Optimal paid job-protected leave policy

Although many countries give workers the right to return to their previous workplace after a temporary leave period ends, little is known about the details of such policies. This study characterizes the optimal paid job-protected leave policy using a model in which a worker has to leave his or her job with a certain probability and chooses whether to return to work after the leave period ends. An optimal policy, consisting of the consumption of a worker, that of a worker on leave, and the length of the leave, maximizes social welfare subject to the resource feasibility constraint and incentive constraint under which a worker on leave voluntarily chooses to return to work after the leave period ends. The present study finds that when the incentive constraint does not bind, the income risk caused by the leave should be perfectly shared among workers and workers on leave and that the leave period balances the marginal welfare gain and loss from a slight increase in the leave period. When the incentive constraint binds, the income risk caused by the leave is not perfectly shared among workers and workers on leave and that workers consume more than workers on leave because the constrained-optimal allocation has to give an incentive to workers on leave to return to work. By lengthening the leave period, another feasible allocation improves social welfare, which implies that the length of the leave period at the constrained-optimal allocation is too short. In addition, the study compares two economies, one that experiences a high discount factor and the other that experiences a low discount factor. In the former economy, the incentive constraint does not bind, whereas it does bind in the latter economy. Comparing the constrained-optimal allocations in these two economies, I find that workers consume more in the latter economy than in the former economy and that {¥it total} consumption during the leave period in the former economy is larger than that in the latter economy. Moreover, as an application of the theory, the study focuses on the paid parental leave policies adopted by most OECD countries. Using a numerical simulation, I conclude that the negative relationship between the replacement rate, namely the ratio of cash benefits during the leave to wages while working, and length of the leave period could result from constrained-optimal allocations.