140 SECTIONAL ADDRESSES 



Rationalisation movement is to increase the areas under one planning and 

 control authority irrespective of whether that area consists of one or a 

 number of plants. And the policy of Planning, so-called, goes as far as 

 to advocate that this area under one control shall be a whole industry. 

 Nor must we forget that here in England we have in the last twenty years, 

 seen the formation of a Big Five in banking, a Big Four in railways, and 

 a Big One in chemicals — not to mention the policy of Bigs Limited in 

 electrical engineering, soap, sugar, matches, cement and so on. 



Economists have unfortunately not always been too clear as to whether 

 they meant firm or plant when they were discussing scale of industry ; 

 sometimes they really meant neither. When they do quote statistics they 

 are usually figures for size of plant irrelevant to the issue. For in dis- 

 cussing the trust problem or the newer problem of industrial leadership 

 or the phenomena of monopoly, duopoly and oligopoly, it is size of firm 

 that is relevant and in particular, the size of one, two or a few firms relative 

 to the whole industry. And in discussing laws of increasing return it is, 

 as we shall see, scope that is relevant — not size of organisation at all. 



It is one of the advantages of a statistical approach that the inquirer 

 must be quite certain what he is discussing. If he wants to know the 

 number of persons in a unit he must, if he is to get a figure, specify whether 

 it is the number in the plant that he wants or the number in the firm. 

 Certainly, if the type of theory most relevant to this issue of size is to be 

 ' checked ' by statistics it is essential to distinguish firm and plant. The 

 type of theoretical model I refer to is that of Mr. E. A. G. Robinson. In 

 his Structure of Competitive Industry, Robinson breaks down the 

 determinants of the optimum size of business unit into five forces : tech- 

 nical, managerial, financial, marketing, and risk. Finance and marketing, 

 he thinks, usually demand a larger, management a smaller size. The size 

 of an actual unit must therefore be a compromise. An industry requiring 

 detailed management may very well have small plants, but the financing 

 and marketing firm controlling several such plants can be large. Is this 

 differentiation of plant and firm a real and common situation ? 



Here the German Census of 1925 is particularly helpful since it deals 

 in separate volumes with plants (Niederlassungen) and firms (gewerbliche 

 Unternehmungen). The distribution of workers among various sizes of 

 organisation looks very different when the organisation referred to is a 

 firm rather than a plant. In German productive industry generally there 

 were in 1925 566,760 persons in large single, plaftts employing over 5,000, 

 but 1,442,039 persons in plants belonging to the large single ^n?z^ em- 

 ploying over 5,000. In my table, 33-6 per cent, of all persons occupied 

 in German productive industry ^^ are shown to be occupied in large plants 

 employing over 500. The corresponding percentage for those occupied 

 in IdLVge firms employing over 500 is 41-6 per cent. In commerce and 

 transport the disparity between the numbers employed according to 

 whether large scale refers to large plant or large firm is still more marked. 

 There were only 23,620 persons in very large commercial and transport 

 plants, e.g. shops and stations employing over 5,000, but, with the same 



2^ Subtracting those in plants employing ten or less. 



