• Clearings and Thickets

Clearings and Thickets
by
Robert Cooter
Comparative Law and Economics Forum (CLEF)
Oxford, England, June 30-July 1, 2011.
Paper revised June 15, 2011.
1
 Robert Cooter1
Clearings and Thickets2
Compounded over a century, 2 percent annual growth (roughly the recent
growth rate of the U.S. economy) increases wealth more than 7 times, and 10
percent annual growth (roughly the recent growth rate of the Chinese economy)
increases wealth by almost 14,000 times.3 Differences in sustained growth
cause one country’s wealth to overtake another country faster than the mind can
grasp.
Like individuals, nations are highly competitive. Poor nations hope to
overtake rich nations, and rich nations fear being overtaken. Besides relative
income, individuals and nations care about absolute income. Most people want to
live a long time and educate their children absolutely. Life expectancy at birth is
83 years in Japan and 66 years in Bangladesh.4 Enrollment in secondary school
is 98 percent among Japanese children of the appropriate age and 42 percent in
Bangladesh.5 Almost everyone would prefer to live at the level of Japan rather
than Bangladesh.
1
Robert Cooter is Herman Selvin Professor of Law, University of California at Berkeley.
2
Robert Cooter excerpted this presentation from two joint writings with Aaron Edlin:
“Chapter 2: Clearings and Thickets” in a preliminary book manuscript tentatively entitled
Introduction to Law and Growth Economics, and “Overtaking” in The American Illness,
ed. Frank Buckley (Yale University Press, 2012).
3
1.02 to the power 100 equals 7.2; 1.10 to the power 100 equals 13,781. Growth in
national income per capita is a better indicator of welfare than growth in national income.
According to the CIA’s World Factbook, U.S. population growth rate is roughly 1percent
(among the highest among developed countries), and China’s population growth rate is .5
percent.
4
Life expectancy at birth, total (years)) in 2008, World Bank.
http://data.worldbank.org/indicator/SP.DYN.LE00.IN. Note that life expectancy is in the
40s in many African countries.
5
Data & Statistics on Education for 2007 and 2008, World Bank,
http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTEDUCATION/0,,content
MDK:20573961~menuPK:282404~pagePK:148956~piPK:216618~theSitePK:282386,00
.html.
2
 Much contemporary policy discussion concerns redistribution of income,
but growth has more powerful effects on human welfare. The lowest wage
earners in the U.S. or any country would benefit far more in the long run from a
faster growth rate in national income than from redistributing existing national
wealth in their favor. Even people who depend on transfer payments would
benefit more from faster growth than income redistribution. A faster growth rate
increases wages and tax revenues available for transfer payments, and this
increase overtakes gains in welfare from redistributing income.6 Returning to
our previous comparison, 1 percent of an increase of 14,000 is 20 times more
than 100 percent of an increase of 7 times.
To appreciate the mathematics of overtaking, consider two sequences of
utility in an infinite number of generations. In the first sequence, income per
capita grows at a faster rate. In the second sequence, growth is slower and
initial utility is higher. The sum of utilities in the first sequence overtakes the sum
of utilities in the second sequence whenever policy makers give weight to future
generations that is similar to the weight given to the present generation.7
Although this “overtaking theorem” concerns infinite time, in practice overtaking
occurs quicker than people imagine, because they fail to appreciate fully the
effects of compound growth. Given overtaking, anyone who values human
welfare should recognize growth’s importance as a policy goal.
In the preceding propositions, “growth” has two attributes lacking in
standard measures of national income such as gross domestic product (GDP).
The first attribute is sustainability. Growth overtakes other welfare considerations
when it is sustainable and continues into the future. Sustainable growth uses
6
The largest effect of income redistribution on human welfare is through growth.
Sometimes more inequality causes more growth by improving incentives for innovation,
as in China after 1980. Sometimes redistributive expenditures promote growth by
improving the education and health of workers as with free primary schools and prenatal
clinics.
7
Robert Cooter and Aaron Edlin, Maximizing Growth vs. Static Efficiency or
Redistribution, Berkeley Law and Economics Working Paper (2010) prove formally that
overtaking always occurs provided the social planner’s preferences for equality are not
too strong.
3
natural resources in fixed supply at a decreasing rate, whereas unsustainable
growth uses them at a constant or increasing rate (or a rate that does not
decrease fast enough). The second attribute is comprehensiveness. A
comprehensive measure of wealth encompasses all valuable goods, including
non-market goods such as public goods and the environment. By contrast, a
partial measure of wealth such as GDP excludes goods that are not sold in
markets. When comparing welfare, partial measures may make little sense,
because a loss in unmeasured wealth may offset a gain in measured wealth.
Thus burning more coal to power air conditioners increases the simplest
measure of market income, but it does not necessarily increase a full measure of
wealth if pollution increases significantly.
The case for increasing the rate of sustainable, comprehensive growth is
compelling, but how is it to be done? Innovation proximately causes sustained
growth in income per capita. In fact, innovation is the only way to have
sustainable growth in a world of finite resources. Education, communication,
transportation, immigration, organization, taxation, and many other factors affect
innovation. This essay takes the legal perspective. From this perspective, the
normative question of growth economics is, “Which laws increase the pace of
economic innovation?”
Property law establishes the ownership of objects by people. Ownership
conveys a bundle of rights that describe what people may do with what they own
-- possess, use, develop, improve, transform, consume, deplete, destroy, sell,
donate, bequeath, transfer, mortgage, lease, loan, exclude, and so forth. By
attaching objects to people, property law supplies a framework for making and
distributing wealth, which can stimulate or suffocate economic growth. Given the
right bundle of rights and effective protection of them, an economy can grow and
enrich the nation. Given ineffective protection or the wrong bundle of rights, an
economy can decline and impoverish the nation.
Sustained growth comes from economic innovation. Discovering new
ideas and developing them usually requires investment by entrepreneurs. An
innovator has a temporary advantage over competitors until they catch up. While
4
the temporary advantage lasts, the innovator enjoys extraordinary profits, which
we call “venture profits.” Higher venture profits imply stronger incentives for
innovation. To increase the pace of innovation, laws should increase venture
profits.
Intellectual property law, which includes patents and copyright law,
establishes the ownership of innovations by people. It conveys a bundle of rights
to creators as determined by rules. Applied to intellectual property law, the
normative question of growth economics is, “Which ownership rules maximize
innovation?” In order to increase the pace of innovation, ownership rules should
increase venture profits. So the question of this essay is, “Which ownership rules
maximize venture profits?”
How Intellectual Property Law Promotes Growth
To answer this question, we begin by explaining how innovations become
property. Legal ownership sometimes goes to the first person to invent a
machine, create a molecule, extract a vaccine, compose a song, write a book, or
record a performance. The first person to embody a new idea in an innovation
may acquire private ownership of it through patent or copyright law. These two
bodies of law convey ownership to intellectual creations, so they are called
“intellectual property law.”
Intellectual property rights sometimes increase investment in innovations.
Thus Figure 2.1 depicts the payoffs to an innovation over its life cycle, from
discovery to obsolescence. Developing a new idea in period 0 costs 8. We
assume that the developer receives a patent for the innovation, which prohibits
others from using it without his permission. Next the innovation is launched and
sold to buyers. When launched in period 1, the innovation has no competitors,
so the innovator is a monopolist who receives a payoff of 7. In period 2, imitators
develop competing products that substitute for the patented innovation, which
reduces the innovator’s payoff to 4. In period 3, the patent expires and
competition intensifies. Others can now produce the same goods at the same
cost as the innovator. Taking competition to its logical extreme, we assume that
5
the market is perfectly competitive in period 3 and the innovator’s payoff is zero.
Summing over the life cycle, the innovator’s net payoff equals +3, so developing
the innovation is profitable to the innovator. If the innovator can foresee this fact
in period 0, it will develop the innovation. Generalizing, the expectation of
positive venture profits causes the development of innovations.
Figure 2.1. Payoff to Innovator with Property Right
" A"
B437" !" 1" "
" "
"
Effective intellectual property law transforms the innovator’s loss in Figure
2.2 into the gain in Figure 2.1. The comparison between the two figures
illustrates that intellectual property rights are necessary to innovation when
development costs are high relative to the payoff from launching an innovation
first. In contrast, some innovations have low development costs relative to the
payoff from launching first. This can occur either because the innovation is
particularly cheap to develop, or because launching it first yields a big advantage
for the innovator. In such cases, intellectual property rights may not be
necessary for innovation.
Figure 2.3 illustrates this possibility. As in Figure 2.2, the innovator in
Figure 2.3 who develops the product spends 8 on development and does not
receive an effective patent. As in Figure 2.2, the launch in period 1 yields profits
of 7. The difference between Figures 2.2 and 2.3 occurs in period 2. We
7
assume that the innovation in Figure 2.3 is a differentiated product that yields a
payoff of 3 in period 2. The imitations in period 2 are not quite as good as the
original. (In contrast, Figure 2.2 assumes that the imitations in period 2 are
perfect substitutes for the original.) Summing over the life cycle in Figure 2.3, the
innovator’s net payoff equals +2, so developing the product is profitable to the
innovator. If the innovator can foresee this fact in period 0, it will develop the
innovation, even though it will not receive a patent. Generalizing, an innovation
is profitable without intellectual property protection when development costs are
low relative to the payoff from launching first.
Figure 2.3. Payoffs With Developer’s Advantage and Without Property
Right
" @"
A437" !" 1" ;"
H7I7-+,37&5" #$%&'(" 2345$6+&" 84%3"
)))))))))))))))))))" )))))))))))))))))))"
*$8B75"C58%'5%87" ))))))))))))))))))) )))))))))))))))))))" )))))))))))))))))))"
" *+&+,+-." 23,7897'5" ?7897'5"
"D"E75",8+F5G" " " :+3,766+&" :+3,766+&"
)J" /0"" /;" @"
" "
"
Besides making an unprofitable innovation profitable, effective intellectual
property law can make a profitable innovation more profitable. Thus a patent in
Figure 2.3 might increase the payoff in period 2 from 3 to 4, so the innovator’s
net payoff summed over the life cycle increases from +2 to +3. This conclusion,
however, does not generalize. Stronger intellectual property rights for a
particular innovation make it more profitable, but stronger intellectual property
rights for all innovations can increase or decrease venture profits overall. (More
on this comes later.)
8
 Intellectual property law differs from one country to another and it changes
with time. Thus U.S. patent duration is 20 years from the date of filing the patent
application. (Before 1995, U.S. patent duration was 17 years from the date the
patent issued.) Besides duration, patents also differ in breadth -- a patent on all
umbrellas is broader than a patent on umbrellas that open automatically. The
duration and breadth of a patent define its scope. The law requires anyone who
acts within the scope of someone else’s patent to license the right to do so from
its owner. Acting within the scope of a patent and without a license infringes it.
The owner can sue to prevent future infringements and to recover damages for
past infringements.
Patent law strengthens by increasing the number, breadth, or duration of
patents. Thus a controversial development recently strengthened US patent law
by allowing inventors to patent new business processes, which were formerly
considered unpatentable, such as Amazon’s patent on “one-click” ordering from
its online catalogue. “Stronger patent law” refers to a change in rules that
increases the scope of private ownership of innovations, leaving less scope for
unowned innovations.
Thus Figure 2.4 divides future innovations into “ownable” and
“unownable.” A new computer chip is ownable (patentable) and a new metaphor
in the English language is unownable (no patent, no copyright). The boundary of
the ownable is labeled β. Strengthening the patent system increases ownable
innovations and decreases unownable innovations, as indicated in Figure 2.4 by
shifting the boundary from β to β`. In effect, stronger patents fence in more of
the common land of innovation.8
Figure 2.4. Future Innovations
8
Increases in the scope of a particular patent do not necessarily increase the scope of
patents as a whole. Rather, the scope of a particular patent can increase by decreasing the
scope of another patent. Instead of fencing more of the common land of innovation, the
fence between patent holders moves to the advantage of one of them and to the
disadvantage of the other.
9
 *+$$
!"#$$
*$$ "%!&'()$
"%!&'()$
!" *+$
#"
*$$
Our discussion of Figures 2.1 to 2.3 explained that stronger patents rights
sometimes make unprofitable innovations profitable and profitable innovations
more profitable. When stronger intellectual property rights make innovations
more profitable, the number of innovations usually increases. To see why, think
of an array of new ideas that differ according to the expected profitability of
developing them. Innovators develop the ideas that they expect to yield positive
profits, and they do not develop the ideas that they expect to yield negative
profits. If stronger intellectual property rights increase the expected profits for
innovations, some ideas will tip from negative to positive expected profits, so
innovators will want to develop them.
Figure 2.5 illustrates this possibility. The horizontal axis indicates an
industry’s array of possible innovations in decreasing order of profitability. The
10
curve R indicates revenues given weak patents. The revenue curve R intersects
the cost curve C at point i. To the left of i, revenues R exceed costs C, so the
innovations will be developed. To the right of i, costs C exceed revenues R, so
the innovations will not be developed. The tipping point is i, which equals the
number of developed innovations. By assumption, strengthening patents in this
industry increases revenues from R to R`. We assume that costs C do not
change. With strong patents, the revenue curve R` intersects the cost curve C
at point i`, which indicates the number of developed innovations. Thus stronger
patents cause the development of innovations to increase from i to i`.
Figure 2.5. Development of Innovations
with Weak and Strong Patents
>!
?!
=#!
=!
!!!!"!!!!!!!!!!!!!!!!!!!!!"#!
$%%&'()&%*!+,,(-./!0-!1,&23(0"4"3-!
!5,&6!7"89!&%!:.;!3&!:&