[Election-Methods] utility theory lesson for a very confused rob brown

daniel radetsky dradetsky at gmail.com
Thu Jan 10 15:31:45 PST 2008


On Jan 10, 2008 1:47 PM, CLAY SHENTRUP <clay at electopia.org> wrote:

> I do not think the fear of risk argument works here, because utility and
> money are not the same thing.  Utility is more like log(money).


I don't know why you're bringing up money here. I didn't.


>  the tendency of people to choose the former suggests a psychological
> component that is commonly thought of as risk aversion.  But this is better
> explained by understanding the relationship between money and utility.


You're missing the point. The decision to take the sure thing is in part due
to the risk-aversion, and in part due to value due to how the initial state
of the person making the decision affects his valuation of the various
payoffs. The behavioral evidence from which you propose to infer a persons
valuation of a prospect does not as far as I can see provide a way for you
to determine how much of his decision was due to valuation of the particular
payoff vs. his valuation of risk.


> There's no reason for a person to prefer less expected utility out of fear
> of "risk".


I'm not sure what the scare quotes here are supposed to indicate about the
notion of risk.

In any case, do you believe that purchasing insurance is irrational? That is
what the above statement implies.


> So it's not really the randomness that gives people negative utility.
> It's the actual negative-ness of a potential outcome that's the detractor.
> Understanding this, risk aversion as understood in monetary systems make no
> sense in utility theory.


I don't see how I'm committed to an understanding of risk aversion where
people are averse to risk simpliciter rather than risk of some particular
thing. Risk implies the risk of something, otherwise it would be called
"randomness aversion," and it isn't. If you think I am, please explain.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20080110/763e64df/attachment-0003.htm>


More information about the Election-Methods mailing list