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.
“Mineral Wells”2 pages
Approved for Release: 2019/07/19 C05210421 Missing intervening years are the most common type of gaps in the SPIOER data. A typical series is fairly complete except for a few years when the relevant production data fail to appear in either an edition of the Narodnoye khozyaystvo or in one of the two editions of the statistical compendium Pr0myshIen- nost’. In general several benchmark observations are available for these series—most often for 1950, 1953, 1955, 1958, and l960—although the precise configu- ration of benchmarks varies from series to series. This problem affects around 50 of the statistical series or about 15 percent of the sample. Almost all of these cases are concentrated in three branches of industry: ferrous metals, processed food, and machinery. In all instances of this kind, years between benchmarks are interpolated by assuming that the average annual rate of growth between benchmarks is constant. Since we have little interest in year-to-year comparisons during the 1950s, this procedure should not introduce a significant distortion. The second type of gap occurs when Soviet statistical authorities fail to extend a newly published series back in time. This is potentially a more serious problem because extrapolation of a data series to complete it involves more uncertainty than interpola- tion. However, this problem is limited to less than 20 series or about 5 percent of the sample. About three- fourths of these occur in the machinery branch. Several different approaches are used to adjust the series depending on the availability of other informa- tion. For example, production data on three different types of boilers are published in different units for earlier years only. A year in common is used to splice the series together, and then extend the desired series backward. The index series for printing machinery and equipment does not extend back to 1950, so quantity production series available in the earlier years for two types of printing equipment are used to extend the full series. For some ruble series and index series in the sample, statistical correlations between GVO branch indexes and product series are used to extend the product series back in time. Metalwares, metal structurals, light industry and processed food industry machinery and equipment, and mineral chemicals are examples where this procedure is used. In a few instances—furniture, logging and paper machinery, machinery repair, and carpeting—the series are merely extrapolated backward assuming a constant average annual rate of growth equal to the rate for the first five years for which data were published. The assimilation of new products into SPIOER poses a different statistical problem whose treatment de- pends on the type of data series used for the new product. In most cases the line items in the sample are so aggregated that a new product merely falls within the rubric of an existing broader classification. Phys- ical product series often cannot be adequately disag- gregated to account for the effect of new products and a changing product mix. This can be a serious handi- cap because the entire output of a given commodity or commodity class—bulldozers in construction machin- ery, for example—has to be given a single average or representative unit price for the entire period. Other examples of quantity series in SPIOER where new products are included in output statistics, but where the average price may not fully reflect them are synthetic plastics and resins, synthetic knits and fi- bers, and scrapers. In those instances where the SPIOER disaggregation accounts specifically for new products—such as trac- tors, automobiles, and trucks—another procedure is used. If the product is considered essentially new, it is added into the sample at its 1967 price or its first permanent price for models introduced after 1967. If the product is considered a modification of an earlier model, production is added to the output of the similar model at the earlier model’s price. The problem of accounting for new products does not apply to the value and index series. Because these series presumably cover all products in a given cate- gory, they include new products and reflect changes over time in the product mix. However, value series also include the value of output of intermediate products as well as the disguised inflation mentioned earlier. All of these problems associated with the treatment of new products are discussed in detail later in the paper. 182 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 above. 399 pages are in the text index: search them above, or from the library's search.