Re: Keynes & New Keynesianism, Davidson's post <PB108928%UTKVM1.BITNET@vaxf.Colorado.EDU>

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

I find Davidson installment #4 problematic for several reasons
specified below. As I have no pretensions to being a Keynes
scholar, I stand ready to be corrected on my claims about Keynes.

On Fri, 21 Jan 1994 00:03:40 -0700 Paul Davidson said:
>A more "general" theory is one that is built on fewer axioms and requ=
>ires fewer
> constraints on the forms of the underlying functions than any other =
>theory
>attempting to explain the same thing.

We might want to use 'general' this way, but I do not think it
matches Keynes's use. I think Keynes saw his theory as "general"
largely because of its field of application: it encompassed the
case of involuntary unemployment as well as full employment,
as opposed to "classical" theory which only applied to the latter.
This reading I believe is supported by chapter 1 and especially
chapter 2 of The General Theory. In addition, Keynes _also_ felt
that his theory gained generality from its specific formulation,
such as dropping the second classical postulate and allowing
labor supply to depend on the nominal wage (again ch.2).

>When I speck of ad hoc restrictions it is exactly because such restri=
>ctions
>on the underlying relationships are cpnstrained to some specific form=
> or forms
>--- rather than being of a general form with no restrictions that I m=
>ean that
>these are not GNERAL THEORIES. When Rosser, Colander, etc can get Key=
>nes-type
>unemployment from a system that does not restrict the form of the rel=
>ationships
>(i.e., applicable to linear and nonlinear systems, statis and dynamic=
> system,
>flexible and inflexible prices, etc.) will they have produceda GENERA=
>L THEORY
>explaining why the system can come to rest at a less than full employ=
>ment
>Marshallian equilibrium.

I basically agree with this. We know that by playing with functional
form we can cause practically any model outcome. I have yet to be
persuaded that this is an interesting research strategy. And yet
it is possible that non-linear dynamic interactions are a crucial
determinant of observable economic behavior, so I am glad some
researchers are exploring this possibility. This tolerance is
rooted in my view that economics is more than an ideological
exercise, that we are in fact struggling to give the best possible
explanations of observable economic behavior.

>  It will then logically follow that what is r=
>equired
>is an exogenous force (governmental policy? trade policy? monetary po=
>licy?)
>to push the system back towards  full employment. But 58 years ago, K=
>eynes
>already produced such an explanation.

I largely agree: Keynes provided the first interesting theoretical
justification of demand management. (Of course, as a pragmatic
matter of policy it was already the order of the day.) Is his
theory the be-all and end-all. No, for many reasons. One reason
is that Keynes did not fully come to grips with the endogeneity
of the money supply (and other aspects of the financial system).
Another is that he did not provide an adequate model of wage
determination. Another is that he did more to show us the
limitations of existing characterizations of expectations
formation than to give us a handle on the actual processes
of expectations formation. Another is that his theory does
not shed enough light on the determinants of economic
growth. etc. So, much was left to do.

In addition, Keynes left us with behavioral descriptions in
need of modification and improvement. For example,
while I am not a promoter of microfoundations for
macroeconomics, Ando and Modigliani's work did help us
improve on Keynes consumption function, even
if it ultimately led to the absurd modern theories of
consumption behavior.

Finally, and crucially to myself and to many other Post Keynesians,
Keynes did not have enough to say about the distribution of income.

>Why do Rosser, Colander, Isaac,=
> and other
>New Keynesians feel the necessity of inventing Specific Theories in t=
>he
>hope of one day generalizing it to a GENERAL THEORY? Why not use Keyn=
>es's
>General Theory, which illustrates tht unemployment equilibrium can b=
>e establish
>ed independent of these rstrictive ad hoc constraints as long as we h=
>ave a
>monetary system used to organize production and exchange activities a=
>nd
>the future is uncertain...

To express interest in and sympathy with a research program is
not to adopt it. I doubt Rosser would characterize himself
as a New Keynesian. I believe my own research has a distinctly
Post Keynesian flavor.



[][][][][][][][][][][][][][][][][][][][][][][][][][][][][][][][][][]
[] |¢   |   |¢   |¢  |       Alan G. Isaac, Associate Professor   []
[] |-¢  |   |-¢  | ¢ |       Economics, The American University   []
[] |  ¢ |__ |  ¢ |  ¢|       Washington, DC 20016        U.S.A.   []
[] FAX:  (202)885-3790       Internet: AISAAC@xxxxxxxxxxxx        []
[] Phone:(202)885-3785       Bitnet:   AISAAC@xxxxxxxxxxxxxxxxx   []
[][][][][][][][][][][][][][][][][][][][][][][][][][][][][][][][][][]