Regression results, coefficient of Routine Task Intensity index of 2003 occupation
| Men | Women | |||||
| Model w/o controls | Model with controls | Sibling model | Model w/o controls | Model with controls | Sibling model | |
| (1) | (2) | (3) | (4) | (5) | (6) | |
| Dependent variable | ||||||
| Employment per 100, 2018 | −2.8 | −2.0 | −1.5 | −5.0 | −2.1 | −1.6 |
| (−3.5 to –2.1) | (−2.3 to –1.6) | (−1.8 to –1.2) | (−6.9 to –3.1) | (−3.2 to –1.0) | (−2.3 to –0.9) | |
| Disability per 100, 2018 | 3.3 | 1.9 | 1.6 | 5.3 | 2.0 | 1.7 |
| (2.5 to 4.1) | (1.5 to 2.4) | (1.2 to 2.0) | (3.2 to 7.4) | (0.6 to 3.4) | (0.7 to 2.6) | |
| Mortality per 1000, 2019 | 4.4 | 2.4 | 2.4 | 3.3 | 1.3 | 1.4 |
| (3.2 to 5.6) | (1.8 to 3.0) | (1.2 to 3.6) | (1.7 to 4.9) | (0.2 to 2.4) | (−0.1 to 2.9) | |
| Observations | 416 003 | 416 003 | 186 369 | 376 413 | 376 413 | 158 524 |
| Number of families | 83 715 | 71 647 | ||||
| Control variables | None | Age, education, civil status, childless | Age, education, civil status, childless, family fixed effects | None | Age, education, civil status, childless | Age, education, civil status, childless, family fixed effects |
| Sample | All | All | Brothers | All | All | Sisters |
Table entries give change in dependent variable from a 1 SD increase in the Routine Task Intensity (RTI) index of 2003 occupation. Higher RTI scores reflect a greater susceptibility to the effects of automation and globalisation. 95% CIs are reported in brackets; SEs are clustered within occupations. Models in columns (2), (3), (5) and (6) include indicator variables for 20 ages and 13 levels of educational attainment, as well as indicator variables for single status and childlessness in 2003. Employment and disability outcomes are conditional on survival until 2019.