The European centrally planned economies (CPEs) have sustained chronic hard currency deficits since East-West trade began to expand in earnest about fifteen years ago. While their outstanding hard currency debts almost doubled over 1975-76 as a result of an inability to adjust quickly to the Western recession-a previously unsuspected vulnerability-other systemic factors rooted in Stalinist central as practiced in the CPEs, have been responsible for the more secular balance-of-payments problems.' I refer to the wide use of direct controls to allocate intermediate products, the prevalence of or over full-employment planning, and irrational domestic pricing. These have several implications for economic performance which are relevant to the CPEs hard currency balances of payments. First, the CPEs tend to produce relatively low quality manufactured products and have a marked inability to sell their products in Western markets. Inability to compete successfully is not due to price, but, to quote a Hungarian economist, Imre Vajda, to deficiencies in performance, reliability, appearance, packing, delivery and credit terms, assembling facilities, after-sale services, advertising, selling itself , primarily factors other than price . (p. 53). This ineptness results largely from lack of competition-the fact that domestic products are distributed by the plan rather than sold and that quantitative goals take precedence over qualitative goals. Further, taut planning results in sellers' markets, additionally weakening managerial incentives to improve quality. Nor does play a significant role in intrabloc foreign trade. This trade is characterized by large state trading agreements, protected markets, and little or no direct contact between the producing enterprise in one nation and consuming enterprise in the other. Second (and related) is the well-known relative weakness of socialist nations in innovation and technological change. This is due to the absence of competition just noted, to rewards for innovation which are inadequate to offset the risks or overcome inertia, and to the dysfunctional organization of R&D establishments and their relations to operating enterprises. Third, the CPEs trade with each other and with the West at roughly world prices, even though these prices usually have no organic or consistent relationship to domestic prices. Their exchange rates serve as units of account but not as real prices. Their currencies are not only totally inconvertible into each other, they are also largely inconvertible into goodsso-called commodity inconvertibility (see the author, 1978). That is to say, foreign importers (exporters) are not allowed to compete freely with local enterprises for products (markets) because this would disrupt the plan. This significantly reduces short-run ad hoc exports-most exports have to be planned long in advance. These factors lead to at least three causes of persistent hard currency shortages: 1) the *Professor of economics, Tufts University and associate, Harvard Russian Research Center. Some of the ideas in this paper appeared earlier in my 1973 article. A much longer current version is available on request. I am indebted to Abram Bergson for incisive criticisms of two earlier drafts. 'Other than systemic factors may also be responsible. For example, the current availability of Western investments and credits on reasonable terms and the present willingness of the CPEs to entertain such relations with the West is one such factor. It should also be noted that the LDCs and some advanced industrial nations also have chronic balance of payments problems. However, I argue that the factors to be mentioned below are unique to the CPEs.

Utilizing a cross-sectional trade flow model of the type developed by Hans Linnemann and Jan Tinbergen, this study attempts to isolate empirically the major forces which have shaped European trade relations over the period 1951-67. We first estimate via the use of dummy variables the impact of the European Economic Community (EEC) and the European Free Trade Association (EFTA) on member trade. For each year of the European integration period (1959-67), a crosssectional equation is estimated ancl used to test for the existence and approximate size of the respective integration effects. The equation is also calculated for the eight years prior to the integration period to obtain a clear picture of the forces which were at work before the formation of the EEC. Secondly, a base year equation is used to make projection estimates of the gross trade creation and European trade diversion effects of the two communities.

In a recent article in this Review (1984), Samuel Hollander suggests that the secular path of real wages in the Marxian model tends toward a subsistence wage at which population growth ceases. He attributes this decline to a higher growth rate in population relative to a positive but decreasing rate of growth in the demand for labor power. From this he contends that, in contrast to the Malthusian prescription, Marx . . is open to the objection that, with no check at all to the population growth rate, the deterioration would have been sharper still (p. 148). The implication of Hollander's argument is clear enough, for if accepted, it would necessarily lead one to dismiss Marx's claim that: ....every special historic mode of production has its own special laws of population, historically valid within its limits alone. An abstract law of population exists for plants and animals only, and only in so far as man has not interfered with them (Capital, I, 1967, p. 632). This comment takes issue with the thesis advanced by Hollander on two major points. First, the secular decline in the value of labor power is a result of the increasing productivity of labor rather than the divergence between the respective growth rates in population and the demand for labor power. Second, this is perfectly consistent with constant or even rising absolute real wages (price of labor power) for the active part of the working class.1 Thus, population control is neither a necessary nor a sufficient condition in assuring against falling real wages for a changing social productivity of labor. Put differently, the supply of labor power to the advanced capitalist sector is the crucial supply variable -not the rate of population growth. I begin by reviewing Hollander's useful distinction between Marx's value of labor power and the classical school's minimum subsistence wage. Next, it is shown that the value of labor power is determined by the social productivity of labor rather than factors exogenous to the system (population). The discussion is brought to a close by focusing on the prime mover of the path of relative real wages in the Marxian model: the endogenously determined supply of labor power (surplus population) to the advanced capital sector.

Originating with James Tobin's initial treatment of money as a second asset in the Solow one-sector growth model, the subject of money and growth has received a great deal of attention in the recent literature. Tobin's emphasis on portfolio balance to determine the equilibrium of the model provides a useful framework for the discussion of the development of two opposing schools of thought among recent writers on the subject of money and growth. The neoclassical approach follows Tobin in his emphasis on portfolio balance and includes contributions by Miguel Sidrauski and Harry Johnson, among others. A cogent and comprehensive statement of the neoclassical viewpoint can be found in the excellent survey by David Levhari and Don Patinkin. The second approach the Keynes-Wicksell approach --as expounded by Jerome Stein in particular, and also including contributions by Hugh Rose and Keizo Nagatani, faults the neoclassical model on two basic and related points. They are the implications of the model for the dynamics of price change, and the lack of independent savings and investment decisions. The neoclassical approach is characterized by the assumption that desired per capita real balances are always held-prices must adjust instantaneously to assure portfolio balance. A given rate of expansion of the nominal money stock combined with the exogenously given rate of population growth serves to determine the equilibrium rate of inflation consistent with asset equilibrium. The division of assets between money and physical capital is thereby determined, and there need be no specification of an independent investment function all physical savings are instantaneously channelled into capital accumulation, and the desired capital stock is always held.' The long-run properties of the neoclassical model allow for the coexistence of nonzero steady-state inflation and goods market equilibrium by specifying that excess demand for goods causes a departure from the steady-state rate of inflation, but is not a necessary condition for a nonzero inflation rate at any point in time. It is contended that, in a dynamic world, there are two forces operating to drive the price level-excess demand for goods and inflationary expectations. In steady state, excess demand is zero; actual inflation equals expected inflation, not necessarily zero; and the possibility of nonzero steady* Queen's University. This paper was written while I was a graduate student at the University of Chicago. My understanding of the issues has been greatly improved by many discussions with Rudiger Dornbusch, Stanley Fischer, and Michael Mussa. I also wish to thank members of workshops at the University of Chicago, University of Rochester, and York University, and to participants in the Chicago Symposium on Trade, Growth, and the Balance of Payments (University of Chicago, December 1970) for helpful comments on an earlier version of this paper. Jerome Stein, George Borts, and an anonymous referee provided very useful comments for which I am most grateful. Of course, I am responsible for any remaining errors. Financial support from the Canada Council is gratefully acknowledged. I A puzzling result of Tobin's initial treatment is that the introduction of money into the barter model lowered the capital intensity and output per capita. I have recently tried to analyze this seemingly paradoxical result elsewhere (see Purvis).

A common finding of virtually all previous studies of black-white earnings differences is that the earnings of white males appear to rise at a more rapid rate with each additional year of work experience (or with age) than do the earnings of black workers. This finding has been noted by economists since at least the mid-1950's when Morton Zeman called

The U.S. policies toward the USSR in the postwar period have been less friendly than those of our NATO allies. These policies have been, however, a largely understandable reaction to the miliary posture and undemocratic domestic policies of the USSR and to the nature of its economic system. But the situation is changing rapidly. The Gorbachev Reform is already a minor social, political, and economic revolution, and may well turn into a major one. These circumstances require a reassessment of our policies toward the Soviet Union. Some reassessment has already taken place as indicated by our signing the INF Treaty. Nevertheless, the overall tone of our administration is dominated by attitudes of skepticism and show me when hardly a week goes by that the Soviets do not show by saying or doing something that two or three years ago was unthinkable. I think it is important for us to adopt a more positive approach toward the Soviet reform. Our major foreign policy goal over the past forty years has been containment of the Soviet Union. This goal has been enormously expensive to implement, and pursuing it bears much of the responsibility for the difficult economic problems this nation faces today. The relentless pursuit of this goal has been based on the assumption that, while uneasy truces between the two camps might be worked out from time to time, the differences between us are irreconcilable over the long run. Now, for the first time in Soviet history, this assumption may no longer be valid. There is little doubt that Gorbachev genuinely wants to eliminate many of those aspects of the Soviet system that we consider objectionable and to end the cold war. His task will be difficult and good relations with the West would simplify it. Gorbachev's policies are in our interest. The opportunity should not be missed-it may not come again for decades.

The purpose of this paper is to settle the controversy surrounding consumer's surplus' and, by so doing, to validate its use as a tool of welfare economics. I will show that observed consumer's surplus can be rigorously utilized to estimate the unobservable compensating and equivalent variations-the correct theoretical measures of the welfare impact of changes in prices and income on an individual. I derive precise upper and lower bounds on the percentage errors of approximating the compensating and equivalent variations with consumer's surplus. These bounds can be explicitly calculated from observable demand data, and it is clear that in most applications the error of approximation will be very small. In fact, the error will often be overshadowed by the errors involved in estimating the demand curve. The results in no way depend upon arguments about the constancy of the marginal utility of income. Consequently, this paper supplies specific empirical criteria which can replace the apologetic caveats frequently employed by those who presently apply consumer's surplus. Moreover, the results imply that consumer's surplus is usually a very good approximation to the appropriate welfare measures. To preview, below I establish the validity of these rules of thumb: For a