Re: Paul D's response

[BILL MITCHELL <ECWFM@xxxxxxxxxxxxxxxxxxx>]

Barkley raised the issue of policy in an uncertain world:

>     To get practical, as you wish us "rigor mortis"
>:-) types to do, don't you see PK policy as imposing
>order and directedness on an otherwise non-ergodic
>economy, pushing for full employment in an economy that
>might not otherwise go there?
>     As I noted in New York, in a multiple equilibrium
>sunspot world it becomes the responsibility of the
>government to direct the outcome if possible.  I suggested
>indicative planning and I note that Keynes in his "The
>End of Laissez Faire" is generally considered to be the
>father of the idea of indicative planning.

Interesting ideas and close to my own heart as a PKT econometrician
(which is one part of my "portfolio of things i do" - studying marx
being another). I am working on a paper about "realism and policy".
it follows some work i have done on "the role of econometrics in a pkt
research program". the latter really confronts "realist thinking",
which in it modern variant is hostile to econometrics due to this
suppostition that keynes said the world exhibits endemic uncertainty
and thus evade the predictive tools of econometrics.

paul follows this line too, and eschews econometrics  b/c the
world is non-ergodic. well it might be and it might not be. how do we know
the exact source of the non-stationarity (or non-ergodocity)? i have
argued this point with paul and others before and nefer got a satisfactory
answer. non-ergodicity is not an homogenous state.

But further, if pkt is about intervention to control and attenuate
the variations in the capitalist economy arising from the changes in behaviour
of cappos who experience all sorts of emotions and outcomes relating to
their profit rate (measuring their surplus value extraction), then
the notion that we cannot do anything b/c of nons-stationarities has to go.

Policy implies order - or a semblance of order within which a standard
error can be defined. and if that is the case you must be  assuming that
for some short to medium horizon, there sufficient stability
among the conjunctions of events you seek to understand and influence,
to warrant intervention.

the next part of the story is htat you need some practical edge on
policy design. it is called having numbers and sensitivities. that
requires converting your "causal" understandings into a medium that
is meaningful - $ estimates. after all competing policies
all spend the same thing - $s.

and i am not nearly as "jumped up" as keynes (in his debates with
tinbergen) who said who needs econoemtrics or statistics, that he
knew the numbers they generated anyway. such humility. but then what
would you expect from a person who calimed to have discoverd a GT, but
who got it mostly of Kalecki, and even then missed the boat by
failing to explain the specific nature of capitalism.

kind regards
bill
*******************************************************************************
 William F. Mitchell            Telephone: +61-49-215027      .-_|\
 Department of Economics                   +61-49-705133     /     \    about
 The University of Newcastle    Fax:       +61-49-216919     \.--._/*<-- here
 Callaghan   NSW  2308                                            v
 Australia                      Email : ecwfm@xxxxxxxxxxxxxxxxxxx

 World Wide Web Home Page: http://econ-www.newcastle.edu.au/~bill/billyhp.html
*******************************************************************************