[extropy-chat] Housing bubble looking good

Eliezer S. Yudkowsky sentience at pobox.com
Fri Mar 10 00:12:57 UTC 2006

Hal Finney wrote:
> You can see how this can apply to markets.  People don't just make
> up their own minds, they adjust their opinions when they see what
> the consensus is.  In fact the other day I argued that from a certain
> point of view, rationality almost forces this.  But, the unfortunate
> thing is that if this happens too much, the market won't work well.
> If everyone just agreed with the market price, then the price would have
> no connection to reality.  So ironically, for markets to work right,
> people have to partially ignore the information they get from the market.
> Markets ultimately depend on people being irrational in this sense.
> This can provide an explanation for market bubbles and crashes, where
> people were relying too much on information from the market price itself
> and not bringing enough of their own insights and information from the
> rest of the world into the market.  The market will consolidate and
> integrate all the information available, but only if people bring that
> information to the market.
> The good thing is that the market constantly dangles that carrot out there
> of great riches, if you do have some private information that contradicts
> the market consensus.  So people still have incentives to take positions
> based on their information and move that market price in the way it
> ought to go.  The question is whether this incentive is strong enough to
> overcome the tendency to be fooled by an apparent market consensus.

If people take into account both private information and market 
information in setting their own buying and selling prices, but have (a) 
no ability to consciously untangle the difference, (b) logarithmic 
utility functions, and (c) generally poorer private information than the 
aggregate market information; then you can see how markets would be 
systematically inaccurate in the direction of "common wisdom" yet people 
would be unable to take advantage of this because of being unable to 
untangle their private opinions from the market wisdom, and hence, 
unable to detect any great divergences.  Which implies that Peak Oilers 
are necessarily irrational (a different matter from being necessarily 
wrong) because they are acting as if they know specifically that a 
market price setting is the result of irrational consensus.  And yet we 
would not be able to trust the markets to say that there would *not* be 
Peak Oil.

Eliezer S. Yudkowsky                          http://singinst.org/
Research Fellow, Singularity Institute for Artificial Intelligence

More information about the extropy-chat mailing list