Skip to main content

Table 10 Effect of interest for graduation period 1998–2002

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

Panel A: Effect of early non-employment in the structural equation:  
Standard errors a Robust Cluster gp Robust Cluster gp
Continuous outcomes: b   (1) (2) (3) (4)
Log earnings coeff –0.0287*** –0.0287*** –0.1406** –0.1406**
  se (0.0045) (0.0049) (0.0666) (0.0597)
  P-val   6.09E-06   0.027298462
  Bootstrap P-val c   0   0.082082082
  Exogeneity test P-val d     0.0306
Log hours worked coeff –0.0215*** –0.0215*** –0.1032** –0.1032**
  se (0.0034) (0.0035) (0.0492) (0.0439)
  P-val   3.23E-06   0.027651497
  Bootstrap P-val   0   0.078078078
  Exogeneity test P-val     0.0318
Panel B: Effect of the instrument in the first stage : OLS  
Outcome: Standard errors: Robust Cluster (g*p)   
Early non-empl.(% hours) Coeff 11.9484*** 11.9484***   
  se (3.4994) (3.4918)   
  P-val 0.00233    
  Bootstrap P-val 0.07007    
  F stat 11.70921    
  Bootstrap F stat e 3.60923    
  1. Standard errors between parentheses. Panel A reports results from estimating β in Eq. (2). β is the effect of one pp increase in \(y^{0}_{it_{1}}\), i.e. the % of hours spent in non-employment at potential experience 0–2 relative to potential total hours if one would work full-time during the whole period. For clustered standard errors, we report the P-value and the wild bootstrap P-value. Column 1–2 (3–4) show OLS (2SLS). In 2SLS the provincial unemployment rate at graduation is used as instrument for \(y^{0}_{it_{1}}\!\). Panel B shows the effect of the instrument on \(y^{0}_{it_{1}}\) in the first stage and the corresponding F statistic
  2. ***p <0.01, **p <0.05, *p <0.1
  3. aRobust indicates heteroskedastic-robust standard errors. Clusters are defined by graduation year g and province of residence at graduation p (G =24 clusters)
  4. bThe outcomes are measured at potential experience 6. For continuous outcomes we add value one before taking the log, so that non-salaried employed at the moment of evaluation are included with outcomes equal to zero after the logarithmic transformation
  5. cComputed according to the wild bootstrap procedure proposed by Davidson and MacKinnon (2010) for 999 repetitions
  6. dWith clustered standard errors, this test is defined as the difference between two Sargan-Hansen statistics: one for the equation where \(y^{0}_{it_{1}}\) 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)
  7. eBootstrap F statistic is the F statistic corresponding to the bootstrap P-value of the t statistic of the instrument: we rely on the equivalence between F and t distribution: t 2(G−1)=F(1,G−1), with G=24