From: hal@finney.org
Date: Thu Nov 02 2000 - 11:12:04 MST
Robin writes:
> I don't think it makes sense to talk about side effects of changing
> discount rates, because that isn't a specific policy which can have
> such effects. You could talk about the side effects of changing
> interest rates, or rates of taxation on capital, or subsidizing
> education, etc. But social discount rates are just a feature of
> social preferences - and preferences don't have side effects, actions
> do.
But the authors do make policy suggestions. In fact I take this to be
the main point of the paper. The abstract concludes, "Our analysis has
immediate implications for public policy: agents discount the future
too much and the government should promote future oriented policies."
They write, in section 4.1,
"Relative to the market evaluation, our analysis tips the scales in
favor of policies with short run costs and long run benefits. In the
area of fiscal policy, it favors subsidies to capital accumulation.
In the area of monetary policy, it favors low inflation. In the area
of natural resource extraction, it favors conservation. In general,
it favors investment and saving at the expense of consumption."
My complaint is that the authors have not shown that these suggestions
will achieve their goal of better satisfying people's preferences.
Making these changes may cause disruptions to the economy that their
analysis did not consider. And this could result in social preferences
being less satisfied than under the status quo.
The burden of proof ought to be on those who propose policy changes
to show that they will achieve their goal. Just because people in the
future might wish they had saved more in the past, that doesn't imply that
making people save more in the past will automatically satisfy more of
the preferences of people in the future. It's like my earlier analogy:
people would prefer to pay less for gasoline, but lowering gasoline
prices will not automatically make their preferences be better satisfied.
Hal
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:31:49 MST