Printer Friendly

The statistics corner: return credibility to statistics.

Editor's Note: For the past seventeen years I have had the opportunity to work with the Central Statistical Office of the U.S.S.R. During the past nine months I have been in six formal meetings with the State Committee on Statistics (Goskomstat); the last three meetings were held with the new Soviet Chief Statistician, Valdim Kirichenko. His candid review of the state of statistics in the Soviet Union should be of interest to all economists in the Western world. His guest editorial follows. A longer version of his views is available from my office. joseph W Duncan, Editor, The Statistics Corner.

INCREASING SOCIAL activeness and persistent efforts to find a solution to adverse critical situations are manifest, among others, in the increased interest shown in statistical information and the growing volume of criticism of it. Under those circumstances the most unwise thing that people professionally involved with the governmental statistical service could do would be to try to save the "honor of the uniform" by proving that, in spite of isolated shortcomings that are being surmounted, generally speaking the situation is not bad. Society will not trust such claims, not without a reason, for they would be inconsistent with reality.

For decades the dominant concept was one of displaying successes and advantages and concealing difficulties and negative phenomena in the development of the country and its various regions. Statistics, like theory, were asked to perform a twisted ideological function: shaping illusions of well-being and infallibility in the activities of the command-bureaucratic system. Such a "glossed" approach was reflected in the very methodology for formulating indicators (rate of social production, price dynamics, level of consumption of goods, efficiency, and losses) and tendentious reviews of the methods used to compile data, which were concealed from society. Comparisons with standards, indicators, and methods for setting the data, used in the statistical practices of developed countries and of international economic organizations, were not provided. Distorted data on the growth rates, levels, and proportions of the country's socioeconomic development did not provide a reliable foundation for the making of most important socioeconomic decisions.


Naturally, many shortcomings in statistical information have to do with figure padding and obvious whitewashing. The struggle against them demands tireless attention. Nonetheless, the main harm caused to the accuracy of statistical data was the imperfection (or tendentiousness) of the methodology of computations and the tendency to make them fit current political tasks. Following are some examples:

Quite recently, even after the beginning of the period of perestroika, in the course of the computation and publication of data for 1985-87 gross national product, the gross social product, the net material product, and the real population income, corrections were made that eliminated the impact of the reduced production and sales of alcoholic beverages on the dynamics of such indicators. Consequently, the growth rates of the net material product for 1985 and 1986 were roughly doubled. Actually, the 1985 net material product produced had increased not by 3.5 percent but by only 1.6 percent, and that of 1986 by 2.3 percent rather than 4.1 percent.

Retail trade should reflect the volume of sales of goods to the population in exchange for its monetary incomes. However, according to the existing methodology, it also includes the volume of sales of durable goods to organizations, establishments, enterprises, and collective farms, meeting their current economic needs, and food products to organizations in the sociocultural area (petty wholesale). This includes all earnings from the marketing of durable goods on a commission basis, although it would be justified to consider only part of the sales of second-hand stores, related to cash turnover.

There are elements of double accounting, above all in computing the volume of paid services. They include services provided not only to the population but to enterprises and organizations as well. The overall volume of double accounting and cashless operations amounts to some 20 percent of the growth of retail trade and trade services.

It is relatively easy to put an end to such obvious absurdities and tendentious methods. It is much more difficult to surmount the disparity between these methods and worldwide statistical standards, a disparity that developed over several decades, and to make the information system consistent with approaches and standards adopted in international practices.


The high road leading to the solution of this problem is clear. We must learn how to use in our macroeconomic computations the most important elements of the System of National Accounts and integrate them with the System of Balances of the National Economy. However, this is much easier said than done, for it will require a substantial restructuring of the entire bookkeeping system at enterprises. However difficult the problem may be, it will have to be resolved, for this approach is consistent with a developed system of commodity-monetary relations and financial-credit instruments, i.e., relations and institutions toward which we too are advancing as we develop radical economic reform.

The first step in this direction was the use in Soviet statistical practices of the GNP indicator, the formulation of which is based on the principles governing the System of National Accounts of the United Nations. It characterizes the overall volume of economic activities in the country, applicable to the combination of material production, and public and private services, and all varieties of income and depreciation. A system of tables has already been formulated, which makes conversion from a Balance of National Economy to indicators of the System of National Accounts possible. We do not consider in any way disgraceful the use of foreign experts, who are specialists in the area of national accounting. It is precisely the possibility of relying on rich international experience that enables us more successfully to solve the most difficult problems of a perestroika in statistics.


The end result of statistical work is characterized not only and even not strictly by specific figures" but by an analytical conclusion that can be derived from the data obtained. The study enables us to identify the reasons for phenomena, the influence of and interaction among different factors, to assess the efficiency of managerial decisionmaking and the possible economic and social consequences of the developing situations. Unfortunately, it is precisely the complex, the multifactor analysis that remains a weak point in domestic statistics. Naturally, there also is the influence of the prevalent concepts concerning establishing an indicator as such, the disparate nature of statistical data, and the lack of coordination among the methodologies used in computing related parameters. Nonetheless, in my view the main reason is the gap in the natural ties linking statistics with the science of economics. Based on my experience in scientific work, I am perfectly aware of the way, over several decades, economists were supplied with very partial information.

Thus, what kind of serious financial analysis could be made if the economic public was unfamiliar with the actual state of the national budget and the size of the national debt? Official publications invariably reported a surplus of state revenue over expenditures. Meanwhile, financial disproportions were growing and becoming increasingly dangerous. The most important types of revenue were declining.

In 1987-88, payments made by state enterprises and organizations based on their profit (income) declined by 10 billion rubles, while the profits left at the disposal of enterprises and organizations increased by nearly 50 million. Compared to 1985, income from foreign economic activities declined by 8.5 billion rubles. With every passing year expenditures increased substantially (by 33.7 billion rubles for the national economy, 25.7 billion for sociocultural projects, and another 11 billion for financing foreign economic activities). Budget subsidies also increased to compensate for price differentials for individual commodities (by 20 billion rubles), and the share of such budget expenditures reached 19 percent (9.6 percent in 1970). This was due essentially to increases in the purchase prices of agricultural commodities and supplements to prices.

The gap between income and expenditures was filled by transferring to the budget general state loan funds. In 1989 alone the USSR State Committee on Statistics published figures that showed the growth of such borrowing during the twelfth five-year planning period: from 18 billion rubles in 1985, borrowing rose to 90 billion in 1988. Finally, the public was informed that the share of the funds borrowed from the states lending fund to cover budget expenditures had risen to nearly 20 percent, compared to 2 percent in 1970 and 5 percent in 1985, as well as the fact that many topical problems, including social ones, were being resolved by borrowing on future revenue, which significantly increased the national debt and the country's foreign indebtedness.


Naturally, we can now blame the economists for failing immediately to notice the difficulty and for sounding the alarm belatedly. However, it would seem odd to demand of a physician an accurate diagnosis and efficient treatment while refusing to inform him of the condition of the patient.

Today the personnel in statistics themselves need the active involvement of science in the practical solution of problems of improving methodology and intensifying analytical work. Studies must be made of many problems that remain "blank spots," without scientific solutions, and that have not been studied by our statistical service.

It is both possible and necessary to discuss the amount of the social funds and each one of their individual elements, and to criticize the specific mechanism for their distribution and choice of priorities. But why distort obvious facts? For even a lie used not for purposes of glorification but in rebuttal, nonetheless remains a lie.

It was thus that in both scientific and practical matters we realized the inadequacy of the traditional value methods for determining the dynamics of economic growth under the conditions of profound quality changes and structural shifts in production, the appearance of essentially new types of goods and services, and the unparalleled increase in the role of intellectual forms of activities. Still prevalent in the theory of balancing economic growth are its natural-physical aspects, which deal with the flow of output and production factors. Yet today we need studies in terms of market balance or imbalance, which would enable us to determine the purchasing power of the population and enterprises, price dynamics, supply of commodities and services, and interaction between price and demand. It is only on this basis that we can obtain a uniform assessment of material-physical and financial-value aspects of the reproduction process.

We need a theoretical and an analytical foundation for the formulation of methods and indicators of statistical observation of inflationary processes. The list of insufficently developed theoretical problems could be extended. One thing is clear: Cooperation between statistical authorities and scientists can become tangible and effective. A major prerequisite to this effect is a radical increase in glasnost in statistical information and the methodology used to obtain it.

Of late we have been able to expand the scope of statistical publications and to reduce the list of information considered classified. In solving this problem, occasionally we come across objections of striking originality. For example, it is being said that the publication of detailed information on foreign economic relations, the amount of the debt, and the cost of servicing it could adversely affect relations with foreign banks. But then, is it not clear that it is precisely such information that is fully accessible to them? Meanwhile, the lack of clarity concerning our reserves in foreign exchange and gold is what the creditors dislike the most. As a whole, in order to make the level of openness of our statistical information consistent with universally accepted international standards, extensive and difficult work lies ahead.

By itself, a statistical indicator proves little if the method by which it was obtained remains concealed. In reviewing the methods, a process which is being quite intensively conducted currently, the State Committee on Statistics intends to make them public. This will also create favorable conditions for making alternate computations of indicators by researchers and organizations unrelated to governmental statistics.

As we know, the practice of alternate computations presumes the observance of certain conditions which guarantee the interests and protect the rights both of governmental statistics as well as of those who undertake to make alternate computations. Above all, this involves a uniform initial statistical base, and the required competence of those who are making alternate evaluations. Unfortunately, against the backgroun of a justified mistrust in official information, the importance of such obvious restrictions is frequently ignored.


Today the computation of price indexes is the main link in the entire chain of problems of accuracy of socioeconomic information and a foundation for a realistic assessment of the dynamics of the most important general economic indicators. Particularly great attention is being paid to price changes in the consumer market. For many years the statistical services claimed that the prices in the consumer market remained virtually stable. Through personal experience, millions of people were aware of the opposite. Given this situation, society is ready to believe virtually any figure characterizing the rate of price increases as long as it is sufficiently high.

Unfortunately, an accurate computation of price indexes and the cost of accurate computation of price indexes and the cost of living is a difficult and very labor intensive statistical operation. In the United States, for example, approximately $65 million is annually allocated for this purpose. That is precisely why I would be somewhat mistrustful of specialists who smoothly make their own assessments of the rate of inflation without any detailed discussion concerning the sources of their data and the methods they use in working with them.

Based on the computations made in 1989 by the USSR State Committee on Statistics for the 197188 period, the average price index, including the collective farms market, rose by more than 40 percent. Between 1986 and 1988 the increase in average purchase prices of goods sold by the state trade system, the consumer cooperatives, and the collective farms market (excluding changes in prices of goods produced by cooperatives and as a result of individual labor activity) was 7.2 percent, including 11.7 percent for food products. Price changes were uneven. Their increase was quite noticeable for some groups of goods, something which was sharply felt by the population. The most substantial increases in average prices were those of processed meat products (18.5 percent), canned meat goods (13.9), potatoes (22.8), vegetables (11.6), bread and bakery goods (22.3), and alcoholic beverages (46.5 percent). In the area of durable goods, the fastest price increases were those of clothing and linens (13.8 percent), woolen fabrics (8.7) and automobiles (7.2 percent).

The list price index that dominated our statistics for many long years takes exclusively into consideration legislatively mandated price changes. It does not reflect the higher prices of new goods and the influence of many other forms of hidden price increases. Nonetheless, we should acknowledge that the average price index, the dynamics of which we discussed, is influenced by the natural process of changes in technical and economic parameters and the quality of output. Price changes consistent with changes in the quality of goods, as we know, are not an inflationary factor.

In order to determine the real price increase more objectively, which characterizes increased or lowered costs to the consumer per unit of consumer value, in 1989 the USSR State Committee on Statistics organized a registration of prices based on a set of representative goods. A total of 650 representative goods sold in state commercial stores in all 150 regional and republic centers were selected.

This method enables us to take into consideration the dynamics of current prices of comparable goods, the appearance of new commodities, and the influence of contractual and temporary prices. It reflects changes in consumer expenditures per unit of the consumer value of low-quality goods and excludes the influence of varietal and structural changes. The consolidated price index of recorded representative goods takes into consideration the influence of prices on sales volumes in the state trade system, the consumer cooperatives, and the collective farms market, as well as in cooperatives and individual labor activities. It is precisely these advantages that explain its extensive use in global practices in computing inflation in the consumer's market.


The statistical study of the structure of inflation highlights the main trends in the struggle against it. Above all, we must concentrate on surmounting suppressed inflation, for it is particularly destructive economically as well as socially. It is a question of the need to increase the production of goods and services, to broaden the range of goods to be sold on the market, to expand paid services, and to introduce truly commercial principles in the organization of trade. A strict tax regulation of income would be required as well. Income could and should grow, but only to the extent to which the end production results increase. Otherwise the increased monetary income of the population becomes meaningless from the viewpoint of any real improvement in the material situation of the people, and of maintaining social justice in our society. Steps must be taken to freeze monetary savings of population and enterprises.

Any major socioeconomic decision affects an entire array of interrelated processes. Without identifying them and without contemplating compensating measures, we cannot solve the problem. For example, the fact that minimal pensions were raised on October 1, 1989 will result in a significant reduction (although not the total disappearance) of the low-income group of retirees. Their number will be reduced by one-half. The increased income of the pensioners will lead to an increased demand for food, inexpensive clothing and shoes, and medicines.

The 1989 census indicated the existence of a substantial number of people of working age who are - about 6 million people (excluding those farming their private plots). The highest share of unemployed population is in Azerbaijan and the republics of Central Asia.

Unfortunately, for the time being, the information we need for a detailed study of the reasons for unemployment, and especially involuntary unemployment, is clearly insufficient. In order to gather the data necessary in pursuing a policy of efficient full employment, we shall have to resolve a number of difficult methodological and organizational problems.


Of late, problems of economic relations among republics have become particularly pressing. Considering the complexity and the economic and political significance of such problems, particular attention should be paid to perfecting the statistical methods used in their study: interregional intersectorial balances and procurement statistics.

Nearly 25 percent of the gross social product is involved in the trade among Union republics. As a rule, the smaller the republic, the higher becomes its share of exports of its output and its consumption imports. The highest share of consumption imports (27-29 percent) is characteristic of the Georgian, Moldavian, Kirghiz, Tadzhik, and Armenian SSRs and the Baltic republic. In 1988 a positive balance and the interrepublic trade (surplus of exports over imports) was found in five republics (Russia, Belorussia, Azerbaijan, the Ukraine, and the Georgian SSR).

A study of foreign economic relations of republics (their exports and imports) indicated that, in terms of foreign exchange rubles, the volume of exports exceeds that of imports in only three republics (Russia, Uzbek SSR, and Tadzhik SSR). In all other republics foreign exchange income from exports was insufficient to cover their imports. For example, in terms of intra-Union prices, Moldavia's actual imports for 1987 exceeded possible imports (based exclusively on foreign exchange income from republic exports) by a factor of 3.8; the respective figures were 4.3 for Turkmenia and 14 for Armenia. It is true that the absolute volumes of imports were small. All in all, for the Baltic republics, this excess was slightly over 100 percent.

The economic results of the trade among republics are influenced by the imperfect correlation between prices of raw material resources and finished goods, the inefficient mechanism of the use of the turnover tax, and subsidies for meat, dairy, and other goods. Many such imperfections would be eliminated with the price reform. However, we can use them as estimates, if commodity trade is assessed in world prices. A great deal is being said and argued who would benefit and who would lose in that case, most frequently citing random and unreliable data.

The USSR State Committee on Statistics has studied interregional flows in terms of world prices. As was to be expected, this improved the situation (net exports) of republics that export fuel and worsened it in cases in which light industry goods account for a significant share of exports. For that reason, in all Union republics (other than the Russia and the Ukranian and Turkmenian SSRS) the general net exports in terms of world prices is worse compared to domestic prices. Based on the conditions prevailing in 1988, a worsening of this balance in terms of world prices, compared with domestic prices in the case of the Baltic republics, Moldavia, and the Georgian SSR, equaled the loss of approximately 10,percent of the net material product.

Thus, should we use world prices, the Moldavian SSR would have to pay more for importing petroleum products, metal, machines, and equipment. It would earn from exporting food products a lesser amount by a factor of 2.7; global prices of wine products are lower by a factor of 1.9; fruits and vegetables, 4.2; tobacco, 4; and meat and meat products, 3.1. The Republic would benefit from imports of chemicals and light industry goods. However, this would not compensate for losses in other import and export items.

Naturally, however, the main task in improving regional statistics is by no means that of assessing who "owes" what to whom. Without a normally operating domestic market, this problem cannot be solved. Computations in terms of world market prices are not a useless analytical method. However, their significance should in no case be exaggerated. As long as domestic prices are unrelated to them, we must apply a large number of conventional assumptions.

It is much more important, rather than engaging in sterile arguments about the redivision of the public pie, to organize balanced statistics on the regional level, which would include drawing up a balance of the social product, a balance of fixed assets, the largest segments of the consolidated financial balance, and, most important, consolidated indications of socioeconomic development, such as net material product, net output by enterprises in sphere of material production, gross national product, overall volume of consumption of material goods and services, and summed indicators of social production efficiency. Only thus can we lay a strong information base for the development of regional self-management and self-financing.

The domestic statistical service enjoyed deserved prestige for its honesty and professionalism until the end of the 1920s. Statistical data were used both by Bolsheviks and Cadets in their efforts to substantiate essentially different views. However, all in all, their accuracy could be relied upon. It is only in such a situation that a normal social debate on economic and political problems and the healthy development of economic theory are possible.

Politicizing of statistics and its conversion into the servant of the power structures dealt a heavy blow at its reputation. I believe that today the most important task of the statistical authorities is to ensure the accuracy of the data and, on this basis, restore the trust in such data on the part of the Soviet and international public. The country can no longer afford to seek the right way with the help of trick mirrors.

* Vadim Nikitovich Kirichenko is Chairman, USSR State Committee for Statistics.
COPYRIGHT 1990 The National Association for Business Economists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kirichenko, Vadim Nikitovich
Publication:Business Economics
Date:Oct 1, 1990
Previous Article:When unemployed business economists call.
Next Article:The PC Corner: the Macintosh.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters