The General Theory of Employment, Interest, and Money

Chapter 19 
CHANGES IN MONEY-WAGES 
I 
It would have been an advantage if the effects of a change in money-wages could have been 
discussed in an earlier chapter. For the classical theory has been accustomed to rest the supposedly 
self-adjusting character of the economic system on an assumed fluidity of money-wages; and, when 
there is rigidity, to lay on this rigidity the blame of maladjustment. 
It was not possible, however, to discuss this matter fully until our own theory had been developed. 
For the consequences of a change in money-wages are complicated. A reduction in money-wages is 
quite capable in certain circumstances of affording a stimulus to output, as the classical theory 
supposes. My difference from this theory is primarily a difference of analysis; so that it could not be 
set forth clearly until the reader was acquainted with my own method. 
The generally accepted explanation is, as I understand it, quite a simple one. It does not depend on 
roundabout repercussions, such as we shall discuss below. The argument simply is that a reduction 
in money-wages will 
cet. par.
stimulate demand by diminishing the price of the finished product, 
and will therefore increase output and employment up to the point where the reduction which labour 
has agreed to accept in its money-wages is just offset by the diminishing marginal efficiency of 
labour as output (from a given equipment) is increased. 
In its crudest form, this is tantamount to assuming that the reduction in money-wages will leave 
demand unaffected. There may be some economists who would maintain that there is no reason 
why demand should be affected, arguing that aggregate demand depends on the quantity of money 
multiplied by the income-velocity of money and that there is no obvious reason why a reduction in 
money-wages would reduce either the quantity of money or its income-velocity. Or they may even 
argue that profits will necessarily go up because wages have gone down. But it would, I think, be 
more usual to agree that the reduction in money-wages may have 
some
effect on aggregate demand 
through its reducing the purchasing power of some of the workers, but that the real demand of other 
factors, whose money incomes have not been reduced, will be stimulated by the fall in prices, and 

---

128
that the aggregate demand of the workers themselves will be very likely increased as a result of the 
increased volume of employment, unless the elasticity of demand for labour in response to changes 
in money-wages is less than unity. Thus in the new equilibrium there will be more employment than 
there would have been otherwise except, perhaps, in some unusual limiting case which has no 
reality in practice. 
It is from this type of analysis that I fundamentally differ; or rather from the analysis which seems 
to lie behind such observations as the above. For whilst the above fairly represents, I think, the way 
in which many economists talk and write, the underlying analysis has seldom been written down in 
detail. 
It appears, however, that this way of thinking is probably reached as follows. In any given industry 
we have a demand schedule for the product relating the quantities which can be sold to the prices 
asked; we have a series of supply schedules relating the prices which will be asked for the sale of 
different quantities on various bases of cost; and these schedules between them lead up to a further 
schedule which, on the assumption that other costs are unchanged (except as a result of the change 
in output), gives us the demand schedule for labour in the industry relating the quantity of 
employment to different levels of wages, the shape of the curve at any point furnishing the elasticity 
of demand for labour. This conception is then transferred without substantial modification to 
industry as a whole; and it is supposed, by a parity of reasoning, that we have a demand schedule 
for labour in industry as a whole relating the quantity of employment to different levels of wages. It 
is held that it makes no material difference to this argument whether it is in terms of money-wages 
or of real wages. If we are thinking in terms of money-wages, we must, of course, correct for 
changes in the value of money; but this leaves the general tendency of the argument unchanged, 
since prices certainly do not change in exact proportion to changes in money-wages. 
If this is the groundwork of the argument (and, if it is not, I do not know what the groundwork is), 
surely it is fallacious. For the demand schedules for particular industries can only be constructed on 
some fixed assumption as to the nature of the demand and supply schedules of other industries and 
as to the amount of the aggregate effective demand. It is invalid, therefore, to transfer the argument 
to industry as a whole unless we also transfer our assumption that the aggregate effective demand is 
fixed. Yet this assumption reduces the argument to an