Tuesday, July 31, 2012

Richard Epstein: Richard Posner Gets It Wrong

President Barack Obama has encountered a torrent of opposition for his use of what the
Wall Street Journal’s Kimberly Strassel shrewdly described as Four Little Words: “You didn’t build that.” The President used these words to show that private success rests on public infrastructure. But in so doing, he slighted the importance of private initiative and innovation. The pushback has been enormous.

Right now, the patent system is also under major attack from a large number of scholars and judges who think that the way to industrial progress lies through an expanded public domain.
The collective war cry here is not, at least for the moment, “You didn’t invent that.” Nonetheless, the clear thrust of the current anti-patent fervor is that, all too often, innovators do not receive sufficient legal protection for their patents even when they did invent them. We need some pushback against this overwrought position.

The anti-patent sentiment has just been fueled by a remarkable opinion by Judge Richard Posner, my long-time colleague at the University of Chicago, sitting as a trial judge in the major case, Apple v. Motorola. The high-profile case concerns five patents—four by Apple and one by Motorola—that are involved in mobile phone technology, and it has drawn more than its fair share of attention. Judge Posner took the extraordinary step of dismissing the claims of both sides with prejudice—meaning, the case cannot be filed again elsewhere—on the grounds that neither side could make good on its argument for either damages or injunctions.

Thus, when the dust settled, there was no reason at all to have a trial on whether either side had infringed the patents of the other. In a subsequent piece written for The Atlantic, grandly entitled “Why There are Too Many Patents in America,” Posner delivered a general critique of the patent system, discussing the broader issues involved in his judicial decision.

Posner’s criticism of patent proliferation reminds me of the famous exchange between the Emperor Joseph II and Mozart on the premier of Mozart’s opera, “The Abduction from the Seraglio,” as recounted in the movie Amadeus:

Emperor Joseph II: My dear young man, don't take it too hard. Your work is ingenious. It's quality work. And there are simply too many notes, that's all. Just cut a few and it will be perfect.

Mozart: Which few did you have in mind, Majesty?

Which patents does Posner have in mind?

To answer that question, it is important to place Posner’s latest sally on patent proliferation into its broader social context. The fashionable academic attack on patents is just one part of the full-scale attack on market institutions that has led to an economy mired in 1.5 percent annual GDP growth. The basic debate in this field can be summarized in one proposition: Strong patent protection is a threat to the overall operation of the legal system, so the appropriate set of remedies calls for narrowing the scope of the patent system at every opportunity.

The current three-part attack on the patent system starts with the proposition that the requirements needed to obtain a patent should be restricted, especially in the areas sof software and business method patents. Second, the ability to license patents should be restricted. Finally, the remedial protections given to patents by way of damages and injunctions should be weakened.

Damages should be harder to get, and calculated in ways that reduce the total levels of compensation required; injunctions against infringement should be issued only sparingly, and not as a matter of course, as was the case prior to the Supreme Court’s watershed 2005 decision in eBay v. MercExchange. The net effect of these combined maneuvers is to increase the cost of patents, which is likely to result in an increase in the size of the public domain relative to the areas of patented technology.

The question is whether this seismic shift is a good thing. As a member of Hoover’s Property Rights, Freedom, and Prosperity Task Force, I have long been on the opposite side the debate, not only in my academic work but also (full disclosure) as a lawyer who has worked for both DataTreasury and Qualcomm on a variety of projects that raise just these fundamental issues. What is strange about this debate is that, often times, the opposition to patents is put forward on the grounds that it will advance economic efficiency by eliminating the administrative costs and expensive disputes that always crop up in this area.

The underlying criticism is that this so-called “patent thicket” places serious obstacles in the path of many innovative companies by allowing a “patent troll” to obtain legal protection for innovations that they do not make, solely to block others from using that technology without first receiving a license from the firm that patents.

The troll objection is wildly overstated. There is nothing wrong with a firm that specializes in developing technologies for licensing to others. That is just a sensible form of market specialization, in which the force for innovation comes from one firm, often with a single idea and modest capital, which hopes to reap hefty profits by issuing non-exclusive licenses to companies that can then incorporate this patent into their business models. An attack on the division of labor seems bizarre, especially when launched in the name of economic efficiency.

Nor is there any obvious global sign of patent malaise in the software industry. Last I looked, the level of technological improvement in the electronics and software industries has continued to impress. The rise of the iPad, the rapid growth of social media, the increased use of the once humble cell phone as a mobile platform for a dizzying array of applications—these do not point to industries in their death throes. It may well be the case that a better patent system could have seen more rapid growth in technology.

What is so striking about Posner’s criticism, however, is that he does not offer any concrete example of how the current patent system has wreaked havoc on society. Moreover, he says nothing about the incredible success that a variety of pooling and licensing devices have had in spreading the use of patents through a wide range of standard setting organizations. Through their umbrella organization, the American National Standards Institute, these organizations have protested efforts embodied in a recent Federal Trade Commission report on the “Evolving IP Marketplace” to weaken their ability to secure rapid innovation and widespread compliance with major industry standards.

Nor does he mention any of the private devices for putting valuable technology into the public domain.  These include not only the well known “open source” software movement, but also decisions by public companies like Google to put some of their own algorithms into the public domain, doubtless to improve the market for its patented technologies.

Instead, Posner’s major move is to recognize, correctly, that patent protection is essential for pharmaceutical products that often involve single molecules (think Lipitor). These patents are easy to protect through the patent system because their chemical composition can be described with exactitude. And they surely need the patent protection in order to secure the major investments that are required not only to isolate the desired compounds but also to run them through the torturous approval process of the Food and Drug Administration. But it is a whopping nonsequitur to assume that because pharmaceuticals are the poster-child for patent protection that software should be kept out of the system.

More specifically, a natural selection device makes it more likely that the most important software innovations make it into the patent system. By any reckoning, patent prosecution—the fine art of getting a patent through the Patent and Trade Office—is not an easy endeavor. Indeed, it is one that on average has been made even more difficult with the passage of the misnamed and convoluted America Invents Act.

Firms that have modest inventions that are easy to invent around will not incur the cost of obtaining patents. It is much more likely that they will attempt to protect them by trade secrets or leave them totally unprotected. We should expect, therefore, that the subset of software innovations that do make it into the patent system are those that are worth enough to protect and to license. The best software patents may not be as good as the best pharmaceutical patents. But that is the wrong test to use. The correct question is whether they still produce positive gains. Should the law, for example, dismantle the suite of patents that protects the complex business of Customer Relations Management because of highly publicized instances of patent folly?

Just these issues were very much at stake in the Apple-Motorola litigation. It’s not clear whether Apple’s so-called “realtime” patent is worth protecting as there is not a mention of how it works or what it does in Posner’s opinion. Instead, Posner took the extraordinary course of canceling the trial on liability because of his deep conviction that the expert witnesses on both sides could not be prepared, even with the guidance of some of the ablest lawyers in the land, to produce reports that met the minimum standard for admissibility on either the damages or injunction question.

What is so striking about Posner’s relentless dissection of the imprecision in these claims was that he could apply it with equal conviction in any patent software dispute. The estimates of damages under the law are not confined to a single standard, but often involve an uncertain choice between reasonable royalties for licensing the patent and actual damages that were incurred because the patents were not licensed. The injunctive relief is (or at least should be) awarded precisely because it is so difficult to figure out what those damages really ought to be.

But Posner said that he would not allow an injunction if the best that the plaintiffs could garner was $1 in nominal damages. That surely seems over the top, because if there is infringement, the one number that is manifestly wrong is $1. A more sensible approach here, therefore, is to mix and marry the two remedies, so that the injunction does not pull the past product off the market, but awards some damages for past losses, while giving the infringer some period of time—say three to six months—to invent around the patent for future output. This then sets the stage for a negotiated license if that is cheaper.

By putting the remedial cart before the liability horse, we have the odd situation that no one can find out anything about the strength of the patent or the potential range of damages. If that is done on a common basis, then we will have knocked out the entire patent system for software, without having the slightest idea of the relative strength of the Apple and Motorola contentions.

The Posner decision looks doubly worrisome against the backdrop of his ominous Atlantic column, which shows his ill-concealed disdain for a complex industry with which he has had no direct engagement. It is an odd way to make patent policy. Right now, a similar Apple-Samsung dispute is before Judge Lucy Koh, which will involve a real trial. The Posner opinion is already on the fast track to appeal before the Federal Circuit, which will give us more information as to whether these submarine assaults on the patent system will take hold. Let us hope that Posner’s mysterious patent adventurism dies a quick and deserved death.

Richard A. Epstein, the Peter and Kirsten Bedford Senior Fellow at the Hoover Institution, is the Laurence A. Tisch Professor of Law, New York University Law School, and a senior lecturer at the University of Chicago.

No comments: