Against Intellectual Monopoly

year, serving to improve performance and provide greater
customer satisfaction. Indeed, the position of the sports leagues with respect
to innovation in their own sport is not appreciably different from that of
the benevolent social planner invoked by economists in assessing alternative
economic institutions.
    Given that sports leagues are in the position of wishing to encourage
all innovations for which the benefits exceed the cost, they are also in the
position to implement a private system of "intellectual property," should
they find it advantageous. That is, there is nothing to prevent, say, the
National Football League from awarding exclusive rights to a new football
play for a period of time to the coach or inventor of the new play. Strikingly,
we know of no sports league that has ever done this. Apparently, in sports,
the competitive provision of innovation serves the social purpose, and
additional incentive in the form of awards of monopoly power do not serve
a useful purpose.
    As always, there is an ironic footnote to this triumph of competition:
some legal analysts in the United States now argue that the government
should enforce patents on sports moves.2'
Profits without Patents
    Patenting is high and growing by historical standards. The number of total
U.S. patents granted yearly has increased by 78 percent, to 113,834 between 1983 and 1995, after which it peaked at 187,015 in 2003 to reach the somewhat lower level of 157,717 in 2005, the last year for which data are available. Similar, albeit quantitatively less pronounced, patterns apply to the
European Union and Japan. Yet it turns out that businesses do not regard
patents as a significant factor in their decisions to innovate. There are
two surveys of research and development directors in which this clearly
emerges. This first is the Yale Survey taken in 1987, and the second is the
Carnegie Survey done in 2000.22 We focus on the more recent and more
detailed Carnegie Survey, but the same facts emerge from the earlier Yale
Survey.

    The Carnegie Survey reported in 2000 that it received responses from
1,118 firms for product innovations and 1,087 for process innovation. The
firms were asked whether particular methods were effective in appropriating the gains from an innovation. The table here shows the percentage
of firms indicating that the particular technique was effective. The numbers in parentheses are the corresponding figures for the pharmaceutical
and medical equipment industries respectively: these are the two industries
in which the highest percentage of respondents indicated that patents are
effective.

    This strongly suggests that legal means, including patents, are regarded as
the least effective method of appropriating rents. Only about one-third
of respondents believed that patents are effective. Secrecy, lead time - the
advantage of being first - and complementary manufacturing were rated as
the most effective. Indeed, in the case of products, being first is viewed as
the most effective means of appropriation. The two exceptional industries,
which report a relatively high importance of patents, are the pharmaceutical
and medical equipment industries. Indeed, these industries, especially the
pharmaceutical industry, are often held up as examples of why it is essential
to have patents. Yet even in these industries, only about half the respondents rated patents as an effective means of appropriation. Also striking is that
in these industries, other means such as lead time, complementary manufacturing, and secrecy are regarded as about equally effective as in other
industries. Hence, while patents are viewed as more effective in these industries, nonlegal means are still quite effective in appropriating rents.

Patent Pools
    In addition to sports leagues, there is another significant and widespread
example of private companies that voluntarily relinquish "intellectual property." These are the so-called patent pools.23 A patent pool is an agreement,
generally by a number of businesses in the same industry, to share patents.
Although it is sometimes the case that when the pool is set up, a company
that has few patents will make a payment to a company that has many
patents, once the pool is operating, there is no payment among companies for patents. Any patent by any company in the pool is freely available
to any other company in the pool. In some cases, patent pools take the
form of