Knowledge that Transforms

IN the postwar discussions of the dollar problem, a number of writers have pointed to differences in national rates of productivity growth as one of the factors which has contributed to the imbalance in the world economy. It is the purpose of the present paper to study the validity of this argument. Clearly, such a study cannot be undertaken without including some consideration of relative movements of money wages. To use a shorthand expression, a country's competitive position in the world economy is, with given exchange rates, determined by the ratio between its productivity and its money wage rates. Consequently, both of these factors must be dealt with explicitly. The discussion of money wages will be undertaken from two points of view. In the first place, we shall analyze different norms for monetary and wage policies under conditions of divergent rates of productivity growth. Secondly, we shall discuss briefly whether the postwar commitments to a full employment policy in a number of countries have introduced a new element into the situation.

QTATE per capita incomes are computed by dividing estimated income payments to the residents of a state by the state's total population at midyear.1 This computation implicitly assumes that the denominator, total population, does not vary from one state to another in any important way. Yet it is known that states vary with respect to family size and composition, birth and death rates, participation in the labor force, and a host of other demographic attributes. It is the purpose of this paper to investigate the extent to which state differences in two of these attributes age composition and participation in the labor force -help to explain differences in state per capita incomes. Attention is confined largely to describing the direct effects on the distribution of per capita incomes among states of using alternative denominators that take account of variations in age composition and labor force participation. The study is confined to the three years, I930, I940, and I950 for which both Census data and official state income payment estimates are available.

Mr. Wright, in his presentation of the Sterling Area's gold and dollar accounts for the period I946-52, finds it convenient to include aid receipts by the United Kingdom and the overseas members of the Sterling Area along with their net balances from all other gold and dollar transactions, since his purpose is to draw attention to individual Sterling Area members' relationships to the centrally-held gold and dollar reserves. This method of presentation, naturally enough, obscures the great changes which have taken place during the postwar years in the Sterling Area's gold and dollar accounts before taking account of its members' receipts of dollar aid.' Data for I953 and I954 have been published since Mr. Wright prepared his tabulation of the Sterling Area's postwar gold and dollar accounts (Table I accompanying his article). When these two years are added to the record, and when Mr. Wright's figures are re-arranged to segregate the Sterling Area's receipts of dollar aid from the gold and dollar transactions exclusive of aid, it will be found as a first approximation that the nine postwar years can be most satisfactorily viewed in terms of the four clearly differentiated periods shown in Table i. Excluding its receipts of aid, the Sterling Area as a whole incurred a cumulative net gold and dollar deficit of $8.3 billion over the four years I946-49.2 The second period covers the year and a half from the beginning of I950 to mid-i95 I, when the Sterling Area had a large gold and dollar surplus (before counting aid receipts) of $I.2 billion, owing mainly to the favorable effects on its balance of payments of the realignment of world currency values toward the end of I949, a tightening of dollar import restrictions, and an inflow of dollars associated with the spectacular rise in Sterling Area raw material export prices which began shortly after the outbreak of fighting in Korea.3 The third period covers the twelve months from

i. The problem. In the mid-I954 forecasting discussion a strong reason for optimism was found in the fact that while late-I953 and earlyI954 sales were down and inventory up, there was no substantial reduction in planned investment in plant and equipment for I954. Here is a behavior pattern that has been studied very little. But the length of the time unit used in induced-investment planning is an extremely important aspect of business behavior, for it can be shown to be crucial for the stability of macro-economic equilibrium. To show this is the purpose of the present paper.

outside the range of attainment once full employment is reached. For even discarding the oft-used but spurious (for the problem at hand) historical average growth rate of 3-3Y2 per cent, the ceiling rate (with little inflation) seems to be in the neighborhood of 6-7 per cent. But if income cannot grow at the (adjusted) required rates, neither can induced investment, which by definition is dependent upon income growth for its achievement. At the very least, then, the accelerator is an extremely doubtful generator of growth. It is with such thoughts in mind that we assert (or reassert) the importance of innovational (autonomous) investment in providing, on the one hand, the principal generating force of periods of sustained growth, and on the other, because of its inherent irregularity, the primary source of unstable growth known as the (major) business cycle.