The aim of this thesis is to explore several aspects of occupation-based inequality
in post-Soviet Russia that have previously been given little attention in
the literature. The data sources for statistical analysis are the Russian Longitudinal
Monitoring Survey (RLMS) and the International Social Survey Programme
(ISSP). Various statistical techniques have been used, such as regression models
with random and fixed effects, nonparametric and semiparametric regression
models, survival models and log-multiplicative models for contingency tables.
First, the thesis looks at the validity of the application of the European SocioEconomic
Classification (ESeC) in Russia. The results show that ESeC classes
in Russia are different in respect to several aspects of their employment contract,
such as the probability of informal employment, the index of fringe benefits and
unemployment risks. This confirms the validity of the ESeC for Russia.
Second, the association between earnings and age is analyzed. The shape
of cross-sectional age-earnings profiles in Russia is different from the shape in
Western countries, especially for men. There is little variation in earnings across
age groups, and younger men have higher average earnings than older men. The
thesis suggests and discusses several explanations for this, such as age segregation
in the labour market and the effect of class structure.
Third, the thesis explores the class gap in mortality. Non-manual classes have
lower mortality risks than manual classes, both for men and women. The size of
the class gap in mortality in Russia is larger than in Western European countries.
Fourth, the thesis constructs an occupational status scale and analyzes its
properties. The scale is based on the information about intermarriages between
occupational groups. The Russian scale is similar to the scales previously constructed
for European countries and the USA.
Overall, the thesis demonstrates similarity in the patterns of occupation-based
inequality in Russia and in Western industrial countries. It also discusses some
technical aspects of class analysis and suggests a more clear separation between
the descriptive and causal logic within it.