14(23)
within the Swedish Social Insurance Agency. If not before, the individual is
then informed of the new SI rules, which could affect his or her incentives
to resume work. Another possibility is that the timing of the effect coincides
with a typical sick-period length stated in the medical verifications. If the
sick reported did not expect to receive compensation beyond the 12th
week, they could have ignored the possibility to apply for further benefits.
The positive effect at 5-8 weeks is followed by a corresponding drop in the
hazard rate in the subsequent 4-week interval, i.e. the interval immediately
before the 91-day assessment. The result could be due to dynamic
selection where a subgroup of sick reported with relatively good health
ended their sick spell at 5-8 weeks, while a group with relatively poor
health remained. Finally, a large (60.7%) positive effect is found at 25-28
weeks, around the 181-day assessment. The positive effect is thus much
larger than the significant effect before the 91-day assessment. This is
expected since the 181-day assessment is much sharper in the sense that
the assessment concerns the entire regular labour market.
Table 3
Effect on the off-SI hazard ratio
No covariates
Full model
Interval (week)
3-4
0.997
(0.037)
0.981
(0.037)
5-8
1.110***
(0.039)
1.104**
(0.039)
9-12 (91-day assessment)
0.892**
(0.058)
0.886**
(0.058)
13-16
1.106
(0.077)
1.104
(0.078)
17-20
1.059
(0.107)
1.052
(0.107)
21-24
1.057
(0.137)
1.049
(0.138)
25-28 (181-day assessment)
1.598***
(0.132)
1.607***
(0.132)
Year 2008
1.019
(0.021)
1.035*
(0.021)
Month July
0.946***
(0.021)
0.967
(0.021)
-2 log likelihood
322 236
320 089
Note: 19,211 observations. Standard errors are within parentheses. */**/*** report
significance at the 10/5/1% levels. The full model controls for full-time/part-time
sickness absence, first day as sick-reported, medical diagnosis, gender, age, educational
level, born abroad, parent born abroad, marital status, number of children under 18,
sick-reported history, labour market attachment, industry, home county and SI benefits.
Effects are estimated as changes in the time patterns in June and July 2008 compared
with the same months in 2007.
