Re: Vote on Values, Bet on Beliefs

From: Robin Hanson (
Date: Wed Aug 16 2000 - 13:13:56 MDT

Nick Bostrom wrote:
> >>Robin says (p. 29) "speculators early in the period may worry that
> >>speculators later in the period will know substantially more, and
> >>bias their estimates thereby." It's not clear to me why this would be a
> >>problem, orwhat the bias referred to would be.
> >This was intended to be a non-technical paper, ...
>But why, then, does not a similar problem arise in plain idea futures (not
>linked to decisions)? ... If there is some technical point here that I'm
>overlooking, could you please give a reference?

OK. Here is an ASCII graphic to the rescue. Let the X axis be the value
of a company's stock price P conditional on dumping the CEO, and let the Y
axis be the stock price conditional on not dumping the CEO. The diagonal
line is where these two prices are equal. Assume that everyone agrees that
the distribution over these outcomes is uniform within the marked region.

      | / The center of mass of this region is to the left of
      | XX/XX the diagonal, indicating that given this info it is
P if | XXXX/XXX better to not dump the CEO. If, however, someone who
not | XXXXX/XX knows where in this space the point is makes the
dump |XXXXXX/X decision, then if the CEO is dumped that means that
CEO |XXXXX/ the point is to the right of the line, and if the CEO
      | / is not dumped the point is to the left of the line.
      | / The center of mass of the right region is more toward
      | / right than the center of mass of the left region
      | / is toward up. Combining these coordinates gives a
      |/ point to the right of the line. Thus market
      ---------------- interpreted naively suggest dumping the CEO.
        P if dump CEO

Robin Hanson
Asst. Prof. Economics, George Mason University
MSN 1D3, Carow Hall, Fairfax VA 22030
703-993-2326 FAX: 703-993-2323

This archive was generated by hypermail 2b29 : Mon Oct 02 2000 - 17:36:08 MDT