Against Intellectual Monopoly

Xerox,
not Microsoft or Apple that invented the graphical user interface based on
manipulating icons on a graphic screen. Still, most of us would agree that
it was socially beneficial that Microsoft imitated and outperformed the
original innovators. Eventually, mature industries reach some kind of longrun equilibrium where there is roughly the correct amount of productive
capacity, the rents earned by the marginal firm are just enough to pay for
its fixed costs, and competitive equilibrium reigns - until, of course, some
other innovator comes along with a new kind of shoe and steps over the
placid equilibrium lake to create the socially beneficial waves of competitive
innovation, which is the source of all progress.
Indivisibility
    Our analogy between shoes and ideas has served us well so far, which is why
it is the right time to drop it and examine the crucial difference between
the economics of shoes and the economics of innovation. The fact that the
innovator will earn a rent means that some ideas will be produced under
conditions of competition. But it does not imply that every socially valuable
idea will be produced: eating the fruit of the indivisibility tree will reveal to
us the limits of competitive innovation.

    Consider again the case of a shoe factory. The standard theory of competition asserts not only that shoe factories will be built but also that the
socially desirable number of shoe factories will be built. The reason for
this is that shoe factories are fairly divisible: we may build smaller or larger
shoe factories. The builder of the first factory, when deciding how large a
factory to build, will not build so large a factory that the rents from the fixed
capacity of the factory will be less than the cost of building it. The builder -
facing competition from imitators building other shoe factories - will wish
to increase the size of his factory as long as the rents from a little more
capacity exceed the cost of adding the capacity. Imitators will do likewise.
This is exactly the condition for maximizing social surplus, and that is why
economists do not argue that owners of shoe factories should be awarded
government monopolies. This pleasant solution does not necessarily apply
in the case of innovations.
    In contrast to shoe factories, even with minimal installed capacity, the
copies of a book that can be made over an extremely short period of time
may be so many as to essentially flood the market, dropping the price to
near marginal cost almost immediately. (We should note that the evidence
suggests that this is not the case.) The resultant difference between price
and marginal cost may be so small that, when multiplied by the number
of copies, it yields an insufficient rent. The rent is insufficient because, say,
the book is very complicated, and it took a long time to complete. There
is no way to offset this combination of excess capacity and large fixed cost
by producing a smaller book that is a good substitute for the complete
book; this is something we can bear witness to. The presence of such an
indivisibility in the innovation process and the fact that initial capacity may
be large relative to the size of the market is a key fact about innovation under
competition.
    Most ideas are not divisible, and there are cases in which the cost required
to come up with the first prototype of an idea is quite large compared to the
size of the market for copies of that idea. Said differently, the capacity the
innovator must install (more often, the capacity that is already installed) is
so large, given the demand for the good, that one is not likely to earn any
rent over marginal cost. In this case, a rational innovator understands that
she cannot recover the initial fixed cost, and she does not even get started.
For a given demand, when these two anomalies - large minimum capacity
and large fixed cost - meet, competitive markets do not function properly.
This is the heart of the economic argument for intellectual monopoly: that
the additional profit achieved by a monopolist may, some of the time, lead to
socially desirable innovations that would not be produced with unfettered competition. Let us be clear: as a theoretical argument this is a sound one
and we would not dream of denying it. In fact, it is a special case of the
very same model we have proposed both here and elsewhere. We are not
arguing the case of large initial capacity and small market