The effect of raising the retirement age on retirement decisions in the Netherlands

The goal of this research is to identify how the labor supply decisions and social insurance usage of older individuals are affected by changes in the official retirement age (the AOW age) in the Netherlands. In the period 2013 to 2017, the AOW age increased stepwise from 65 years to 65 years and 9 months. As a result, individuals were subjected to different AOW ages depending on their date of birth. This setting provides an opportunity to employ a Regression Discontinuity Design. Using administrative data, I find an average decrease in the retirement rate of 28.5 and 50.2 percentage points in the period before treated individuals reach the new AOW age for one-month and three-month increases, respectively. Individuals seem to finance this period in which they normally would have retired by retaining their original work state. As a result, the employment rate, unemployment rate and the share disabled increase due to the one-month (three-month) increases by on average 4.8 (15.8), 1.0 (3.2) and 8.2 (10.5) percentage points, respectively. Of the proceeds of the reform in terms of additional income tax revenue and decrease in AOW benefits paid approximately 18.5 percent is lost to the increase in unemployment and disability benefits paid.