Month | 1 | 2 | 3 | 4 | 5 | 6 |
|---|
Sanction | 0.0092 | 0.1347 | 0.1689 | 0.4475 | 0.5604∗ | 0.6766∗∗∗ |
(based on Z1) | (0.2015) | (0.2383) | (0.2436) | (0.2795) | (0.2927) | (0.2607) |
Sanction | 0.2802∗∗ | 0.2402∗∗ | 0.3594∗∗∗ | 0.4496∗∗∗ | 0.4950∗∗∗ | 0.5802∗∗∗ |
(based on Z2) | (0.1124) | (0.1216) | (0.1249) | (0.1353) | (0.1437) | (0.1396) |
Observations | 15,361 |
- Remarks: The upper part of the table refers to the estimation in which Z1 (sanction strategy of welfare agencies) is used as instrument; the lower part of the table refers to the estimation in which Z2 (observed sanction rates within welfare agencies) is used as instrument. The dependent variable in each estimation and for each month is defined to be 1 if an individual is employed and no longer receives welfare benefits. Otherwise, the variable is 0. Displayed are the estimated sanction effects and standard errors in brackets. The results refer to the first six months after the (hypothetical) sanction date. ∗∗∗denotes p < 0.01,, ∗∗denotes p < 0.05, and ∗denotes p < 0.01. The standard errors take into account clustering at the agency level. Detailed estimation results for Z1 and Z2 including all covariates are displayed in Additional file 1: Table S6 and Table S7.
