Re: Fw: Epistemology & Economics

["John O'Donnell" <jackodonnell@xxxxxxxx>]

paul davidson wrote:

<<SNIP>>

> In a nonergodic system science means only "explanation"
> and eschews statistically reliable quantitative predictions.

As poor as "nonergodic" is to describe uncertainty, its use
in describing an economy may be part of the reason
economists have such great difficulty understanding the
difference between controllable variables and uncontrollable
environmental variables. The simple truth is that there are
many uncontrollable and unpredictable environmental
influences on the outcome of changes made in control
variables that do have predictable effects on those
variables.

In physical systems the most easily recognized unpredictable
and uncontrollable variables are the weather. [Although
prediction does seem to be getting better every day.] In
controlling the power output of an internal combustion
engine, the humidity, barometric pressure, temperature, etc.
are all neither controllable nor predictable but each has an
effect on the power output. When they cause change in the
desired outcome, the control variables must be adjusted to
compensate.

In economic systems there are also certainties that can be
used to affect economic variables that are affected by the
"nonergodic" environment. One such controllable variable is
the quantity of money [i.e. -- The monetary base.] with a
predictable effect on the value of money. The outcome of the
effect in any particular instance [i.e. -- An increase in
quantity will reduce its value and vice versa.] may be
overcome by the uncontrollable factors. Such uncertainty in
outcome only means that the instituted change must be
adjusted continuously in response to the changes in value,
not that the effect is "nonergodic" and incapable of
controlling the outcome.

The presumption that statistical correlations can
demonstrate the certainty of either the effectiveness or non
- effectiveness of a control is a blatant misapplication of
statistical correlations. However, correlations do provide a
very good indication of places one should apply rational
analysis to discover the reality of cause and effect and the
variables that can be used to have predictable effects on
controlled variables. They also provide rational indications
that any assumed relationship, such as the assumptions of
the effects of money quantity and/or nominal interest rates
on economic activity, that do not provide expected effect
are ripe for reevaluation or outright rejection of the
assumptions. In short, there is no such thing as "an
exception that proves the rule" as is so often expressed to
justify continued use of a failed assumption.

<<SNIP>>

--
			-- jbod

		Tax Privilege, Not People
___________________________________________________
Come visit and see a new economic perspective --
       http://www.geocities.com/CapitolHill/1067
           Comments/arguments welcome.
.