We quantify and study state-level economic policy uncertainty. Tapping digital archives for nearly 3,500 local newspapers, we construct three monthly indexes for each state: one that captures state and local sources of policy uncertainty (EPU-S), one that captures national and international sources (EPU-N), and a composite index that captures both. EPU-S rises around gubernatorial elections and own-state episodes like the California electricity crisis of 2000-01 and the Kansas tax experiment of 2012. EPU-N rises around presidential elections and in response to 9-11, Gulf Wars I and II, the 2011 debt-ceiling crisis, the 2012 fiscal cliff episode, and federal government shutdowns. Close elections elevate policy uncertainty much more than the average election. VAR models fit to pre-COVID data imply that upward shocks to own-state EPU foreshadow weaker economic performance in the state, as do upward EPU shocks in contiguous states. The COVID-19 pandemic drove huge increases in policy uncertainty and unemployment, more so in states with stricter government-mandated lockdowns.