Re: S=I, an old debate.

[Bruce McFarling <ecbm@xxxxxxxxxxxxxxxxxxx>]

At 05:54 PM 17/08/1998 +1000, I wrote:
>At 00:26 17/08/98 -0700, Chas wrote:
>
>>I could, say, use part of my cash holdings to purchase a new financial
>>asset.
>
>	Then net savings have not changed, they have just changed hands.
>(BTW, readers will note that this is one of the first issues covered in that
>dusty old GT).

        Slipped up on this, in that I did *not* read "new financial asset"
as *newly created* financial asset.  I can't decide on whether to blame
fatigue or illness, so instead I'll blame my silly decision to send a first
draft to pkt.  Just because a message is short doesn't mean it shouldn't
be edited before posting!
        Newly created financial assets are central to the GT D1/D2/C/S
interaction:

        Expenditures which are *not* financed out of income are financed
by the creation of new financial assets
        As income received is spent, any residual remaining is saving.
Therefore, in the period of income receipt, before the income begins to
be spent, it starts out as all saving, and this declines as expenditures
are made.
        As E=Y, [E less expenditure financed out of income] equals [Y
less income devoted to finance expenditure]. [See note 1]
        If there is a stable relationship between C and Y, such that
we are instrumentally validated [See note 2] in writing C=f(Y), then we can
use this to model the income generated by a given level of expenditure
financed by creation of new financial assets. So, the saving generated by
the interaction of:
        expenditures financed with creation of new financial assets,
        and the relationship between consumption and income,
will always be sufficent to fund the expenditure financed by newly created
financial assets.

        Of course, this sufficieny does not guarantee that the
expenditure financed by newly created financial assets is in fact
adequately funded: there may well be many slips between cup and lip --
banking crises, panics in the stock market, etc.  And to my mind
[See note 3], if there is a cyclical regularity to susceptibility to
these problems -- a la the Minsky Hypothesis -- this adds an
interesting medium-term dimension to the essentially short-term model
in the GT.

[Note 1]  Of course, if income is measured as aggregate factor income,
then indirect taxes will be omitted: for the theoretical relationship,
we would have to either include or exclude indirect taxes in both E
and Y.

[Note 2] This term is not from the GT, but rather from Veblen's writings.

[Note 3] Note that there is a transition from definitive mode to personal
perspective.  For example, I have a hunch that Paul Davidson would
dispute the characterisation of the GT as an essentially short-term
theory.
Virtually,

Bruce McFarling, Ourimbah, NSW
ecbm@xxxxxxxxxxxxxxxxxxx