Re: Keynes and competition

[Harry Veeder <eo200@xxxxxxxxxxxxxxxxxxx>]

Alan Isaac wrote:

> On Fri, 25 May 2001, Harry Veeder wrote:
> >  I think John is saying that many economists  engage in poor analysis
> >  because they don't recognize the difference between the stuff of
> >  economic theories and the stuff of economic reality. They manipulate
> >  variables according to a specific explanatory theory without really
> >  understanding what the variables represent or whether such variables
> >  even correspond to anything in the real world.  I wish economists
> >  would stop and think about the nature of their variables.
>
> Why not read what they are writing and discover (lo and behold!)
> that they are quite aware of this problem? Then go out and try
> to do some empirical work yourself and learn something about the
> pragmatic decisions that must be made in trying to understand
> the world.

Sorry for not making my self clearer.
In this case I am being critical of the theoretician, not the empiricist.
It goes without saying that it is an on going concern of the empiricist.

> PS For example:


from _Growth Theory: An Exposition_ by Robert Solow.
> ``There are always aspects of economic life that are left out of any
> simplified model.

There is a difference between the simple and the simplified.
Data quality aside, here are three ways of categorizing of models:
1)simple models,
2)super models made from simple models.
3)simplified models which are incomplete super models
because the best model would be too complex and costly
to construct.

Some parts of economic life are simple and require simple models.
Macroeconomics is the science of simple economic relations
and assumed to be universal.

> There will therefore be problems on which it throws
> no light at all; worse yet, there may be problems on which it appears
> to throw light, but on which it actually propagates error. It is
> sometimes difficult to tell one kind of situation from the other. All
> anyone can do is to try honestly to limit the use of a parable to the
> domain in which it is not actually misleading, and that is not always
> knowable in advance.''

In matters of policy it is best to focus on the simple models.
Simplified models  should be avoided, and one should
not to get carried away by the allure of super models.

Harry Veeder