An economic illusion

[From Rick Marken (2000.01.25.0900)]

Now that the "baseball" paper is in the publication queue I have
decided to relax for a while with my favorite reading material:
the Statistical Index os the US;-) I've been working on extending
TCP's analysis of the US economy (as described in his book
"Leakage", http://www.benchpress.com/Leakage1.htm) which included
data that went only to 1988. I have now included data to 1998
and there is one interesting surprise. To understand the surprise
you have to know something about the findings reported in "Leakage".

The main finding of interest here is that for the last hundred
years (at least) capital investment (I) in the US (corporate and
government) has been almost exactly 20% of GNP; that is I/GNP
is basically a constant: 0.2. In fact, the lowest I/GNP ratio
reported in "Leakage" occurred in 1951 when I/GNP was 0.18.

TCP reports I/GNP ratios (as percentages) through 1987 (the
latest values available to him from the Statistical Index at
the time he was writing "Leakage", in the early 1990s). Here
are the I/GNP ratios (as percentages) from 1988-1998:

1988 19.4
1989 18.6
1990 17.3
1991 16.5
1992 16.0
1993 16.5
1994 17.8
1995 17.7
1996 17.8
1997 19.6
1998 19.4

The values for 1995-1998 will probably be revised slightly, but
not by much. What's interesting is that starting in 1990 we have
the lowest levels of capital investment (as a proportion of GNP)
ever seen in this country. We seem to have recovered by 1998. But
the numbers for the early 90s are so low that I checked and rechecked
my calculations. I thought I might be getting the wrong data from
the Index. But I'm now pretty confident that the data is correct;
there was a huge decline in capital investment starting in 1990 and
continuing for at least five years.

I think this decline might be a reflection of the banks becoming
very conservative in the early 1990s about loaning money for
any capital investment given all the bad loans and bank failures
at the end of the 1980s. Anyway, these data create an interesting
illusion. If one looked only at the relationship between I/GNP and
average rate of economic growth (dGNP/dt) over this period one would
conclude that capital investment does, indeed, drive economic growth.
Here are the growth date (dGNP/dt) numbers for the 10 years (1988-98)
corresponding to the 10 years of I/GNP shown above:

3.3
1.2
-0.9
2.6
2.3
3.3
2.2
3.3
3.8
3.7

(Note the recession of 1991 = growth rate a negative .9).

The correlation between capital investment (I/GNP) and growth
rate (dGNP/dt) over this 10 year period is .61, not huge but pretty
good. If one looks at the correlation between capital investment at
year t and growth at year t+1 the correlation goes up to .72. So it
looks like, over this 10 year period, capital investment is strongly
related to economic growth, just as conventional economists have
always believed. Indeed, conventional economists think of capital
investment as one of the main independent variables in the economy;
increase capital investment and you increase growth.

But the fact is that the period 1988-1998 is an aberration; if one
looks at the relationship between capital investment and growth over
the entire last half of the 20th century (1951-1998) one finds
that the correlation is -.01 (virtually zero); the lagged correlation
(capital investment at year t with growth at year t+1) goes up to
-.09 (still virtually zero).

In fact, the variable that controls economic growth -- the main
independent variable in the economy -- is leakage (unspent GNP).
I/GNP was correlated with growth from 1988-1998 simply because
the availability of capital was correlated with the Fed's monetary
policies during that period. The illusion is similar to the
"behavior illusion" in psychology and I think it illustrates the
importance of understanding data in terms of working models.

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates mailto: rmarken@earthlink.net
http://home.earthlink.net/~rmarken

[From Phil Runkel on 2000 Jan 25 at lunch time in Oregon]

Responding to Rick Marken's of 2000.01.25.0900.

1. Thanks very much for the additional data on the GNP and cap invest.
    And by the way, could somebody explain to me in one sentence or less
    the difference between GNP and GDP? And in another sentence or less
    what is being done about adjusting either of those statistics for loss
    of natural resources? I know I should go look it up myself, but I
    don't want to take time from my book, and I'll be grateful to anybody
    who can yank the info out of his or her head and type it here.
    Thanks.

2. You said your baseball paper was "in the publication queue." Does
    that mean actually accepted? If so, good, good!

--Phil R.

[From Norman Hovda (2000.01.25.1420 MST)]

At 12:56 Philip wrote about Re: An economic illusion on 25 Jan 00,

1. Thanks very much for the additional data on the GNP and cap invest.
    And by the way, could somebody explain to me in one sentence or less
    the difference between GNP and GDP?

None. GDP is newer usage of the former GNP.

nth

[From Norman Hovda (2000.01.25.1425 MST)]

At 12:56 Philip wrote about Re: An economic illusion on 25 Jan 00,

And in another sentence or less
    what is being done about adjusting either of those statistics for loss
    of natural resources?

Gain or "loss" depends on the reference level of the beholder. <g>

Lies, damn lies and...
nth

[From Rick Marken (2000.01.25.2140)]

Phil Runkel (2000 Jan 25 at lunch time in Oregon) --

And by the way, could somebody explain to me in one sentence
or less the difference between GNP and GDP?

The difference is small, both numerically and in terms of
definition: GNP is the dollar value of all goods and services
produced by US firms or citizens _wherever they may be_;
GDP is the dollar value of all goods and services produced by
by US _or_ foreign firms and citizens _living in the US_. So GDP
is stuff produced in the US by anyone; GNP is stuff produced
anywhere by US people.

And in another sentence or less what is being done about
adjusting either of those statistics for loss of natural
resources?

Not much. I don't think the statistics can really address
this problem; people will have to address it eventually.

You said your baseball paper was "in the publication queue."
Does that mean actually accepted?

Yes, indeed! And I hope it is in print by the time the CSG
meeting rolls around in July.

Best

Rick

···

---
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/

[From Mike Acree (2000.01.26.0847 PST)]

Bruce Nevin (2000.01.26.2140 EST)

This comment is from my nephew after a very superficial exposure to
_Leakage_:

............................. begin quote .............................
It still doesn't make sense to me. First, what I really objected to about
the Leakage book was that it suggested that people shouldn't save money.
Obviously, it's important for people to save money in order to protect
themselves and plan for their retirement. It would be very irresponsible
for anyone to advise individuals not to save money, even if doing so
slightly lowers the economy's growth rate, because it would be disastrous
for anyone who took that advice. Second, money that is saved isn't
"unspent." In fact it is _invested_. People deposit money in banks, and
banks either loan it out to other people or invest it in some other way.
So, far from being "leakage", it is pumped straight back into the economy.
.............................. end quote ..............................

I have only read part of the book, and haven't understood it well enough
to
explain it to him. Can anyone give me some help?

        These criticisms, especially the second, were made by a number of
people on the Net 3 years ago. Bill and Rick defended the concept of
leakage, but I didn't have the sense that anyone else was convinced on this
point.

        Mike

[From Rick Marken (2000.01.26.1020)]

Quick note: I've discovered (with the help of Gary Cziko) that
the year 2000 archives of CSGNet posts _is_ (and has been) available
at the usual place, ftp://postoffice.cso.uiuc.edu/csgnet/
They are just at the top of the page (starting with csgnet.log001a).
This is my first experience with a minor, but annoying, Y2K bug.

Also, I still need to know who is coming to CSG 2000. If you are on
CSGNet or reading CSGNet posts and know you are coming to the
Boston conference (and have not yet informed me) please send me an
e-mail (rmarken@earthlink.net) ASAP. Thanks!

Mike Acree (2000.01.26.0847 PST) to Bruce Nevin (2000.01.26.2140 EST)

These criticisms [of the economic model described n "Leakage"],
especially the second, were made by a number of people on the
Net 3 years ago. Bill and Rick defended the concept of leakage,
but I didn't have the sense that anyone else was convinced on
this point.

You might want to take a look at the paper I presented at the 1999
CSG conference ("Economics: The cooperative control of perception")
which is at http://home.earthlink.net/~rmarken/ConfEcon.html

The paper was cobbled together in some haste but I think it gives
a reasonably good, high-level description of TCP's view of the
economy and it's amazing consistency with WTP's view of human
nature.

I think it would really be great if those who don't agree with TCP's
model of the macro economy would describe their alternative.

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates mailto: rmarken@earthlink.net
http://home.earthlink.net/~rmarken

[From Bill Powers (2000.01.26.1404 MST)]

Bruce Nevin (2000.01.26.2140 EST)--

(Nephew speaking:)

It still doesn't make sense to me. First, what I really objected to about
the Leakage book was that it suggested that people shouldn't save money.
Obviously, it's important for people to save money in order to protect
themselves and plan for their retirement.

I think we can take this as a given. People prudently save money for their
old age (if they are in the small group of people who are rich enough to
have any savings), and when they are old, they spend it again. What TCP
realized, however, is that while some people are saving, in their youth,
other people are withdrawing and spending, because they are old. The
_composite_ consumer, at any given moment, consists of people of all ages,
and to a first approximation the rate of saving equals the rate of
withdrawing in the dollar values of the moment. The NET saving rate has to
be much closer to zero than one might guess just by looking at a "typical
family." Many economists who follow Keynes (and probably others) apparently
make exactly this mistake, thinking that the whole economy is just like a
scaled-up family. They forget that when you consider all people together,
they are at all stages of the economic life cycle at the same time. From
the national perspective, the rate at which some people are saving is
largely offset by the rate at which other people are spending their savings.

Over the past hundred years, TCP found from the public record, the
difference between total income and total expenditures, across business and
individual affairs, has averaged around 7 per cent of the GNP. Economists
have taken this difference (without proof, as far as I know) to be the
total savings rate. But when you consider how few people have any savings
at all, and when you realize that among those who do have savings, there
are always roughly as many people drawing down their savings as building
them up, the idea that the _average_ whole-population savings rate could be
as high as 7 percent per annum becomes utterly impossible to believe. The
true _net average_ savings rate must be much closer to zero than to 7 percent.

If that is the case, then what is this 7 percent difference between
earnings and expenditures? TCP says it is leakage: money that disappears
from the ideal circular flow and represents a loss of buying power from the
economy as a whole. Some of this loss is permanent, some is reversible.
Since the 1960s most of the loss has apparently been permanent. People who
save money early in their lives and then spend it all before they die cause
reversible leakage: by the time they spend it all and die, the reversal is
complete. Those who spend their money in ways that do not eventually return
it to the economy (for example, by creating bad debts which are then
written off, or by building up huge fortunes that are never spent from one
generation to the next) cause permanent leakage.

When people who have extremely large incomes save money, they remove a
considerable amount of buying power from the circular flow. If they spent
it all, the money they spent would be income to other people, who would in
turn spend it, and thus keep the circular flow in balance. But people in
the top 5 or 10 percent of the income range find it essentially impossible
to spend all they earn, and anyway many of them want to build up family
fortunes so their heirs will also be rich and powerful. Since these people
control probably half of the money that potentially circulates, they don't
have to hold back much on their expenditures to create a lot of leakage --
to move a lot of buying power out of circulation. This inevitably creates
inflation and unemployment, while slowing down the growth of the economy,
as TCP's model clearly shows.

Money that is invested does indeed get spent, and does not create leakage.
It is essentially a gift from the individual saver to the business that
borrows it (bankruptcy makes it a gift). However, that money is no longer
available for the giver to spend on the products of the company in which he
invested (remember that when we speak of _composite_ entities, there is
really only one consumer and one producer). So we have this paradox, which
baffled Keynes and still baffles many economists, even those who will not
admit to being baffled (they are more baffled than the rest). How can
investment be said to produce economic growth, when the money that is
invested is not available for buying the increased production? What most
people seem to forget is that ALL of the money people receive from the
system is given to them by the producers of goods and services, either as
wages or as capital income. ALL of the money. And that is the only money
that is available for the purchase of goods and services. If the producer
wants to sell everything that is produced, he must distribute as wages and
capital income exactly enough money to buy it all at the prices he has set,
which determine his income. Otherwise he will have more goods than
customers with money to buy them, or he will sell all he makes but his
costs of production will exceed his income from sales. In the _composite_
picture, equilibrium is the only viable state. And that can be achieved
only when costs equal income.

Economists have believed in magic for a very long time. Somehow money can
create more money without inflation nullifying the gain; somehow some
people can win while nobody else loses. Money that is borrowed magically
multiplies itself so the borrower can pay back the loan with interest and
still make a profit. In the aggregate or composite picture, this is all
total nonsense, of course, but the system is set up in such a complex way
that the self-contradictions are buried many layers down; furthermore,
those who benefit the most from the legerdemain are busy with their shovels
burying the truth even deeper.

All that really produces economic growth is human ingenuity. People have
ideas about how to make life easier, and put them into practice. As a
result, we either work less to maintain the same quality of life, or we
work the same amount to increase the quality of life. That's what it says
here, in the Handbook of Life. But if that's true, why do so many people
feel they are working harder and longer than ever and suffering a continual
decline in the quality of life? The answer is: because a few people are
working a LOT less and making a LOT more money than before. And this is
caused by the economic system under which we live, a system that is run by
people who don't understand it and, indeed, whose interests are best served
if _nobody_ understands it. Money cannot make money; you can only get more
money by making sure that someone else gets less money. The circular flow
is real; it can't be denied. That is what TCP was talking about. He didn't
believe in magic any more than I, his son, do.

Best,

Bill P.

[From Bruce Nevin (2000.01.26.1832 EST)]

Bill Powers (2000.01.26.1404 MST)]

Thank you, Bill, that is very helpful indeed! I have passed it on to Doug
(Douglas Singsen, my wife's brother Bill's boy, at Haverford), somewhat
embarrassed that I had not understood this fundamental property of the
model (aggregate entities) well enough to explain the misunderstanding when
it came up at Christmas time. (I had given _Leakage_ to Bill and to another
brother, Gerry, and Doug was looking at his father's copy. I gave MSOB to
the third brother.)

        Bruce Nevin

[From Bill Powers (2000.01.27.0503 MST)]

Bruce Nevin (2000.01.26.1832 EST)--

Thank you, Bill, that is very helpful indeed! I have passed it on to Doug
(Douglas Singsen, my wife's brother Bill's boy, at Haverford), somewhat
embarrassed that I had not understood this fundamental property of the
model (aggregate entities) well enough to explain the misunderstanding when
it came up at Christmas time.

The problem with the book _Leakage_ is that it was written, in the end, by
a man in his 90s who could no longer remember what he had written on the
previous day. It was only through tremendous effort and continuous
re-reading of his own words that he was able to complete it. All the work
that led up to it was done when he was in his 70s and early 80s when his
mind was still functioning well, but when the time came to produce the
final drafts of the book, he was failing and he knew it. Thus the quality
of the ideas in this book far exceeds his ability to express them toward
the end. The repetitions you find in the book are simply TCP trying to
maintain his own train of thought as the span of his memory shrank.

Understanding what he produced requires not so much a sympathetic reading
as a patient one. The basic concepts are all there, and in places are
expressed with his normal brilliance. His argument is basically factual --
what actually has happened in the American economy as opposed to what
economic theoreticians require to have happened in order for their theories
and beliefs to hold up. His basic model is simple -- a few equations that
express some simple relationships of kinds with which every physical
scientist is familiar. But its implications are profound, and in the end I
think they will prove unassailable. Economics has been a matter of ideology
and bluster; TCP has shown how it can become a science.

Five to ten years back, I wrote some letters to the Clinton administration,
recommending that they check up on Greenspan's claims about the effects of
meddling with the money supply on inflation (the effect of tightening the
money supply as shown in the statistical record has been to slow economic
growth, increase unemployment, and increase inflation, and TCP shows why
this has to be the effect). For a while afterward, Greenspan stopped
tightening money every time he imagined that inflation was about to occur,
and during that time, it seems to me, the economy started to take off. Did
my letters about Dad's theory actually reach the right eyes, and did
someone demand that Greenspan prove that his interventions actually helped?
That is all I said in my letters: look at the historical record, and see
what the effects of restricting the money supply actually have been. Maybe
they did. Somebody had better remind them again, because Greenspan is
trying to sell his bear-scarer again.

Best,

Bill P.

[From Bruce Nevin (2000.01.27.1045 EST)]

I see that my nephew's misunderstanding of _Leakage_ is akin to the
misapplication of group statistics to individuals.

        Bruce Nevin

[From Rick Marken (2000.01.27.1000)]

Bill Powers (2000.01.27.0503 MST)--

Five to ten years back, I wrote some letters to the Clinton
administration, recommending that they check up on Greenspan's
claims about the effects of meddling with the money supply on
inflation...For a while afterward, Greenspan stopped tightening
money every time he imagined that inflation was about to occur,

You must have written to them right near the start of the
administration (92). Here are the leakage (alpha), growth (dGNP/dt)
and estimates of Fed money tightening policies (rho) from 1988
(the last year of TCP's data) through 1998. It looks like Greenspan
came to his senses as soon as Clinton became president (1992). I
guess he was rebounding from the recession he created in 1991 (when
the rate of Fed-caused reduction in growth -- rho -- was over 8%).
The Fed has pretty much behaved itself since 1992. I believe
that the Fed did start tightening up again in 1998 -- and this
shows up as an increase in rho in 1998.

year alpha (%) growth(%) rho (%)

88 4.2 2.9 5.92
89 4.8 3.3 4.91
90 5.6 1.2 6.12
91 5.8 -0.9 8.12
92 6.4 2.6 3.96
93 5.8 2.3 4.97
94 5.4 3.3 4.24
95 6.0 2.2 4.80
96 5.5 3.3 4.20
97 5.1 3.8 4.07
98 3.9 3.7 5.35

Note, by the way, that the alpha leakage levels are fairly
low during this period (always less than 7%). However, alpha
leakage (rather than the Fed's monetary policies) seems to
be responsible for the sudden slowdown in growth in 95.
Alpha leakage is otherwise steadily declining from 92 to 98.
The Statistical Index shows that most of the decline in alpha
leakage is due to a steady, precipitous decline in what is
called "personal savings". This decline is somewhat offset
by a steady increase in "undistributed corporate profits".
I bet both are explained by the movement of unspent GNP
(alpha leakage) into the stock market.

By the way, now that I know how the Fed works (sort of),
I think it's possible to get an estimate of actual savings
levels (the amount of money people have on deposit in
banks) by looking at the Fed's data on "Aggregate Reserves
of Depository Institutions". These reserves (adjusted for
changes in reserve requirements) should be proportional
to savings (the required reserve rate being the
proportionality factor).

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates mailto: rmarken@earthlink.net
http://home.earthlink.net/~rmarken

[From Bill Powers (2000.01.27.1046 MST)]

Bruce Nevin (2000.01.27.1045 EST)--

I see that my nephew's misunderstanding of _Leakage_ is akin to the
misapplication of group statistics to individuals.

Nice try.

I see that you're really determined to piss me off, so if I do get angry
with you, it will be your fault. I'm obviously too mature and too nice to
get angry without really severe provocation. If you drive me to extreme
reactions, you will be calling the consequences down on your own head. Who
knows, you might even give me a heart attack, and then won't you wish you
had listened to me? You just have to learn to behave more responsibly, by
doing what _I_ want instead of selfishly doing what _you_ want. You must
learn to defer to other people's rules, and in this case, _I_ am the "other
people."

I'm glad we agree on that and I'll expect you to remember it next time. For
now, I forgive you.

:slight_smile:

Best,

Bill P.

[From Bill Powers (2000.01.27.1121 MST)]

Rick Marken (2000.01.27.1000)--

You must have written to them right near the start of the
administration (92).

Appended is the letter: (Jan 31, 1994).

How do you compute rho leakage? My impression is that TCP computed it as
(estimated growth-rate potential (13%/year) minus actual growth rate) -
alpha leakage . Alpha leakage and actual growth rate come from data, but
estimated growth-rate potential and rho leakage don't. It's dangerous to
use rho leakage as an indicator of tightening the money supply, because
that assumes that the theory is right. It would be better to look at actual
interest rates, which do objectively reflect the Fed's actions. Even better
would be to look for the actual announcements of changes in the rediscount
rate and the other ways of applying the screws.

I was kidding about the influence of my letter on the Clinton
administration. But it's clear that events since 1992 support TCP's
conclusion that if the Fed allowed the money to flow more freely, the
result would NOT be inflationary, and it _would_ reduce unemployment and
increase the growth rate. TCP said that his analysis shows that the U.S.
has _never_ experienced inflation due to too much money being available (or
perhaps it was only once, near the beginning of the last century). The
primary cause of inflation other than leakage has been industry-wide
collective bargaining, in which wages go up in an entire sector, and the
affected producers can non-competitively pass the increased costs on to
consumers by raising prices (for example, if auto workers get raises across
the whole industry, all cars can become more expensive because no auto
maker has to fear that another one will _not_ raise prices. Labor's net
gain in buying power from such wage increases has, in fact, been zero).

Best,

Bill P.

···

========================================================================
January 31, 1994

Dear President Clinton,

The news today is that Allen Greenspan is considering a rise in short-term
interest rates. If you will look at the economic record, you will see that
the theoretical reason for doing this is factually invalid. When the money
supply is tightened, three things actually happen:

1. Unemployment increases
2. The growth of the economy slows (and sometimes goes
   negative).
3. Inflation remains the same or INCREASES.

According to the economic theory in which Greenspan believes, tightening
the money supply should curb inflation. There is absolutely no support for
this theory in the Statistical Abstracts. I urge you to have someone do a
little research on the facts, and then confront Greenspan with the results.
An increase in interest rates or any other move to make money less easily
available will have a deleterious effect on real growth, may well worsen
inflation, and could wreck your whole program -- as it has wrecked other
programs of other Presidents in the past.

Respectfully yours,

William T. Powers
73 Ridge Place, CR 510
Durango, CO 81301
powers_w%flc@vaxf.colorado.edu

[From Rick Marken (2000.01.27.1110)]

Bill Powers (2000.01.27.1121 MST)--

How do you compute rho leakage? My impression is that TCP
computed it as (estimated growth-rate potential (13%/year)
minus actual growth rate) - alpha leakage .

Yes. That's what rho is in my table.

It's dangerous to use rho leakage as an indicator of tightening
the money supply, because that assumes that the theory is right.

I agree. That's why I want to get Fed data on this. I was just
basing the reasonableness of the rho numbers on my memory of what
the Fed was doing in the past. As I recall, Fed interest rates
did start inching up in 1998. But interest rate is not the only
(or even the main) way the Fed regulates the money supply.

It would be better to look at actual interest rates, which do
objectively reflect the Fed's actions.

Yes. The problem is that interest rate change is only one of
the ways that the Fed puts on the screws (there is also change
in the reserve requirement, for example). I will try to get data
from the Fed that give at least a proportional indication of how
much money they have taken out of (or put back into) circulation
each year.

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates mailto: rmarken@earthlink.net
http://home.earthlink.net/~rmarken

[From Bruce Nevin (2000.01.27.1719 EST)]

Bill Powers (2000.01.27.1046 MST)]

Bruce Nevin (2000.01.27.1045 EST)--

I see that my nephew's misunderstanding of _Leakage_ is akin to the
misapplication of group statistics to individuals.

Nice try.

I see that you're really determined to piss me off, [...]

Whoa! I have no idea what my words meant to you to elicit that!

I guess I'm way off base, but this is what I had in mind: The behavior of
an aggregate entity does not predict the behavior of an individual in the
aggregate.

And no, I have never yet aimed to piss you off, or anyone else here. Though
obviously I've achieved that unintended side effect now.

        Yours in apologetic but really innocent ignorance,

        Bruce Nevin

···

At 10:59 AM 01/27/2000 -0700, Bill Powers wrote:

[From Bill Powers (2000.01.27.2148 MST)]

Bruce Nevin (2000.01.27.1719 EST)]

I see that my nephew's misunderstanding of _Leakage_ is akin to the
misapplication of group statistics to individuals.

Nice try.

I see that you're really determined to piss me off, [...]

Whoa! I have no idea what my words meant to you to elicit that!

I guess I'm way off base, but this is what I had in mind: The behavior of
an aggregate entity does not predict the behavior of an individual in the
aggregate.

Of course that's right. But your starting out "I see that my nephew's
misunderstanding ..." was just too apropos to the other thread to resist.
Sorry to have shocked you.

Best,

Bill P.