The General Theory of Employment, Interest, and Money

ignoratio elenchi
. For, whilst no one would 
wish to deny the proposition that a reduction in money-wages 
accompanied by the same aggregate 
effective demand as before
will be associated with an increase in employment, the precise question 
at issue is whether the reduction in money-wages will or will not be accompanied by the same 
aggregate effective demand as before measured in money, or, at any rate, by an aggregate effective 
demand which is not reduced in full proportion to the reduction in money-wages (i.e. which is 
somewhat greater measured in wage-units). But if the classical theory is not allowed to extend by 
analogy its conclusions in respect of a particular industry to industry as a whole, it is wholly unable 
to answer the question what effect on employment a reduction in money-wages will have. For it has 
no method of analysis wherewith to tackle the problem. Professor Pigou's 
Theory of Unemployment
seems to me to get out of the classical theory all that can be got out of it; with the result that the 
book becomes a striking demonstration that this theory has nothing to offer, when it is applied to 
the problem of what determines the volume of actual employment as a whole. 
II 

---

129
Let us, then, apply our own method of analysis to answering the problem. It falls into two parts. (i) 
Does a reduction in money-wages have a direct tendency, 
cet. par
., to increase employment, '
cet. 
par
.' being taken to mean that the propensity to consume, the schedule of the marginal efficiency of 
capital and the rate of interest are the same as before for the community as a whole? And (2) does a 
reduction in money-wages have a certain or probable.tendency to affect employment in a particular 
direction through its certain or probable repercussions on these three factors? 
The first question we have already answered in the negative in the preceding chapters. For we have 
shown that the volume of employment is uniquely correlated with the volume ofeffective demand 
measured in wage-units, and that the effective demand, being the sum of the expected consumption 
and the expected investment, cannot change, if the propensity to consume, the schedule of marginal 
efficiency of capital and the rate of interest are all unchanged. If, without any change in these 
factors, the entrepreneurs were to increase employment as a whole, their proceeds will necessarily 
fall short of their supply-price. 
Perhaps it will help to rebut the crude conclusion that a reduction in money-wages will increase 
employment 'because it reduces the cost of production', if we follow up the course of events on the 
hypothesis most favourable to this view, namely that at the outset entrepreneurs 
expect
the 
reduction in money-wages to have this effect. It is indeed not unlikely that the individual 
entrepreneur, seeing his own costs reduced, will overlook at the outset the repercussions on the 
demand for his product and will act on the assumption that he will be able to sell at a profit a larger 
output than before. If, then, entrepreneurs generally act on this expectation, will they in fact succeed 
in increasing their profits? Only if the community's marginal propensity to consume is equal to 
unity, so that there is no gap between the increment of income and the increment of consumption; or 
if there is an increase in investment, corresponding to the gap between the increment of income and 
the increment of consumption, which will only occur if the schedule of marginal efficiencies of 
capital has increased relatively to the rate of interest. Thus the proceeds realised from the increased 
output will disappoint the entrepreneurs and employment will fall back again to its previous figure, 
unless the marginal propensity to consume is equal to unity or the reduction in money-wages has 
had the effect of increasing the schedule of marginal efficiencies of capital relatively to the rate of 
interest and hence the amount of investment. For if entrepreneurs offer employment on a scale 
which, if they could sell their output at the expected price, would provide the public with incomes 
out of which they would save more than the amount of current investment, entrepreneurs are bound 
to make a loss equal to the difference; and this will be the case absolutely irrespective of the level of 
money-wages. At the best, the date of their disappointment can only be delayed for the interval 
during which their own investment in increased working capital is filling the gap. 
Thus the reduction in money-wages will have no lasting tendency to increase employment except 
by virtue of its repercussion either on the propensity to consume for the community as a whole, or 
on the schedule of marginal efficiencies of capital, or on the rate of interest. There is no method of 
analysing the effect of a reduction in money-wages, except by following up its possible effects on 
these three factors. 
The most important repercussions on these factors are likely, in practice, to be the following: 
(1) A reduction of money-wages will somewhat reduce prices. It will, therefore, involve some 
redistribution of real income (a) from wage-earners to other factors entering into marginal prime 

---

130
cost whose remuneration has not been reduced, and (b) from entrepreneurs to rentiers to whom a 
certain income fixed in terms of money has been guaranteed. 
What will be the effect of this redistribution on the propensity to consume for the community as a 
whole? The transfer from wage-earners to other factors is likely to diminish the propensity to 
consume. The effect of the transfer from entrepreneurs to rentiers is more open to doubt. But if 
rentiers represent on the whole the richer section of the community and those whose standard of life 
is least flexible, then the effect of this also will be unfavourable. What the net result will be on a 
balance of considerations, we can only guess. Probably it is more likely to be adverse than 
favourable. 
(2) If we are dealing with an unclosed system, and the reduction of money-wages is a