This is the basis of the doctrine of the ‘foreign trade multiplier’, according to which the production of a country will be determined by the external demand for its products and will tend to be that multiple of such demand which is represented by the reciprocal of the proportion of internal incomes spent on imports. This doctrine asserts the very opposite of Say’s Law: the level of production will not be confined by the availability of capital and labour; on the contrary, the amount of capital accumulated, and the amount of labour effectively employed at any one time, will be the result of the growth of external demand over a long series of past periods, which permitted the capital accumulation to take place that was required to enable the amount of labour to be employed and the level of output to be reached which were (or could be) attained in the current period.
Keynes, writing in the middle of the Great Depression of the 1930s, focused his attention on the consequences of the failure to invest (due to unfavourable business expectations) in limiting industrial employment below industry’s attained capacity to provide such employment; and he attributed this failure to excessive saving (or an insufficient propensity to consume) relative to the opportunities for profitable investment. Hence his concentration on liquidity preference and the rate of interest, as the basic cause for the failure of Say’s Law to operate under conditions of low investment opportunities and/or excessive savings, and the importance he attached to the savings/investment multiplier as a short-period determinant of the level of production and employment.
On retrospect I believe it to have been unfortunate that the very success of Keynes’s ideas in connection with the savings/investment multiplier diverted attention from the ‘foreign trade multiplier’, which, over longer periods, is a far more important and basic factor in explaining the growth and rhythm of industrial development. For over longer periods Ricardo’s presumption that capitalists only save in order to invest, and that hence the proportion of profits saved would adapt to changes in the profitability of investment, seems to me more relevant; the limitation of effective demand due to oversaving is a short-run (or cyclical) phenomenon, whereas the rate of growth of’external’ demand is a more basic long-run determinant of both the rate of accumulation and the growth of output and employment in the ‘capitalist’ or ‘industrial’ sectors of the world economy.
Capitalism and industrial development: some lessons from Britain’s experience, Camb. J. Econ. (1977) 1 (2): 193–204, link