ABSTRACT
Some results of great importance in economic theory can be established as empirically valid at any rate over the period studied:
(i) There is a numerical relationship between changes in the value of output of investment goods, and changes in the national income as a whole, corresponding to a multiplier of about 2, subject to a fairly rapid secular increase in the propensity to consume.
(ii) In view of the above, and in view of the fact that the principal source of savings is companies' undistributed profits, it is interesting to find that the "propensity to pay dividends" on the part of companies is 0·46 of each increment of profits, the remainder being saved.
(iii) Outlay by the working classes, subject to a fairly constant rate of saving, closely follows their income. But expenditure by the more prosperous classes shows a delay of nearly two years in following movements of income, and does not fall so violently as income during a period of slump, owing to the maintenance of their customary standard of living by spending capital.
(iv) It is generally held that during a period of recovery the price of investment goods rises relative to that of consumption goods, and that it falls during a period of slump. The exact opposite is the case.
(v) The average real labour costs of production per unit of output in industry fall strongly with increasing output, the proportion of overhead labour being high. Marginal labour costs, however, also show a tendency to fall with increasing output. About 1930 the whole curve of labour costs per unit of output was shifted to a lower level, which has since been maintained.
(vi) In non-manufacturing industry constant and not increasing returns prevail.
xix(vii) The amount of investment in fixed capital largely depends on the ratio between the current level of profit per unit of output and the price of new capital goods multiplied by the rate of interest, subject to a lag of about one year.
