Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
An evolving outside option is introduced into a stochastic directed search model with skill loss during non-employment. Using multi-spell data from the SIPP, I show that average reemployment wages are only mildly sensitive to unemployment duration while the job finding probability is highly sensitive to duration, with evidence of true duration dependence in both variables. Though untargeted, the model produces a quantitatively accurate decline in the job finding probability and starting wage, improving over a model with a fixed outside option. The addition of aggregate shocks leads to an nonlinear response of the unemployment and participation rates during and after recessions, with more severe recessions resulting in stronger hysteresis.