1. Early retirement programs Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press.

2. What are we talking about? From a historical perspective large-scale retirement of workers rather recent phenomenon Today, retirement = extended period of self- financed independence & leisure Forced retirement – mandatory retirement age Defined benefit (DB) – defined contribution (DC) programs Early retirement programs – offers that cannot be refused Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press.

3. Overlaps with other institutions Unemployment benefits Employment protection legislation Working time Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press.

4. Decline in fertility as global phenomenon

5. Pension reforms in the G10

6. Outline 1.Measures and cross-country comparison 2.Theory 3.Empirical evidence 4.Policy issues 5.Why do early retirement programs exist? 6.Review questions Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press.

7. Measures Pension wealth: present value of stream of expected pension benefits Benefit accrual – implicit tax/subsidy Earliest retirement age – related to early retirement programs Standard retirement age Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press. Cross-country comparisons

9. Cross-country comparison Standard and earliest retirement age  Most common standard retirement age: 65 Variation in retirement incentives Net replacement rate for public old-age pensions –<40% in Ireland, Japan, Mexico –>100% in Greece, Hungary, Turkey Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press. Cross-country comparisons

10. Theory Continue to work if expected present value of continuing work is greater than expected present value of immediate retirement Option value of work: positive – postpone retirement; negative – retire Eligibility for early retirement: downward shift in option value – offer you can’t refuse Incentives may depend on DB or DC benefit systems Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press. Theory

11. Empirical evidence – age and employment Average age of transition to inactivity dropped substantially; 1970-2002: 14.8 years for Turkish men Employment rates prime-age and elderly men and women: more variation for women Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press. Empirical evidence

13. Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press. Empirical evidence

14. Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press. Empirical evidence

15. Box 6.1 Stimulating early retirement in Norway Bratberg, Holmås and Thøgerson (2004) Norway: introduction of early retirement scheme (AFP) for some firms 65 to 64 years; labor market 3 months after 64 th birthday BeforeAfter AFPyesno  yesno    Work82.683.81.264.786.021.3–20.1 AFP–––26.0––26.026.0 Other17.416.2-1.29.314.04.7–5.9 Total100 0 00 Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press. Empirical evidence

16. Empirical evidence – age and productivity Older workers: more reliable, better skills Older workers: high health costs, low flexibility, less suitable for training Age-productivity profile not exogenous to institutions Age-productivity relationship difficult to establish but employers have strong opinions about the productivity of older workers Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press. Empirical evidence

17. Policy issue 1: Should mandatory retirement age be increased? Lazear (1979): delayed compensation contracts: age-earnings profile upward sloping to prevent workers from shirking Issue of selection: least productive workers most likely to retire first Mandatory retirement not a unique instrument to end long-term relationships Increase mandatory retirement age may be neither necessary nor sufficient to increase labor force participation among older workers Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press. Policy issues

18. Box 6.2 Elimination mandatory retirement age US Ashenfelter and Card (2002) Exemption from general elimination of mandatory retirement age for college and university professors expired in 1994. Effects of mandatory retirement: YesNoDiff. Probability to stay to age 70 (%) From age 6026.125.40.7 From age 6539.238.60.6 Employment outcome if work at age 70 (%) Leave at 7076.629.647.0 Empl. at 7123.470.4–47.0 Empl. at 728.451.6–43.2 Empl. at 736.339.4–33.1 Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press. Policy issues

19. Policy issue 2: Should early retirement programs be phased out? In many countries early retirement policies introduced as a short-term policy response to combat unemployment Lump-of-labor fallacy May have affected perception of employers and workers themselves – vicious circle of perceptions that lack a solid empirical basis Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press. Policy issues

20. Why do early retirement programs exist? Lump-of-labor fallacy not valid: countries with high employment of elderly workers have a low youth unemployment rate Argument in favor of early retirement programs: health problems However: life expectancy increased substantially & health of older individuals improved greatly Little reason not to abolish early retirement programs Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press. Policy issues

21. Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press. Policy issues

22. Review questions 1.Why is early retirement not a good instrument to reduce youth unemployment? 2.How would an increase in the standard retirement age affect the behavior of employers and workers? 3.Explain why many firms employ older workers, but few firms hire older workers. Source: Tito Boeri and Jan van Ours (2008), The Economics of Imperfect Labor Markets, Princeton University Press.