Technological Innovation and Intellectual Property

Friday I heard a talk by Brad Smith, Senior Vice President and General Counsel of Microsoft. He compared the recent problems of the US patent system to those during the period from 1865 to 1885, when patent “sharks” threatened farmers and railroads. There was rising discontent with the patent system and a push for legislative reform (some other countries actually abolished their patent systems during this period). But in the end, in the US, it was the courts, notably the Supreme Court, that changed the patent system then.

Brad argues that history is repeating. The Supreme Court has already fixed most of what Microsoft felt was wrong, including legal treatment of injunctions, non-obviousness, willfulness, and foreign patenting. Apparently, some other IT firms feel they have gotten much of what they were asking for. This has prompted Bob Armitage, General Counsel at Eli Lilly, to ask “Now That the Courts Have Beaten Congress to the Punch, Why Is Congress Still Punching the Patent System?”

Of course, what the IT firms asked for and what they really need might be two different things. It will be interesting to see how this plays out. Shortly after 1885, the patent system went into a nearly century-long decline in importance–the number of patents granted per capita decline until 1984, when the rate reversed and began rising sharply (see figure). Will 2007 mark the start of a similar decline in patenting rates and, more importantly, in litigation rates?

The initial indications do not seem to support the analogy. Patent application rates continue to rise despite these changes and despite a growing backlog at the patent office. And although litigation rates may have leveled off (see US Patent Lawsuits Increase Again), there is no evidence of a decline yet.

Indeed, Brad reported that after several years of defending 30-odd patent lawsuits each year, Microsoft is now defending over 50 lawsuits. Time will tell whether recent trends have reversed and these will indicate whether the problems of the patent system have indeed been fixed.

–Jim Bessen

I was thrilled to read that Eric Maskin won the Nobel prize for economics along with Leonid Hurwicz and Roger Myerson. They won the prize for mechanism design theory, an elegant and very general mathematical theory for analyzing the best way to align incentives between parties. This has been applied to problems of designing optimal contracts between two private parties and also to designing effective government regulation.

Eric is a friend and, informally, a teacher to me. As my college roommate, he introduced me to economics and has provided guidance and encouragement since. And we are co-authors of “Intellectual Property on the Internet: What’s Wrong with Conventional Wisdom?” (published in German in Open Source Jahrbuch 2005: Zwischen Softwareentwicklung und Gesellschaftsmodell) and “Sequential Innovation, Patents and Imitation” (an improved version forthcoming in the RAND Journal of Economics). The latter paper surprised us and became quite controversial; in it we question whether patent protection might not be the best policy in technology fields that were already highly innovative, such as software. This paper has been widely seen as influencing European legislative deliberations regarding software patents.

But Eric’s contribution to economics goes well beyond mechanism design and intellectual property. Eric has played a seminal role in bringing mathematical analysis, especially game theory, to economics. David Warsh, writing in the Boston Globe several years ago, saw the significance of Eric’s role symbolically: when Eric went to the Institute for Advanced Study in Princeton from Harvard, he moved into the house where Albert Einstein had lived.

Best wishes on a well-deserved honor.

–Jim Bessen

See also, Philippe Aigrain’s blog

A few weeks ago we reported that the number of patent lawsuits increased again. Of course, the total number of patents in force has also been increasing, so an interesting question is whether this increase just reflects the larger base of patents. Earlier research (Lanjouw and Schankerman 2004) showed only a modest increase in the rate of litigation per patent in the early 1990s.

As the following shows, the probability that a patent is in a lawsuit within 4 years of issue date has been rising more rapidly since then:

Moreover, this increase is not because lawsuits are happening more quickly after issue—we checked the pattern of lawsuits over time and it has not changed significantly since the 1980s. Part of this increase arises because lawsuits these days involve more patents, probably because lawsuits are increasingly in “complex” technologies such as computers and software. But even correcting for the number of patents in each lawsuit, the trend is still up.

This trend implies an increasing “litigiousness” on the part of patent holders and it suggests that patent holders are incurring greater litigation costs per patent. But it also means that prospective innovators have to wend their way through a minefield with many more patents and many more of those patents are likely to “explode” into lawsuits.

Lanjouw, Jean O. and Mark Schankerman. 2004. “Protecting Intellectual Property Rights: Are small firms handicapped?,” Journal of Law and Economics 47:45-74.