Causal effects of vaccine uptake and population mobility on COVID-19 cases and deaths in Russia

2025. No. 1.

Ivan Gordeev

This paper analyzes short-run COVID-19 dynamics in Russia using a weekly panel of 12 regions over 54 weeks. Growth rates of cases and deaths are modeled as functions of vaccination uptake and behavioral activity, with region fixed effects and a set of controls. Identification uses three sources of exogenous variation: the timing of regional QR-code mandates as an instrument for Retail & Recreation mobility, news about potential WHO approval of Sputnik V interacted with the mandate indicator, and a short post-mandate window as instruments for vaccination. First-stage diagnostics show strong relevance (the homoskedastic partial F-statistics are 519 for mobility and 98 for vaccination), while event-study and placebo checks detect no pre-trends or reverse-timing effects (future WHO-related news terms do not predict current vaccination). TSLS estimates indicate a clear protective role of vaccination (particularly full immunization) and a positive causal impact of mobility. The six-week vaccination lag reduces future growth in cases and deaths, with the effect more than twice as large as naïve OLS suggests. The impact of mobility on subsequent cases and deaths is about twice as large as suggested by OLS and remains robust across different specifications. Overall, under standard IV assumptions, completing full vaccination and reducing high-contact leisure activity substantially reduce short-run transmission and mortality.

Ключевые слова: Russia, instrumental variables, vaccination, panel data, COVID-19, Two-Stage Least Squares, behavioral response