JNFLATION is colloquially described as a situation in which the flow of purchasing power is increasing faster than the flow of goods and services with consequent price increases. Most remedies for inflation stress contraction of the flow of purchasing power. In this article I shall argue that these remedies often need to be supplemented by and sometimes even replaced by measures to increase the flow of goods. In the United States, inflation is intimately bound up with the wage-price spiral, and persistent inflationary tendencies appear to present a difficult, if not insoluble, dilemma. Undoubtedly wage and price increases can be prevented by a sufficient contraction of demand. But contraction of demand may have to go so far as to conflict with the political imperative to maintain reasonably full employment. The authorities may thus face a serious conflict between the objectives of full employment and price stability. They may be forced to turn their remedies on and off in the hope that by pursuing first one objective and then the other a reconciliation may be achieved. I believe that the dilemma is less acute than it appears and that a resolution may be found if the inflation problem is handled in the context of a growing economy. But to do this a combined use of the available control instruments, and possibly of additional ones, is needed.'

shall escape a serious wage-price spiral, leading to collapse, within the coming year and a half. In effect, we believe, this is to assume that throughout this period union wage demands are kept within very moderate limits. No doubt it is incorrect to say that the inflationary pressure on wage levels observable at full employment results exclusively from union behavior; we know from the experience of previous booms that competition will generate it in markets, and the evidence of the past fifteen months suggests that this will be true even under price ceilings. But powerful unions enhance the pressure immeasurably in at least two ways: ( i ) they tend to force increases much larger than those which would take place under competitive conditions in factor and product markets; and (2) they intensify the speed and magnitude of any reaction of wages to a rise in prices, when the latter are free to rise. With price controls rapidly going by the boards, they will now be able to move along both avenues. At the present writing (November, I946), among the agents making for rapid short-term inflation they seem to be the most important as they are the least controllable. If they run wild, the inevitable collapse might create a need for a strong reflationary policy.

IN THIS paper some of the implications of the approach to the problem of interest in a capitalist economy will be examined. A discussion directed toward interest may seem an anachronism at a time when, in discussion of the recent past, the present, and the future, the determination of the size of the national income has assumed a position of central importance, and instruments for increasing income are available which are admittedly more potent than manipulations of interest rates. But the magnitude of interest rates may have an important bearing on the extent to which private enterprise, and in particular private investment, will occupy the stage in the future. Although conditions are conceivable under which the interest burden of the national debt would not hamper private enterprise, it seems improbable that they will be realized, and I think it can be safely assumed that changes in the interest cost of the national debt will have a significant effect on the rate of private investment. Further, although Dr. Tinbergen I has found that long-term interest rates have had but a moderate effect on investment in the past and have been overshadowed in importance by profit fluctuations in the case of industry and by the shortage and abundance of houses in the case of residential building, it may well be that if a stable national income is achieved in the future, interest rates may be a factor of prime importance in determining the rate of private investment. We need not dwell for long on the question whether interest is a monetary phenomenon in the sense that to assume the absence of money is to assume away one essential determinant of interest. That question has been conclusively answered in the affirmative by the work of Professor Schumpeter and of Lord Keynes.2 Professor Schumpeter has shown that uncertainty is inherent in the process of capitalistic development; that profits are the prizes of the few who plunge successfully into the unknown; and that if the future were certain not only would the genius of the entrepreneur not be required, but existing profits would be encroached upon and finally disappear. Furthermore, in order to exploit their opportunities, entrepreneurs require command over money which enables them to obtain resources for investment either by raiding the current flow of production or by increasing it. Lord Keynes' analysis complements Professor Schumpeter's on the supply side: in order to take advantage of unforeseeable possibilities of profit or to guard against unforeseeable risks, people desire to hold liquid resources, and will surrender them only if they receive sufficient compensation in the form of interest. Since uncertainty is an ever present phenomenonif all else were assumed away, the uncertainty of human mortality 3 would remain liquidity preference is an essential ingredient of any economic situation, and therefore of any valid theory of interest. To assume the absence of money as a store of effectively, uncertainty must be assumed away, since, as Keynes has shown, people will use substitute stores of value if deprived of dollars. And if uncertainty is absent, we are confronted with Schumpeter's compelling argument that, under those conditions, there is no room for the phenomenon of interest. Since interest is a monetary phenomenon in the sense of the last paragraph, it follows that there must be some relation between the rate of interest and the quantity of money, and that, within limits, permanent effects on interest and output can be produced by changes in the quantity of money. The determination of these effects will be the subject of the following sections. Section II will follow the argument of the General Theory, but will, I hope, escape the stigma of vain repetition by suggesting some improvements of Keynes' model and by