| During the 1990s, the social security earnings test was reformed for individuals aged 65–69, eventually leading to its abolition in 2000. Despite the Delayed Retirement Credit (DRC), the earnings test is widely viewed as a work disincentive to the elderly and it is presumed that loosening or abolishing the earnings test would increase their working hours. However, the impact of the earnings test rule changes on working hours may be theoretically ambiguous due to an offsetting income effect. In addition, when we consider the DRC, the impact of the earnings test rule change might be small. To identify the policy impact, I use 3 natural experiments provided by the earnings test rule changes after 1990 and the difference-in-difference method. The regression results indicate that after the earnings test removal in 2000, male working hours for individuals aged 65–69 increase significantly, relative to the younger age group. This can be explained by the combined effect of an increase in working hours for individuals aged 65–69, relative to individuals aged 70–74 and a decrease in working hours for the younger group, relative to individuals aged 70–74. However, female working hours do not increase significantly. Low earnings of women, longer life expectancy of women and spouse's benefits could be used to explain the different behavior between men and women. |
