From: hal@finney.org
Date: Wed Nov 01 2000 - 17:14:43 MST
Robin writes, quoting Hal:
> >Another part of the paper that I disagree with is sections 4.2 and 4.3,
> >on procrastination and addiction.
> >... discounting as typically understood cannot
> >be considered to play a significant role in these phenomena.
> >The problem is simply that the time frame is too short ...
>
> That paper is just referring to a large economic literature that does
> try to explain procrastination via non-exponential discounting.
> A common model is "hyperbolic" discounting. If you could persuasively
> show such models to be inconsistent with experiments, you could publish
> that and win fame among economists. So far such models do a reasonable
> job of accounting for a wide range of otherwise puzzling phenomena.
I will try to learn more about hyperbolic discounting. One abstract I
found at http://www.opioids.com/abstract/discount.html generally supports
the hyperbolic model, but it closes with:
Consistent with previous findings, delayed heroin rewards were
discounted at a significantly higher rate than were delayed
monetary rewards.
This conforms to my observation that in order to explain addiction via
discounting, you must invoke a significantly higher discount rate than the
kind of rates I am familiar with from monetary markets. My conclusion was
that discounting didn't fit, however I see now that another explanation
is to invoke different discount rates for different sources of happiness.
Still, if people discount different pleasures by different amounts,
the model in the social discounting paper is incomplete at best.
Hal
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:31:48 MST