FIGURE 5-12 Tax force to retire, proportion of men aged 55–65 out of the labor force and proportion of youth aged 20–24 unemployed, circa 1995. SOURCE: Gruber, Milligan, and Wise (2009).
labor force behavior of older workers and (2) the relationship between the impact on older workers and the impact on younger workers.
A reform in Denmark provides a clear example (see Figure 5-13). In 1979, the Danish government introduced the Post-Employment Wage program, a labor market policy program intended to redistribute physically demanding jobs from older workers to unemployed younger workers. Before this reform, the youth employment rate had been rising while the youth unemployment rate changed little since 1975. Hence it seems unlikely that the reform was precipitated by a decline in the employment rate or an increase in the youth unemployment rate. Between 1978 and 1983 the employment rate for men aged 61–65 decreased by nearly 23 percentage points, a drop of 35 percent. During the same time period the employment rate for people aged 20–24 decreased by about 4 percentage points and the youth unemployment rate increased by about 4 percentage points. This natural experiment belies the fixed number of jobs proposition.6
6This example does not specifically address the issue of the business cycle, which may or may not have had an impact on the results. The study does, however, discuss endogeneity issues and notes that “economic shocks to the economy are likely to induce parallel movements in both the employment of the old and the employment of the young. [The study sought] to evaluate the effect of precipitating events that seek to induce older persons to leave the labor force, without a contemporaneous influence on the employment of the young (unlike macro economic shocks that tend to affect both simultaneously)” (Gruber, Milligan, and Wise, 2009, p. 23).