Final Report Summary - PRIVMORT (The Impact of Privatization on the Mortality Crisis in Eastern Europe)
Employing an innovative design, we were able to quantify and compare inequalities in all-cause mortality caused by policy intervention, during the 1990s in the urban population of Russian mono-industrial towns. We clarify the uncertainty regarding the link between the pace of privatisation in the republics of the former Soviet Union and the increasing mortality, and laid to rest the controversy surrounding this issue. This study is unique in using individual-level data collected specifically to study the health effects of privatisation and thus fills gaps in the earlier literature. Methodological contributions include the use of the propensity score matching method to isolate the effect of privatization by closely matching the settlements for a prospective survey prior to collecting the data via survey. Indirect demographic techniques were used to show how mortality levels in Russia changed over time. We used the methodology of establishing a convenience cohort study, based on the Brass indirect method that surveys random population samples to collect data on deaths of respondents’ relatives, so as to estimate key population mortality parameters. Most studies focusing on this period examined macro data, while high-quality data at the level of the individual remain scarce. Even though the design of the study was dictated by the need to test the privatization thesis, this large new empirical base provides rich data for individual-level health risk analysis.
The method of isolating the effect of privatisation by closely matching the settlements prior to collecting the data is a novel approach in the study of distal determinants of health. We deployed propensity score matching on the baseline covariates that are most important to health outcomes. A systematic literature review of the causes of privatisation shows that the primary determinants of privatisation were political, rather than underlying economic weakness or social conditions. The political choice of the speed of privatisation was made at the level of enterprises or local and regional authorities, thus it can be claimed that it was likely to be orthogonal to mortality outcomes.
Clearly there are multiple factors that contribute to excess mortality in Russia, acting at different points along several causal pathways, with alcohol playing a key role in several, including that related to speed of privatization. Another possible component of several causal pathways involves social protection. Elevated mortality could be related to variations in the provision of social benefits at settlement level. However, local expenditures are strongly dependent on the health of city-forming enterprises. It is essential to recognize that the health effects of the transition were complex and defy mono-causal explanation. We do not claim that mass privatisation was the only cause of elevated mortality levels. In fact, it would be remarkable if one single policy or factor explained all the variation in the post-communist mortality crisis. Nevertheless, we believe the findings provide strong evidence for the hypothesis that rapid privatisation contributed to elevated levels of working age male mortality. Fast privatisation was consistently positive and statistically significant for men but was not usually significant for women. Robustness checks did not significantly change the magnitude or the direction of the covariates of mortality.
The method of isolating the effect of privatisation by closely matching the settlements prior to collecting the data is a novel approach in the study of distal determinants of health. We deployed propensity score matching on the baseline covariates that are most important to health outcomes. A systematic literature review of the causes of privatisation shows that the primary determinants of privatisation were political, rather than underlying economic weakness or social conditions. The political choice of the speed of privatisation was made at the level of enterprises or local and regional authorities, thus it can be claimed that it was likely to be orthogonal to mortality outcomes.
Clearly there are multiple factors that contribute to excess mortality in Russia, acting at different points along several causal pathways, with alcohol playing a key role in several, including that related to speed of privatization. Another possible component of several causal pathways involves social protection. Elevated mortality could be related to variations in the provision of social benefits at settlement level. However, local expenditures are strongly dependent on the health of city-forming enterprises. It is essential to recognize that the health effects of the transition were complex and defy mono-causal explanation. We do not claim that mass privatisation was the only cause of elevated mortality levels. In fact, it would be remarkable if one single policy or factor explained all the variation in the post-communist mortality crisis. Nevertheless, we believe the findings provide strong evidence for the hypothesis that rapid privatisation contributed to elevated levels of working age male mortality. Fast privatisation was consistently positive and statistically significant for men but was not usually significant for women. Robustness checks did not significantly change the magnitude or the direction of the covariates of mortality.