¹ In our example there is an inverse relationship between price ratios and quantity ratios. Such a relationship is a common characteristic of comparisons involving two economies at a point in time, or the same economy at different points in time. The explanation lies in the tendency for commodities produced on a relatively large scale to be cheaper than those produced on a relatively small scale. But an inverse relationship between price and quantity ratios is not necessary for a divergence in the product ratios. For this, what is necessary is that both the price ratios and the quantity ratios be different for the various products.

² Since there are some important things wrong with Soviet prices as measures of relative opportunity costs, it may be argued that the comparison in ruble prices is misleading. But whatever the inadequacies of Soviet prices, no one would claim that US relative prices are proper indicators of Soviet opportunity costs. Second, it is possible to make certain adjustments in Soviet prices which make them more meaningful for these purposes.

³ That is, there is no economic meaning which can be attached to the average ratio. Attempts have been made to attach such meaning but, to the writer's knowledge, none of these efforts has been successful.

Since it is sometimes claimed that the average, in effect, neutralizes the counteracting biases of the primary ratios, it is necessary to say a word here about the nature of these “biases.” Without going into technical detail, suffice it to say that such biases do exist and they result from our unavoidable need to approximate the rates at which resources are mutually substitutable at capacity levels in a wide range of alternative possible output combinations by means of relative prices which refer to only one historical point in that range of combinations. Nevertheless, this does not certify the correctness of the average ratio. In the first place, in this context, the average ratio still has no claim to economic meaning but only to statistical expediency. Second, there is no assurance that the average ratio can fulfill even this limited role of statistical compromise, for we know little about the shape of the curve of alternative output combinations in either economy and, hence, of the direction and magnitude of the biases to which the primary ratios are subject.

⁴ For statistical reasons, the usual procedure is to employ a geometric rather than arithmetic average. However, the same objection applies to use of either average.

⁵ “Comparison of Soviet and United States National Product,” Comparisons of the United States and Soviet Economies, Washington, D.C., 1959, 11, p. 385.

⁶ For example, Allen W. Dulles, head of the Central Intelligence Agency, testified in hearings before the Joint Economic Committee of Congress on November 13, 1959, that Soviet GNP in 1958 was estimated by CIA as 45 per cent of US GNP in the same year (Hearings Before the Joint Economic Committee, 86th Congress, 1st Session, November 13–20, 1959, Washington, D.C., 1960, p. 4). I want to point out that Professor Bornstein's understanding and treatment of the index-number problem in Soviet-American comparisons are above reproach. I have used his estimates as the basis of the discussion, not as an example of the practices I am criticizing.

⁷ Ibid., p. 9.

⁸ Three per cent is the approximate US average rate of growth over the past half-century, as well as of the last nine years; 7 per cent is CIA's estimate, according to Mr. Dulles, of the Soviet growth rate for 1950–1958.

⁹ The Soviet claim is that the national income of the USSR is now about 60 per cent as large as ours, both defined according to the Marxist concept, and that growth rates of 8 per cent for the USSR and 2 to 3 per cent for the United States mark the crossover dates as 1968 for total output and 1970 for per capita output. (Index-number problems rarely intrude on Soviet-made projections.) See, for example, Aganbegian, A., “Dogna' i peregna' SShA po urovniu proizvoditelnosti truda” [Overtake and surpass the USA with respect to the level of productivity of labor], Sotsidisticheskii trud, No. 4 (1959), p. 22.Google Scholar

¹⁰ Heymann, Hans Jr, “Problems of Soviet-United States Comparisons,” Comparisons of the United States and Soviet Economies, 1, p. 10.Google Scholar

¹¹ Howard C. Peterson (Committee on Economic Development), “Soviet Economic Growth and United States Policy,” ibid., 11, p. 519.

¹² “Putting First Things First,” Foreign Affairs, XXXVIII, No. 2 (January 1960), p. 200.

¹³ This point is explicitly recognized at an earlier point in the Brief, only to be ignored in the passage under discussion.

¹⁴ Bornstein estimates that, in ruble prices, defense consumed only 4 per cent of the US GNP in 1955. Advocates of larger US defense budgets might be interested in this comparison for public relations purposes.