Skip to main content

Table 7 Complete estimations on continuous outcomes

From: Scars of early non-employment for low educated youth: evidence and policy lessons from Belgium

  Second stage on continuous outcomes First stage
Outcomes a Log earnings Log hours worked Early non-empl.
  OLS 2SLS OLS 2SLS OLS
Clustered standard errors: gp gp gp gp gp
  (1) (2) (3) (4) (5)
UR_grad      5.4615***
      (1.6848)
Early non-empl. –0.0269*** –0.1002*** –0.0203*** –0.0723***  
  (0.0040) (0.0291) (0.0029) (0.0207)  
UR_pe6 0.4823** 0.4172** 0.3341** 0.2880** 0.3779
  (0.1904) (0.1919) (0.1382) (0.1392) (1.6071)
lin_grad_year –0.1588 0.0469 –0.1139 0.0319 3.5329
  (0.2945) (0.2995) (0.2134) (0.2191) (2.2946)
lin_grad_year |trend>3 0.9033** 0.6407 0.6676** 0.4814 –4.4281
  (0.4003) (0.4242) (0.2915) (0.3109) (2.8479)
lin_grad_year |trend>6 –0.4604* –0.4394 –0.3335* –0.3186 3.9169
  (0.2582) (0.2852) (0.1866) (0.2036) (3.1378)
d_province2 –1.3812* –1.2603 –0.9479* –0.8622 3.6574
  (0.7107) (0.9544) (0.5041) (0.6684) (8.6769)
d_province3 –2.1753*** –2.9226*** –1.5774*** –2.1070*** –11.7452
  (0.6066) (0.9315) (0.4344) (0.6703) (8.9621)
d_province4 0.0841 0.1824 0.0740 0.1436 0.7495
  (0.4801) (0.4983) (0.3504) (0.3570) (4.7080)
d_province5 0.8990 0.9210 0.6674 0.6829 –15.1893**
  (0.6099) (0.6721) (0.4495) (0.4866) (6.6042)
lin_calend_year_prov2 0.0498 0.0410 0.0349 0.0287 0.0774
  (0.1386) (0.1413) (0.1019) (0.1024) (0.9931)
lin_calend_year_prov3 0.0723 0.1638 0.0588 0.1236 2.2624
  (0.1136) (0.1441) (0.0811) (0.1040) (1.7325)
lin_calend_year_prov4 –0.0946 –0.1524 –0.0684 –0.1094 –1.0198
  (0.1267) (0.1477) (0.0927) (0.1051) (1.2618)
lin_calend_year_prov5 –0.0813 –0.0760 –0.0593 –0.0555 1.8419
  (0.1067) (0.1194) (0.0795) (0.0868) (1.1761)
avg_UR_pe3-6 –1.2836*** –1.4348** –0.9652*** –1.0724*** –0.3037
  (0.4509) (0.5863) (0.3262) (0.4140) (4.4775)
Min_UR_pe0-6 –0.5718 –0.4848 –0.3128 –0.2511 –6.4742
  (0.6219) (0.6613) (0.4391) (0.4664) (7.1097)
Birth cohort76 0.7043 –0.3300 0.5421 –0.1909 –13.2641***
  (0.5632) (0.7686) (0.4125) (0.5583) (3.6802)
Birth cohort78 0.4261 –0.0504 0.3188 –0.0190 –5.9156**
  (0.3677) (0.4687) (0.2691) (0.3409) (2.5815)
Live in single-parent 0.3453 0.7897 0.2731 0.5880 6.1392
  (0.4632) (0.5207) (0.3466) (0.3840) (4.1354)
Not live with parents 0.4272* 0.5203 0.3476* 0.4136 1.0913
  (0.2431) (0.3501) (0.1792) (0.2530) (2.5611)
HH members aged 0-11 –0.0177 0.0846 –0.0462 0.0263 1.3007
  (0.1123) (0.1472) (0.0878) (0.1069) (1.1570)
HH members aged 12–17 0.1680 0.1467 0.1261 0.1110 –0.2429
  (0.1177) (0.1473) (0.0883) (0.1087) (0.9030)
HH members aged 18–29 0.0112 0.2054 0.0073 0.1449 2.6587**
  (0.1164) (0.1561) (0.0870) (0.1134) (1.0291)
HH members aged 30–64 –0.0334 –0.0801 –0.0112 –0.0443 –0.4836
  (0.4065) (0.4287) (0.3022) (0.3129) (3.2574)
HH members aged 65+ –0.0680 0.0308 –0.0289 0.0411 1.2468
  (0.3639) (0.4196) (0.2683) (0.3079) (3.1042)
Father education 0.0011 0.0294 0.0000 0.0201 0.3816
  (0.0255) (0.0330) (0.0188) (0.0240) (0.2531)
Mother education –0.1053*** –0.0485 –0.0765*** –0.0363 0.7800**
  (0.0356) (0.0432) (0.0269) (0.0317) (0.2925)
Years of delay in sec. edu –0.0545 0.3485* –0.0423 0.2433 5.4123***
  (0.1112) (0.2094) (0.0808) (0.1510) (1.1119)
Technical edu 0.4431 –0.4557 0.3563 –0.2807 –11.8663***
  (0.3274) (0.4733) (0.2470) (0.3399) (2.9533)
Vocational edu 0.4572 –0.3520 0.3675* –0.2059 –10.6291***
  (0.2772) (0.4516) (0.2101) (0.3233) (3.3428)
Apprenticeship/PT edu –0.1879 –1.1341* –0.1093 –0.7798* –12.5300**
  (0.4530) (0.6234) (0.3375) (0.4500) (4.7050)
Constant 15.0564*** 18.1335*** 10.7790*** 12.9598*** 24.5592
  (3.7529) (5.2571) (2.6949) (3.7277) (49.3992)
Observations 1,902 1,902 1,902 1,902 1,902
R-squared 0.0895 –0.2401 0.0895 –0.2099 0.1070
F stat of first step b      10.51
Exogeneity test P-val c   0.00970   0.0112  
  1. Standard errors between parentheses. Columns 1–4 report the results from estimating Eq. (2) by OLS (odds columns) and 2SLS (even columns). Column 5 reports OLS results from estimating the first stage. All estimations report cluster-robust standard errors by graduation year g and province of residence at graduation p (G=44). Column 5 reports the F statistic of the first stage and even columns report the exogeneity test for early non-employment \((y^{0}_{it_{1}}\!)\)
  2. ***p <0.01, **p <0.05, *p <0.1
  3. aContinuous outcomes are measured at potential experience 6; early non-employment is measured at potential experience 0–2. For continuous outcomes we add value one before taking the logarithm, so that those who are not salaried employed at the moment of evaluation are included with outcomes equal to zero after the logarithmic transformation
  4. bThis statistic is not corrected for the problem of few clusters. The corrected value resulting from the bootstrap procedure is 9.25 (see Table 1)
  5. cWith clustered standard errors, the exogeneity test is defined as the difference between two Sargan-Hansen statistics: one for the equation where \(y^{0}_{it1}\) is treated as endogenous and one for the equation where \(y^{0}_{it_{1}}\) is treated as exogenous. Under the null that \(y^{0}_{it_{1}}\) is exogenous, the statistic is distributed as χ 2(1). This statistic is not corrected for the problem of few clusters