Re: Software Patents

2000-11-17 Thread Alex Tabarrok

  I do not want to get into a debate on the morality of patents - there
is plenty of that in the Armchair archive.  On the economics here are
some notes and references written for another purpose which may be of
interest.


   Patents can *reduce* investment in RD in *theory* as well as in
practice.  There are several ways in which this can occur.  First,
consider a single firm in an industry with many firms and suppose that
the firms are competitive in the sense that patents compete with
patents, i.e. the latest technology replaces the previous technology. 
Now consider a policy which makes patents harder to get.  On the one
hand this reduces the return to RD since fewer discoveries will be
patentable but on the other any patented discover is likely to dominate
the market for a longer period of time.  That is, making a patent harder
to get reduces the probability of getting a patent but it increases the
economic life of the patent.  Recall, the idea relies on a structure in
which your patent competes with future patents.  Thus, making patents
more difficult to get reduces the probability that you will get a patent
but also reduces the probability of competition from a competitor's
future patent, conditional on you getting a patent.  It can be shown
that the second effect can dominate the first so that making patents
more difficult to get actually increases RD.

See Hunt, Robert. 1999. Patent Reform: A Mixed Blessing for the U.S.
Economy? Federal Reserve Bank of Philadelphia, Business Review. Nov/Dec.

and the same author's FRB of Phil. Working Paper.  Both available on the
net.  For the first see the link below, for the second search under
working papers at the FRB of Phil. site.

http://www.phil.frb.org/files/br/brnd99rh.pdf

Patents can also reduce innovation when one patent builds upon
another.  This is especially possible with business method patents. 
(Note the author refers to "Internet patents" while actually internet
patents are merely the most obvious example of the larger class of
"business method" patents.)  Consider why ideas such as Newton's law of
gravity are not patentable, one reason is that such ideas generate a
host of other secondary implications, ideas and inventions which would
certainly be reduced in number should the primary idea be patentable. 
Business method patents may be more like ideas than specific
inventions.  More could be said on this.

See Bessen and Maskin "Sequential Innovation, Patents and Imitation"
Working Paper available at

http://www.researchoninnovation.org/patent.pdf

and the short version at

http://ksgwww.harvard.edu/iip/econ/bessen.html

For evidence on the strategic use of patents, see the paper "The
Patent Paradox Revised: Determinants of Patenting in the US
Semiconductor industry," by Bronwyn Hall and Rosemarie Ziedonis
available on Hall's web page at

http://elsa.berkeley.edu/~bhhall/index.html

More generally, it would be interesting to think of what sort of
industries benefit from patents and what sort of industries are harmed
and why?  It's hard to see a large pharmaceutical industry existing
without some patent protection.  On the other hand there seems no reason
why semi-conductors could not exist without patent protection as they
did up until the mid 1980's (see the Hunt article).  What accounts for
these differences?  Are there any principles involved which could help
to improve the awarding of patents?  Is it really a case of all or
nothing?  What improvements to the current system might the author
recommend?  One modest suggestion comes from Robert Merges see his paper
"As Many as Six Impossible Patents before Breakfast" available at:

http://www.law.berkeley.edu:80/institutes/bclt/pubs/merges/

In addition to public policy changes the author might want to
think about market mechanisms.  A few thoughts can be found in:

"The Contractual Alternative to Patents" (with Roger E. Meiners),
International Review of Law and Economics 1 (1981), pp. 227-231.

On the issue of reform the author may also want to examine the excellent
paper by Michael Kremer

Kremer, Michael. 1998. Patent Buyouts. QJE. (Nov): 1137-1167.


Another benefit of patents is the diffusion of information.
Trade secrecy is an alternative to patents.  Thus the gains from getting
rid of patents may be less than believed if firms then invest
more in keeping their ideas secret - thus, there will still be
monopolies although not government granted monopolies.  Second, it could
even be the case that the public nature of the patent increases
information diffusion enough to make patents superior to no-patents. 
Certainly, on the margin, this idea suggests that we should patent ideas
which can be kept secret if not patented but not other processes, eg. we
should not allow a patent on one-click buying since this idea could not
be kept secret if there was no patent.


Alex
-- 
Dr. Alexander Tabarrok
Vice President and Director of Research
The Independent 

Re: Software Patents

2000-11-17 Thread Chris Rasch


The League for Programming Freedom has a dated bibliography relevant to software
patents:

http://lpf.ai.mit.edu/Links/prep.ai.mit.edu/index

Some of the more relevant entries include:

On the complex economics of patent scope
by Robert P. Merges and Richard R. Nelson
Columbia Law Review, May 1990, V90 #4, pp 839-916
Argues that product patents and broad basic patents slow progress.
Contains some good case study information.

Forbes, 3 September 1990, p 46. article on how patents can have a chilling
effect on development (p 46, 3 September 1990)

Fritz Machlup and Edith Penrose, "The Patent Controversy in the Nineteenth
Century," Journal of Economic History 10 (May 1950)--generally critical of
patents

Eric Schiff, _Industrialization without National Patents: The Netherlands,
1869-1919, Switzerland, 1850-1907_ (Princeton University Press,
1971)--suggests that those countries did just fine without a patent system

America By Design
By David F. Noble
Knopf, 1977
Not a basic text but an account of the history of US patent law, among
other things.  Noble argues that patent law has been converted from
protecting inventors to protecting the corporations inventors work for.


An economic review of the patent system:  A study of the Subcommittee on
Patents, Trademarks, and Copyrights of the Committee on the Judiciary,
United States Senate
by Fritz Machlup
Forward dated June 30, 1958
Critical of patent system:  it protects that which would become public
knowledge anyway, rewards to inventors are slight, incentives to research
not otherwise done questionable.  There is no empirical evidence to decide
the conflict of theories, however, so safest policy conclusion is to muddle
along without changes.

Imitation costs and patents:  an empirical study
by Edwin Masfield, Mark Schwartz, and Samuel Wagner
The Economic Journal, 91 (December 1981), pp 907-918
From the conclusion:  "Contrary to the assumption of many economic models,
a patent frequently does not result in a 17-year monopoly over the relevant
innovation.  Patents do tend to increase imitation costs, particularly in
the drug industry, but excluding drugs, patent protection did not seem
essential for the development and introduction of at least three-fourths of
the patented innovations studied here."

The use of patents for the protection of technological innovation:  A case
study of selected Swedish firms
by Ove Granstrand
UNCTAD/ITP/TEC/13, 18 September 1990
From the "Summary and Conclusion":
"(viii) For both large and small firms, it was confirmed that patents were
regarded as more appropriate for product innovation, while secrecy ws
considred to offer better protection for process innovation."  "For small
firms, patents were of no significant relevance to prevent imiatation.
Such firms rely more on secrecy and technologicla lead time.  The purpose
of patenting for the small firms engaged in patenting (5 out of 11), was
mainly to create bargaining power in negotiations relating to financing,
licensing or co-operation.  For large firms, patents are often viewed as an
alternative to secrecy and the propensity to patent is a function of
expectations.  The firms tend to apply for patents in cases when reverse
engineering is believed not to be costly and infringements could be easily
detected.  Another resaons for patenting is when the enforcement of the
patent law is regarded as effective.  Still other reasons for patentign are
government requirements for ublic disclosure (chemicals, agro-industry) and
the collective behaviour of the firms, which causes firms to patent if one
firm starts patenting in an area that was not previously covered by
patents.
(ix) Regarding the effectiveness of patents, the firms regretted the lack
of international standards and the limited effectiveness of patent
protection.  ...
(x)  ...  Small firms have a higher number of patents per employee and per
Swedish crowns spent on research.  However, they also have a lower share of
commercially exploited patents and a higher proportion of patenting costs
of the total R and D expenditures.  This indicates that small firms are
relatively more innovative.  ...
(xii) Patents as an instrument to stimulate innovative activities appeared
to be of little relevance for small firms.  It was found that no
significant changes in the R and D behaviour would take place if the patent
protection time were reduced or extended.  Aslo, for large firms, the R and
D behaviour seems to be rather independent of the availability of patenting
protection.  The survey showed that increased patent protection is likely
to provide, at most, a modest stimulus for R and D activities.  Chemical,
and particularly pharmaceutical, firms appear to be more sensitive to such
changes."

International association for the study of common property maintains some links
that may be of relevance:

http://www.indiana.edu/~iascp/information.html

Brian Martin's article Against Intellectual Property may have some references of
interest: