02.06.2014 Views

Evaluation of the Australian Wage Subsidy Special Youth ...

Evaluation of the Australian Wage Subsidy Special Youth ...

Evaluation of the Australian Wage Subsidy Special Youth ...

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

83<br />

included in <strong>the</strong> employment model. This is <strong>the</strong> Heckman selection correction model<br />

where <strong>the</strong> lambda is <strong>the</strong> inverse mills ratio. The employment model had <strong>the</strong> variables age,<br />

sex, year left school (but not qualifications), post-school formal training (any post-school<br />

qualifications), a post-school job (pre-programme), weeks unemployment (weeks in 52<br />

week year before programme entry), multiple programme participation, dismissal,<br />

voluntary withdrawal, 10 programme dummies with SYETP private as base, state<br />

dummies and <strong>the</strong> LAMBDA correction factor. The response model used a subset <strong>of</strong> <strong>the</strong>se<br />

variables, so that <strong>the</strong> exclusion restriction variables were year left school, post school<br />

formal training and post school job (pre-programme).<br />

No additional results were presented showing <strong>the</strong> sensitivity <strong>of</strong> <strong>the</strong> employment model to<br />

whe<strong>the</strong>r or not nonresponse was accounted for, so it is not clear whe<strong>the</strong>r <strong>the</strong> additional<br />

Heckman selection model performed better than a simpler model without nonresponse.<br />

Baker estimated predicted probabilities for employment for various subgroups using <strong>the</strong><br />

model outcomes. Predicted probabilities were estimated for <strong>the</strong> eligibility period <strong>of</strong> 17<br />

weeks unemployment in <strong>the</strong> past 52 weeks, for a participant who left school in year 10/11,<br />

did post-school training, took part in no o<strong>the</strong>r programme and completed <strong>the</strong> programme<br />

<strong>of</strong> participation. They found that SYETP private generally performed better than<br />

education based programmes, but did no better than o<strong>the</strong>r SYETP, and was significantly<br />

worse than extended SYETP. The results are shown below for SYETP in Table 2.19. It<br />

can be seen that <strong>the</strong> ‘out <strong>of</strong> work’ outcome had a very low probability for all types <strong>of</strong><br />

SYETP.<br />

Table 2.18 Baker (1984) Post-programme full-time employment outcome<br />

Continuous<br />

fulltime work<br />

Non-<br />

Continuous<br />

fulltime work<br />

Retained Not-retained Total<br />

21.1 14.3 35.5 33.7 30.8<br />

SYETP<br />

Commonwealth<br />

SYETP private 34.9 6.3 41.1 38.4 20.4<br />

2nd SYETP 24.3 7.3 31.6 41.2 27.2<br />

Extended SYETP 37.1 5.5 42.6 30.6 26.8<br />

Source: Baker (1984) p19 Table 5.2 May 1982 post-programme Survey <strong>of</strong> participants<br />

Out <strong>of</strong> work

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!