We model automatic trigger policies for unemployment insurance by simulating a weekly panel of individual labor market histories, grouped by state. We reach three conclusions: (i) policies designed to trigger immediately at the onset of a recession result in benefit extensions that occur in less sick labor markets than the historical average for benefit extensions; (ii) the ad hoc extensions in the 2001 and 2007-09 recessions compare favorably ex post to common proposals for automatic triggers; (iii) compared to ex post policy, the cost of common proposals for automatic triggers is close to zero.