Did Mass Privatisation Really Increase Post-Communist Mortality? Web Appendix

In this document, we describe more fully the methods and results summarized in our letter to The Lancet. We begin by presenting our replication of the cross-country analysis reported in Stuckler, King, and McKee (henceforth SKM).1 We then proceed to an alternative research design, where we explore the relationship between privatisation and mortality across Russian regions. Finally, we revisit the question of whether privatisation increased mortality, the only mechanism for which SKM provide evidence by which privatisation might have increased mortality. Cross-Country Analysis As discussed in the letter, we begin by reexamining the cross-country correlation between mortality and privatisation, focusing on the 15 former republics of the former Soviet Union (FSU), as SKM find a positive relationship between privatisation and mortality only in that sample. The table in the letter presents these results. We first perform pure replications using the SKM regression specifications. The dependent variable in SKM and our replications is the natural log of the age-standardized mortality rate for males aged 15–59. SKM use two alternative measures of privatisation in different specifications: a “mass privatisation ” indicator and the “average EBRD privatisation index, ” the latter the average of two widely used measures of progress in privatisation published in the annual Transition Report of the European Bank for Reconstruction and Development.2 We discuss both measures further below. In all specifications, SKM include country fixed effects, and they control for various time