Re: The Aggregate Supply Function (fwd)

["Alan G. Isaac" <aisaac@xxxxxxxxxxxx>]

On Mon, 3 Sep 2001, Paul Davidson <pdavidson@xxxxxxx> wrote:
> Or as Keynes (GT, p. 151)  noted in explaining the hourly reevaluations on
> the stock market; "In practice we have tacitly agreed, as a rule, to fall
> back on what is, in truth, a CONVENTION. The essence of this convention --
> though it does not, of course, work out quite so simply -- lies in assuming
> that the exiting state of affairs will continue indefinitely, except in so
> far as we  have specific reasons to expect a change.  This does not mean
> that we really believe that the existing state of affairs will continue
> indefinitely. We know from extensive experience that this is most unlikely".
> This description seems quite applicable to Alan's "normal times".

Yes.

> The problem that Alan does not seem to understand is that econometricans
> believe (at least they use to) that their econometric relations would be
> the normal for the entire future!  Recently with the concept of
> "hysteresis" , even that idea seems silly -- so what are we left
> with?  Unless we think something will change, we assume nothing will
> change--- does not take high-tech econometrics to "prove" that.

I simply do not meet such econometricians.
If you mean that many make use of theory (e.g., classical 'trix)
that is rigorously justified only when applied to stationary
stochastic processes, yes.  But the econometricians I chat with are
pragmatists, and it is easy enough to find discussions of using
'trix to provide ``windows'' on the data to aid understanding
and, if we are lucky, to help with policy making.

> But it is the accidents that policy is (should be) designed to prevent!
> Using "normal times" econometrics does not help us to understand what
> causes such accidents if the sample is always drawn from "normal times"--
> and if so of the data is drawn from abnormal times-- then the sample is
> being drawn from two different universes.

I agree that we should devise policy to deal with the ``accidents''.
Absolutely!  This means devising policies that are robust when our
econometric models break down. (E.g., automatic stabilers.)

But this is not the *only* thing we should be doing.

> But what is the tool for if it is usable only in "normal times". and we al
> recognize the ephemeral possibilities of normal times?

For using in normal times, of course!
Which is where we spend most of our time, after all.

Alan Isaac