Measurement of inequality and poverty of population in Russia

During the transition in Russia’s economy and social sphere a number of new phenomena and processes arose there, that did not obtain sufficiently adequate and complete reflection in statistics, and gave an origin to dramatic discrepancies of opinions and estimates in analytic papers. The official estimates by Goskomstat foster the impression that income inequality and poverty of the Russian population, after achieving their maxima correspondingly in 1994 and in 1992, was decreasing (with some fluctuations, in particular, with some poverty increase in 1995) up to 1998’s crisis, which enlarged poverty, but did not considerably affect the inequality in income1. That is in sharp contradiction with several alternative estimates by independent analysts. Such estimates of inequality and poverty of the population in 1994-98 typically are considerably over the estimates by Goskomstat, and represent the tendency of inequality and poverty changes in an opposite manner. In particular, by the results of The Russia Longitudinal Monitoring Survey 1992-96 (RLMS, see T. Mroz et al., 1997), inequality and poverty rapidly rose from 1993 up to 1996. In fact, the inequality and poverty measures used in estimates by Goskomstat and in the RLMS analysis were incomparable: Goskomstat’s estimates were related to per capita money income, while the RLMS estimates were related to total household income with equivalence scaling (the total income includes, besides the money income, monetary estimation of household natural consumption resources in current regional prices). But inequality and poverty estimates, based on scale-equivalent total income of household’s member, as a rule are lower than the estimates, based on per capita income. Therefore the contrast between the results by Goskomstat and by RLMS, if represented in comparable measures, would be striking. In this paper we shall show that both the estimations by Goskomstat and by RLMS were valuably shifted, the former to underestimate and the latter to overestimate inequality and poverty, but such shifts had different causes, related to methodological differences in data processing and modeling, and to high biases in sample data.

Our project aims to: a) develop new analytic tools in order to explain the sources of discrepancies in estimates of inequality and poverty in Russia; b) provide accurate measures of inequality, poverty and standard of living in order to obtain a transparent picture of dynamics of social conditions in Russia and its regions during reforms; c) analyze the determinants of inequality and poverty, and draw policy conclusions in order to reduce poverty and make income distribution more equitable.

The problem of constructing an accurate and transparent picture of socioeconomic conditions in Russia during reforms is very complicated: it concerns the initial data structure, methodology of data processing, methodology of modeling the distributions of standard of living characteristics in the general sample, choice of a system of measures in order to represent and correctly interpret the results. Let us consecutively discuss these basic aspects of the problem, while indicating the crucial points of our approach to solve the problem.

• http://catalog.hathitrust.org/api/volumes/oclc/174965787.html