F.— ECONOMIC SCIENCE AND STATISTICS. 113 



in the volume of labour directly employed, there are not necessarily any 

 resources available by which that labour can be indirectly absorbed. The 

 consumer pays the price which brings the maximum aggregate return. 

 Unless this price is lower than the price previously ruling, he cannot 

 increase his exjjenditure in other directions. The price need not be lower, 

 because, though prime cost may be lower, supplementary cost, for the 

 volume actually sold, may be so much higher as to lead to no general 

 lowering of cost at all. In other words, rationalisation undertaken on 

 technological grounds without taking into account demand conditions 

 may not increase the aggregate national dividend and so may create an 

 unemployment problem which it cannot solve. And we have no right to 

 assume that the race of rationalisers never makes a mistake. 



Returning now to the general problem, we necessarily employ concepts 

 which are familiar to all students of economics. 



(o) If the demand for an article has an elasticity greater than unity, 

 a reduction in its price results in a more than proportionate increase in 

 the quantity demanded. Thus, even in a rationalised industry, in which 

 labour cost has been reduced, the greater the elasticity, the greater the 

 derived demand for labour, and the greater, therefore, the opportunity 

 for reabsorbing labour and of adding to the total quantity of labour 

 employed. But how much labour will be needed, depends not only on the 

 state of demand, but on the technical conditions in that industry. 



(&) Even if the derived demand for labour in this industry has an 

 elasticity of less than unity, yet provided that the demand for the product 

 of the industry has an elasticity greater than unity, the indirect derived 

 demand for labour may have an elasticity greater than unity. For the 

 machinery and other equipment used by the industry has itself to be 

 created by means of labour, and, if the output of the industry is expanding, 

 it requires an expanding plant. Thus, the increased demand for labour 

 in equipment industries which marks the first stages of a rationalisation 

 niovement calling for large quantities of new equipment, may continue 

 after the first stages have been passed. But too much must not be based 

 upon this. For if rationalisation is a continuous process, it will affect not 

 only the industries supplying consumers' goods in the narrower sense, but 

 also the industries subserving these industries. 



Optimistic interpretations of the rationalisation process will generally 

 be found to be based upon the assumption that what is true in some cases 

 is necessarily true in all. The demand for certain popular luxuries is no 

 doubt highly elastic, but it is equally clear that the demand for agricultural 

 products, for example, is not. There is no reason whatever to suppose, 

 therefore, that an all-round cheapening of products already available will 

 necessarily absorb all the labour unemployed in consequence of technical 

 changes, though no doubt that wall be the case to some extent. But to 

 what extent is unfortunately unknown. 



S. 7. Nevertheless, is it not legitimate to argue with Babbage in the 

 passage already cited that ' as each new machine adds to these gratifica- 

 tions, new luxuries will open to his view ' ? or with Professor Robbins,^^ 



" Economic Effects of Hours of Labour, by Prof. L. Robbins (Economic Journal, 

 March 1929, p. 25). 



1930 I 



