Davidson seminar -A final response (fwd)

[Paul Davidson <pdavidso@xxxxxxxxxxxxxxxxxx>]

Here is a final response , summing up and some thoughts about Japan's bubble.

---------- Forwarded message ----------
Date: Mon, 25 Sep 95 15:55:12 LCL
From: Paul Davidson <PB108928@xxxxxxxxxxxxxx>
To: pdavidson@xxxxxxx
Subject: Davidson seminar -A final response

SUBJECT: DAVIDSON'S SEMINAR -- A FINAL RESPONSE

Enclosed is the responses to the last commentators. This is
followed up by responding to Ric's request for a summing up and
what I learned from this interchange -- and finally some
statement about comments on the bubble and Japan.

DAVID ANDREWS:

David asks for an interpretation of a quote on p. 39 of the GT
involving "the causal sequence of economic events, which are
clear cut and determinate" and asks whether Keynes is trying to
have it both ways in emphasizing uncertainty  and "clear cut and
determinate".

This quote comes from the chapter on units  where JMK argues that
"the quantitative indeterminacy" of the units used to measure and
compare "one real output with another" are not useful for such
comparisons (although "quantitative" enough for the
econometrician to spin stories about all sorts of "stable"
empirical facts). Moreover, these real output "quantitative"
measures are not useful or important for a theory that explains
employment decisions in the private sector. In any explanation of
private sector employment decisions it is "long-period
expectations" regarding NOMINAL sales revenue vis -a vis- nominal
contractual costs of hiring workers plus buying materials that
are the "clear-cut and determinate" concepts that are "the causal
sequence of economic events" that explains private sector
employment this period!

The "quantitative indeterminacy" of real output measures is
easily illustrated (not intentionally however) by an article in
the October 1995 issue  of THE ATLANTIC MONTHLY entitled "In the
GDP is Up, Why is America Down?". Keynes's answer is that
comparing GDP statistics between periods is not a very useful
occupation for economists (see Keynes on Qcomparing Queen
Victorisa and the first Queen Elizabeth p. 40 GT).

GDP (or earlier GNP) statistics are often (mis)interpreted a
"quantitative fact" that measures comparative well-being as well
as a determinate of employment. Using these "real output"
quantifications are not useful in a theory of employment, while
future nominal sales and nominal costs are clear cut determinates
of employment. In a world without forward contracts for sales of
goods and services, then only expectations of nominal sales are
relevant, but as I argue in MRW and PKMT (and Hicks argues in
VALUE AND CAPITAL -- pp. 135 ff.-- in his concept of a
complete"forward trading system") in a world where production and
sales are organized on forward contracts, then it is the nominal
values of these forward contracts that are the complete and
clear-cut determinants of employment this period.

Please note that the a vari ant essness of GDP statistics for
comparison purposes can be used against Gelles's continual
calling what people buy as "trinkets", (others in this seminar
use the more vulgar terminology of "sh-t"), or others denigrating
"military Keynesianism".  As a human I may make all sorts of
value judgments about what is "good" to consume and what is
"bad", but as an economist trying to explain employment -- I try
to keep these views under control.

MITCHELL

Finally, David's response to Bill is excellent, the fact that you
find "empirical regularities" over some limited time horizon (or
as Bill puts it between "shocks and discontinuities... things are
very stable") and then you expect the unexpected I.e., a change
in regime" is exactly the pattern one might expect in an ontolo.
uncertainty with some human institutions  (e.g., forward nominal
contract) creating some inertia in the system.

Moreover, Bill  still does not explain why when he tried the
random number generator experiment I suggested, several times (I
think 4 out of 20 if I remember() bill found a second difference
that appeared to have a "structure" (What Yaglom calls an nth
difference structure function) despite the fact that the original
series of numbers were obtained by a random number generator.
Bill if you can get 4 out of 20 with a 2nd difference function,
how many out of 20 tries would you have found a structural
equation in n tries where n---> infinity. And if the number of
"structural" difference equations discovered increases with n,
then how do you decide whether this an "empirical" fact or not?


Bill and I agree on the usefulness of terms such as Preference
function as a short-hand way of specifying peoples "exogenous
choice" to the problem at hand.

HENWOOD:

Doug still talks about fundamentals in the stock market e.g.,
price/earning ratios, price/book ratios, -- but these are as
fundamental as Purchasing Power Parity IN A FLEXIBLE EXCHANGE
RATE system -- about all that is certain is that the actual value
will either exceed or fall short of the calculated time series
average value and the time series will be heteroskidastic. Of
course for any series of numbers, even a random generated series
one can always use the equations to "calculate" the average value
of the series, the question remains is there an intrinsic real
value that is exerting a gravitational force.

Regarding Tobin's-Q ratio as a real determinant: Doug you might
look at my 1968 ECONOMETRICA article -- writing and accepted in
1965 with a three year delay  in publication --(criticizing
Tobin's investment theory and suggesting that the proper thing
was to make an investment theory that revolved around the ratio
of Spot to Forward market prices of each investment good. You
will note that what provides an anchor for this ratio (which is
the forerunner to Tobin's Q) is the contractual nominal cost
function of producing I goods -- and especially nominal
contractual wage contracts.
(This bottom half of the ratio is not fixed, like Avogardo's
number by fundamental real determinants that can not be changed
by human decisions, Doug!)

Doug also thinks he stepped in to the looking glass when I argued
that the money rate of interest rules the real roost, and that
real is a metaphysical concept. No Doug, the problem is that you
are a captive of the classical view dichotomy where money is
neutral and real things rules the roost! By accepting the
neutrality of money -- at least in the long run where real
centers of gravity are assumed to be a "universal truth" not
needed to be proved -- you are living on the wrong side of the
professional mirror -- at least in the long run~!

QUINN:

Kevin asks how policies  about how policies defeat  Keynes's ch.
12 "dark forces of time and ignorance". Answer: in building
institutions that "guide policy" in the direction that society
believes is desirable-- which in an entrepreneurial world is to
give every one who wants to work at the going wage, the
opportunity to earn income

MCCLINTOCK:

Brent raises the different historical experience of New Zealand
and the US in the 1920s. But Brent one could say the same thing
about the UK who had a great unemployment problem throughout the
1920s -- only in one year 1927 did unemployment fall below 10%.
In 1927 it was 9.7%. As I argue in CONTROVERSIES IN POST
KEYNESIAN ECONOMICS that may explain why Keynes was way ahead of
US economists in trying to explain unemployment as not a short-
run departure from classical full employment equilibrium. Why did
the US have such a different experience? Partly it was that the
US, via the Dawes loan plan, took the part of at least
temporiarily exporting a portion of its current account surpluses
to Germany which helped expand the export component ( and the
investment in the export sector) of US GNP (See my PKMT).

WHEELER:

Kevin raises the old Hobbesian worry of the LEVIATHAN to solve
the problem of self-interest conflicts .  But as my son Greg (as
senior author) and I explain in ECONOMICS FOR A CIVILIZED SOCIETY
people in a social setting are motivated by two types of
incentives -- external incentives (market valuations) and
internal incentives (what Aristotle called virtu --  or we call
civic values). What is required are policies that play both on
the self-interest of market values and the internal incentives of
civic values.)

GLEVY:
Jerry asks about my comments on Guttman's book. Although Robbie
Guttman was a colleague of mine at Rutgers (I think I was at
least partly responsible for his being hired) I am sorry to say I
had not heard of Guttman's book until now. So no comment
possible.

PERELMAN:

Michael talks about the "inflated bubble in real estate prices in
the period 1982-1986. Why the bubble burst? Think about the
impact of the Tax Reform Act of 1986 which, while previously the
marginal tax bracket was 70% for property income, changed, with a
stroke of the pen, the time that people could amortize their
costs of acquiring real estate from 7 years to 27+ years. Is that
reform?

Michael raises the questions of the exchange rate being like a
derivative rather than directly bearing on some fundamental
relationship. Derivative of what? I suspect that Michael has a
purchasing power parity (PPP) view of the determination of the
exchange rate so that it is a"derivative" of the "real costs" of
producing a market basket of goods in Country A vis-avis the
"real costs" of the same market basket of goods in B. If so we
find that the Us $ is always undervalued (relative to its PPP) or
overvalued -- it is apparently never "valued" at PPP. Why?
Because people are merely speculating of the exchange rate a day
later or so and not on the PPP. That is what my Tobin Tax paper
is and why a marginal exchange rate tax does little to stop
speculation in a world of flexible exchange rates but can do a
lot of damage to aggregate international trade. (More on this in
my comment on Japan below).

Finally Michael suggests that my rhetoric turns people off. Sorry
but I guess that's me. I like to discuss with people who are
willing to take positions rather than talking to what Truman
defined as an economist. When Pres. Truman was asked what he
thought of the CEA (organized first in his Administration) he
indicated he would like to get a one-armed economist to be its
Chair. The reason was when he asked economists for solutions they
tended to reply "on the one hand....., but on the other
hand....."

GUSTELL:

Stephen still confuses "probabilistic risk (even if it is fuzzy)
with uncertainty. When insurance companies (Stephen's example)
dump "uncertainty onto someone else" , Stephen it is because they
believe they "know" the objective probablistic risks -- and have
no uncertainty about it. A probability of unity means one is
certain that the event will occur at least 95 times in 100 tries.
While a probability of ZERO means one is CERTAIN that the event
will not occur at least 95 times out of 100.

Stephen, I think you will find a group of empirical studies on
"Catastrophe Insurance" by Wm. Kunraether and colleagues at the
University of Pennsylvania very instructive. They  find in cases
of low probability- high costs events, people tend to
"underinsure" -- even when they are educated to the actuarial
probabilities, e.g., flood insurance, earthquake insurance (Yes
Gary D. even in California) until a catastrophe hits the
headlines -- and then they rush to try to overinsure relative to
the actuarial probabilities used by insurance companies.  Even
Kenneth Arrow has noted in a preface to one of Kunraether's
studies that these empirical finding jeopardize the foundations
of classical theory -- but this represents either epist.
uncertainty (at best) or a non=classical specified preference
function.

Fuzzy probabilities are defined by Stephen as"being more
uncertain [than actuarial probabilities?] because we do not KNOW
what the odds of an event REALLY are." But this is a case simply
of epist. uncertainty -- where human knowledge of the REAL odds
is defective.


RESPONSE TO RIC's request.

The main points that came out in the discussion are

1. Most of the discussant's accepted my argument (to either a
greater or lessor degree) that the basic economic problems of
employment, etc are due to ONTOL. uncertainty.  Most however
preferred to use a different dictionary. So Roger K (representing
the Austrians) and Barkley R representing the technologically
advanced) used different words (and often different models) but
ended up close to what Barkley called "EFFECTIVE ontological
uncertainty".

2. If the difference in modelling approach is more semantic than
fundamental, then why the debate?  Ultimately the question is do
these different variants lead to different POLICY initiatives.
Here I think one can draw a crucial distinction.

Austrians (and most mainstream New Classicals) think that
government policy -- in a world of uncertainty (defined as they
want to define it) -- deliberate government policy is likely to
be a LEVIATHAN (thereby crushing political rights) and also
economically inefficient (thereby destroying significant economic
gains that a free market would insure at least in the long run).
To oversimplify, government is the problem, and the free market
is (almost) always the best solution obtainable. They want micro-
deregulation of the financial system -- even the complete
deregulation of the banking system to the point that there is no
franchise necessary to start a depositary institution.

Barkley's high techies and New Keynesians in general see
unemployment as a necessary short-run cost of operating the
system and see government macropolicies as merely short-run pump
primers, jump starters, or information coordinators ("Build an
information superhighways and they will come" ). Over an extended
period of time, the government should not get involved -- one
should not permit government to generate permanent deficits (in
an earlier era these people would be for balancing the fiscal
budget over the cycle -- though nowadays I expect there is a
division over this cyclical balanced budget view -- but certainty
in the long run, the budget should be balanced).  Moreover the
logic of most New Keynesians would force them, if they ever
thought about an open economy context, to be for a free market
exchange rate and free global financial markets.  In the long-run
they believe in -- as do the neoRicardians-- some persistent
centers of gravity, i.e., some stable long-run parameters that
because the relationships are non-linear, may generate patterns
which prevent humans from reliably forecasting the future unless
they guess (or know from the outside) the very close to precise
values of these parameters.

I suspect that most of these people would be for deregulation of
the financial system -- or at least their logical models do not
provide a rationale for a regulated financial system.

3. There are some economists who believe that the major economic
problem of unemployment in an entrepreneurial money-using market
system is due to ontological uncertainty. For us, as for Keynes,
there is a need for a permanent positive role for the government
in both micro and macro-areas of the financial system and the
regulation of effective demand----what Heilbroner and Milberg
call the VISION for the 21 century (in their forthcoming book THE
CRISIS OF VISION IN MODERN ECONOMIC THOUGHT) --. (Keynes talked
about the permanent socialization of Investment (or what I call
D2) spending -- for that D2 in an open economy context involves
exports as well as domestic gross investment spending. Calling
for an annual, cyclical or even long-run balanced government
budget PER SE IS NOT A GOAL FOR A CIVILIZED SOCIETY. The budget
is only one means to an end!

(Heilbroner and Milberg insist that the old "respective rank
orders of private and public activity must give way to a much
more considered approach... The enhanced legitimacy of the public
sector...we see as central to a new vision will require not only
a change in the nature of economic analysis but also a
transformation of the status of analysis in economic inquiry in
general".)

The real test of policy differences comes on what long-run
(persistent - not pump priming) policies do the various models
advocate for the global economic model of the 21st century where
national states currently exist with multinational corporations.
(a)Does that make nation states impotent? (b)Can or should we
permit free international global capital movements? (c) Is there
any role for national central banks? and if (d) so should they be
"independent"?

My answers to these queries are (a) No -- but that requires MNC
not playing one major national interest off of another --
therefore cooperation of the national states are required for if
they all don't hang together they will hang separately. (b)NO.
(c) Yes (d) No.

Any one of the issue policy issues above are possible subjects
for future seminars.


Finally , let me thank all those who took the time to provide me
with comments. I found them always stimulating and even when I
did not agree it required me to think about why I didn't agree.
Hopefully both the author and the audience learned something. I
know I did!


JAPAN AND BUBBLES

Ric raises the issue of why did the Bank of Japan allow the
"bubble economy" of the 1980s to occur. An excellent question.
And Ric suggests a similarity with the "S&L Fiasco"

My answer is that during the preceding decade "academic
scribblers" influenced men (very few woman) in authority that
deregulation was desirable because of some "bad" regulatory
decisions made in the late 1960s and early 1970s.

My response is not from hindsight. I am on the printed record as
to trying to  explain why financial deregulation at the end of
the 1970s was unwise for the S&Ls.  On p. 611 of the second
volume of the Collected Writings of Paul Davidson (edited by
Louise Davidson) in a speech to the Federal Home loan Bank Board
Examiners (on April 7, 1983)entitle "The Future of the Three Rs:
Regulation, Reserves and Reaganomics" I pointed out that:

"When I addressed this group in 1978, I suggested that the then
new interest deposit forms had permitted the thrifts to trade a
disintermediation problem for a profits problem.  I close today
by suggesting we may have via our recent economic policies
created a problem where the authorities will have to judge how
many special cases in the banking system there are -- or is the
problem of nonperforming loans becoming systemic?"

As events proved, at both the 1978 and 1983 meetings, the
audience enjoyed and digested the food for their stomach much
more than the food for thought that I offered.

Ric, the answer is that there is no substitute for good
regulation -- but that in a world of uncertainty people make
mistakes (including Authorities). The solution, however, is not
to look back and say that since mistakes have been made, we
should have no regulation. Rather the solution involves looking
at the current predicament, seeing where you want to go and then
trying to devise all sorts of public and joint public-private
institutions that play on peoples internal and external incentive
to build a corridor of policy to guide the economy in the
direction we want. Just as in evolution where those who adapt
tend to have a better chance of survival, so we should be
eternally vigilant to make sure that our human institutions are
adopting to a reality and affecting this transmutable reality
into the best possible corridor we can engineer.

And be modest like Dentists!