Finnish pension system

Part-time pension program

The part-time pension program was introduced in 1987 for private sector workers to increase the flexibility at the end of their careers. The age limit was set to 60 while old-age retirement age was 65. The public sector work-ers have been eligible for part-time pension since 1989 but their age limit was originally set to 58. In 1994 the eligibility ages were harmonized for both sec-tors and set to 58. In July 1998 the age limit was temporary reduced to 56 until the end of 2002 when the age was set back to 58.

The government lowered the age limit in 1998 in order to support the labour force participation of the elderly. The target was to reduce labour mar-ket exits through other pension programs. However, the reform was made temporary in order to receive information on whether this policy was effec-tive. The reform in 1998 also added a clause which obliged employers to arrange part-time work whenever possible. (Government Bill HE 13/1998).

However, employers were not monitored or sanctioned if part-time work was not made available.

Workers who fulfilled the working, pension accrual and age condition were eligible for part-time pension if part-time work was available. For a private sector worker the working condition required 12 months of full-time work during the 18 months preceding the part-time pension take-up and for the public sector the requirement was 6 months of full-time work during the pre-ceding 18 months. The worker also needed to have been accruing pension rights for 5 years during the preceding 15 years in the private sector and 3 years during the preceding 5 years for the public sector. Also the employer needed to consent to the part-time work arrangement.

The pension provider monitored the income received during the part-time pension spell. The hours worked and the earnings needed to decrease in the same proportion being in the range of 30-75 percent from the previous levels.

Work hours had to be at least 16 but at most 28 hours per week. The decrease in working time is not observable from the data. However, comparing the

wage levels before and after part-time pension it seems that the reduction in hours is about 45 percent. This is also in line with the results from surveys done to the part-time pensioners (Takala, 2004).

The pension received during the part-time retirement was determined as half of the difference in earnings between full-time work and part-time work, however it couldn’t exceed 75 percent of the accrued pension. The reduc-tion in disposable income was not proporreduc-tional to the decrease in earnings because the average tax rate for part-time pensioners was lower than for full-time pensioners or full-full-time workers conditional on income⁴. The difference also accrued future pension rights, with an accrual rate of 1,5 %. In full time work (and for the wages earned as part-time pensioner) accrual rates were 1,5% for the individuals below the age of 60 and 2,5% for the individuals who were 60 or older. All in all, the effect of part-time pension on life-time earnings was modest.

The part-time pension system has been evaluated in Kyyrä (2015). As in this study, he also exploits the changes in the eligibility ages. His findings sug-gest that part-time pension did not lengthen the work career and if anything slightly reduced the risk of unemployment or exit out of the labour market.

However, these results hinge on the eligibility effect and not on the actual take-up of part-time pension. The current study can also look at the ones who have taken the part-time pension.

Other pension programs

The pension system at the end of the 1990s and early 2000s included mul-tiple labour market exit pathways. Before the early old-age retirement age was achieved, individuals exited through disability, an individual early re-tirement scheme or unemployment pension. Figure 5.1 presents the eligibility ages by cohort for the different pension schemes⁵. For the cohort born in 1940 the eligibility age for individual early retirement had the same age limit as part-time pension. Individual early retirement was granted for individuals

4From the data, the ratio of net incomes between one year before the start of the part-time pension spell and one year after the starting year, conditional on being in the part-time pension for the whole year, is 88%.

5As part-time pensioners had to fulfil the work condition, the changes in the unemployment tunnel or unemployment pension are not shown or discussed.

Figure 5.1 Minimum eligibility ages for different pension schemes.

with reduced work ability but the medical conditions were more lenient than in the disability pension, which can be applied at any age. The pension re-form in 2005 abolished the individual early retirement pension. However, it was also legislated that individuals over the age of 60 were to be granted dis-ability pension with similar conditions as previously in the individual early retirement.

The individual early retirement scheme or the disability pension and the part-time pension were most likely partial substitutes. Individuals with strong preferences for working but limited work ability probably seek part-time pen-sion possibilities. This fact already makes the part-time penpen-sioners a selected sample from the whole population of elderly workers. However, not all in-dividuals were entitled part-time work by their employers. This is another source of selection bias.

The major pension reform in the year 2005 affected the cohorts under study.

The 2005 reform decreased the full old-age retirement age from 65 to 63. This reform might have had independent effects on health outcomes and it cer-tainly affected labour market exit decisions. For this reason I focus on the period of 1995-2004 in this study.