Table 4.
Effect on transitions from employment and unemployment by gender
| Status at t: | Employment (1) | Unemployment (2) | Disability (3) | Residual category (4) |
|---|---|---|---|---|
| Men | ||||
| A. Employed at t – 1 | ||||
| I(age<ERA) | 27.52*** (0.78) | 2.21*** (0.13) | 0.49*** (0.12) | −0.90*** (0.16) |
| R2 | 0.095 | 0.038 | 0.019 | 0.025 |
| Pre-policy mean | 70.40 | 1.00 | 0.75 | 2.46 |
| #Obs. | 2,077,166 | |||
| #Individuals | 202,544 | |||
| B. Unemployed at t – 1 | ||||
| I(age<ERA) | 0.72 (0.57) | 80.36*** (1.75) | 0.15 (0.28) | 1.53*** (0.18) |
| R2 | 0.091 | 0.233 | 0.050 | 0.014 |
| Pre-policy mean | 2.39 | 21.58 | 0.92 | 1.63 |
| #Obs. | 557,806 | |||
| #Individuals | 82,880 | |||
| Women | ||||
| C. Employed at t – 1 | ||||
| I(age<ERA) | 11.37*** (0.26) | 0.93*** (0.07) | −0.10** (0.04) | −0.33*** (0.06) |
| R2 | 0.056 | 0.027 | 0.003 | 0.027 |
| Pre-policy mean | 88.10 | 2.21 | 0.87 | 1.92 |
| #Obs. | 4,375,311 | |||
| #Individuals | 325,905 | |||
| D. Unemployed at t – 1 | ||||
| I(age<ERA) | 0.45** (0.20) | 44.36*** (1.95) | −0.15 (0.12) | 1.83*** (0.17) |
| R2 | 0.075 | 0.140 | 0.011 | 0.017 |
| Pre-policy mean | 5.85 | 55.35 | 2.08 | 4.30 |
| #Obs. | 792,910 | |||
| #Individuals | 117,777 | |||
Notes: This table displays estimates of equation (1). Standard errors, in parentheses, are clustered at the year-quarter of birth. Coefficient and standard errors are multiplied by 100 and should be interpreted as percentage points. Controls in all specifications include dummies for age in months, dummies for year-quarter, experience in last 15 years, blue-collar status, number of insurance years, annual earnings, average earnings in best 15 years, expected UI benefits, dummies for weeks of UI eligibility (20, 30, 39, 52, 78 weeks), expected retirement benefits, number of sick leave days between ages 45-49 for women and ages 50-54 for men, dummies for industry, dummies for birth cohort at a quarterly frequency, dummies for birth cohort interacted with a second-order polynomial in calendar time, and dummies for age in months interacted with a second-order polynomial in calendar time. The time period is 1997-2010. Reported pre-policy means are for men aged 60-61 and women aged 55-58.2 in 2000. Significance levels:
1%
5%
* = 10%.