Economic Blues

[Richard_Marken2]

[From Rick Marken (2003.02.04.1420)]

Bill Powers (2003.02.04.1051 MST)

I consider TCP's main contributions to have been his clear vision of the
difference between composite entities and individuals, which economists
have clearly not seen (in my limited experience), and his attempts to work
up a system of realistic equations and interpret them in terms of available
data. He had the basic concept of system analysis and the inclinations of a
research scientists to back it up. But he did not know about simulation,
nor did he have any concept of a human agency being important in the
workings of the economy.

I agree with this appraisal of TCP's contributions completely. My _H. economicus_
paper (in _More Mind Readings_) describes my (very) modest efforts at implementing
TCP's model as a simulation. Doing this required adding some "composite" human
agency in the form of control systems that control for balancing the books and
consuming a reference amount of goods and services. It was pretty simple stuff but
it helped me understand how to approach understanding the macro economy from a
control perspective.

Best regards

Rick

···

---
Richard S. Marken, Ph.D.
Senior Behavioral Scientist
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

[Richard_Marken2]

[From Rick Marken (2003.02.06.1120)]

Bill Williams (2003.02.05)

If you assume money makes the circuit
go then it seems entirely

reasonable to attempt to follow the
money because the money is

something of a causal force

Bill Powers (2003.02.05.1629 MST)
I suppose that makes
a certain amount of sense, but from the PCT

standpoint, what makes the system
go is the fact that people want things

and make whatever efforts are necessary
to get them. If nobody wanted or

needed anything there would be no
economy. The motor is in the human

agents, not in the goods and not in
the money.
Exactly. And that is what my H. economicus
model does. It puts in the human agency that makes the circular flow of
money flow.
This seems like a nice place to try
to explain my perspective on economics. To some extent I have already done
that in my H. economicus paper, which is reprinted in More Mind Readings.
The H. economicus paper is basically the same one I presented at the
2000 CSG Meeting in Boston. At that meeting Bill Williams criticized the
work described in that paper in very general terms. I think the main criticism
was that, because I am not a real economist, I have no business sticking
my nose into the study of economics. Bill Powers has also expressed (to
me personally) his dislike of the paper after it was published in More
Mind Readings. I don’t recall exactly what Bill Powers didn’t like
about it. But I think it would be nice if both Bills would re-read the
paper and post their criticisms on CSGNet. Perhaps the paper can then serve
the useful purpose of uniting Bill Williams and Bill Powers in their
efforts to build a PCT based model of the economy. I just reread
the H. economicus paper myself and I think it’s pretty good. Not perfect
but OK. Indeed, I think it’s a nice first step in the direction of
building a PCT based economics. But if it’s not, I think it would be useful
to know why.

Anyway, here is my perspective on economics.
I had no perspective on economics before I read “Leakage”. But that book
made me realize that an economy is just control of input carried out at
the aggregate level. It’s a collection of people cooperatively controlling
inputs (goods and services) for themselves. The relation of economics to
control theory suddenly became obvious to me when I read TCP’s discussion
of the aggregate producer returning GNP to itself in the role of aggregate
consumer. I don’t think TCP thought of it this way. But it jogged me out
of my dogmatic slumbers.

After finishing “Leakage” my control
theory thinking about the economy was further jogged by my reading of the
first few chapters of Smith’s Wealth of Nations, particularly his discussion
of specialization. I realized that an economy is just a cooperative
extension of what people used to do all by themselves: control for inputs
such as food and clothing. The first people had no economy, really.
Hunting and gathering were individual control behaviors; controlling for
food and clothing. Individuals worked on their own to produce the goods
and services they wanted. Hunting, for example, is controlling. When done
by an individual, it is a person acting – by moving, hiding, waiting,
attacking, killing, butchering, and cooking – to produce the desired
input – food in the mouth.

Of course, some specialization was
probably taking place right from the start. Each member of a tribe was
probably producing things that all members needed. Some hunted, some made
cooking pots, etc. This low level of specialization with small groups
of people could probably be handled by sharing; the hunter let the pot
maker have some of the food that was cooked in the pot maker’s pot.
But as social organizations became larger and specialization became more
detailed there had to be some formal basis for sharing the fruits of specialization.
If the level of specialization is not too great teen barter works as the
basis for sharing. But eventually people developed money as the way to
share. The problem was to figure out how to distribute the money in a way
such that people who made only pots could use money to get what they needed
from the people who made only food, clothes, etc. and vice versa.
I imagine bartering for goods turned into bartering for money; the market
emerged.

The market determines how much money
the producers get for producing their specialty and how much those same
producers, as consumers, have to pay for the specialties produced by others.
I think Smith was mainly interested in how this market process worked.
My conception of the economy assumes that the market has already done it’s
work. It has determined what consumers must pay for what others have
produced and what producers get from consumers for what they produced.
Of course, in even more sophisticated economies the producers are not only
specialized but some producers specialize in making capital producers (machines,
training equipment) that increase the productivity of other producers.
But, from my perspective, what is ultimately produced is GNP; that collection
of capital and non-capital goods and services that everyone in the population
wants.

I have come to a view of the economy
as “human nature writ large”. I think of the total economy as a huge
reflection of the controlling done by that one original hunter. The
hunter controls for all his desired inputs on his own. If we smush all
the hunters inputs together and represent all his outputs as a single action
variable then the hunter can be represented as a single control loop:

–>r–>|C|—>action

  >

  v

<-----input<—disturbance (eg. prey dodging weapons)

Looking at the modern economy as a
whole and doing what we did with the hunter, smushing inputs together (as
gross consumption, GC) and action together as GNP (the gross output of
the economy), then the economy can be represented as a single control loop
too:

–>r–>|C|—>GNP

 >

 v

<------GC<—disturbance (eg. leakage)

The reference (r) for the hunter represents
the hunters collective wants; the r for the economy represents the population’s
collective wants.

Of course, a single control loop can’t
capture the behavior of the hunter and a single control loop can’t capture
the behavior of the macro economy. But a simple control model (two or three
loop) can capture some aspects of the behavior of the hunter (we could
probably model spear throwing with a couple of control loops). What I tried
to do in my H. econonicus paper is model the behavior of the economy,
as described by TCP in Leakage, using a simple control model, just a couple
of loops, if possible.

TCP described the economy as a circular
flow of money. But he was also, I believe, describing the economy as a
system controlling two major variables: gross consumption (GC), which is
the portion of GNP that is actually consumed, and “balanced books”, which
is keeping the amount spent to produce GNP (PQ’) equal to the amount received
for purchase of that portion of GNP that is purchased (P’Q’). I was able
to model TCP’s “circular flow” as the result of the operation of two aggregate
control systems, one controlling for consumption of PQ’ and one controlling
for keeping PQ’- P’Q’ = 0. The model included disturbances to both controlled
variables, with “leakage” being the main disturbance to control of consumption.

The model controlled for balanced books
(PQ’-P’Q’=0) by charging more for the products sold than it cost to produce
them, resulting in auto inflation. This control mode works as long as money
can flow into the system to make up for the money lost from leakage. In
the model, this just happens. The model doesn’t say how money is produced
or lost. But this is a feature that could be added to the model.

Anyway, the only goal of the model
was to introduce a working control system model of the behavior of the
macro economy. I know it needs work but I still think it’s a reasonable
start. But if people disagree, substantive criticism is most welcome.
As I said before, the model exists as a spreadsheet program which I am
happy to send to anyone who wants to critique the model in detail (unfortunately
the spreadsheet model is not well documented but it might be a fun challenge
for those who want to get into spreadsheet modeling).

Best regards

Rick

···

–

Richard S. Marken, Ph.D.

Senior Behavioral Scientist

The RAND Corporation

PO Box 2138

1700 Main Street

Santa Monica, CA 90407-2138

Tel: 310-393-0411 x7971

Fax: 310-451-7018

E-mail: rmarken@rand.org

[Richard_Marken2]

[From Rick Marken (2003.02.06.2200)]

Bill Powers (2003.02.06.1456 MST)

Rick Marken (2003.02.06.1120)--

>The _H. economicus_ paper is basically the same one I presented at

the

2000 CSG Meeting in Boston. ... Bill Powers has also expressed (to me

personally) his dislike of the paper after it was published in _More

Mind

Readings_. I don't recall exactly what Bill Powers didn't like about

it.

But I think it would be nice if both Bills would re-read the paper and

post their criticisms on CSGNet.

I'm not sure you really want that.

Looks like I got it anyway. But I appreciate it.

I assumed that this was a very
introductory paper, not to be taken too seriously since it left so

much out

and was pretty undeveloped.

I'd say that's basically right. It wasn't so much introductory as an
effort of limited scope. It was simply an attempt to simulate the agency
needed to maintain the circular flow in the face of leakage, as
described analytically by TCP.

While it's possible to make a model too
detailed, it's also possible to make it so general and abstract that

it's

incomprehensible. Figure 2 is more of a conceptual diagram than a

picture

of a working system -- nobody could figure out what your model is from

that

diagram.

OK. I imagine the diagram could be better.

And just who or what is capable of perceiving PQ' - P'Q' is not
explained. What I didn't like about this model was mainly that I

couldn't

make sense of it. I don't think there is a composite GNP controller.

That

implies a person or group who has a goal for GNP, monitors actual GNP,

and

converts the difference into an action that can increase or decrease

GNP. I

don't think there is any such thing, group, or person. Maybe you can
convince me there is.

I guess one thing I was trying to do was model at the aggregate level,
the same level at which TCP was describing the behavior of the circular
flow. The variables whose behavior I am trying to model are aggregate
variables, like GNP. So the agents are really conceptual agents.
Individuals compare income (PQ' at the individual level) to expenses
(P'Q' at the individual level). My PQ' - P'Q' controller is a
representation of the aggregate control over their balance sheets
exerted by many individuals. Same for GNP control. The GNP controller is
a representation of an aggregation of a set of individuals, each one of
whom is controlling for an amount of goods (Q) "perceived" in terms of
their cost (P). Individuals control for having things like cars (Q) but
the model represents that car as, say, $20,000 (PQ) so that every good
and service is perceived in the same dimension: dollars.

One problem with your model is that you draw quantitative conclusions

from

it well before you have given anyone reason to believe the model is

working

properly and that its behavior is to be believed. We need to see

detailed

evidence that this is in fact a proper model and that everything going

on

inside it is plausible. The discussion is much too sketchy to support

your

conclusions.

The evidence I gave that the model is working properly (given what I
wanted the model to do) is presented in Table 1, p. 169. This table
shows that the model does what it was designed to do: produce behavior
(of Q/Q' and inflation, at least) that matches the behavior of these
variables, according to TCP's model, as a function of leakage. I'm
sure I could have done a better job of
validating the model. But the model did do what I built it to do. I
agree that the discussion might have been too sketchy. But I don't know
what detailed evidence (other than what I presented in Table 1) you
would want to prove to yourself that this is a "proper" model.

Your conclusions about the relationship of rate of growth to
leakage, for example, must be based on something you didn't talk

about,

because I see nothing in the model that can "grow." You simply put in

an

increasing reference level for GNP, which naturally makes GNP grow (or

stay

the same or decline) in whatever way you make the reference level grow

or

remain constant or decline. So of course leakage has no effect!

Neither

does anything else in your model have an effect on growth. At least

this is

how it looks to me -- have I missed something vital?

I don't think your conclusion that "of course leakage had no effect" is
correct. The increasing reference for GNP corresponds (somewhat) to
TCP's intrinsic growth rate; I think it represents the growth in demand
for GNP that results simply from population growth. I constructed the
model to mimic the behavior of TCP's model. And my model succeeds at
doing that. I thought that the effect of leakage on growth rate would
"fall out" of the model. But it didn't, and that's how I discovered that
TCP had put the effect of leakage on growth rate in as an assumption. In
fact, _change_ in leakage does affect growth rate. So the model will
allow you to find variations in leakage that would produce observed
fluctuations in growth rate. Then one could go and look for evidence
that those kinds of variations in leakage actually took place. Of
course, to do that, one would have to figure out how to measure leakage.

One big plus you
bring to the discussion is that you have some experience with Delphi,

which

I assume we'll need when array sizes go beyond what Turbo Pascal can

handle.

I really have very little experience with Delphi. And I really don't
have any time to work on that model; I'm juggling 3 projects at work and
3 other PCT related projects at home right now. And I don't think I'm
particularly good at economic modeling, anyway. So I think I'll just let
you guys proceed.

>I just reread the _H. economicus_ paper myself and I think it's

pretty

good. Not perfect but OK. Indeed, I think it's a nice first step in

the

direction of building a PCT based economics. But if it's not, I think

it

would be useful to know why.

I think it's a giant leap, but not in the right direction. You became
satisfied with it before you had done nearly enough work on it.

Approving

too much of one's own productions is risky. Better to let others do

the

judging -- another name for overly favorable self-evaluation is

bragging.

Back in grammar school, I thought the height of humor was when someone

told

someone else who was bragging, "Don't break your arm," meaning, I
discovered, " ... patting yourself on the back."

I'm not patting myself on the back. I publish this stuff to show others
what I'm doing and and to get help and suggestions. The feedback I got
on this paper before I published it -- when I presented it at the
meeting and discussed it on CSGnet -- seemed OK. I do appreciate your
criticisms, though I don't really understand some of them. It doesn't
seem like a giant leap in the wrong direction to me. The model did what
I wanted it to do (simulate the behavior of the circular flow as per
TCP), and I learned something from it (that TCP put the effect of
leakage on growth rate in as an assumption). I wasn't trying to solve
all the problems of economics. I was just trying to do what I said in
the paper: develop a mechanism that could produce the results predicted
by the circular flow analysis.

As to the rest of your post, I think it's much too early to be drawing

sweeping conclusions about economics. Of course you might get lucky

and be

able to claim that we "heard it here first," but what usually happens

when

people do this is that they end up hoping everyone has forgotten their

hasty guesses. We will do better taking it step by step and making

sure of

our ground as we go along.

Maybe you are right. I think by "step by step" you mean from individual
interactions up to the economy as a whole. Maybe that's best. I was
approaching it from the macro level as a first step. My step by step
would work down, increasing the resolution of the model as that seemed
necessary. The economic data is macro data so I'm building models at
that level. But you may be right; the best approach may be to go from
the individual level up. I do look forward to seeing the behavior of
your model and seeing how it explains fluctuations in macro variables
like Q/Q', unemployment, growth rate and inflation. I suspect, however,
that your approach may be like trying to account for tracking behavior
by modeling individual cells and their interactions with each other and
with external forces. But I really hope it is successful.

Best regards

Rick

···

---
Richard S. Marken
MindReadings.com
marken@mindreadings.com
310 474-0313

[Richard_Marken2]

[From Rick Marken (2003.02.07.1330)]

Bill Powers (2003.02.07.1037 MST)

Ah. This was very helpful. Thanks.

Rick Marken (2003.02.06.2200)--

>OK. I imagine the diagram could be better.

Then make it better. It's not much help as it stands. It is hiding errors.

I will do that. And I will try to present a better description of the model so
that you can get a better idea of what it is.

>I guess one thing I was trying to do was model at the aggregate level,
>the same level at which TCP was describing the behavior of the circular
>flow. The variables whose behavior I am trying to model are aggregate
>variables, like GNP. So the agents are really conceptual agents.

That's not an answer to my objection. If you want to see what the model
predicts about the growth of GNP, you don't make GNP a controlled variable
and give it an exponentially-rising reference level. If you do that, you'll
get out of the model precisely what you put into it, an
exponentially-rising GNP. That is exactly what TPC did: declared that
economic output depends on exponents, and then gave the impression that he
had proven that it rises exponentially. In other words, the way you're
going to prove that GNP rises exponentially is to make sure it rises
exponentially. What kind of modeling is that?

Actually, that's not quite what I did. It's true that the reference for GNP rises
at a constant rate but I wasn't trying to show that GNP rises at a constant rate.
I was trying to see if, within the context of the model, leakage would have an
effect on growth rate. It could have, just as a constant disturbance will result
in a constant offset of controlled variable from it's reference. In fact, it
turns out that d leakage/dt does have an effect on dGNP/dt in the model.

It would be like proving that
the outfielder runs along a curved path by giving him reference levels for
direction of running in X and Y that correspond to the curved path you
observe. In your baseball model, you at least show how the directions of
running are functions of error signals in some _other_ control systems than
the direction-of-running systems. The fact that they are curved _emerges_
from the operation of the model; there's nothing in the model that
specifies the curvature directly, as your rising reference level for GNP
specifies how GNP is to behave.

What bugs me is that you know better than this. You have never, to my
knowledge, made this mistake before.

Well, no wonder you're pissed. You just expect great things of me;-)

If I actually did this (putting conclusions in as assumptions) -- and I suppose
it's possible that I did -- then that certainly would be a huge mistake. I don't
_think_ I did it. I certainly don't think I came to any conclusions about the
behavior of the model after putting those conclusions into the model as
assumptions. But you never know. I might have. If so, it was the result of pure
stupidity, not nefarious intent.

Since you don't have Excel, I will try to write out the model as procedural code
and we can go over it and see if I did, indeed, put in my conclusions as
assumptions. I'm pretty sure I didn't; that's why I find your reply so
encouraging. I suspect I may be more guilty of poor communication than of stacking
the deck. But if I goofed up then I'm happy to correct the error.

>Individuals compare income (PQ' at the individual level) to expenses
>(P'Q' at the individual level). My PQ' - P'Q' controller is a
>representation of the aggregate control over their balance sheets
>exerted by many individuals.

So are they trying to make PQ' = P'Q'? That denies leakage right there; if
there is actually leakage of money from the system, it is impossible for
PQ' to equal P'Q', if I interpret those symbols correctly. Your model
contains no source of money to replace the leakage disturbance. so the
books clearly cannot balance.

Actually, money does enter the system automatically to make up for leakage. But
it does sneak in. Inventory also builds up until enough money comes back in to
make up for leakage. But I'll present the details of the model so you can judge
for yourself (and then you can explain to me what's wrong).

>The evidence I gave that the model is working properly (given what I
>wanted the model to do) is presented in Table 1, p. 169. This table
>shows that the model does what it was designed to do: produce behavior
>(of Q/Q' and inflation, at least) that matches the behavior of these
>variables, according to TCP's model, as a function of leakage.

But it does that only because you specifically made it do that. And there
has to be an error in your model, because Table 1 shows the rate of growth
increasing at a constant 13% a year, while in fact the amount of money
available for spending is decreasing at the assumed leakage rate. There is
no way that the rate of spending can be increasing according to the
indicated inflation rate while the rate of growth is remaining constant in
terms of goods purchased. The amount of goods may be increasing year by
year, but inflation is increasing faster, so there is a net loss of output
in constant dollars -- exactly as the Circular Flow model predicts.

I'll have to look at this. There is a net loss of output relative to potential
output in constant dollars. I'll have to explain what the model did.

> I'm sure I could have done a better job of
>validating the model. But the model did do what I built it to do.

So why didn't you just make GNP increase at 13% per year, so you could say
the model is correct, since GNP increases at 13% per year? Essentially
that's what you did by making the reference level increase 13% per year.

I wasn't trying to show that GNP increases at any particular rate per year. I was
looking to see if leakage would have an effect on the rate of increase in GNP.

> I
>agree that the discussion might have been too sketchy. But I don't know
>what detailed evidence (other than what I presented in Table 1) you
>would want to prove to yourself that this is a "proper" model.

Look at Econ004.pas. It's not complete and it may even be mistaken, but it
is a proper model. No behavior is simply assumed to happen and then made to
happen that way.

Again, I don't believe I did that. If I _did_ do it, then it was a result of
stupidity and I obviously have some important things to learn about modeling. But
I'll try to describe the model in more detail and then you can point out where I
did that. But I will say that I didn't put the increasing reference for GNP into
the model because I wanted to show that GNP increases at a constant rate. I put in
the increasing reference because I assumed that demand increases with population
growth. I wanted to see how the behavior of the demanded variable (GNP) would be
affected by variations in leakage. If anything, I assumed that the behavior of
the controlled variable would by affected by leakage. It was, but not in the way
described by TCP.

>I don't think your conclusion that "of course leakage had no effect" is
>correct. The increasing reference for GNP corresponds (somewhat) to
>TCP's intrinsic growth rate; I think it represents the growth in demand
>for GNP that results simply from population growth.

None of that is in your model, and your model can't really work for that
reason. Leakage has no effect in your model because you have not provided
any way for it to have an effect.

But leakage does have an effect -- on Q/Q' and inflation rate (as shown in Table
1).

> I constructed the
>model to mimic the behavior of TCP's model. And my model succeeds at
>doing that.

Except for the effect of leakage. Imitating the behavior is trivial. I can
construct a model for stopping a car by saying its velocity decreases
exponentially to zero. That duplicates the behavior of the real car. But
the whole point is how that result is achieved; the essence of the model is
in the how, not the what. Your model contains no how, or no valid how.

I _think_ you are wrong about this and I will try to prove it to you. But you may
be right. If so, that would be fine with me. I'm just trying to do it right.

> I thought that the effect of leakage on growth rate would
>"fall out" of the model. But it didn't, and that's how I discovered that
>TCP had put the effect of leakage on growth rate in as an assumption.

But you did the same thing: you assumed a constant growth rate by making
the reference level grow at a constant rate. Of course there was no way
leakage could have an effect on growth rate when you specifically set up
the model so _nothing_ could materially affect growth rate.

But d leakage/dt _does_ materially affect growth rate in the model

> In
>fact, _change_ in leakage does affect growth rate.

How could it? You show the growth rate in Fig. 1 as constant at 13% per
year.

There is a _change_ in growth rate (above and beyond that demanded by the
reference) that results from a _change_ in leakage. If leakage changes
continuously then growth rate changes continuously (with an average rate of 13%,
if that is the rate at which the reference for GNP changes).

But if your home projects
with modeling are to fare any better than the economic model, you should
consider very carefully what we're discussing here.

Well, I haven't given up on my economic model yet but of course I'll consider
carefully what we are discussing here. I want to learn. If I did what you say then
I've made a terrible mistake and I want to correct it. So if I give you a better
description of the model I hope you well help me understand what I did wrong. I'm
really not trying to put one over on anyone. This was an honest effort to build a
working model based on TCP's concept of the circular flow of goods/services and
money between aggregate producer and consumer. If I put conclusions in as
assumptions then it was a huge mistake. I don't think I did that but if I did then
please just explain it to me.

Your model works by
magic, and that is not how you have normally made models in the past. If
you don't figure out exactly where you went wrong, you could make the same
mistake again.

Great. At the moment I honestly have no idea where I went wrong (or _that_ I went
wrong). I don't think the model works by magic. I don't think I put conclusions in
as assumptions. But if I did I would like to learn about it so I don't do this
again.

Thanks

Rick

···

--
Richard S. Marken, Ph.D.
Senior Behavioral Scientist
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

[Richard_Marken2]

[From Rick Marken (2003.02.07.1350)]

Bruce Nevin (2003.0207 13:27 EST)

Rick Marken (2003.02.06.2200)--
>I suspect, however,
>that your approach may be like trying to account for tracking behavior
>by modeling individual cells and their interactions with each other and
>with external forces. But I really hope it is successful.

For this analogy to be valid, there must be a "superorganic" control system
of which humans are the "cells." Are you asserting this?

Not at all. I'm just not sure that it's feasible, in this case anyway, to model
the macro phenomenon (behavior of macroeconomic variables) by modeling the micro
phenomena of which it is presumably a result. But that's just a guess. Who
knows?

Best

Rick

···

At 01:05 AM 2/7/2003, Richard Marken wrote:

--
Richard S. Marken, Ph.D.
Senior Behavioral Scientist
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

[Richard_Marken2]

[From Rick Marken (2003.02.10.2220)]

Rick Marken (2003.02.07.1330)--

>Since you don't have Excel, I will try to write out the model as

procedural code and we can go over it and see if I did, indeed,
put in my conclusions as assumptions.

Bill Powers (2003.02.07.1917 MST)--

Fine, I'm willing.

OK. First I'll say that, after reviewing my arcane little model in some detail, I
think that what is mainly wrong with it is that one of the controllers controls
P'Q', which is GNP measured in terms of the current price per good, P'. This is
ridiculous. It means that this controller could get what it wanted if we just
raised the current price per good (P' in the model) and left the amount of goods
consumed, Q', constant. I selected P'Q' as a controlled variable for one system
because that was the variable going round the circular flow in _Leakage_. Having
the system control this variable _may_ be an example of putting my conclusion
(that constant leakage has no effect on growth rate) in as an assumption. But I
don't think so. What's wrong with the model is that one of the agents should be
controlling for Q' (the stuff), not P'Q' (the money).

The spreadsheet model works something like this. The GNP (P'Q') controller
controls for a reference amount of P'Q' (P'Q'*) by increasing (or decreasing)
production of Q':

Q' = Q'+.0001*[100*(P'Q'* - P'Q')-Q'] (1)

(I'll just use the numbers I used for gain and slowing in the model; dt is 1 so I
think of each iteration as a year. That makes the computations a tad clunkly but
the systems are stable; they control).

The cost to the producer of producing Q', PQ', is

PQ' = P*Q' (2)

Where P is the production cost per unit Q'.

P'Q' is the proportion of PQ' which is received by the composite consumer and used
to consume GNP (P'Q'):

P'Q' = PQ' - L*PQ' (3)

where L is leakage. If L = 0 then there is no leakage and all of PQ' is received
by the consumer. When there is leakage, P'Q' <PQ' and P'Q' will be less than P'Q'*
and, by equation (1) the composite GNP controller will act to increase what it
receives by increasing production of Q'.

At the same time that the GNP controller is trying to control for P'Q' by
increasing or decreasing production of Q', the composite manager is trying to keep
P'Q'= PQ' by adjusting the price of Q' to the composite GNP controller. The
output of the composite manager is

P' = P+.000001*(0-(P'Q'-PQ') (4)

When the cost of production, PQ', is greater than the amount received from the
consumer for purchase of what was produced, P'Q', the composite manager raises the
price of the goods produced. So during each iteration of the model, goods are
produced at a price per unit of P and sold at a price per unit of P'. When there
is no leakage, P' always equals P. When there is leakage, P' is slightly greater
than P. This, of course, is where money has to be pumped into the system. The
model assumes that when P is increased to P', the composite GNP controller is able
to borrow exactly enough money to make up for leakage. Note also that increasing
P (as P') affects the value of the quantity controlled by the GNP controller,
P'Q'. When there is leakage, the quantity controlled by the GNP controller is
affected both by that controller's own efforts to increase P'Q' (by increasing
production of Q') and by the composite manager's efforts to keep P'Q' = PQ' (by
inflating the price, P', of the Q' that is produced).

I can see that this is not a pretty model. But, as I said, it produces the
behavior described by TCP's equations. That is, the effect of L on Q'/Q (I
calculate Q, potential production, in a parallel GNP control loop in which P'Q' is
not affected by leakage) and on inflation rate (dP/dt) is exactly what is
predicted by TCP. L has no effect on growth rate (dP'Q'/dt), not because I can
make P'Q'* change at a particular rate, but simply because it doesn't. TCP said L
does have an effect on growth rate but that conclusion of _his_ was really an
assumption. On page 101 of _Leakage_, TCP simply assumes that growth rate is the
algebraic sum of intrinsic growth rate and leakage rate. The relationship
described in equation 2-29 on p. 101 is not a derived conclusion; it is an assumed
conclusion. I would not have noticed this if I hadn't gone through the modeling
exercise.

Best regards

Rick

···

--
Richard S. Marken
MindReadings.com
marken@mindreadings.com
310 474-0313

[Richard_Marken2]

[From Rick Marken (2003.02.11.1330)]

Bill Powers (2003.02.11.1130 MST)--

OK. These are certainly reasonable criticisms of the model. And some are not.
But I don't think it's worth arguing about the merits (or demerits) of the model.
It obviously has to be changed and I will work on that. But I do still think that
the basic idea behind the model is correct, the idea being that the aggregate
producer must continuously be paid back by the aggregate consumer the amount paid
to the aggregate consumer as the cost of producing Q. I'll just have to build a
model that captures this fact in a realistic way -- not by controlling for weird
variables like P'Q'.

Best regards

Rick

···

--
Richard S. Marken, Ph.D.
Senior Behavioral Scientist
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

[Richard_Marken2]

[From Rick Marken (2003.02.11.1630)]

Bill Powers (20034.02.11.1537 MST)

Rick Marken (2003.02.11.1330)--

> > Bill Powers (2003.02.11.1130 MST)--
>
>>... I'll just have to build a model that captures this
> fact in a realistic way -- not by controlling for weird variables like P'Q'

That would be good. I suggest creating intermediate variables: cash
reserves and inventories, for both producer and consumer...
to producer reserve. I tried to talk TCP into doing this but he saw no use
in it.

I think you'll find that I have considerably more respect for your suggestions
than did your Dad.

I'll try to work on this this weekend.

Best

Rick

···

--
Richard S. Marken, Ph.D.
Senior Behavioral Scientist
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

[Richard_Marken2]

[From Rick Marken (2003.02.03.0820)]

Rick Marken (2003.02.01.1800)

Bill Williams (UMKC 1 Feburary 2003 4:00 PM CST)--

Rick,

As I pointed out quite some time ago in Boston quoting Joan Robinson,
"Economics is a serious subject." When I read T.C. Power's manuscript
some years before it was published, it was quickly apparent that what
he had done was to repeat with some variations a version of an old
tradition in economics known as "underconsumptionism."

That's interesting. But could you explain what's wrong with the
"underconsumptionist viewpoint".

Hi Bill

I don't recall seeing your reply on this. I really would like to know what is
wrong with the "underconsumptionist viewpoint"? And what is wrong with the
notion of "leakage" to which the "underconsumptionist viewpoint" is
equivalent?

Thanks

Rick

[Bill_Williams1]

[From Bill Powers (2003.02.03.1328 MST);

Rick Marken (2003.02.03.0820)--

>I really would like to know what is

wrong with the "underconsumptionist viewpoint"? And what is wrong with the
notion of "leakage" to which the "underconsumptionist viewpoint" is
equivalent?

The short answer is that there are two categories of income. Investment goods and consumption goods. The underconsumptionist variety of theories focus on a short fall of consumption. A short fall relative to the potential production which, as an engineering matter, an economy is capable of producing. Underconsumptionist theories focus upon this short-fall, and in competition with other theories which focus upon investment decisions, _typically_ argue that if policies are managed so that there is an adaquate volume of consumption investment will take care of itself. TCP makes a point about the volume of investment being relatively stable. But, this part of his arguement is based upon using an accounting figure that measures payments made to capital accounts intended to pay-off debts ( either real or implicit ) which have in the past been taken on when capital equipment has been purchased. The proceedures within firms to make payments to these "sinking funds" are surprizingly (at least to me stable). However, the actual creation and purchse of "real" capital equipment is more variable than consumption. If you look at time series for the creation, sale, purchase of investment goods, it would appear that to account for the total level of economic activity it is necessary to explain _both_ the factors that determine the level of consumption but also the factors that determine the level of investment.

If you look at the onset of the "Great Depression" during 1932 investment fell from an index number of 120 to about 60. While consumption expenditures fell from about 110 to 100. Then in 1933 consumption increased by about 5 percent and investment rebounded from about 65 to 125. NOw, I'm not an expert on economic time series, there are people who think only about different ways to construct and criticise the merits of various indexes. But, what ever index you choose to use a consensus has been reached that investment expenditure is more variable than is consumption expenditure. (The numbers I mention above are reported by Peter Temin 1989 _Lessons from the Great Depression: The Lionell Robbins Lectures for 1989_ MIT PRess. Termin descibes the numbers as representing heavy equipment sales and department store sales. p. 100. Looking at such numbers economists typically have come to the conclusion that an explaination of changes in the level of income and employment ought to include both the determinates of investment and consumption. Further, there is a consensus that the two types of purchases are determined by quite different motives. Prior to the 1930's there was a competition between economists who thought that fluctuations in investment explained the business cycle and those who thought that consumption behavior was the source of the business cycle. During the 30's at least three theories emerged a Swedish school, a Polish, and the Keynesian theory that integrated both an explaination of investment and consumption into a single explaination. As far as I can tell the three versions while they differ in style are logically quite similiar. In the last 20 years there has emerged a Marxist version under the caption Post-Keynesian Macro Economics. The conservatives or maybe better the reactionaries ( Chicago school folks ) after complaining bitterly about the new conception finally generated a version of their own. As a MA student 1969 _Equation and Equlibrium in Marshall and Keynes_ I worked my way through this literature and came to the conclusion that it was possible to simplify matters if one defined the income relationships in terms of a system of symultaneous equations of time rates. And, then stuck to a consistent practice in which the equations were treated as equations and equals signs were consistently treated so that the left and right sides were actually equal. Strange as it may seem there has been and still is a widespread conception in econmics that it is possible to explain economic behavior in terms of dis-equilibrium conditions-- which in turn are thought of as conditions in which the equational relationship ( the equals sign ) does not hold. Thinking about it in this way, then fluctuations are explained as adjustment processes in which there are forces which eventually will drive the economy toward if not precisely into equilibrium. By attempting to think about the economic process in this way, things get very confused, because the basic structure of explaination-- the fundamental equations aren't thought of as equations in the same sense that they are most areas of science in which dynamic relationships are important. When I applied the same sort of criteria to psychological theory, it appeared to me that the "scientific" analysis of behaviorism had the same defects as the orthodox thinking in economics. Behavorists treat casuation as a sequence in time in which an external forces which is prior in time "impacts" upon a subject and the result is a "response" which takes place later in time. Behaviorists think of this as "the method of physics." But, it isn't. Properly handled force and causation in physics are continuous in time rather than sequential. John Dewey seems to have understood something about this. In his _Logic the Theory of Inquiry_ 1938 there is a chapter "Sequence and Causation" which seems to me to represent an understanding of the importance of consistency in handling time, equations, causation in a consistent manner. Kurt Lewin had an idea about field theory in psychology which may be similiar to the idea that I had. (I haven't looked at Lewin lately!) THat is an attempt to get away from the sequence notion of causation. Veblen had a similiar idea ( that equations were really equations ) which he talked about under the caption "the machine process." Veblen also had the idea or the anticipation that it might be possible to combine a description of a process as a system of simulataneous equations with an analysis of purpose. At the conclusion of a sequence of lectures at Harvard in 1900 "The Preconceptions of Economic Science" Veblen concludes by anticipating the emergence of an analysis of teleological or purposeful behavior which in no way the logical rigor characteristic of the material sciences. But, nobody has been able to figure out what VEblen was talking about. Dewey seems to have had a similiar idea when he expressed in his critical paper on the "Reflex Arc." What does this have to do with the Under-consumptionist economics you may ask? Well for one thing if there are two main kinds of decisions being made investment and consumption decisions, what makes sense to me is to proceed as if, which I'm convinced is the case, these decisions are going on simultaneously. It's not that decisions about consumption are primary and investment decisions are seconday, or vis versa. Instead _both_ are going on concurrently and they are equally important. The task then is understand the process involved in their interaction. And, that interaction can as a practical measure be understood, and only understood in a reliable way, as an analysis specified in terms of simultaneous equations of time rates. After working the question over in the MA thesis, I went at it again in a dissertation _Mathematical aspects of Veblenian Economics_ 1972. It was One of Veblen's principle arguements that economics had to be reconstructed in terms of "the life Process." Veblen argued that the orthodox economic conception of behavior in which human action was the result of an external force amounted to treating human behavior in terms of a passive substance when "the life process" human behavior included was active in character. Economic man he argued was a conception that ignored the biological locus of human behavior. My dissertation was a failure in the sense that I hadn't been able to put together Veblen's insistence upon reconstructing economics as a "life process"
with my idea that only when the theory was specified properly as a system of simultaneous equations would it be possble to keep the analysis in proper order. The consideration of the Giffen paradox appears a sentence or two in terms of how the arguements in the thesis might be extended. After completing the degree I purchased an osciloscope and set out to learn the nuts and bolts of cybernetics. Along the way I got a job designing the electronics for a control system for a soybean harvester. And, it was while I was working on the soybean harvester board that I encountered first Glasser's Stations of the mind and then Bill's BCP. And what Bill had to say about human behavior confirmed what I'd observed student pilots sometimes doing. But, I couldn't put what I understood ( intuitively ) about the Giffen paraodox with what I understood about control theory. I'd made some efforts to get into computing, but before PC's became availible what most computer science courses amounted to wasn't what I had in mind. So it wasn't untill I met up with Bill that the ideas I had about the Giffen effect became a working model. Which brings me to the current situation. None and I emphasize NONE of the contemporary schools or traditions in economics are based upon working models. I would go further and say that NONE of the existing bodies of theory in economics are based upon a self-consistent format. I could be wrong but I reached the conclusion thirty years ago that only by specifying economic theory in terms of set of differential equations of time rates would it be possible to develop a reliable foundation for economic analysis. And, it was in terms of this conclusion which I read the appendix to BCP I said to my self this is the ticket! He's set this up the proper way. I had a friend who went through BCP and found some mistakes in the derivations. However, when I looked in Bill's copy he'd written in corrections that match what my friend had found. So, I'm not such a ninny that I reject theories based upon trival or correctable errors.

However, in contrast to Bill's work when I look at what T.C. Powers has attempted to do in economics what I find is quite different. I'm not saying anything about TCP's work in structural chemistry which is distinguished and has been recognized as such. But, when TC Powers attempted to construct a novel analysis of the economic process his experience as a structural chemist didn't provide him with the sort of scholarly skills to make a judgment about where to begin. And, at the age he began his research the task of maintaining a consistent analysis seems to have been beyond him. Most extended works in mathematical economic theory contain errors. A casual attitude toward equations on the part of most economists contributes to this. Keynes for example proceeds to define the propensity to consume and the multiplier relation in such a way that definitions are only true at the conclusion of an infinite sequence of adjustments. So the definitions are violated during the entire process with which the analysis is concerned. As I remember it, when Bill Powers sat down an examined his dad's work, he found that in places A was equal to B, and A was equal to C, but B and C were not equal to each other. ANd, commenting on this difficulty said, "This is going to be difficult to explain." Its not that I have a bias against TCP's stuff that has resulted in my having rejected what he did out of hand. I looked at the manuscript and saw what I saw. I made the best judgement I could. TCP wasn't remotely inclined to listen to what I said. And, I didn't think anything further about it untill you and Bill were pushing what TCP had done as revolutionizing economic theory. Now my criteria for reasonableness in discussion has something to do with how errors are handled. I've pointed out what I percieve as mistakes such as the confusion between payments to a capital sinking fund, and expenditures upon capital equipment. It isn't true that my objection to TCP work is non-substansive in character. You can look it up. Payments to sinking funds are surprizingly stable, while capital purchases are variable.And, it is the purchases of equipment that are associated with the employment required to create the equipment. I would suggest that before, and/or if this goes any further you check this out. This isn't an original observation on my part it is a standard feature of the historical accounts of how the contemporary income/expenditure model was developed. If TCP's conception of the economy were to be corrected by taking into account the effect of capital expenditures rather than the accounting for sinking fund transfer payments, and the details were corrected so that eveything was expressed in terms of simultaneous equations, then I don't see offhand that I would have any fundametnal objections. However, it is my impression that when an attempt was made several years ago to do a line by line check of what TCP had done the conclusion which was reached was that a lot of inconsistencies were found.

So, I'm satisfied that Bill's description of the way he is constructing the Econ model is correct. Wolfgang's report that the simulations match the results obtained by the analog board are reassuring. I don't see any reason at this point to bring in conclusions reached by Bill's dad, Keynes, or french marxists for that matter. Suggestions from any source may prove helpful or not as the case may be, but the point, or the potential point can be approach from an analysis based upon first principles carefully considered.

I won't try to respond for Bill W., but my answer is from another point of
view anyhow. Neither underconsumption nor leakage is defined in terms of a
working model (though leakage would be easier to define), so we can't say
what effect either of them would have, or if they are the same thing.

In a discpline in which just about every possible explanation has been
offered for everything (with good or bad justification), it is always
possible to say that any idea is "just ****ism," fill in the blank. That
certainly holds true in psychology and philosophy. Perhaps it is also true
in economics. But nothing is "just" anything else. It's an irrelevant
criticism.

My criticism of your dad's stuff was never "it's just something or other." However, when I saw resemblences between what he'd done and the underconsumptionist work, it suggested looking at particular feataures of his work to see if errors which had been identified in the underconsumptionist literature were also present in what he done. And, the confusion between payments to sinking funds and current payments for capital equipment is a principle example.

Maybe I ought to explain the work I do here at UMKC. My principle task is to construct a control theory based economic analysis. In addition I work with graduate students. As a typical example a student a French Marxist recently ask me to review a paper she was going to give at a confrence in Lyon France.
She knows I don't believe Marxism is a productive way to approach contemporary economic problems. But she also knows that I can, despite that identify problems in a paper which is based upon assumptions very different from those that I would use. In a fifteen minute presentation I found about 12 things that I thought were mistaken in some way. And, there was an issue about whether it was politic to state one argument explicitly. As it turned out she accepted what I had to say without exception. (It doesn't always work that way.) The point being if I can work with a student who's approach is as different from mine as is a French Marxist then it seems to me implausible to think that my criticism of TCP's stuff is based upon some non-substansive bias. After all she listened to what I had to say knowing that I had fundamental objections to featurs of the tradition she's choosen to adopt. And she did so because in her opinion I wouldn't let those differences get in the way of assisting her in improving the paper.

Lets take another example. I'm working with a student, an Airforce Officer who was until recently a B2 pilot out at Whiteman. He's ABD in the program and currently serving as a miltary attache in China. He's expressed an interest in the question of cultural differences regarding risk-- in particular the Chinese conception of, and approach to risk. Now I don't plan to write his dissertation. But I'm doing my best to think of suggestions that would be helpful. There's the example of the Allais paradox which is a violation of the economic expected utility analysis. I think it may be possible to use control theory with priority levels to explain what's happening, but I haven't gotten very far myself in thinking about the issue.
But, I'm trying hard to come with stuff that he could use. You don't get to fly a B2, or serve as a military attache in China, let alone both without being an extremely capable individual.

So, I've got work here to do. The sort of work where my biggist worry is if I can do the sort of job that will contribute to making this place a successful doctorate program. I've also got the task in front of me of getting a book out. At long last I think I've got things figured out sufficiently to generate the sort of text I think is required to introduce control theory in a way that has some reasonable chance of convincing people that _this is the way to do it_. GIven the load I'm experiencing I'm not that interested in continuing an argument about whether TCP work provides the basis for the economics of the future. Bill Powers and I seem to be working together now in a productive way based upon approaching issues from a first principles standpoint. Disagreements and conflicts can be expected, but I'm not pushing Keynes, he's not pushing his dad's approach. Maybe by taking it slowly, and I agree with bill that keeping it simple is a good idea, it will be possible to get to the point where some control theory based agents can be inserted into the system. I've got a model of the Veblen-Duessenberry effect that I'd like to see ( eventually ) integrated into such a model. There's also some recent work on the "stickiness" of prices that reformulated as a control theory proposition could be explored. There's some basis for thinking that prices are more than just exchange ratios-- that is people attempt to control prices for reasons to do with the symbolic nature of prices.

I'm not happy, however, with some of the stuff Rick has said in the last few posts. THere are things he's said that aren't true. In the past I've made "substansive" objections to TCP's work, but what I've said has been brushed off for various reasons-- including personal attacks upon my character. But, I'm willing to continue a discussion about TCP's work with Rick under some conditions. I'm not interested right now in an exchange consisting of refuting accusations. So, RIck if you want to continue the discussion why don't you as a first step check on TCP arguement about the place of investment in the economy. It appears to me that what TCP says is not consistent with reports by Termin and others see Peterson 1992 _income, Employment and Growth_ p. 297. that investment is more variable than consumption?

Best,

Bill P.

···

-----Original Message-----
From: Bill Powers [mailto:powers_w@EARTHLINK.NET]
Sent: Mon 2/3/2003 2:40 PM
To: CSGNET@listserv.uiuc.edu
Cc:
Subject: Re: Economic Blues

[Bill_Williams1]

[From Rick Marken (2003.02.04.1220)]

Rick Marken (2003.02.03.0820)--

>I really would like to know what is
>wrong with the "underconsumptionist viewpoint"? And what is wrong with the
>notion of "leakage" to which the "underconsumptionist viewpoint" is
>equivalent?

"Williams, William D." wrote:

The short answer is...

Well, I'm certainly glad you didn't send the _long_ answer;-)

I'm pressed for time just now. So, I'm not responding to your post-- for the time being. But, I would like to report on a realization or sort of gestalt switch I had thinking about "leakages." For me the sequence which is characteristic of TCP's stuff or the underconsumptionists seems peculiar. Money is constantly being created and and destroyed. Governments do it by printing money and collecting taxes. People do it by declaring bankrupcy. BAnks do it when the make and collect loans. Banking is sort of a strange business, but there really isn't any mystery about it. THere's a fair amount of US currency that becomes in effect the reserve currency for other nations, and for either formal or informal "dollarization." So, there are all sorts of "leaks" and what I've term "fountains" in which money comes into existence and then disappears. Figuring out quantitative what's going on with all these different activities can be quite a job. But, its puzzled me why would one want to trace the flow of money around in terms of a circuit. But, walking to the plazz to get something to eat, I conducted a thought experiment ( deganken ? ) and attempt to trace the flow of money around in a circuit from producer to consumer and back. It was a little bit surprizing, but the result was that I came up with questions such as how to account for what happens when a consumer doesn't spend all that is recieved as wages, and what happens when a producer takes the money recieved from a consumer and pays off a loan. Now I wasn't seriously in danger of becoming trapped in an alternative world, but it did in a sense illuminate for me how TCP reached the conclusions he did. If you assume money makes the circuit go then it seems entirely reasonable to attempt to follow the money because the money is something of a causal force.

Then I decided when I'd had enough of chasing money around the circuits, to find out what I had to do to convert the gestalt I was in at that moment to the Income/expenditure model. It took me a while, but what I found was that the INcome/expenditure model isn't about a circuit. And it isn't about money as such. Instead, like its title indicates its about EXPENDITURE. What happens to money is of minor or no concern-- except in so far as what's going on has an effect upon expenditure-- expenditure upon the current production and consumption of real goods. So the resale of art is not a part of the analysis-- except inso far as such transactions might have an effect on initial sale and consumption of newly produced items. The focus of the analysis ought to start with real economic actitities and then add money to the analysis rather than starting with money and considering employment and production as an effect of money. Money isn't a causal agent. Now texts don't do a good job of making this point. They retain the circuit illustrations which are misleading, but people have taught the course for the last two centuries have made the circuit stuff a part of their course so why change now? However, if you look at a circuit analysis isn't money being treated as a causal force? If money doesn't make the wheel of circulation go around, what does? And, there's another problem with the wheel of circulation driven by money. It encourages thinking about the process in terms of a causal- sequence, rather than a causal process. Using the circuit concept you trace sequentially what happens if A and then as a result B and then C and implicitly what usually ( almost always ) happens is that A is assumed to exist in t1, and B in T2 et ceteria ... et ... When what's required is thinking about what's happening in terms of A/dt .

best

Bill Williams

···

-----Original Message-----
From: Richard Marken [mailto:marken@MINDREADINGS.COM]
Sent: Tue 2/4/2003 11:19 AM
To: CSGNET@listserv.uiuc.edu
Cc:
Subject: Re: Economic Blues

[Bill_Williams1]

[From Bill Powers (2003.02.05.1629 MST)]

I offer a number of ideas here as flat assertions; please understand they
they are all proposals, not pronouncements.

Sure.

I may have created a misleading impression in my comments about my thought experiment. One of the things I have to do, or at least one of the things I regard as a principle task is explain the control theory methodology to economists. In order to explain the advantages of control theory its helpful to understand how they think in enough detail to identify in a very convincing way where the current proceedures go wrong. I should have made it clear that most of my comments were about adopting the mindset of some concepts like "the wheel of circulation" and other stuff to see what things looked like from that standpoint. And, I did genuinely get to a point where some of what I think of as the traditional problems looked like very real issues. But, it was only an effort on my part to see if when I adopted for the moment ideas like "the wheel of circulation" I could understand better what things look like from that starting point. Sorry if reporting on this exercise created the impression I was going way too far round the bend.

You say regarding TCP that

[He] had no way to construct a model of the ideas in his head, and
the complexities are such that the only way to handle them properly is with
a model.

Everybody's been in this position.

As you will see in Econ004, there is no "flow" of money. Money
disappears from one account and pops up in another, while a quantity of
goods passes the other way according to their price. Price determines how
much money is needed to obtain a given amount of goods to fulfill the
consumer's needs and wants; wages and other distributions determine how
much money the consumer has with which to satisfy those needs and wants. Of
course any variable in such a system can be expressed in terms of all the
other variables, so we could eliminate hours worked, or wages, or money, or
goods and services, from the equations without losing anything but
intelligibility. But there is no reason not to retain all the variables
with which people actually operate; the computer doesn't care.

>If you assume money makes the circuit go then it seems entirely
reasonable to attempt >to follow the money because the money is something
of a causal force.

I was posing this as a retorical question. I don't really think of money as a causal agent.

I suppose that makes a certain amount of sense, but from the PCT
standpoint, what makes the system go is the fact that people want things
and make whatever efforts are necessary to get them. If nobody wanted or
needed anything there would be no economy. The motor is in the human
agents, not in the goods and not in the money.

Then I decided when I'd had enough of chasing money around the circuits,
to find out what I had to do to convert the gestalt I was in at that
moment to the Income/expenditure model. It took me a while, but what I
found was that the INcome/expenditure model isn't about a circuit. And it
isn't about money as such. Instead, like its title indicates its about
EXPENDITURE.

I take it that you mean the exchange of money for something non-monetary of
value. You will find that in Econ004, income and expenditures take place,
as do labor and consumption. What is an expenditure for one party is a
sale, and income, for another. What is income for one party becomes that
which is available for expenditure and thus income for someone else.

  What happens to money is of minor or no concern-- except in so far as
what's going on has an effect upon expenditure-- expenditure upon the
current production and consumption of real goods.

Let's not assign relative importances to links in a chain.

Well, the links I'm most interested in consist of employment and production.
Its employment and production that provide the means of life-- the fuel, the fiber and corn we have to have to exist.

There's no need to.

I'm not so sure about this. It seems to me that what models like the Giffen effect do, in part, is point out that economics is a kind of "life science." And, because it is a life science things like the requirement for calories matters. If you look at the neo-classical approach you won't find any place where calories enters into the analysis.

I'm not sure I understand what you intend to say in the following, so my comments may not be responsive to what you intended to say.

If you focus
on one aspect of the system at the expense (pun not intended) of another,
the picture will become distorted according to your subjective feelings
about relative importances.

Whatever my "subjective feelings" about things, including calories, I don't see that insisting at least tentatively upon treating economic questions in terms of a "life process" neccesarily "distorts" the analysis. It seems to me that rather than "distorting" the analysis to instead focus the analysis upon the questions that will be productive. Despite differences in nominclature, I'm not sure there is actually a disagreement between us regarding "first principles."

  Money isn't a causal agent. Now texts don't do a good job of making
this point. They retain the circuit illustrations which are misleading,
but people have taught the course for the last two centuries have made
the circuit stuff a part of their course so why change now? However, if
you look at a circuit analysis isn't money being treated as a causal force?

It's treated as if it had physical momentum -- once you start the flow
going, it keeps going until something stops it. I'm certainly with you on
this point.

If money doesn't make the wheel of circulation go around, what does?

Since there is no wheel of circulation, nothing is needed to make it go
around. Instead, we have some people using money as a means of making goods
come into their posession, while they and other people use labor and goods
as a means of making money come into their posession. The human agents are
making everything happen. Without them, any apparent flows would stop
instantly.

I was asking a retorical question above. We're in full aggreement about the above.

And, there's another problem with the wheel of circulation driven by
money. It encourages thinking about the process in terms of a causal-
sequence, rather than a causal process.

I agree. I double-agree, if there is such a thing. I'd even try to drop the
use of the word cause except in an informal sense.

Maybe. Except I think I'd have a hard time explaining how things work if I couldn't say "A happens because B, etc etc .... " You mayhave some alternative ways of expressing the same relationship??

At this point despite some continuing differences in modes of expression I think we are in enough aggreement to move on to considering how the program functions. In the long-run I think discussion concerned with the models behavior will smooth out questions about nominclature and useage.

best
bill Williams

I don't mean to ignore, not permanently at least issues I haven't responded to as yet in your's and Ricks posts, but I'm planning on paying more attention to stuff here for the next few days.

···

-----Original Message-----
From: Bill Powers [mailto:powers_w@EARTHLINK.NET]
Sent: Wed 2/5/2003 6:14 PM
To: CSGNET@listserv.uiuc.edu
Cc:
Subject: Re: Economic Blues

[Bill_Williams1]

[From Bill Powers (2003.02.05.1629 MST)]

I offer a number of ideas here as flat assertions; please understand they
they are all proposals, not pronouncements.

I had written out a much more extended commentary on your "not prounouncements" when something went wrong here and what I'd written went into the "great bit bucket." While what I said might have contributed to an improved understanding as a preliminary to exercising the model, it probably wasn't really neccesary. I am confident that we are in sufficient agreement about first principles. My post about my thought experiment, however, may resulted in some minor misconceptions. Such as,

we won't get far if we start with the most complex possible picture. That's a good way to discourage an orderly attempt to make sense of things.

I fully aggree, and I have difficulty understanding how you might think I would be in favor of doing anything other than starting by examining very simple relationships. My approach has been to tackle comparatively simple problems that can be completed in a week or two. You are the one who's in favor of taking on what is a more complex task. I'm not saying we shouldn't attempt it, just I don't see how you think of me as being inclined to begin by starting "with the most complex possible picture." Not a big deal, but somewhat puzzling.

You say:

And, there's another problem with the wheel of circulation driven by
money. It encourages thinking about the process in terms of a causal-
sequence, rather than a causal process.

I agree. I double-agree, if there is such a thing.

Good.

I'd even try to drop the use of the word cause except in an informal sense.

I'm not so sure about this. Hasn't the whole intent of all the arguments about behavior been to clarify, maybe without the sucess that could be hopedfor, where the causal agency is located, and how it is exercised? Hasn't the purpose of these discussion been to arrive at some clarity about the distinction between behaviorism's locating agency of a sort external to the organism, while descretion and agency are located within the organism in PCT? Maybe you have in mind adopting a more precise nominclature such as independent and dependent variables, or some such phrasing??? I'm not saying everyone who wants to understand control theory ought first become profient in the calulus of variations. But, it seems to me that control theory, _as theory_, needs to be formally defined. I'm not adaquately prepared to handle control theory in all its aspects in such terms, regardless of that I have doubts about whether "informality" is going to deliver the goods.

If I had the power to build a curiculm from the ground up I think I would teach people about control theory intially by an extended course with op-amps and actual control circuits-- with as little use of language as possible. Then they could be exposed to computer simulations, and the mathematical representations. Following all that, when the behavior of control loops was understood, I'd begin by assigning names and concepts to the phenomena they had been introduced to non-verbally. Then it seems to me the non-verbal understandings would "fill-up" the linguistic symbols. Instead what we have now is a situation in which we use verbal symbols with prior extraneous associations and as a result inadvertantly apply these extraneous associations in a way that confuses our understanding of control theory. The result is, of course, a lot of confusions. But, we're to some extent stuck with the resulting problems and attempts to work free of the problems in verbal discussions can be very tiresome. But, is there any alternative?

best

bill williams

···

-----Original Message-----
From: Bill Powers [mailto:powers_w@EARTHLINK.NET]
Sent: Wed 2/5/2003 6:14 PM
To: CSGNET@listserv.uiuc.edu
Cc:
Subject: Re: Economic Blues