We investigate whether unemployment insurance (UI) policy affects the drinking behavior of the unemployed. Using NLSY data supplemented with Geocode data, we estimate the effect of benefit replacement rates on changes in individual alcohol consumption following job loss. Identification relies on variation in replacement rates across states and over time. Results indicate that a 100% increase in benefit replacement rate, roughly equivalent to a state moving from the lowest to the highest replacement rate, would, on average, result in unemployed individuals consuming 19.1 additional drinks a month. Looking at the change in an individual's binge drinking upon job loss, individuals receiving the highest level of benefits are 14.7% more likely to increase their binge drinking than those receiving the least generous benefits. We find that individuals’ responsiveness to changes in replacement rates vary based on drinking history, industry labor market conditions, education, and age.