Against Intellectual Monopoly

that, for a long time, the software industry was
free of patent protection. The long-standing tradition of free competition
and lack of intellectual monopoly began to crumble in 1981, with the
Supreme Court decision in Diamond v. Diehr, and collapsed completely
with the publication of new examination guidelines by the U.S. Patent
and Trademark Office in 1996, which made computer programs fully and
clearly patentable. This change in the property-rights regime in the software
industry was relatively fast; it constitutes, therefore, an interesting case study
to test competing hypothesis on the determinants ofpatents and their impact
on productivity. After carrying out a careful econometric analysis of the
microeconomic evidence from the software industry, Bessen and Hunt reach
three interesting conclusions. The first is that the shift in legal standards for
patenting software was a potent incentive to increase expenditure in patents.
It may, in fact, be one of the key factors behind the dramatic increase in
the number of patents we reported earlier in this chapter. As we noted, the
increase in the number of patents in the U.S. economy was not accompanied
or followed by an equally visible increase in TFP or in any other economic
measure of effective innovation and productivity. The second finding by
Bessen and Hunt supports and reinforces this assertion:
    Thus, our analysis appears to decisively reject the incentive hypothesis during the
1990s. Software patents may have complemented R&D during the early [19] 80s -
when patenting standards were still relatively high - but they substituted for R&D
during the 1990s. Regulatory changes increased the amount of patenting, but they
are also associated with lower R&D. We can reject naive arguments that more
patents, relaxed standards, or lower patenting costs lead to more R&D.29
    Notice, in particular, that patenting is found to be a substitute for research
and development, leading to a reduction of innovation. In the authors'
calculation, innovative activity in the software industry would have been about 15 percent higher in the absence of patent protection for new software.
Finally, and most interestingly in our view, Bessen and Hunt point out that
one of the channels through which relaxed patenting criteria and a judicial
system more prone to entertain claims of patent infringement, negatively
affect innovative activity is by increasing the risk of the return on innovations. Stephen P. Fox, associate general counsel and director of HewlettPackard highlights that "pervasive uncertainty about legal rights, both in
terms of ability to enforce one's own patents and ability to avoid rapidly
escalating exposures to infringement claims by others. And that uncertainty
heightens risks surrounding innovation investment decisions."30

    According to Cecil D. Quillen Jr., former general counsel at Eastman
Kodak:
    If the uncertainties are such that you cannot be confident that your products are
free and clear of others' patents you will not commercialize them, or a higher return
will be demanded if you do to compensate for the additional risk. And this probably
means you will not do the R&D that might lead to low return (or no return)
products.31
Submarine Patents
    A particularly egregious method of patent abuse is the submarine patent.
Until recently, the length of patent term was measured from the time at
which the patent was awarded; prior to the award, the existence of the
patent is secret, and it is possible to continually defer the award of the
patent by filing amendments. Although the patent term was measured from
the date of award, prior art and the validity of the patent are measured from
the day of submission. Hence, the submarine patent - the filing of a useless
patent on a broad idea that might, one day, be useful. The existence of the
filing is secret (thus, the term "submarine"), and the application process
is dragged out until some actual innovator invests the time and effort to
make the idea useful. At that time, the amendment filing stops, the patent
is awarded, and the submarine surfaces to demand license fees.
    This form of legal blackmail was pioneered by George Selden, who
patented the idea of a "road engine" in 1895. He first applied for a patent in
1879 and used all possible legal means to delay approval for sixteen years.
This took place while the American automobile industry was developing and
the technology of