law — also must be clear. If the courts’ determinations of infringement are variable, uncertain, or unpredictable, then a loser pays scheme cannot effectively internalize the social costs of frivolous litigation. It would change plaintiffs incentives to sue at the margin, but only a small amount unless attorney fees are a large fraction of the financial stakes of the case, which seems highly unlikely.
Unsurprisingly, this has made for considerable merriment in the Apple v. Samsung case. In the Washington Post, Dominic Basulto asks Siri who would win:
Here’s a transcript of our exchange:
Me: Did you read the pre-trial coverage from the Washington Post?
Siri: Sorry, Dominic. I can only read incoming text messages.
Me: What is the best phone?
Siri: The one you are holding.
Me: What is your opinion of Samsung?
Siri: Dominic, you are not supposed to ask your assistant such things.
Me: What’s your favorite iPhone design?
Siri: I don’t really like talking about myself, Dominic.
Me: What is your favorite iPhone feature?
Siri: Your interest flatters me, but… is there something I can do for you?
Me: Is it true that Samsung phones are “not as cool” as Apple’s?
Siri: I can’t answer that. But I could search the Web for you…
Me: Have lawyers been asking you a lot of questions, Siri?
Siri: I found a number of lawyers… 22 of them are fairly close to you.
Me: Where is Tim Cook?
Siri: I haven’t a clue.
Me: What’s Apple’s current stock price?
Siri: AAPL – Apple, Inc. is up so far… $595.79
Me: What’s $2.5 billion in damages divided by Apple’s stock price?
Siri: 4.213 million.
The main problem with any loser pays scheme is that Siri’s prediction of a case’s outcome under patent law may be just as good as anyone else’s. In a lot of cases, the stakes will be way too large for a loser pays rule to make any difference.