Table 4: Marginal effects on the probability of working full-time, part-time and choosing to retire, sample of full-time workers 2001–2005, by age and gender (Percentage points difference in estimated probability).

Age 50–59 Age 60–69 Male Female Male Female Earnings ($10,000 increase) Full-time 1.79*** 4.55*** 2.58*** 8.52*** Part-time −1.61*** −4.32*** −2.01*** −7.79*** Retirement −0.18*** −0.23*** −0.57*** −0.73*** Lifetime earnings ($10,000 increase) Full-time 0.01 0.07 −0.16 −0.38 Part-time −0.01 −0.06 0.13 0.35 Retirement 0 −0.01 0.03 0.03 Wealth ($10,000 increase) Full-time −0.05*** 0.01 −0.22*** −0.08 Part-time 0.04*** −0.01 0.17*** 0.07 Retirement 0.01*** −0.00 0.05*** 0.01 One-year accrual ($10,000 increase) Full-time 0.04 −0.17 −0.09 0.26 Part-time −0.03 0.16 0.07 −0.23 Retirement −0.01 0.01 0.02 −0.03 Health shock Full-time −1.73 −10.27** −10.54*** −13.68*** Part-time 1.64 9.51** 7.74*** 12.86*** Retirement 0.19 0.75** 2.80*** 0.82*** Immigrant Full-time 0.89 3.38*** 4.05* 3.86 Part-time −0.80 −3.22*** −3.19* −3.55 Retirement −0.09 −0.16*** −0.86* −0.31 Working spouse Full-time 3.08*** 0.05 6.53*** 3.66 Part-time −2.76*** −0.05 −5.06*** −3.35 Retirement −0.32*** 0.00 −1.49*** −0.31 Dependent children Full-time 1.33*** 0.59 −1.08 13.06* Part-time −1.12*** −0.56 0.83 −12.27* Retirement −0.21*** −0.03 0.25 −0.79*

Other control variables (education level and year dummies) are included in ordered probit regressions. Marginal effects are significant at 99% (***), 95% (**), and 90% (*), not significant otherwise.