In the present paper we try to investigate both statistically and analytically the problem of the relative share of manual labour in the national income. There are two reasons why we do not consider the total share of labour, although it would be more interesting from the social point of view. (i) The statistics of national income include in the salaries the incomes of directors, managers, etc., which should rather be placed under the heading of profits. In this way, what statistics give as the total share of labour does not correctly represent the distribution of the product of industry between profits and interest on the one hand and wages and salaries on the other. (ii) The relative share of manual labour in the national income is more suitable for theoretical analysis.
It must be added that we shall deal here with the relative share of manual labour not in net but in gross income, by which is meant the income before deduction of maintenance and depreciation. (Gross income = net income + maintenance and depreciation.) The reasons for this are again both 'technical' and theoretical. (1) The statistics of gross income are much more reliable than those of net income; the latter are based on the net incomes of firms whose allowance for depreciation1 is certainly more or less arbitrary. In adding to the net national income aggregate depreciation, as given by the firms' accounting, we obtain gross income free from this arbitrariness. (2) It is the share of manual labour in the gross national income which--as we shall see below--ex natura rei can be more easily treated theoretically.
1. The figures for Great Britain are based on Professor A. L. Bowley's The Change in the Distribution of the National Income,____________________