Acceso usuarios
Ignacio Moral Arce, Concepció Patxot Cardoner, Guadalupe Souto Nieves
In this paper the scope for measures delaying retirement age in Spain is analyzed using the continuous working life sample published by the Social Security administration (MCVL). In particular, retirement probabilities of workers are estimated using a limited dependent variable model, controlling for the possibilities to opt for early retirement. The results indicate that, despite the attempt to focus on the relevant subsample –those who can opt for delayed retirement– the incentive measures explicitly governed by legislation have a limited impact on retirement decision, this being mostly determined by age.
When age is specified as single year dummies it captures most of the significance, but the option value still turns significant. Hence, there is some room for policy simulations.
We first simulate the impact of the last reform introduced in Spain fostering delayed retirement, finding a small impact. This is not surprising, given that it affects a small share of pensioners and it is a marginal change. Other measures often discussed in order to foster the Bismarckian nature of the system are also analyzed to the extent that they can affect retirement age. Despite their impact on pension rights, the response of retirement probability is also limited.
Only a direct increase in the normal retirement age would produce a sizeable increase in retirement age. Despite the small impact on the probability of retiring of the change in incentive, there is a 1.1 increase in retirement age in the male subsample when age dummies are switched accordingly.
