Vladimir G Treml, "Soviet Dependence on Foreign Trade," in NATO, Economics Directorate, EXTERNAL ECONOMIC RELATIONS OF CMEA COUTRIES, Brussels, Belgium, 1983 pp. 35-47 Footnotes, bibliography and appendix omitted 1. Introduction: Scope of the Paper The purpose of this paper is to summarize briefly the changing role of the foreign trade sector in the Soviet Economy as perceived from various measures in internal values. The main conclusions and estimates are derived from the ongoing work on the Soviet economy in the framework of input-output analysis conducted at the Center for International Research, US Bureau of the Census, and at Duke University. In the last 10-15 years the foreign trade sector proved to be the most dynamic element of the Soviet economy-in terms of rates of growth, which have by far exceeded the rates of growth of national income and of other national aggregates, in terms of significant shifts within export and import flows, and in significant changes in the relations between the foreign trade sector and the rest of the economy. As will be shown in greater detail below, the overall trade participation ratio, i.e., the ratio of the sum of merchandise exports and imports to national income (Net Material Product) measured in established prices rose from 12.3 percent in 1960 to 14.9 percent in 1970, and to about 27 percent in 1980. We can expect that as a result of the 1982 price reform and continuous fast growth of trade the participation ratio will rise to well above 30 percent in the early 1980s. Important shifts, rather unfavorable from the Soviet long-term perspective, occurred in. the commodity structure and in relations between exports and imports. Thus, from the position of net exporter of grain and with an almost balanced trade in meat products in the 1960s, the USSR has become a major net importer of both. An equally significant shift was effected in metals, where the USSR had long been a major net exporter: in the mid1970s the country became a net importer of ferrous metal products, and by the late 1970s the net importer's position was extended to the entire metal group, i.e., ferrous and non-ferrous metals and ores. Lastly, it should be noted that while the USSR was always a net importer of machinery, the extent of dependency on imports has also risen: in the early 1970s the ratio of imports to exports, measured in either external or internal prices, was about 1.4, and by 1980 it had risen to over 2. 1. All of these developments were counterbalanced by a rapid increase in Soviet exports of oil, gas and military hardware. The Soviet economy is also becoming a more open one in a different sense as the traditional complete separation of external and internal prices has gradually been changing. While the Soviet ruble remained inconvertible, direct and indirect links have emerged between world market and domestic prices on both the export and the import sides. All of these developments have, in my opinion, changed the Soviet Union from a position of an almost totally closed economy with a minimal exposure to world markets to a relatively open one with a high degree of dependence of foreign trade. This study is focused on the role of total Soviet foreign trade without differentiating between trading blocs such as CMEA or the "West". The reason for this is that the arguments of this study are offered in terms of various measures in domestic Soviet prices; we simply do not have the necessary data to separate Soviet foreign trade data in domestic values into trade with different partners. Pilot studies suggest that as a rule Soviet imports from developed Western countries comprise a higher share of total imports when measured in domestic prices than when measured in external prices, but much additional research will be required before we can separate Soviet foreign trade flows by blocs. Trade data differentiated by countries would clearly be needed for any analysis of Soviet vulnerability, to denial measures or economic sanctions. But such analysis would be incomplete and inconclusive without some understanding of secondary Soviet dependence, i.e., dependence on imports from CMEA countries which in turn are dependent on imports from the West. For instance, in 1980 the USSR imported about 15 billion rubles worth of machinery of which about 35 percent came from the West and 65 percent from CMEA countries. However, CMEA countries have, in turn, imported more than 5 billion rubles worth of machinery from the West, a large share of which consisted of components and parts. Since we do not know what proportion of these components was necessary to sustain the CMEA imports to the USSR we cannot completely assess Soviet dependency on Western technology. 11. Soviet Foreign Trade in Domestic Prices and Trade Participation Ratios Western and Soviet studies and analysis of Soviet foreign trade and its relation to the rest of the economy have been long handicapped by the absence of necessary data. Soviet foreign trade handbooks, Vneshniaia torgovlia, annually publish a fairly comprehensive set of foreign trade statistics in physical units and in value terms. Unfortunately, the latter are given in so-called "foreign trade" values, i.e., prices in foreign currencies actually received or paid by the USSR converted to rubles by means of officially set foreign exchange rates. The true domestic prices, that is, prices received by Soviet producers of goods for export and Soviet buyers of imported goods have, as a rule, no fixed relationship to "foreign trade" prices. The ratio of domestic to "foreign trade" prices has ranged in the past from 0.2 to 10.0. In an early pioneering study Franklyn Holzman explained the differences between "foreign trade" and domestic prices and estimated the latter for a sample of Soviet exports and imports for the late 1950s, producing aremarkably accurate estimate of the domestic value of Soviet foreign trade. However, no comprehensive effort has since been made to estimate these values or to inquire into the changing rules governing the setting of domestic prices to Soviet exports and imports. As a result Western authors continued to believe that the Soviet economy is largely self-sufficient and that the trade participation ratios range from 3 to 5 percent each for exports and imports. Table 1 summarizes the data on Soviet foreign trade measured in foreign trade and in domestic prices and the shares of exports and imports in domestic prices as a percent of Soviet national income. Table I Soviet Exports and Imports in Foreign Trade and Prices, Selected Years (in millions of roubles) Exports Imports Foreign Domestic % Foreign Domestic % prices prices of prices prices of NMP NMP 1955 3,084 2,960 3.00 2,755 5,344 5.43 1960 5,007 5,307 3.67 5,066 12,000 8.28 1965 7,357 8,387 4.34 7,253 15,740 8.14 1970 11,520 18,300 6.31 10,559 24,919 8.60 1971 12,426 18,639 6.11 11,232 26,508 8.69 1972 12,734 17,819 5.68 13,309 31,375 10.01 1973 15,802 20,227 5.99 15,544 37,927 11.21 1974 20,738 23,227 6.56 18,829 41,612 11.70 1975 24,034 22,900 6.30 26,671 54,400 14.97 1976 28,022 24,659 6.39 28,733 60,914 15.79 1977 33,255 27,269 6.72 30,093 65,001 16.01 1978 35,670 28,893 6.84 34,554 78,438 18.51 1979 42,426 30,100 6.83 37,881 82,600 18.71 1980 49,634 31,800 6.88 44,463 92,300 19.98 Sources and Notes: 1. Data on exports and imports in foreign trade prices and on national income underlying the calculations of shares are from standard Soviet sources. 2. Exports and imports in domestic prices are from Treml and Kostinsky (1982, p. 15) with estimates for 1979 and 1980 made on the basis of the methodology developed in this study. 3. Shares of exports and imports in domestic prices measured as percent of national income. -------------------------------------- The values in domestic prices were estimated on the basis of a number of Soviet sources of a varying degree of reliability and were verified, whenever possible, against other data. By and large these estimates appear to be reasonably accurate, except for the years after 1975, where the margin of possible error is probably greater. Several comments would be helpful in interpreting the data. It should be noted that the shares shown in the table are based on the Soviet definition of national income, i.e., Net Material Product. Were we to use the Western concept of Gross National Product the export participation ratio for 1980 would be 5.00, and the import ratio would be 14.51. Rules governing the setting of domestic prices of foreign trade goods, i.e., domestic prices paid by final buyers of imports and received by Soviet manufacturers of goods for export are not fully understood by Western specialists and much more research will be required. A discussion of Soviet practices would be too long to be included in this paper but a brief summary of what we do know would be interesting particularly as it shows the increasing "openness" of the Soviet economy to'external market forces. The standard "textbook" description of setting of domestic prices for exports and imports is quite simple: producers receive the same price regardless of whether their products are earmarked for export or for domestic markets, and imported goods are valued at the level of prices of similar domestically produced goods. However, these rules are observed only in few cases and the implied independence of the internal Soviet price system from world markets is a fiction. It appears that domestic prices of a large share of intermediate goods and of all machinery and machinery components are set by applying fixed "adjustment coefficients" to the so- called "foreign trade", or "valiuta" prices. Thus, the domestic prices of these imports do not correspond to prices of similar domestically produced goods and vary proportionally to changes in external prices. The rules covering prices of imported consumer goods are apparently differentiated by product. Domestic prices of most basic imported staples such as sugar, meat, butter, flour and the like are set at the level of similar domestically produced goods and do not change regardless of changes in external prices. On the other hand, prices of most consumer soft goods, consumer durables, and prices of most processed foods such as coffee, cacao, wines, canned goods are changed to reflect changes in external prices although changes in these prices are sometimes introduced with a time lag and are not necessarily proportional to changes in external prices. Prices of most manufactured goods produced for export are not set at the level of prices of goods produced for domestic consumers for the simple reason that Soviet export goods are not identical to goods produced for domestic markets and are, as a rule, of higher quality. Accordingly, producers of export goods are compensated for higher costs associated with higher quality by specially fixed "export price supplements". Most of these rules or modifications of older rules have been in effect in the 1970s. The changing conditions in world markets are thus transmitted, directly or indirectly, to the Soviet economy through prices of foreign trade goods making it much more open to external influences than in the past. Of course, foreign trade is still very much a state monopoly in the USSR and Soviet authorities could always change these rules or block a particular change in external prices from affecting the domestic prices. An interesting aspect of the data shown in Table I and related to the price setting of foreign trade goods is the chronic imbalance between exports and imports when measured in established domestic prices, with imports being roughly twice the level of exports. A partial explanation lies in the Soviet accounting rules, under which exports are recorded in enterprise prices, i.e., prices net of turnover taxes, while imports are measured in final purchaser's prices including turnover taxes when applicable. However, an analysis of import values in 1972 showed that turnover taxes accounted for not more than 5 to 6 percent of the total value of imports and thus could not explain persistent and relatively high import surpluses. The low quality of most Soviet goods and the resulting strong demand for higher quality imported goods should be mentioned as another explanation for the imbalance. The main explanation probably lies with the inconvertibility of the ruble and the traditional Soviet policy of balancing foreign trade flows in external currencies. It can be shown that theoretically such a policy would always produce an excess of imports over exports in domestic values, which would be true regardless of whether the exchange rate is fixed by the USSR overvalues or undervalues the true parity ratio between domestic and world prices . III. Dependence on Trade: Sectoral Evidence and Leadership's Perceptions The degree of dependence of the Soviet economy on external markets measured in terms of a high and rising ratio of imports to national income is fully consistent with the dependence on imports observed in different sectors and branches of the economy. A comprehensive micro survey of this aspect of the Soviet economy is beyond the scope of this paper, but we can form a fairly complete picture of the role played by imports in large aggregates, such as investment, and in a sample of sectors and branches. The various ratios and shares presented here are based on statements and quantitative references found in the Soviet technical literature. It appears that around 1980 more than 20 percent of capital investment in machinery and equipment was of foreign origin, and the share of imports was at about the same level in the total installed stock of machinery. The degree of dependence on imports varied, of course, from sector to sector. Particularly dependent on imports was the chemical industry, where more than 60 percent of specialized machinery was of foreign origin, and the paper industry with more than 50 percent; in such industries as woodworking, textiles, food processing, and in ferrous metallurgy the share of imported machines was around 25 percent.' More than 60 percent of the tonnage of the Soviet merchant marine is of foreign origin and imports play a significant role in providing the USSR with buses and railroad rolling stock. But the degree of Soviet dependence on imported machinery cannot be properly assessed without full evaluation of the composition and type of machinery exported and imported by the USSR. It appears that the Soviet machine building industry, or at least its non-military component, has been falling behind their counterparts in the developed West and even the CMEA countries in terms of design, quality, durability, versatility, and other performance characteristics. This can be clearly seen in trends and in the composition of Soviet foreign trade in machinery. In a comprehensive study of the technological level of the Soviet industry Amann, Cooper, and Davies noted that as a rule unit values of machine tools exported by the Soviet Union are much lower than unit values of imported machine tools. This observation can be extended to a large share of all traded machinery. In 1977, for instance, out of 35 machinery groups which were both exported and imported and for which quantities and values were available, in the case of 30 groups unit values of imports were on the average four times higher than unit values of comparable exports, and only in 5 groups was the opposite true. In some cases low unit values of Soviet exports can be probably explained by export subsidies, political reasons, or discounting necessary to penetrate new markets but the overall numbers suggest significant differences in the technological sophistication and quality. We can conclude, therefore, that the Soviet machinery industry does not have the necessary capacity and technology to produce a whole range of more sophisticated, higher quality final machinery products, components, parts, and instruments, or is producing these in insufficient quantities. Thus, according to one study, "almost one-half of numerically-programmed machine tools is equipped with imported components"; while another report indicated that domestic industries could supply only onethird of the needed scientific instruments and gauges. " Analysis of Soviet import data suggests that this is true not only of "state-of-the-arts" high technology but of a variety of products such as ball bearings, telephone and electronic equipment, control mechanisms, pumps, compressors, and the like. Imported consumer goods and raw materials for their production play an equally important role. In the late 1970s-early 1980s imported final consumer goods comprised between 14 and 15 percent of the total value of goods sold in state retail trade. Imported processed foods and agricultural raw materials accounted for well over 20 percent of the total calorie intake of the Soviet people. It is rather difficult to evaluate how the dependence on imports has been perceived by the Soviet leadership in the last few years. The relative decline of the share of machinery in total Soviet exports, the rapid expansion of the export of fuels, and increasing reliance on imported Western technology were becoming sources of embarrassment to the leadership. Expressions of concern with the fact that the USSR was apparently reverting to the foreign trade pattern characteristic of a developing rather than a powerful industrial nation were becoming sufficiently vocal to require some explanation. Such an explanation was provided by the newspaper Sovetskaia Rossia in a special section titled, "They say that . . . but what is the true story?" in which the paper addresses and attempts to explain policy issues too sensitive for an official government statement. In one issue the the paper reported that "one hears talk of our country's inability to export technology and being therefore forced to export raw materials and fuels . . ." and another issue posed the question, "whether, indeed, as some say, we cannot avoid importing Western technology." Raising these issues in public was significant in itself. First and rather vague notes of concern about excessively large imports of machinery and equipment were sounded by Brezhnev and others in 1980. 15 It appears that the concern with dependence was reinforced by the fact that the rapid growth of machinery imports, particularly of Western imports, was not providing the expected payoff, or at least was not providing it in sufficiently visible terms.16 Among other contributing factors we should note that the economy was developing some problems in effectively absorbing the imports, and the stock of uninstalled imported machinery was growing faster than the stock of uninstalled domestic machinery. Soviet specialists were by no means unanimous on the issue of an optimal strategy for machinery imports. Some expressed concern with the growing imbalance between machinery imports and exports and with the fact that Soviet machinery exports do not do well in foreign markets. Others, however, argued that machinery imports should continue to grow in order to close important gaps in domestic technology." One Soviet economist has gone even further by stating that a curtailment of machinery imports from the West and a corresponding expansion of the capacity of the do mestic machinery industries was impossible. The theme of significant dependency on Western technology transfer was stressed by the leading Soviet specialist in foreign trade, Academician 0. Bogomolov. Concerned with "a might wave of [Western] discriminatory policies" in trade Bogomolov urged CMEA nations to strive towards technological independence by changing the mix of imported machinery away from specialized final products to complete plant equipment and technology. Expressions of concern with the growing dependence on foreign, and particularly Western, markets for grain and other foodstuffs have also been voiced recently by the Soviet leadership. Thus, Tikhonov complained about the "decreasing level of self- sufficiency in agricultural products in CMEA countries, and Brezhnev called for the reduction of dependence on the West for imports of foodstuffs in the introduction of his "20-year food program." The reaction of Soviet foreign trade officials to the statement of the Secretary-General is rather interesting. Referring to the statement, an official of the Ministry of Foreign Trade explained that the USSR must be basically self-sufficient in foodstuffs and livestock feed and criticized the US for "transforming a common commercial deal into a political weapon". However, lest the Western business community were to take Brezhnev's statement literally, the author hastened to add that it should not be taken to mean that the USSR is planning to cut food imports in the future. An additional aspect of foreign trade policy which was not discussed in this paper concerns the high and rapidly growing budgetary revenues derived from trade. Earnings of foreign trade, which accounted for some five percent of the state budget in 1970, rose to over ten percent in 1975, and probably reached 20 percent in 1980. The high level of these earnings thus mitigates against any drastic reforms of foreign trade, particularly against reduction in imports which provide the bulk of these earnings. Thus, starting in the late 1970s we witnessed some lowering of expectations of great benefits to be derived from the international division of labor, particularly in the Soviet search for assistance in shifting to an intensive path of economic development. A sobering note was introduced by the inability of the Soviet industry to break through to wider markets for machinery exports and by the need to expand oil and gas exports instead. It appears that at the same time Soviet leaders were becoming increasingly concerned with the rapid growth of dependency on machinery and food imports. However, as in many other policy issues during the Brezhnev era, considerations of economic expediency and short-run advantages prevailed and Soviet imports continued to grow constrained only in their hard currency component. IV. Measuring Trade Participation A legitimate question can be raised concerning the best way to measure the role of foreign trade in the Soviet economy. Soviet prices are administratively fixed and are not true scarcity prices, and, therefore, trade participation ratios based on Soviet established prices may be said to distort the true opportunity cost of resources and the dependence on foreign trade. Accordingly, Dr. Vanous suggested that a preferred method of measuring Soviet dependence on trade would be in comparable world market prices, as these prices are "scarcity prices indicative of opportunity cost of different commodities on a worldwide basis. However, since we do not have the necessary statistics to convert Soviet national income data to comparable world market prices Dr. Vanous proposed to use the CIA-estimated Soviet GNP in current dollars. Using the actual dollar transaction prices of Soviet exports and imports (adjusted for export subsidies in intra-CMEA trade), he estimated the share of exports as 6.0 percent of the Soviet dollar GNP and the share of imports as 4.1 percent. Calculated in this manner the overall trade participation ratio is approximately half of the ratio based on internal Soviet values, which leads Dr. Vanous to support the long-accepted position that the USSR is a relatively closed economy with minimal exposure to the outside world. I believe that this approach is unacceptable for both theoretical and empirical reasons. One of the main theoretical arguments against the use of dollars or world market prices lies in the index number problem. It will be recalled that both theoretical and applied studies suggest that, as long as prices and quantities are inversely related, comparisons of national aggregates (GNP, NMP) of two different countries will vary depending on whether prices of one or the other country were used. Thus, the ratio of national incomes of countries A and B will be higher when measured in B prices than when measured in A prices. For instance, the ratio of Soviet to US GNP for 1976 was estimated as 73.7 percent when measured in dollars and only 49.5 percent when measured in rubles. In order to partially compensate for the index numbers effect, CIA analysts convert the ruble components of Soviet GNP (consumption, services, etc.) into dollars by applying the "average purchasing power ratio of the dollar and the ruble", calculated as the geometric mean of individual ruble-dollar ratios weighted by both US and Soviet expenditure weights. The final effect, in terms of the example given above, is that the Soviet/US GNP ratio in dollars or in rubles (with this approach there is no difference) is 60.4 percent. This adjustment is justified for the purpose of US- Soviet comparisons, but is undesirable for the dollar measure of Soviet GNP used by Vanous. The degree of overstatement of the Soviet GNP produced by the conversion to dollar prices is reduced, but, on the other hand, the resultant dollar prices are even further away from world market prices. But, in order to evaluate fully the distortive effect of the index number problem on the dependency ratio of the foreign trade/national income type we must look at the numerator as well as the denominator. The proof is rather cumbersome, but it should be intuitively clear that the logic of the index number effect applies to foreign trade. Countries export commodities which are relatively abundant at home and that are, therefore, relatively inexpensive and import commodities which are scarce and expensive at home. Since the outside world buys the exports and sells the imports to the country we are examining, the relative scarcities and prices on the world market must move, therefore, in opposite directions from domestic scarcities and prices. Here we must introduce a complicating factor. The Soviet ruble is inconvertible and the USSR is trading in world markets in terms of external prices, which are different from domestic prices. Historically, the USSR tried and almost succeeded in balancing her commodity trade in external prices which also meant that the foreign trade was not balanced in domestic prices. I will return to this point later, but in the meantime we note that it can be shown that, for a country which balances its trade in external prices, the foreign trade dependency measure calculated as the ratio of the sum of exports and imports and national income will always be lower when expressed in external prices as compared to the ratio based on internal prices. As a rule, therefore, the distortive index number effect is compounded in the case of foreign trade/national income measurement, as the denominator is overstated and the numerator is understated. Bearing in mind these problems, CIA analysts specifically ask the readers of their statistical compendia not to use the estimated Soviet dollar GNP figures in conjunction with the Soviet foreign trade data in dollars. And, lastly, we should note that Henry Rowen, the Chairman of the Intelligence Council of the CIA, reported in a recent statement that imports comprised between 12 and 13 percent of Soviet GNP. It thus appears that CIA specialists are now using a measure of trade participation based on established ruble prices Dr. Vanous's calculations of Soviet trade participation based on foreign trade flows measured in world market prices and on Soviet GNP converted to dollar prices rest on the assumption that world market and US internal dollar prices are either identical or close. This assumption is difficult to accept. In the first place, a large share of US goods and services are not traded outside of the country and thus dollar prices of these goods and services cannot be taken to reflect world-wide scarcity ratios. But the assumed identity of US domestic and world market dollar prices does not necessarily hold even for traded goods. Possibly more research will be required to clarify this issue but at least one study by a group of competent Soviet foreign trade specialists concluded that in the 1960-1978 period internal US and world market prices were seldom close and often moved in different directions. It is interesting to note that Soviet authors reported that most internal prices in the US and West Germany were actually higher than world market prices for comparable goods. Thus, the use of US dollar prices in Soviet GNP will introduce a downward bias in foreign trade/GNP ratios. There are other compelling empirical reasons against the use of dollar prices in the assessment of the Soviet dependency ratio which center around the exact nature and possible biases of the CIA-estimated Soviet GNP in dollars. Without doubt these estimates are the best available both in terms of the methodology and statistical and classificational conventions used and in the meticulously careful manner in which the original Soviet data are interpreted and adjusted. Nevertheless, they are still estimates and, given the lacunae and methodological shortcomings of Soviet statistics, they are rather rough estimates with benchmark years being somewhat better and the estimates for interim years affected by priceadjustments and other problems. One specific problem with dollar estimates of the Soviet GNP emphasized by CIA analysts is our inability to adjust the individual ruble-dollar ratios for the well-known but unmeasureable differences in the quality of US and Soviet goods and services. An example which can throw some light on the possible differences can be seen in the case of machinery. In a 1976 study of Soviet GNP, the CIA authors estimated the ruble-dollar ratio for machinery as ranging from 0.266 based on Soviet weights, to 0.434 based on US weights. On the other hand, the ruble-dollar ratio estimated by me for 1976 on the basis of internal Soviet prices at which imported machinery is purchased by Soviet users and dollar prices at which this machinery was acquired abroad is 0.764, that is, twice as high as the average CIA-estimated ratio. Similar discrepancies would probably be found in binary comparisons of other goods and services. We can, therefore, reasonably assume that the dollar values of Soviet GNP are upward biased and trade participation ratios based on these GNP values would therefore be understated. Are there any other alternative methods of measuring the role of the foreign trade sector in the Soviet economy? The most accurate measure would, of course, be derived by a recomputation of Soviet national income accounts and of foreign trade flows in true equilibrium scarcity prices, but this method is simply not feasible. Alternatively we could consider the methodology developed by Bergson and Becker for recomputation of Soviet national income accounts in terms of so-called adjusted factor cost prices. For some time now the CIA has been constructing Soviet GNP accounts on the basis of this methodology and the series are now available for a number of years in current and constant prices. 32 These estimates are clearly superior to Soviet national income series in established prices, particularly for the purposes of analysis of national income distribution patterns, as well as for intertemporal comparisons. However, a recomputation of Soviet foreign trade data from established to adjusted factor cost basis presents a number of highly complex theoretical and applied problems. One could conceivably "clean up" export and import values by removing turnover taxes from imports and adjusting exports and imports for subsidies, but a correction for profits and rent-like elements in prices of imports presents almost insurmountable problems. A more promising approach to the problem of measuring the changing role of foreign trade in the Soviet economy lies in the construction of Leontief's "skyline diagrams" and of the supporting dependency indexes, particularly as this approach combines micro and macro measurements and allows for the substitution of physical units for value terms for certain homogenous commodity groups such as oil or grain. The necessary data in the form of input-output tables for the USSR are available for several years. Under the circumstances, the measurement of Soviet trade dependency based on exports and imports in domestic prices and on official Soviet figures for national income appears to be the only available one. In this measure, the foreign trade flows are valued in actual final transaction prices, i.e., prices at which they are recorded in national income accounts. Exports are measured in prices received by export manufacturers and imports are recorded in prices at which they were purchased by final consumers. All the shortcomings of the Soviet prices system and the idiosyncrasies of Soviet national income accounting notwithstanding, we know what we are dealing with and we are operating with value magnitudes which are used in the USSR. Soviet prices are, after all, not just random numbers and totally irrational. Domestic prices, particularly domestic prices of exports and imports, regardless of how much they depart from scarcity prices, are set by Soviet authorities and therefore reflect the priorities of these authorities. It is now quite apparent that the Soviet domestic price system is no longer fully insulated from world market forces. In the 1967 and in subsequent price reforms, Soviet price authorities attempted, albeit with mixed results, to move Soviet price relatives closer to world market prices. The shares of foreign trade flows in Soviet national income measured in this manner are also fully consistent with other measurements of national aggregates such as consumption or investment and contributions made by imports to their formation, or the state budget and its share of earnings of foreign trade.