Nick Bostrom wrote:
>> Buying my groceries today during rush hours, I had to wait some
>> five minutes in a queue in order to pay. During this time I
>> thought of the following simple way of cutting the queues: Instead
>> of having every customer pay for his groceries, let everybody
>> activate a randomizing machine when they are about to exit. With 50%
>> chance a green light flashes, and with 50% chance a red light
>> flashes. If green light, then you just pass without paying; if
>> red light then you pay twice the amount you'd normally do.
This is an interesting suggestion, worth exploring further.
Curt Adams replies:
>It's a cute idea, but I think few would go for it, because people don't
>correctly value losses vs. gains. Most people are substantially
>risk-averse even for small amounts.
I'm not so sure this is the reason people resist this.
>The grocery stores wouldn't like it for a different set of reasons.
>Sometimes people change their minds in the checkout line (they
>don't have enough money, they notice the box is damaged, they
>decide to stick to their diets after all, etc.). If the store lets you
>opt out after the doubling, it will obviously be abused; but otherwise
>the customers will end up unhappy from time to time.
The policy could be that damaged and otherwise returned goods are
exchanged for the regular price, not the doubled or halved price.
I see other problems:
1) Why should people believe the light is really 50/50 chance of
red/green? I know there are crypto ways of doing this, but
convincing ordinary customers that a real light is immune to
corruption sounds hard.
2) What if I don't have any assets on me when I'm supposed to pay
double, or what if I run for the door? The store really doesn't
want to hire a security guard to handle these cases. This could
be dealt with by requiring people to swipe their credit card
before entering the red/green light randomizer.
3) Will current contract law will enforce damages against a person
who refuses to pay the double price, and walks away from the
grocery cart? Liquidated damages rules limit damages to the
actual damages in a particular case; has the store suffered
any actual damage here?
Robin Hanson
hanson@econ.berkeley.edu http://hanson.berkeley.edu/
RWJF Health Policy Scholar FAX: 510-643-8614
140 Warren Hall, UC Berkeley, CA 94720-7360 510-643-1884
Received on Tue Dec 1 19:27:07 1998
This archive was generated by hypermail 2.1.8 : Tue Mar 07 2006 - 14:45:30 PST