Economics

[Bozo_the_Clown]

From [Marc Abrams (2006.06.22.0842)]

I have always found Bill P's ideas about economics a bit odd and skewed.

He talks about how "inefficient" our economic system is and my first thought is; compared to what? What else do you have in mind?

It is obvious that Bill does not understand that 'capitalism" is actually a consumer driven economy. The consumers decide what gets produced and what survives in the marketplace by their purchasing decisions they make and don't make. "Prices" are set not by a single company or individual but by supply and demand.

When you buy your house for $100,000 and can sell it two yearsl ater for $500,000, are you "price gouging"? Or are you getting what the market will allow? In selling your house how would you expect to be able to buy a new or better one if you had to sell for only $150,000 while the prices of new houses were $500,000?

Price controls don't work if you can't control the costs as well. California experienced blackouts not because there was a "shortage", but because the prices were set arbitrarily by government and no one was willing to supply the energy at those prices (i.e. by losing money)

When I was involved in the micro-computer marketplace in the early '80's we had a standing joke about Texas Instruments and its computer entry the Ti-88. They lost money on every unit sold but they would make it up in volume.

Bill has also had a very odd take on entrenpenuers. Those are the folks who bring innovation to the market. That is, they are the ones who either invent something new, or reconfigure something old in the hopes of gaining interest from folks who they believe would be willing to purchase it. What is important to understand about this is that failure is just as important as success. It is failure that redirects the use of resources and man power from unproductive (i.e. unwanted goods) to wanted goods. As anyone who has tried to bring a new product to market (like PCT) knows, it is often quite difficult to do.

Of course the problem here is that none of this happens instantaneously. It takes time and money to see if a product will or will not succeed and of course the uncertainty this all brings. A new product innovation can wipe out entire industries and that always creates havoc even if it ultimately is for the good. Anyone interested in an 8-track tape deck? How about a Wang word processor? You get the picture.

The major prblem with capitalism from a control standpoint is the uncertainty involved. Nothing is gaurenteed to take hold and certainly nothing is guarenteed to last forever. So dealing with the enevitable change through retraining is extremely important and of course no one wants to be that person who's job becomes obsolete. But discarding a system that has brought so much wealth to so many, seems like throwing out the baby with the bath water because it is not quite "perfect".

You have often scolded people who come on to CSGnet and want to "change" PCT before they even understand it. Why do you see your position vis a vie economics any differently?

Yes, there are crooked business men, just as there are crooked politicians and crooked everything else, but again, do you throw out the baby with the bath water?

Regards,

Marc

···

________________________________________________________________________
Check out AOL.com today. Breaking news, video search, pictures, email and IM. All on demand. Always Free.

[Tracy_Harms1]

(2006.06.22.09ish Pacific)

To follow up on the comments made by Marc
Abrams(2006.06.22.0842), I want to emphasize a
difference that seems to not be so well appreciated in
this forum on the topic of economics as it is on
control systems.

Economics is best viewed as a science, an inquiry into
how-things-are. Insofar as economics is a science it
has no normative implications. Accurate insights of
economics do not, by themselves, imply actions to be
taken in the market, nor in changes to government
policy. This should be a familiar insight, as control
system theory does not tell us what to do with
air-conditioning systems or caged pigeons, physics
does not tell us whether or not to build bombs,
biological theory does not include claims about
sending weponized anthrax through the mail, and so
forth.

As interested, curious, and ambitious individuals we
naturally will extend our ideas beyond the austere
range of scientific inquiry. This is a good thing;
I'm all for noticing problems and working at solving
them. As we involve ourseves in questions of "should"
and "ought," what science can do is help us keep
grounded in what "is," and what "can be."

Economics, for example, can inform us of the
effectiveness of the price system. This has exactly
the implication Marc mentioned: If somebody proposes
doing away with market pricing, we must ask "What else
do you have in mind?" What is to substitute for it,
or what is to compensate for the consequences of its
elimination? The inability to answer those questions
adequately has plainly sunk hard socialism, which has
no intellectual credibility in this regard. Its
normative proposals run in contradiction to scientific
knowledge.

I'm not saying that Bill, or anybody else here, has
advocated the wholesale elimination of market pricing.
I will, however, say that I have encountered
proposals in this forum that are not only contrary to
economic knowledge, but also confuse matters of fact
with matters of value. Perhaps it is easier to fall
into such confusion when the topics involve economics?

Tracy Harms

···

__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com

[Bozo_the_Clown]

From [Marc Abrams (2006.06.22.1317)]

Tracy Harms (2006.06.22.09ish Pacific)

Very well said Tracy.

Economics, for example, can inform us of the
effectiveness of the price system. This has exactly
the implication Marc mentioned: If somebody proposes
doing away with market pricing, we must ask "What else
do you have in mind?" What is to substitute for it,
or what is to compensate for the consequences of its
elimination? The inability to answer those questions
adequately has plainly sunk hard socialism, which has
no intellectual credibility in this regard. Its
normative proposals run in contradiction to scientific
knowledge.

I might add the notion of "competition" to the list along with market pricing. Competition provides us with our choices, reduces costs, and improves the products and services we receive by forcing providers to give the most for the least. Of course this does not hold for government sanctioned monopolies. Yes, competition can often be very inefficient over the short term, unpleasant, and nasty to some, but as Tracy ask's; "what are the alternatives?"

I will, however, say that I have encountered
proposals in this forum that are not only contrary to
economic knowledge, but also confuse matters of fact
with matters of value. Perhaps it is easier to fall
into such confusion when the topics involve economics?

An extremely important point.

Regards,

Marc

···

________________________________________________________________________
Check out AOL.com today. Breaking news, video search, pictures, email and IM. All on demand. Always Free.

[Bill_Powers1]

[From Bill Powers (2006.06.22.1104 MDT)]

Tracy (long time no see!) Harms (2006.06.22) --

Economics is best viewed as a science, an inquiry into
how-things-are.

In principle, yes. But I am far from the first to point out that economists are extraordinarily reluctant to look at how things are, relying almost completely on abstractions and unproven (indeed, untestable) theories and assumptions. When someone points out that according to the historical record, economic growth is unrelated to capital investment, the outrage is palpable, and in fact some economists use that finding to "prove" that the available data are unreliable. The same goes for the clear evidence that raising the prime rate is normally followed by an increase in inflation. Present-day economics is more like a religion or a political system than a science, certain basic assumptions being so untouchable that they are used as the basis for deciding what evidence will be accepted. If the facts are at variance with theory. the facts are simply "reinterpreted" to make them come out right -- assuming there is any reference to them at all.

Probably the most pertinent example for us control theorists was a report in Samuelson's text (I forget what edition) about an "anomalous" behavior on the part of African laborers. They did not want to work a full eight-hour day, so in the attempt to entice them the contractor rewarded any improvement by raising their pay. The result was that they went home even earlier, having earned all the money they wanted. Instead of viewing this behavior as a possible reason for revising the model of economic man, Samuelson relegated it to a footnote as a curiosity, like a zebra with horizontal stripes, and forgot about it. Economic man does not "want" things. Supply creates demand, everyone knows that. Show 'em a Sears catalog and they'll work their heads off to get all those goodies (a "fact" once proposed by a member of the CSG. He didn't actually say "Show those jigaboos a Sears catalog," but the smirk said it for him).

The way to test economic theories is with models which are constructed according to the existing assumptions and then run so their behavior can be compared with what actually happens. There is a movement called "agent-based economics" in which a few small steps in that direction have been taken, but nothing else has come to my attention. I proposed a simple model some years ago, but it was not taken up or as far as I could see understood by anyone.
By the way, in this model several "supply and demand" phenomena were clearly visible, simply as outgrowths of the underlying control processes and not because they were assumed in the beginning. There was nothing magical about them, or even basic. They did not offer any guarantees of optimum performance, or even survivable performance. It all depended on other things in the system. The "dead hand" is not an active agent -- it's a consequence of other relationships, and not always a beneficial one.

Best,

Bill P.

···

  Insofar as economics is a science it
has no normative implications. Accurate insights of
economics do not, by themselves, imply actions to be
taken in the market, nor in changes to government
policy. This should be a familiar insight, as control
system theory does not tell us what to do with
air-conditioning systems or caged pigeons, physics
does not tell us whether or not to build bombs,
biological theory does not include claims about
sending weponized anthrax through the mail, and so
forth.

As interested, curious, and ambitious individuals we
naturally will extend our ideas beyond the austere
range of scientific inquiry. This is a good thing;
I'm all for noticing problems and working at solving
them. As we involve ourseves in questions of "should"
and "ought," what science can do is help us keep
grounded in what "is," and what "can be."

Economics, for example, can inform us of the
effectiveness of the price system. This has exactly
the implication Marc mentioned: If somebody proposes
doing away with market pricing, we must ask "What else
do you have in mind?" What is to substitute for it,
or what is to compensate for the consequences of its
elimination? The inability to answer those questions
adequately has plainly sunk hard socialism, which has
no intellectual credibility in this regard. Its
normative proposals run in contradiction to scientific
knowledge.

I'm not saying that Bill, or anybody else here, has
advocated the wholesale elimination of market pricing.
I will, however, say that I have encountered
proposals in this forum that are not only contrary to
economic knowledge, but also confuse matters of fact
with matters of value. Perhaps it is easier to fall
into such confusion when the topics involve economics?

Tracy Harms

__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com

--
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.9.2/372 - Release Date: 6/21/2006

[Acree_Michael]

[From Mike Acree (2006.06.22.1701 PDT)]

Rick Marken (2006.06.22.1645)--

I think you understood Bill's point about theory vs. data, but not mine
about data. You take the subject matter of psychology to be individuals
and the subject matter of economics to be aggregates. My point was only
that both of those premises are contested. Conventional (research)
psychologists take their subject matter to be aggregates--all their
statistical models are modeling averages of hypothetical populations;
and some economists insist on the methodological individualism of PCT.
They would accord macroeconomics the same status that PCTers accord
statistical research in psychology. So it's not a trivial matter to be
"guided by the data," when what counts as data is itself heavily
theory-dependent.

Mike

[Acree_Michael]

[From Mike Acree (2006.06.22.1748 PDT)]

Rick Marken (2006.06.22.1740)--

Do economists who insist on methodological individualism ignore

aggregate data?

In the same way that PCTers "ignore" the aggregate data of psychological
research.

Would such an economist not count variables like GNP, dGNP/dt,

aggregate aggregate

capital investment and the relationships between them as data?

She or he would not.

If not, would such an
economist also eschew making statements about aggregate phenomena, like

the relationship >between dGNP/dt and aggregate capital investment, for
example?

Yes.

Mike

[Richard_Marken2]

[From Rick Marken (2006.06.22.1802)]

Mike Acree (2006.06.22.1748 PDT)

Rick Marken (2006.06.22.1740)--

Do economists who insist on methodological individualism ignore
aggregate data?

In the same way that PCTers "ignore" the aggregate data of psychological
research.

I don't ignore it. I think much of it is very good policy research.

And just ignoring aggregate data doesn't guarantee that you're going to get things right, really, does it? B. F. Skinner studied only individual behavior and look where it got him!

Would such an economist not count variables like GNP, dGNP/dt,
aggregate aggregate capital investment and the relationships between
them as data?

She or he would not.

They must not be very much fun in economic policy debates;-)

If not, would such an
economist also eschew making statements about aggregate phenomena, like
the relationship between dGNP/dt and aggregate capital investment, for
example?

Yes.

So what is such an economist good for? They can't consult on policy. And unless they're using the TCV and control modeling they haven't learned much about individual human nature. So what so they do?

Best

Rick

···

----

Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

[Bozo_the_Clown]

From [Marc Abrams (2006.06.22.2148)]

[From Rick Marken (2006.06.22.1802)]

  > So what is such an economist good for? They can't consult on policy. And unless they're using the TCV and control modeling they >haven't learned much about individual human nature. So what so they do?

Good questions, and you should have taken Tracy's post seriously. You are not the only one asking them.

First, the notion of "macro" vs. "micro" economics is a relatively new invention. The divergence originated with Keynes and frankly has been a huge failure. That is, Keynesian economics. Samuelson was a disciple of Keynes. Keynes was a big believer in government intervention and became the most important economist for FDR.

Unfortunately his policies of tariff's and trade restrictions just deepened and widened the depression which did not really end until the end of the Second World War and our econmoy took off after all the regulations were abandoned and the four years of pent up buying.
let loose.

FDR was not the great economic savior everyone makes him out to be.

Keynes like Watson in psychology thought that economics could be made into a "science" by mathematizing it. Unfortunately the very foundation of economics is based on some very flimsy ideas about human rationality.

Bill Williams understood this and was in the process of putting this all in a book. Unfortunately most folks thought Bill was sime kind of eccentric and poh=poohed his ideas. Even his supposed best friends.

Bill understood what Nobel Lauterate Simon did not understand and that is we are controllers. We do not have "bounded" rationality and we do not satisfice or optimize, _WE CONTROL_.

Mises and the Austrian school of economics comes closest to understanding all this but even they do not fully understand the importance of control and the significance it places on rethinking our notions of rationality. In economics this is known as the "expected Utility Theory" and is the very foundation of economic thought

I am currently working on this issue (rationality, or the lack of it) with several modelers in the Psych Chapter of the SD Society. If anyone is interested in this kind and type of modeling let me know and you are more than welcome to join the effort. No membership in the SD Society is required.

Of course the Chapter is open to all discussion about modeling psychological phenomenon but I am heading up a specific project with control in mind.

It has started slowly but I have two fine modelers who are extremely interested and I'm hoping this initial work will get the interest going for others, at least that is my hope. I'm not however holding my breath.

As Mike ACree has pointed out so well, macro economics faces the same aggregated problems that most psychologfical research does and when you start taking a hard look at things like the GNP you realize how utterly maningless these indexes are.

But none of this is news. I've been saying this same thing for years but it seems Marken believes that Macro economics (whatever it is) has some actual value in policy making.

As Tracy has pointed out, when used normatively the use of macroeconomics has caused more problems and hardship than it it has heloed. Economics and people work just fine without a whole lot of interference from the government or anyone else.

Regards,

Marc

Regards,

Marc

Best

Rick

···

----

Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

________________________________________________________________________
Check out AOL.com today. Breaking news, video search, pictures, email and IM. All on demand. Always Free.

[Richard_Marken2]

[From Rick Marken (2006.06.22.2130)]

Marc Abrams (2006.06.22.2148)--

Rick Marken (2006.06.22.1802)--

So what is such an economist good for? They can't consult on policy. And unless they're using the TCV and control modeling they >haven't learned much about individual human nature. So what so they do?

Keynes was a big believer in government intervention and became the most important economist for FDR.

Unfortunately his policies of tariff's and trade restrictions just deepened and widened the depression which did not really end until the end of the Second World War and our econmoy took off after all the regulations were abandoned and the four years of pent up buying. let loose.

On what basis do you come to this conclusion. You must be basing it on something other than macroeconomic measures like GNP, which are, as you say below, "meaningless".

As Mike Acree has pointed out so well, macro economics faces the same aggregated problems that most psychologfical research does

Really? Can you describe just one of those problems and explain how it is the same for both macro economics and psychology?

I believe that aggregate data is a problem only if you are trying to study individuals. So the use of aggregate data is a "problem" for psychologists who are trying to understand individuals. But I don't see it as a problem for sociologists, economists and other social scientists who are trying to understand groups rather than individuals. Seems like a good way to study aggregates is by looking at aggregate data.

and when you start taking a hard look at things like the GNP you realize how utterly maningless these indexes are.

OK, GNP is meaningless. Then on what basis do you conclude that FDR's policies did not move us out of a depression? Indeed, how do you know that there even was a recession? After all, at an individual level, some people were doing just fine in 1929-33. There was no depression for them.

Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

[Richard_Marken2]

[From Rick Marken (2006.06.24.0800)]

Marc Abrams (2006.06.23.1443)

Rick Marken (2006.06.23.1015)

What is it in the historical record that is the basis of your conclusion? If it is just what other people have said, then what is the basis of what they said? In other words, what _data_ did they use as the basis for the conclusion that 1) there was a depression that started in 1929 and 2) that FDRs policies deepened and widened it.

The GNP does not tell the full story.

GNP doesn't tell any story if, as you have said, it is meaningless.

You need to understand the monetary policies, trade polocies, and other government interventions both before and after that magical year of 1929 in order to understand why what happened, happened.

What happened? You said GNP and other aggregate measures of economic performance are "meaningless". So what is it you are trying to explain?

In order to make this dialog a bit more comprehensible to me I'd appreciate it if you would answer two simple questions: How do you know that there was a depression in 1929? How do you know that FDR's policies widened it? That is, what _data_ is the basis for making these claims?

Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

[rsmarken]

[From Rick Marken (2008.06.07.0900)]

Boy, those Republicans sure know how to run an economy, eh?

Lower taxes, put us in debt, have an unnecessary and criminal war and
continue to lower taxes so we go even further into debt, make the
dollar worthless, drive energy costs through the roof, don't invest in
communal capital projects (universal healthcare, education,
transportation systems) that actually promote growth and development,
focus on short term gains and top it off by scoffing at the economies
of Western Europe that actually do these things so that they now have
a much higher quality of life the we do. Heck of a job, Bushies;-)

I realized, however, the things will eventually get better because
people are control systems and they will _eventually_ reorganize and
start acting in ways that produce results that work better for
themselves. Things eventually will get better. Gosh, we're actually in
a position where we might get a thoughtful, intelligent, decent person
as President (Obama, of course). But it's too bad that so much
unnecessary pain, waste and destruction has had to occur in the
meantime. Ah, well.

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com

[Richard_Marken2]

[From Rick Marken (2000.08.28.1400)]

Kenny Kitzke (2000.08.26.1200EDT)--

Your comments about Clinton and harangue by Republicans
strike me as inane as your attempt to explain how the
real economy works based on your model. Sorry.

No need to apologize. It would be nice, however, if you
could tell me _why_ you think my comments are inane. What,
for example, is inane about my comments about how the real
economy might work? That is, what is wrong with my model?

I was actually planning to give you (and the rest of
CSGNet) more (unintended) inanity by commenting on an
economics column that appeared in the LA Times yesterday.
So maybe you can just tell me what's inane about my economic
comments that follow.

···

On the front page of the Opinion section of the Times, Walter Russell Mead had a column in which he discussed the current economic boom. He argued that the current boom is really just a return to normal growth levels that is the result of a return to normal levels of productivity. Mead says:

"An economy's long term growth depends on two factors: growth
in the work force and growth in productivity... Productivity
is driving us back to those long-term growth levels [of the
past]."

This caught my attention because it struck me that productivity
(above some minimal level) should actually have _no effect_
on growth in a closed loop economy. My reasoning was as
follows: if productivity (size of output per size of workforce
= Q/N) increases then there is more GNP (PQ) being produced by
a smaller proportion of the total population. But whatever is
produced must be bought with the money used to pay for production
of PQ. So no matter how efficient production of Q is (in terms
of amount of goods and services produced per workforce size) it
is impossible to produce more Q than can be bought.

I tested this idea using my spreadsheet model. In my model,
productivity is the gain of the system that produces Q as
the means of controlling P'Q' (which is purchased goods and
services). As expected, changes in productivity (above some
minimal level) had no effect on the rate of economic
growth produced by the model. Below this minimum level, the
composite controller is unable to keep P'Q' growing at the
reference rate. If productivity becomes too _high_ there is
marked instability that can lead to catastrophic oscillation
(and eventually a runaway condition) where the composite
controller loses control of the controlled variable (P'Q').
So, if the closed loop model is correct, productivity can
be too low (in which case people are not able to produce
what they can consume) _and_ too high (in which case people
are producing more than they can consume -- ie. more than
they can pay themselves back for producing).

By the way, this lack of effect of productivity on the amount
of growth in the economy is equivalent to the lack of effect
of output gain on the size of the output produced by a control
system: doubling the output sensitivity of a control system
(output/error) does _not_ double the amount of output; it just
decreases the size of the error driving the output. Economists
(like psychologists) have to learn to think in circles -- causal
circles, that is.

Mead also suggests that increased productivity is responsible
for the fact that the economic expansion has occurred without
inflation. In Mead's words "...rising productivity is holding
inflation in check despite high rates of growth". In fact,
productivity (gain of the P'Q' control loop) has absolutely
no effect on the inflation rate that occurs in a closed-loop
economic model.

Mead's conclusions about the effect of productivity on economic
growth (dGNP/dt) and inflation seem to be based on economic
lore. The presumed effects of productivity are apparently what
_would_ be expected if the economy worked as economists assume
it works: as an open loop system. There is certainly no clear
evidence that productivity drives economic growth. Even if there
is an observed association between measures of productivity
and growth (and inflation) rate it would not be convincing
evidence of a _causal_ influence of productivity on growth
(and/or inflation) rate without a model to explain how the
influence works.

Perhaps Mead's conclusions are based on applying micro-
economic thinking to macroeconomics. At the microeconomic
level -- say at the level of a company that produces
product X -- increases in productivity will definitely
lead to increases in the growth of the company (if the
market is there for product X). It will also lead to
decreases in the cost of producing the product so the
selling price of product X can remain constant or even
decrease. So, at the micro level, productivity does
lead to growth without price inflation. But this is not
true at the macro level. A company can be treated as an
input-output component at the mirco level but it must
be seen as part of a closed loop system at the macro level.

I believe that macroeconomics has failed because it has
used the behavior of micro level economic entities, like
individuals, families and companies, as models of macro
level economic phenomena, like economic growth and inflation.
This is why behavioral psychology has failed as well; the
failure of psychology has resulted, in part, from taking
the behavior of micro level behavioral entities, like
neurons and hormones, as models of macro level behavioral
phenomena, like operant behavior and reflexes.

Best

Rick
--
Richard S. Marken Phone or Fax: 310 474-0313
MindReadings.com mailto: marken@mindreadings.com
www.mindreadings.com

[Bruce_Gregory]

[From Bruce Gregory (2000.0829.1246)]

Rick Marken (2000.08.28.1400)

This caught my attention because it struck me that productivity
(above some minimal level) should actually have _no effect_
on growth in a closed loop economy. My reasoning was as
follows: if productivity (size of output per size of workforce
= Q/N) increases then there is more GNP (PQ) being produced by
a smaller proportion of the total population. But whatever is
produced must be bought with the money used to pay for production
of PQ. So no matter how efficient production of Q is (in terms
of amount of goods and services produced per workforce size) it
is impossible to produce more Q than can be bought.

That's bothered me for some time. Think of an economy with only two
workers. A growth in productivity makes it possible for all the output
to be produced by one worker. The second worker is laid off. Half the
purchasing power in the economy has disappeared. Unless you double the
salary of the first worker, you have simply lowered aggregate demand by
50%. That doesn't sound like growth to me.

BG

[Richard_Marken2]

[From Rick Marken (2000.08.29.1000)]

Bruce Gregory (2000.0829.1246)--

That's bothered me for some time. Think of an economy with only two
workers.

Exactly!! If the economy produced only one commodity that everyone
consumed the closed loop problem would be obvious. But an economy
is made of a zillion producers; productivity benefits each individual
producer but it's (nearly) irrelevant to the economy as a whole; ergo
the problem of applying micro economic facts (productivity produces
growth) to the macro economy (where productivity can't possibly
produce growth).

Thanks. Good example!

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
MindReadings.com mailto: marken@mindreadings.com
www.mindreadings.com

[Bruce_Gregory]

[From Bruce Gregory (2000.0830.1024)]

Bill Powers (2000.08.30.0324 MDT)

Rick Marken (2000.08.29.1000)--

Rick, could you walk us slowly through the details of your model?

I would like to second that request. I promise to pay close attention.

BG

[Richard_Marken2]

[From Rick Marken (2000.08.30.0820)]

Bill Powers (2000.08.30.0324 MDT)--

Rick, could you walk us slowly through the details
of your model?

Bruce Gregory (2000.0830.1024)--

I would like to second that request.

I will try to write something up ASAP. But I do have
a few things to do here at work (productivity matters
to me if not to the macro economy;-)). I would like to
politely suggest, however, that the CSG meeting (to attend
which I used my goddamn vacation time!) would have been a
good place to have asked me to do this (walk people
through the details of my model). What the heck is the
meeting for anyway? Oh, I remember. Teaching people that
they should defer to the authority of famous economists.

I'll be Bach.

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
MindReadings.com mailto: marken@mindreadings.com
www.mindreadings.com

[Richard_Marken2]

[From Rick Marken (2000.08.30.2210)]

Ok. Here's some details of my control model of the economy.

First, let me say that my goal was to implement a working
version of TCP's circular flow model of the economy. This
required making some assumptions about how things actually
worked. For example, TCP's analysis says that the composite
producer's income (from the composite consumer) must equal
it's costs. When there is leakage, TCP assumed that this
equalization happened by a process he called "autoinflation".
But TCP didn't explain how auto inflation might work, nor
did he explain how the circular flow "knows" when enough
money has been added (via autoinflation) to keep income equal
to costs. It sounded to me like TCP assumed that there is
a control process going on so my model includes a (composite)
control system that controls for zero difference between
the composite producer's income (P'Q') and costs (PQ') by
increasing prices (autoinflation).

The point of saying the above is simply to point out that
TCP's circular flow analysis of the economy (as described
in "Leakage") is missing a description of some of the
mechanisms that make the analytical equations "work", so
to speak. What I have done is try to design a working
model that provides a mechanism to explain how those
equations could work. For example, my model explains how
the equalization via autoinflation process described by
TCP's equations might work; my model also describes a
mechanism that could produce a ratio of actual to potential
production (Q'/Q) as a function of leakage that is exactly
what is predicted by TCP's equations.

Now for the details of my model (which I dubbed H. Economicus
at the meeting). The model features two control loops. One
loop (which I call the composite manager) controls the
difference between composite producer costs (PQ', which is
GNP) and income (P'Q', which is the amount _spent_ on Q';
it is _not_ necessarily the same as GNI). The reference input
to this loop is a constant, zero. The composite manager keeps
PQ'-P'Q' equal to the reference value (zero) by varying P'
(the cost of goods to the consumer). The model assumes that
the composite consumer can pay P'Q' back to itself (as composite
producer). But the model doesn't say where the money comes from
when there is leakage; that is a hole in my model but it's a
hole in TCP's analysis as well. TCP never says where the money
comes from that makes up for leakage. In my model, if money
does not come in to replace leakage (if there is no
autoinflation) then production of Q' (by anothr loop) goes down;
if money never comes in to replace the lost money Q' goes to
zero. Actually, this could be a parameter of the model; the
ease with which money can be borrowed to replace money lost
to leakage; if money gets too hard to borrow (replace) there
is recession or depression.

The second control loop (which I call the composite controller)
controls P'Q' by varying Q'. P'Q' is a strange variable; it's
the (possibly marked up) cost of Q' to the consumer; but since
it is an input variable, I think of it as the _stuff_ the
composite producer/consumer (composite controller) wants to
consume (Q') perceived in terms of it's cost. The reference
input to this system is, then, the reference for the amount
of stuff (perceived in terms of dollar cost) desired by the
composite controller. This reference is constantly _increasing_.
This increase presumably results from an increase in population
(as the number of people increases the total amount of stuff
desired increases) and, perhaps, an increase in living standards.
This reference for stuff corresponds to TCPs economic growth
rate (g) which is the product of the derivative of what TCP calls
national productivity (z=PQ/N) and the rate of increase in
population (N). TCP assumes that g drives growth of stuff (Q).
On p. 101 he give the equation as:

(dQ/dt) = (g-alpha)Q' (2-30)

I couldn't see how to put this open loop expression for growth
in Q' into my model; Q' was already part two control loops. So I
assumed that g is the rate of change in the reference for Q' as
perceived (in terms of money); ie. P'Q'. I also took alpha
(leakage) out of the equation that determines the reference
for P'Q' because it brings the effect of leakage on growth into
the model deus ex machina.

The rate of growth (g) in the reference for stuff (P'Q') is, thus,
the rate of growth of the composite controllers' desire for the
stuff produced by the economy. This rate of growth in P'Q' will
be produced to the extent that the composite controller has
enough gain to produce the increases in Q' needed to maintain
this rate of growth. In other words, the composite controller
must have enough gain to bring P'Q' to the ever increasing reference
for stuff. This gain of the composite controller system is what
I think of as _poroductivity_.

The gain of the composite controller system converts the
difference between the desired and the actual amount stuff into
output, Q'. This difference is the _error_ in the system. This
error can be thought of as being proportional to the size of the
workforce devoted to producing Q'. When error is large, not enough
stuff (as perceived) is being produced so more people (error) are
employed to produce Q'. As the error decreases, fewer people are
need to work to produce Q'. When the gain of this system is high,
it always takes less error (a smaller workforce) to produce a
given amount of Q'. So when gain is high, productivity is high; a
small workforce (error) can produce enough Q' to keep P'Q'
equal to the increasing referecne for P'Q'.

In my model, growth of the economy (growth of Q') is mainly
determined by the change in the reference for stuff (P'Q') as
perceived. There is also an effect on growth when leakage
changes. But when leakage is constant, the rate of growth of
the economy follows changes in the reference for stuff,
regardless of the size of the leakage. Productivity, measured
as output of Q' per input of error (workforce), obviously
has no effect on growth of Q', at least if the gain if
productivity is above a certain minimum vakue.

As I said at the meeting, my control model, as currently
constructed, reproduces some of TCPs more important analytical
findings, ie. the effect of leakage on inflation and relative
productivity (Q/Q'). It does not reproduce TCPs prediction
regarding the effect of leakage on growth rate (equation 2-30
above) but it could if leakage were simple inserted into the
equation that determine the rate of change in the reference for
P'Q' in the composite controller.

I hope this helps.

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: marken@mindreadings.com
mindreadings.com

[Bruce_Gregory]

[From Bruce Gregory (2000.0831.0700)]

Here's an example of increased producivity leading to growth in a closed
system. Imagine an economy where everybody is a subsistence farmer. No money
is involved, only barter. If every farmer can feed her family by only
working half as much as she used to, she can use her new found leisure to
make pots, baskets, and palm computers to trade with other farmers. Clearly
the "wealth" of the economy has increased.

BG

[Richard_Marken2]

[From Rick Marken (2000.08.31.0800)]

Bruce Gregory (2000.0831.0700)

Here's an example of increased producivity leading to
growth in a closed system...

Yes. I'm beginning to think that I made a mistake by turning
TCP's intrinsic growth rate into a reference signal. It may be
that the economy is open loop with respect to the production
of Q. The economy just keeps producing as much Q as it can;
productivity increases would then have a direct effect on
the rate of growth of Q. What is closed loop is the system
that pays for production of Q; the cost of producing Q must
be continuously returned to the composite producer by the
composite consumer.

I'll try changing the model during this labor day weekend
(how appropriate).

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
MindReadings.com mailto: marken@mindreadings.com
www.mindreadings.com

[Richard_Marken2]

[From Rick Marken (2000.08.31.0930)]

Bruce Gregory (2000.0831.0700)

Here's an example of increased producivity leading to
growth in a closed system...

Me:

Yes. I'm beginning to think that I made a mistake by turning
TCP's intrinsic growth rate into a reference signal.

I take that back. I tried an "open loop" version of the model
and quickly learned that leakage does not have the expected
effect on Q/Q' (relative productivity), of course. In fact, it
has _no effect_ on Q/Q'. I don't know how to let production
of Q be open loop and still have the model work correctly. But
as I was thinking about this, I realized that production of
Q can't possibly be open loop. I think my model is moving in
the right direction; there must be a reference for Q. It
can't be open loop.

I think your example of increased productivity leading to growth
leaves out one important consideration. Let's go with your
imaginary economy where everyone is a subsistence farmer.
Subsistence means that the farmers are producing _exactly_
the amount of Q they want.

Now say productivity increases so that "every farmer can feed
her family by only working half as much as she used to". All
this means is that half the labor is needed to produce the
desired Q. But then you say "she can use her new found leisure
to make pots, baskets, and palm computers to trade with
other farmers". So productivity has increased the size of Q. But
you are also assuming that the increase in Q was _wanted_. That
is, you are now assuming that the Q produced by the subsistence
farmers was _less_ than what they wanted; the farmers actually
wanted _more_ Q than was being produced by subsistence farming.
So when productivity (and Q) increases, the new Q is absorbed
into the circular flow of PQ (where P is 1 because it's a barter
economy).

But suppose that productivity grew to the point where the
subsistence farmers can produce thousands of "pots, baskets,
and palm computers" for every person in the economy (including
themselves). Obviously, there is not going to be a market for
all this stuff because people don't want or need it. In other
words, Q can get _too big_ in the sense that it can exceed the
wants and needs (the reference for goods an services) of the
composite consumer.

In my model, there is an explicit reference for Q (actually, for
PQ since this is a model of a money based economy). The reference
makes sense to me for the reasons given above; people have limits
to how much they want. At the composite level this reference for
PQ grows because the number of people in the economy grows; it also
grows because people _learn_ to want more and higher quality stuff.
People in the US want desktop computers now because they have
learned to want them -- they have learned what these commodities
can do for them. I think this process of "learning to want better
stuff" is an important contribution of education to the growth of
GNP in western societies. So, according to my model, education
(learning what to want and how to use it) drives growth, _not_
productivity.

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
MindReadings.com mailto: marken@mindreadings.com
www.mindreadings.com