Keynes’s Taxonomic Attack on Say’s Law

Keynes’s Taxonomic Attack on Say’s Law
When Keynes became convinced that the vocabulary of orthodox economics
was not sufficient to explain why an economy might become mired in unemployment,
he developed an expanded classification and new definitions to
demonstrate that Say’s Law ‘is not the true law relating the aggregate demand
and supply functions … [and hence] there is a vitally important chapter of
economic theory which remains to be written and without which all discussions
concerning the volume of aggregate employment are futile’ (Keynes,
1936, p. 26).
Keynes’s General Theory is developed via an aggregate supply–aggregate
demand function analysis to achieve a point of effective demand (Keynes,
1936, pp. 25–6). The aggregate supply function (Z) relates entrepreneurs’
expected sales proceeds with the level of employment (N) entrepreneurs will
hire for any volume of expected sales receipts. This aggregate supply (Z)
function indicates that the higher entrepreneurs’ sales expectations, the more
workers they will hire. The aggregate demand function relates buyers’ desired
expenditure flows with any given level of employment (see Davidson, 1994).
Say’s Law specifies that all expenditure (aggregate demand) on the products
of industry is equal to the total costs of aggregate production (aggregate
supply) including gross profits. Letting D symbolize aggregate demand and Z
aggregate supply, if:
D = fd(N) (8.1)
and
Z = fz(N) (8.2)
then Say’s Law asserts that:
fd(N) = fz(N) (8.3)
‘for all values of N, i.e. for all values of output and employment’ (Keynes,
1936, pp. 25–6). In other words, in an economy subject to Say’s Law, all
costs of production are always recouped by the sale of output. There is never
a lack of effective demand. The aggregate demand and aggregate supply
curves coincide. In a Say’s Law economy, there is no obstacle to full employment.
The aggregate demand and supply functions will be coincident only if
money is neutral, everything is a good substitute for everything else (gross
substitution) and the future can be reliably predicted in terms of probabilities
(the ergodic axiom).
To challenge the applicability of Say’s Law to the real world in which we
live, Keynes had to develop a model where the aggregate demand and aggregate
supply functions, fd(N) and fz(N), were not coincident. Since Keynes
accepted the normal firm short-run flow-supply function developed in
Marshallian economics as the micro-basis for the aggregate supply function,
he could therefore differentiate his approach only via the concept of aggregate
demand. Keynes divided aggregate demand into two classes, that is,
D = D1 + D2 (8.4)
where:
D1 = f1(N) (8.5)
and
D2 f2(N) (8.6)
D1 represents all expenditures which ‘depend on the level of aggregate income
and, therefore, on the level of employment N’ (Keynes, 1936, p. 28).
D2, therefore, represents all other expenditures which are not related to income.
Even if D2 is related to aggregate income (that is, D2 = f2(N) so long as
f1(N) + f2(N) fz(N) for all values of N, then Say’s Law is not applicable.
Explicit recognition of the possibility of two classes of demand expenditures
must make Keynes’s analysis a more general theory than the orthodox
theory since the latter recognizes only a single demand class. Classical theory
is ‘a special case’ (Keynes, 1936, p. 8)
for all values of N.
By proclaiming a ‘fundamental psychological law’ associated with ‘the
detailed facts of experience’ where the marginal propensity to consume was
always less than unity (Keynes, 1936, p. 96), by decree, Keynes declared that
f1(N) would never coincide with fz(N) in the real world, even if D2 = 0. Say’s
Law could not be applicable to ‘the facts of experience’.