[From Bill Powers (2006.08.17.0500 MDT)]
Rick Marken (2006.08.16.2145) –
In economics, in particular, I
think you have a problem modeling aggregate behavior with individual
agents not only because many different individuals are involved but also
because each individual plays different roles at different times. As I’ve
noted, for example, I am both a producer and a consumer and these roles
overlap (and alternate) in time (and in function) in complex ways. It
seems like that kind of realism would be tough to incorporate into an
individual level model. Even Bill’s Econ models are not pure individual
interaction models Some of the individual agents in these models, for
example, are “households”, which in real life often
consist of more than one individual.
I think that all economic “laws” are aggregate concepts.
Insofar as modeling an economy is possible, it has to be done in terms of
properties common to large sectors, or you’d just end up with as many
models as there are individuals, which would be useless for a social
theory.
The trick is to find properties that are sufficiently representative that
you can find ways to check the model against reality. And of course, to
avoid setting up as controlled variables things which are really emergent
from control of the actual variables. For example, it would be a mistake
to say there is a control system that creates a balance between supply
and demand. Forcing that balance to exist would conceal what really
causes it. This is why I objected to some features of your model. For
example, by making growth a controlled variable that tracks a rising
reference signal, you kept it from being something that emerged from the
variables that are actually under control. And you can’t say what the
effect on growth of other factors (like population growth or increases in
productivity) is, because if growth is under direct control, other
factors can’t affect it. If you just make growth match the observed rate
of growth, there’s no longer any way to test the model. It has to
produce the observed rate of growth – or, with equal justification, any
other rate of growth you please.
In my model I tried to think of how an actual manager or consumer would
control things. I think it’s safe to say that most consumers work in
order to make money, and they make money partly to spend on wants and
needs, and partly to provide for the future (the two goals in my model).
What’s different about consumers is how much they want and which things
they want but they all want some amount of some things and of money. So
it’s relatively safe to sketch in a model with only one good and one bank
account controlled by one consumer. That consumer represents a whole
household, all households. And we can, later, expand that one consumer
into multiple consumers with different reference levels for different
goods and different saving habits, once we get the basic model
running.
The same goes for plant managers. I realized that one thing any producer
will pay attention to is whether unsold goods are piling up or whether
production is running behind sales so inventory is shrinking. A lot of
classical economic puzzles involve what to do about inventory – the
newstand problem, for example, which involves how many newspapers the
proprietor should order. If you keep an eye on inventory, one thing you
can do to control it is manipulate prices; have sales on things that are
selling poorly, and jack up the prices of really popular items. I think
that’s exactly what is done, so that’s in my model.
If it turns out that prices have to be lowered a lot to keep inventory
from accumulating, there may not be enough income to pay for production
as well as providing for profit and other capital expenses and
distributions. The first thing that can be done is to adjust how much of
the income is distributed to investors and owners, and how much is used
to pay for production and upkeep. with wages and productivity being
constant in this incarnation of the model, only the capital distributions
are available for a manager to manipulate. My model uses that as a way of
keeping cash reserves constant. Keeping cash reserves constant is simply
a way of assuring that the business remains viable – neither piling up
undistributed income (and taking money out of circulation) nor running
out of cash to pay expenses. I think that’s very realistic, in that when
sales lag, businesses have to reduce or skip dividends and take other
economy measures that reduce capital expenses, and when sales are
booming, capital distributions are happily increased.
From the standpoint of owners and investors (and wage-earners, too), the
business is a means of creating a flow of money, not goods and services,
so there is an interaction between the amount taken out of the business
as capital income, and the amount spent to produce the goods that are
sold to create the income. That interaction happens at the same time
there is an interaction between inventory and sales based on prices, and
at the same time there is an interaction among hours worked and savings
and consumption, and between consumer income and production costs. All
these loops are in my model and they all operate at once, with the entire
system coming to equilibrium in a state that nobody could predict just
from looking at the system one part at a time (as David Friedman seems to
be preparing to do in the parts I’ve read).
It’s perfectly true that my model oversimplifies the system, but I think
that is where we have to start. When we run the model, we can see where
the oversimplifications create deviations from reality, and add just
enough complication to take care of those deviations. In this way the
model will acquire more and more detail, but no more than necessary to
make it predict correctly. At the end of this process, we hope, will be a
model that is nowhere near as complex as the real system, yet which
correctly predicts the effects of widespread policy changes and changes
in other system-wide conditions.
Of course it will never predict individual behavior, but that’s not the
point of a macroeconomic model. The macroeconomic model should tell us
things like the effect of government-imposed price or wage controls, or
the effect of raising interest rates, or the effect of outsourcing as a
widespread practice, or the effect of population growth or decline at
various rates, or the effect of collective bargaining. That’s what macro
means: system-wide properties and effects. Economists claim they know the
effects of these things, but the statistical abstracts do not bear them
out. The reaction to that is what the Austrian school recommends: ignore
observations; the data are misleading. So without data exactly how DO
they know their theories aren’t just fantasies or systematic delusions?
My conclusion is that they don’t. Economic theory is a *folie a
deux-*to-the-nth for the simple reason that economists have never used
proper system analysis.
It seems to me that the primary use of economic theory is to justify
whatever system one has already bought into. Capitalists, socialists,
communists, anarchists, fascists, all have their theories and somehow
their theories always turn out to say that what they’re doing is not only
OK, but the only rational way to do things, while all the other theories
are hogwash. They’re trying so hard to justify themselves that they have
no time simply to ask how the damn system really works. Economic theory
is a shower of random sparks from axes being ground.
The cure for that is to make a working model that operates according to
the properties you’ve openly given it, and according to clearly-stated
rules. When you’ve laid out your theory as a working model, you’re
committed to it: the model either runs as you claim it will run, or it
doesn’t. The reasons for success or failure are in the open for all to
see. That means that your own model could actually show that you were –
gasp – WRONG about what you believed would happen. Heavens to
Betsy, what a disaster. But wouldn’t it be better to know than just to
have faith?
Best,
Bill P.
···
My inclination is to think that
it’s OK to develop aggregate economic models, with aggregate entities,
like an aggregate producer/consumer, that control “virtual”
macro-economic variables . I think the validity of such models will
ultimately be tested by their ability to account for and predict
aggregate data. If an aggregate level model of the economy can predict
the behavior of economic variables as well as aggregate models of
electronic circuits (like Kirchhoff’s laws) predict electrical
variables, I think the question of their legitimacy will become moot —
maybe.
Best
Rick
Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310
729-1400
–
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.10.9/417 - Release Date:
8/11/2006