For the last twenty years Russia has confronted the Mortality Crisis- the life expectancy of Russian males has fallen by more than five years, and the mortality rate has increased by 50%. Alcohol abuse is widely agreed to be the main cause of this change. In this paper, I use a rich dataset on individual alcohol consumption to analyze the determinants for heavy drinking in Russia, such as the price of alcohol, peer effects and habits. I exploit unique location identifiers in my data and patterns of geographical settlement in Russia to measure peers within narrowly-defined neighborhoods. The definition of peers is validated by documenting a strong increase of alcohol consumption around the birthday of peers. With natural experiments I estimate the own price elasticity of the probability of heavy drinking. This price elasticity is identified using variation in alcohol regulations across Russian regions and over time. From these data, I develop a dynamic model of heavy drinking to quantify how changes in the price of alcohol would affect the proportion of heavy drinkers among Russian males and subsequently also affect mortality rates. I find that that higher alcohol prices reduce the probability of being a heavy drinker by a non-trivial amount. An increase in the price of vodka by 50% would save the lives of at least 40,000 males annually. Peers account for a quarter of this effect.
