Subject: Re: Tom W. Bell paper
From: <>
Date: Mon, 4 Sep 2006 12:47:02 +0900

Executive summary: devices to reallocate risk do not change the
supply of investible funds to a first order approximation.

Don Marti writes:

 > We see this today in the oil market.  Terrorists who
 > attack oil facilities can participate, but that
 > doesn't keep hedgers out.

They *can* participate, but I doubt that they do on a large scale.
Current terrorist financing is not done like a Ludlum story, with the
terrorists controlling the world financial system.  Instead, it works
like samizdat.  In order to make this work, the terror networks would
have to surface the money into the financial system where it can be
tracked---and trapped.  And guess who is going to be holding that
money?  By and large, companies who had hundreds or thousands of
employees in the World Trade Center.

But what Bell is advocating is a system where we *encourage* the
insiders to participate.

 > (The market extremists need their own version of
 > Burning Man -- instead of a temporary market-free
 > zone, a temporary all-market zone where you can
 > let your inner Homo Economicus get out and play for
 > a week.)

It's called "the world that we live in."  You can do anything you like
in the privacy of your own home.  (Well, maybe playing Monopoly with
real money is a crime in the DPRK.)

The problem comes when you start handling *other people's money*.

 > But determining whether or not a patent is bogus
 > is more work than the SPEX.

Really?  I don't think so.  What is more work is doing the legal
beagling that is required to make your contention stick aginst an
issued patent.  But that's an institutional, and theoretically
correctable, bias in the system, not a feature of patents per se.

 > And investing in startups is pretty much useless as a hedge against
 > your exposure to lack of innovation.

So what?  I'm not arguing against prediction markets; I rather like
them.  What I'm arguing is that they do not provide significant
incentives for innovation.

 > Why am I selling claims here?

*For the money.*  The claim is that prediction markets can be used to
incentivize, ie fund, R&D.

 > An investor looks at my qualifications and proposed
 > Web 3.0 research, and funds my lab and takes the "yes"
 > side of the contract.  You invent Web 3.0 ahead of me,
 > and my investor still wins.

Yes, but *you* get *less funding*.  I haven't done the math to know
how likely the scenario is, but in the plausible scenario that the
investor wants to bet on Web 3.0 but only has the opportunity to
invest in individual firms, appearance of a pure Web 3.0 play will
result in a complete transfer of investment from your project to a
hole in the ground (as you see it).