Knowledge that Transforms

THE allocation of resources in a planned economy is a problem which has engaged the attention of many writers in recent years.2 Theoretical treatment of the question has emphasized mainly the static solution, i.e., the organization of production to maximize the satisfaction of wants, or the output desired by the State, given a fixed volume of resources, labor force, and level of technology. It has generally been held that the rate of capital formation either depended on an arbitrary decision by the planning authorities, or on the time-preferences of the society. Given some rate of capital formation, Western writers have agreed that investment could most effectively be allocated by equalizing the marginal net productivity of capital in all directions.3 Section II of this paper presents an abbreviated translation' of an analysis of this last problem by a Soviet writer, Professor Khachaturov, published in a I946 textbook 5 to be used in technical schools for engineering and operating personnel in railroad transportation. The general approach is ostensibly quite different from that referred to above, and should be of interest as a sample of Soviet economic thought.6 Statements in the text indicate that this approach has actually been used by Soviet planning organizations. The author of the text also proposes certain conceptual innovations. It appears doubtful whether they have been, or will be, accepted in the USSR.7 The following section contains the gist 8 of the argument as presented by Khachaturov. In sections III, IV, and V, below, an attempt is made to relate the Soviet approach to familiar capitalist concepts, analyze the proposals made, and assess their operational feasibility.

IN THE three and a half years since Hiroshima, considerable loose talk about atomic power has circulated among both scientists and laymen. At first people envisioned virtually costless power. Upon more sober reflection they realized that even if the cost per kilowatthour of the atomic fuel were to approach zero other important items in the cost of generating power would be far from zero.' But though this realization has become general, what discussion there has been of the cost of atomic power has proceeded on an elementary and unrefined basis. As yet the estimates which have been made of the costs of generating electricity in atomic stations are highly tentative. All existing nuclear piles have been constructed for non-power purposes, and there is great uncertainty about both capital and operating costs of the pile, chemical and metallurgical plants, and heat exchanger of the future atomic plant. Little if any significance, therefore, should be attached to the numerical estimates of costs of atomic power used in this paper. Our purpose is to discuss the implications of what can now be perceived about the probable characteristics of the costs of atomic power plants and to bring out certain economic relations which will be of importance when, through accumulated experience, nuclear engineers are in a position to make fairly exact cost estimates.

Merton Miller produce some interesting propositions concerning the degree to which cyclical fluctuations may be mitigated through the automatic response of tax revenues to changes in national income.' In the case of the individual income tax, it is possible to go further and derive certain upper bounds to the degree to which the yield of such taxes may be made responsive to income fluctuations, and thus demonstrate even more severe limitations

R EFERENCES to the velocity of circulation of money sometimes imply that movements of V are accurate guides to shifts in the aggregate-expenditure function. Among the varieties of economic analysis in which identity of velocity changes and aggregatedemand-schedule changes has been implicitly assumed are controversy concerning the validity of the lack-of-investment-opportunity theory of business downturns2 and speculation concerning the behavior of velocity in a Iooper-cent-reserve monetary system.3 As various economists have pointed out, however, changes in V are not necessarily accurate guides to changes in the aggregate-demand schedule. In a fractional-reserve system, movements of the investment and consumption functions may be expressed by expansion and contraction of M as well as by rises and falls in V. All that is required in the case of M's expansion is that excess reserves exist; and not even that is required when 11f falls. The implications that these conclusions have for statistical arguments concerning the lackof-investment-opportunity theory of recessions are obvious. Even if records were to show that