F.—ECONOMIC SCIENCE AND STATISTICS 113 
tinue to grind. We know that a rise or fall in the wages of one group 
will not be permanent unless it is followed by a corrresponding change in 
the wages of the other group, or unless there has been a change in the 
nature of the persistent forces to which I have referred. It is precisely 
this sympathy in wage movements that gives significance to the conception 
of an average wage and to movements in that average. 
If it be true that the relationship between individual wages is not 
arbitrary, it is also true that the relationship between individual prices is 
not arbitrary. In the long run prices are governed by costs, and costs 
ultimately mean wages. Even economic rent, in the last resort, is a 
function of the wage average. If prices are governed by costs and costs 
by wages, and if relative wages obey a law of distribution, it follows that 
actual prices also tend towards a ‘ normal’ arrangement or distribution. 
If a house of ten rooms could be purchased for the same sum as a hundred 
tons of coal, everybody would recognise the existence of some abnormal 
influence which could not fail, ultimately, to bring a correcting influence 
into play. A rise or fall in a large group of prices will not be permanent 
unless either a similar change takes place in the remaining group or a change 
has occurred in the real costs, and therefore money costs, of supply. 
It is precisely this sympathy in prices that gives significance to changes 
in the price average or general level of prices. As in the case of wages 
so, too, in the case of prices : the ‘ short period,’ during which deviations 
from the ‘normal’ distribution may continue, tends to grow longer ; 
but in the long run the effect of the persistent force of competition 
(broadly interpreted) becomes evident even in a constantly changing 
world. 
These elementary facts seem to me to provide the true foundation of 
a theory of money. The supply of money needed by a community, and 
the supply of money that can be absorbed by a community, is a function 
of the price average. If every pound of wages or of prices were called 
ten pounds, the community would merely be using ten times as much 
money as before. Conversely, if the supply of money is fixed, the price 
average must conform to that supply, and in a state of equilibrium the 
wage average and the price average will reflect the normal distribution of 
individual wages and prices. But a change in the supply of money 
produces intermediate effects before the final state of equilibrium is 
reached. Nor is it necessary to stress the practical importance of these 
intermediate effects, which will presently be considered. At the present 
stage, however, it is desirable to confine our attention to the characteristics 
of a community in a state of equilibrium in the sense of being free from 
the intermediate disturbances of a process of change. 
I have referred to the existence of a normal distribution of wages and 
of prices. The statements that I made are applicable to every community 
in which order is maintained, either through the force of competition or 
by legal enactment. But the normal relationship of wages or of prices 
is not the same in all communities: each has its own characteristics. 
Thus, for example, the relative rates of remuneration of school teachers, 
coal-miners and railway workers may not be the same, under normal con- 
ditions, in Great Britain as in Germany. ‘The normal distribution may 
