Is Early Retirement Encouraged by the Employer? Labor-Demand Effects of Age-Related Collective Fees
In Sweden, employers pay non-wage costs for their workforce in the form of legislated
employment tax and collective fees. For parts of the workforce, the collective fees are
progressive with respect to the employee’s age and wage. The objective of this paper is to
examine how non-wage costs affect voluntary early retirement. To this end we use a large
longitudinal employer–employee matched data set with administrative records of the private
sector in Sweden. We exploit the variation in collective fee costs across companies to identify
employer incentives to encourage early retirement. The results from the instrumental variable
estimator suggest that a 1 percentage point increase in non-wage costs in relation to wage
costs increases retirement by 6 percent. Further, given the wage sum and workforce structure,
large firms spend more on non-wage compensation than small firms. The share of non-wage
costs in relation to the wage sum is also positively linked to net employment growth.
In the economic literature, retirement is usually seen as a choice of the individual, with a focus on the
individual’s economic incentives, health status and earnings (see e.g., Gruber & Wise, 2004; Hurd,
1990, and the references therein). However, it is also highly likely that the performance of firms and
firms’ economic incentives to hire or give older employees early retirement offers are important for
the individual decision. Withdrawal from the labor market may in fact be a joint decision on the part
of the employee and the employer (the relevant literature is presented below). For the demand of older
workers the labor costs are often seen as the main disadvantage relative to that of younger workers
(see Munnell et al, 2006; and references cited in Skirbekk, 2008). In particular, older workers have
higher wages on average but are also linked to substantial non-wage costs such as pension premiums
and health insurance expenses that on average are higher than those for younger workers.
In Sweden the non-wage costs, or social costs, consist of legislated employment tax and collective
fees, the latter being stipulated by collective agreements between unions and employers’ federations.
Almost all employees are covered by an agreement but the agreements vary across different parts of
the labor market. The collective fees consist mainly of contributions to the second-pillar pensions (the
occupational pensions) but also – to a lesser extent – other collectively agreed insurances. While the
employment tax is proportional to the wage sum, the collective fees are, interestingly, for some groups
of workers, progressive in relation to the employee’s age and wage. This causes an age-related
variation in workforce non-wage costs, which implies that the wedge between the wage costs and the
gross wage becomes much broader for the older workforce.
Given the higher costs, employers may have financial incentives to offer early retirement to their
older workers and substitute them with younger personnel. Furthermore, a solution involving an early
retirement pension for an older worker is in many cases financially favorable for the employer
compared with continued employment (described further below). Thus, although older employees in
Sweden are protected via seniority rules (‘last-in first-out’), employers have in the case of redundancy
great financial possibilities to persuade an older worker to retire ‘voluntarily’ by the use of early
retirement packages. However, younger workers may be imperfect substitutes for older workers.
There may also be positive educational and experience spillover effects of older workers. The
composition with respect to age and education may be a result of an intricate dynamic pattern that
differs with respect to the specific surroundings and context in which a firm operates.