Tuesday, September 23, 2003
The Summer 2003 Journal of Economic Perspectives arrived yesterday
with an article about the research of Berkeley economist Matthew Rabin.
Among many other things, Rabin has worked on models of choices that
display "dynamic inconsistency." Would you prefer $100 today or $112
one week from today? Many people would take the immediate $100.
Would you prefer $100 one year from today, or $112 one year and one
week from today? Many of those same people who liked the immediate
$100 in the first choice are willing to wait one more week for $112
when the choice concerns the same outcomes, but delayed for one year.
These people are "dynamically inconsistent". Why? Because in the second
situation, what happens after one year has passed? The choice is now the first
situation, an immediate $100 or a one-week delayed $112. But their choice
in the first situation tells us that now they don't want to wait that week
for the additional $12. There is some evidence that many people tend to
exhibit this form of dynamic inconsistency, a greater impatience displayed
in choices concerning the near future than the more distant future. In my
case, introspection also lends support to the hypothesis of dynamic
What does this have to do with vice? Dynamic inconsistency is
important with respect to potentially addictive goods. From the point of
view of the more patient version of our inconsistent consumer,
the person deciding today on whether to consume cocaine or
alcohol is excessively indulgent. And potentially addictive goods are
reinforcing, use today encourages more use tomorrow. So a person could
end up consuming large amounts of alcohol, for instance, even though
her more patient persona would consume very little, and regrets the
fix that her short-term decision-maker has got her in.
The policy implications for this sort of behavior can be extremely
significant, as Rabin pointed out in a recent paper with Ted O'Donoghue,
"Studying Optimal Paternalism, Illustrated by a Model of Sin Taxes."
Specifically, a "sin tax" on addictive goods can be very valuable for
someone who is dynamically inconsistent, while at the same time imposing
few costs on fully rational, consistent consumers. A similar approach
has been taken to cigarette taxes by MIT professor Jonathan Gruber;
Professor Kip Viscusi of Harvard Law has a different perspective.
The rationality of choices concerning potentially addictive goods, and its
implications for vice control, will undoubtedly be a topic that this blog
will return to....