Subject: Re: economic efficiency of free software
From: "Stephen J. Turnbull" <stephen@xemacs.org>
Date: Wed, 14 Jan 2004 18:19:39 +0900

>>>>> "Kragen" == Kragen Sitaker <kragen@pobox.com> writes:

    Kragen> Here's a naive economic argument that spending money on
    Kragen> free software is more efficient than spending it on
    Kragen> proprietary software.  Perhaps there are some good ideas
    Kragen> in here, but I don't know enough about economics to be
    Kragen> able to judge.

The argument, as far as you go, is fine.  Where it is naive is that
it's basically an argument from hindsight.  You see that Microsoft
made a lot of money, and it could have been done at a lower price, eg
by a benevolent government.  [Russ: Just for the sake of argument.]
This is called "comparing to the first best".

However, to work in the real world, the mechanism has to fund
development _before_ the profits roll in.  As Ben Tilly remarks, most
of the known methods for figuring out whether or not to produce a
given public good can manage to fail to produce any value whatsoever
no matter how large the opportunity is.  But we know that monopolies
will do a pretty good job of getting a large amount of (gross) value
distributed to many customers.  It doesn't mean that you're wrong, but
you do have more work to do to show that we can do better than
proprietary products _in practice_ (ie, you need to specify a way to
find the "second best" subject to real-world constraints on knowledge).

A second point is more important.  And that is that it is _not_
obvious what software needs to be developed from the point of view of
non-developer users.  Yes, after Microsoft manages to gouge $500
million out of customers for a Word upgrade, you know that there was a
lot of value to be had by developing it.  But if you're not getting
paid by the copy, how do you know how much your customers really value
what you're giving them?

In the worst cases you get arrogant developers saying "I know how to
save the world, and all you have to do is fund my ivory tower to the
tune of $190,000/year."  My Lord, that's a generous offer, if the
world is truly going to be saved, but I don't think it's true.  Or
maybe even worse are the various major forks, with _two_ sets of
arrogant developers claiming unique insight into user needs.

    Kragen> The two effects I mentioned earlier --- more efficient
    Kragen> allocation of money due to elimination of monopoly rents
    Kragen> (which we trade for the free-rider problem, which is a
    Kragen> smaller inefficiency in some cases), and more users ---
    Kragen> have a multiplicative effect that means that a dollar
    Kragen> spent to fund an improvement to a piece of free software
    Kragen> is a far more efficient allocation of resources than the
    Kragen> same dollar spent to buy a license for proprietary
    Kragen> software.

I think enough money has gone into Cygnus, Red Hat, SuSE, the X
Consortium, Apache, Python, etc to demonstrate that this is
empirically false, at least for the software producers.  They simply
have failed to achieve World Domination, despite many millions of
"seed" money to get that multiplicative effect going.

There might be a social mechanism to do better, but certainly from the
individual standpoint the multiplicative effect doesn't look huge.


-- 
Institute of Policy and Planning Sciences     http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
               Ask not how you can "do" free software business;
              ask what your business can "do for" free software.