"Even in countries with stringent job protection, workers typically only benefit from job security once they have worked at their employer beyond a minimum qualifying (or probationary) period. This paper analyzes how such a probationary period influences firms' behavior and workers' outcomes. I specifically examine a 1999 British reform that lowered from two years to one year the tenure necessary for a worker to be able to sue their employer for unfair dismissal. I first construct a model based on the assumption that firms learn about match quality over time. The model predicts that, after the reform, the hazard of firing workers between 1 and 2 years tenure decreases relative to the hazard
beyond 2 years in all cases. Moreover, if, to avoid keeping lemons beyond the shorter qualifying period, firms react by recruiting workers more carefully, the hazard between 0 and a few months is predicted to decrease relative to the hazard beyond 2 years; an increase in monitoring has the opposite effect. Cox proportional hazard regressions show that the reform decreased the firing hazard between 0 and 2 years relative to the hazard between 2 and 4 years by about 30%. The calibration of the model reveals an increase in both recruitment and monitoring efforts, hence match quality. Consistent with an increase in match quality, I find that low tenure workers are more likely to receive
training after the reform. Lastly, the reform has no detectable impact on unemployment duration, wages or employment."

Support: Digital