(*) Professeur à l'Université de Melbourne.
Assessing the change in the level of Soviet welfare in the first half of the
Twentieth century presents many problems. There are the problems associated with the
reliability and accessibility of Soviet statistics, and there are those associated with
problem of understanding the peculiar nature of the Soviet situation in which rapid
secular improvements in welfare and life expectancy were accompanied by massive short-term
welfare and mortality crisis. These problems are made even more complicated by the intense
politicization of this question. This paper addresses all of these problems.
This paper argues that Soviet statistical data are generally far more reliable than are
normally presumed. In many respects they are unique in offering a sophisticated first
world level of evaluation (and statistical recording) of third world phenomena. There is
an obvious problem of avoiding propagandistic materials and getting behind the limitations
of censorship. But now that Soviet archives have been opened the prospects are good. Data
on food consumption, per capita GNP, and mortality are all available in surprisingly great
detail (surprising considering the extent of the contemporary social crises). Some
anthropometric data are also available and more will undoubtedly become available. So far
these latter sources have been relatively little used, with the notable exception of the
work of Professor Boris Mironov.
This combination of a rapid secular improvement in welfare over period in which massive
short-term welfare and mortality crises are occurring appears highly unusual, and it is
not well recorded in data for other societies. An explanation of this phenomenon may well
have some general significance for our understanding of the relationship between mortality
and nutrition, and between anthropometric indicators and nutrition in other societies with
less well developed statistical services.
This paper begins by considering the nature of the Soviet anthropometric indicators and
official attempts to manipulate the picture that they present. Then it presents various
sets of welfare indicators (nutrition indicators, mortality indicators as well as
anthropometric data) that characterize both the secular trends and the local crises. The
paper then attempts to explain the relationship between the trends and the crises, and how
these compare with the trends and crises observed in other societies. Finally the paper
considers some of the problems raised by the currently available anthropometric data and
the prospects for the further use of these data.