On 7/18/2010 2:09 PM, Colander, David C. wrote:
> I will start a new thread since it is not directly on Larry's question, but it is related.
>
> I do not believe that we should stop dividing up micro and macro, but I believe that they should be divided somewhat differently than they currently are in the standard courses. The reason is that they can, and should refer to different modes of analysis and cannot necessarily be integrated.
>
> I believe micro includes what we currently teach in micro and in graduate macro--which is equilibrium based micro analysis. Then there is what Samuelson described as non-equilibrium dynamic analysis, if I remember correctly--this included the multiplier-accelerator model and other dynamic models, which today would be described under the heading complex system analysis.
>
> The reason this cannot be micro based is that emergence of new properties is possible here, which could not have been deduced from any micro analysis, but which nonetheless has sufficient regularities to warrant study.
>
> After Keynes wrote the GT some economists, such as Goodwin and Strotz starting investigating along these lines but the math was too complicated for most in the field, which stuck with comparative static models with implied micro foundations. Ultimately, that led to the synthesis and the fading away of neoKeynesian economics.
>
> In terms of teaching, if in macro we teach the Keynesian type models as simple examples of the type of turbulence that can occur in non-linear dynamical systems, and give up the pretence of a well-specified equilibrium micro foundation, then we can continue to teach much of what we have in intro macro courses since they are in large part definitional and institutional.  Then in intermediate macro theory we can go into some non-linear models and introduce students to complex systems--possibly through agent based models and simulations.   In graduate macro theory, the students can really get into complex systems analysis and explore the interconnection between the two realms.
>    

David, a problem I have with this justification is that of getting a 
handle on what you mean by different "modes of analysis." If this is 
what you really mean, then you could point to a particular phenomenon 
that is being analyzed and demonstrate how the two modes that you have 
in mind differ. But this is not what you do. So it seems to me that you 
either mischaracterized your post or that you are using the term "mode 
of analysis" in a special way.

I have never taught microeconomics as a totally deductive system, 
although I have taught it as a logical system that is related to a 
totally deductive system. I have always taught about Schumpeterian 
entrepreneurship, for example, the precise consequence of which is not 
deducible. Neither are a trade cycle or bubble, precisely deducible. Yet 
they may be partly consequences of entrepreneurial speculation. under 
certain conditions. Both the Schumpeter and equilibrating entrepreneur 
are part of a microeconomics that distinguishes between (1)  the 
mechanical interaction that can be represented by a mathematical model, 
and (2) the distinctly human interaction that everyone with ordinary 
powers of perception knows characterizes real market interaction.

The multiplier-accelerator model makes certain implicit assumptions 
about behavior that do not account for either the equilibrating or the 
Schumpeterian entrepreneur role. So what kind of real economic 
interaction could Samuelson have been characterizing? Which raises the 
question not whether micro and macro should be divided up but whether 
macro, as you characterize it, is economics -- i.e., about real economic 
interaction.

It occurs to me also that you may want to base the distinction on the 
mathematical distinction between either equilibrium foundations and 
non-equilibrium foundations or well specified and not well specified 
foundations. I must admit that I do not understand what you mean. I 
cannot think of a good reason to teach Keynesian type models without 
using the concept of an entrepreneur. I can see some benefit in teaching 
institutional details. However to label a course that teaches about 
institutions "macroeconomics" and to include the teaching of models and 
definitions for their own sake seems either disingenuous or deceptive.

Have I misunderstood?

  --

Pat Gunning
Professor of Economics
Groton, Connecticut
http://www.nomadpress.com/gunning/welcome.htm