Documents / FOIA release
This Joint Committee Print, dated December 8, 1982, gathers studies that the Central Intelligence Agency's Directorate of Intelligence prepared for the Joint Economic Committee of the US Congress. It gives estimates of Soviet gross national product for 1950 to 1980, with indexes of industrial production, agricultural production and consumption, and explains the methods and data behind them. The foreword says official Soviet statistics inflate growth and that the CIA measures offer a more accurate picture. The document contains no UFO material.
“Gross”60+ pages
Approved for Release: 2019/07/19 C05210421 Western countries—remains unresolved. Much de- pends on the direction and degree of technological change. If materials use or intensity has generally declined over time, SPIOER’s growth would be exces- sively slow. Increases in materials intensity would cause errors in the opposite direction. Biases in the Quantity and Value Data. While the SPIOER sample covers a large portion of Soviet industrial production in 1972, other sample properties could bias the synthetic indexes. The quantity series employed in samples of sector production on balance may understate growth. For example, a bulldozer in a given model category produced in 1950 is counted the same as a bulldozer produced in, for example, 1962 despite any model improvements that may have been introduced. Ideally, the SPIOER methodology should account for changes in average product quality, but this would require engineering studies of every product. Quantity series also do not measure changes in product mix, which could reduce the reliability of these series as measures of production. Returning to the earlier example, the capacity of the average bulldozer has fluctuated over time as the product mix has changed. As a result, the fixed price weight used for the sample becomes unrepresentative. If the Soviet Union is gradually moving toward higher capacity equipment, the sample price becomes lower than the true average in later years, higher in earlier years, and growth is understated. Similar biases could occur in other sectors as well. The magnitude of this quantity series bias depends on the nature of the product involved. Products that are less complex, more homogeneous, and less vulnerable to innovation, such as many industrial materials and basic consumer goods, probably are less vulnerable to this bias. Bias should be greatest in technically com- plex, heterogeneous products such as machinery. Some notion of the possible extent of the bias that might arise in the machinery branch from using fixed price weights can be obtained by drawing on a recent CIA study.” This study has collected for several types ” CIA, An Analysis of the Behavior of Soviet Machinery Prices, I 960-73, ER 79-10631, December 1979. of machinery produced during 1960-73 a sample of model numbers, prices, and physical specifications. The purpose of this study was to determine what price changes have occurred that cannot be explained by changes in product quality or mix." A result of this research was the derivation of price indexes for two types of machinery represented by quantity series in SPIOER-—construction machinery and hoist-trans- port equipment (cranes). By using these price series to deflate the final demand component for these two sectors in the 1959, 1966, and 1972 input-output tables to 1972 producer prices, we derive series simi- lar conceptually to two quantity indexes in SPIOER, except that average unit prices are allowed to vary due to changes in product mix. Comparison of the growth rates of the series over the periods covered by the input-output table should give a rough indication in at least two quantity series of the scale of the bias that results from using fixed price weights. The results of this test shown in table 16 suggest that the SPIOER quantity series do tend to understate growth. Although the fixed price series grow more rapidly in the 1960-66 period than the variable price series, the downward bias becomes evident when examining 1967-72 and both periods combined. Over- all, the series with fixed price weights grow 1.1 and 0.4 percent less rapidly than the series with variable price weights for hoist-transport machinery and con- struction machinery, respectively. We have attempted to limit the effect of this bias on the machinery index by disaggregating the product sample as much as possible. Trucks, cars, and buses are disaggregated by model numbers, for example. By increasing the level of product detail in this way, the sample implicitly adjusts for changes in product as- sortment and partly for quality improvements. This approach cannot be followed for most sectors, howev- er, because the necessary production and price data are not available. " The cited study estimates regressions for each type of machinery covered. By regressing prices on physical parameters and time, it is possible to discover how prices have increased because of changes in product specifications and inflation. 212 Approved for Release: 2019/07/19 C05210421
Not linked to a story yet.
FOIA release, from the cia-readingroom collection. The PDF is mirrored here; the original link is under it. 399 pages are in the text index: search them above, or from the library's search.