Dear Basil and others,
I can agree that Keynes was confusing (as distinct from wrong), and that
his confusion had to do with his not assuming endogenous money. He was
confusing because, when writing about the elasticity of production, he
mixed a discussion of variations in the volume of labour with a discussion
of variations in the quantity produced of an asset.
As I said earlier, in chapter 17 he *defines* the elasticity of production
in terms of a change in the volume of *labour*: 'the response of the
quantity of labour applied to producing it [money] to a rise in the
quantity of labour which a unit of it will command'(Keynes, 1936: 230). The
point seems to be that, if money becomes more attractive as an asset, this
will not significantly increase the volume of employment. Strictly
speaking, this definition, per se, does not mean that the supply of money
is fixed (for example, it could be used together with the assumption that
money is endogenous).
Confusion arises when then, on the same page, Keynes discusses changes in
the quantity of *an asset* (as distinct from labour), because this is what
matters for his discussion of changes in the asset's own-rate. In the case
of money, he then argues that the supply if fixed (abstracting from 'a
deliberate increase in its supply by the monetary authority' and, as Basil
points out, from endogenous money).
By the way, confusion increases when, in chapter 20, he offers a different
definition of the elasticity of production, now in terms of a change in
output in response to a change in demand. This is the definition that Basil
and Kazuhiro could use to make their point about *the elasticity of
production*, not the definition given in chapter 17, p. 230 - but see my
comment below on one of the essential properties of money. With exogenous
money and a fixed supply, these two definitions are quite similar (the
quantity of labour employed and the quantity of money supplied will not
increase much when the demand for money increases), but with endogenous
money this is not the case.
Now, if we want to consider endogenous money, we should distinguish the
issue of changes in labour employed to produce money from the issue of
changes in the quantity of money. With exogenous money the distinction is
not so important as with endogenous money, and perhaps this is why Keynes,
assuming exogenous money in the GT, was not very concerned with making the
distinction clear or even perhaps aware of its importance.
I insist that on p. 230 he defines the elasticity of production in terms of
the first issue, although he is confusing about its relation to the second
issue. However, regardless of the semantic issue of how to define the
elasticity of production, it seems to me that one of the essential
properties of money is that not much more labour can be employed to produce
money when the demand for it increases. This is also an essential property
of other very liquid assets. Therefore, when liquidity preference increases
this does not lead to an increase in employment. The recognition of this
property, *regardless of whether it is called a low elasticity of
production or not*, is an essential part of Keynes's theory and, I would
argue, of a good Post Keynesian theory. In contrast, assuming exogenous
money is not essential. The introduction of endogenous money does not
change that essential property, although it does change the possibility of
increasing the money supply.
Hope this contributes to our dialogue. Cheers,
David