Table 6

Heckman’s (1979) two-stage analysis

Predicted signFirst stageSecond stage
Dependent variable = INVOPDependent variable = Firm performance (PERF1)
Model (1)Model (2)Model (3)Model (4)
INVOP+–0.083***0.073***0.119***
  (5.251)(4.656)(5.128)
INVOP × FEMOWN+––0.031**–
   (2.309) 
INVOP × CONTROL––––−0.049**
    (−2.373)
FEMOWN?–0.001−0.022*0.001
  (0.319)(−1.888)(0.328)
CONTROL––−0.005−0.0050.031**
  (−1.574)(−1.553)(1.992)
SIZE+0.131***0.007*0.007*0.007*
 (6.516)(1.746)(1.722)(1.693)
FAGE?−0.087***−0.034***−0.034***−0.034***
 (−5.913)(−8.308)(−8.308)(−8.314)
FOROWN+0.003−0.000−0.0000.000
 (0.066)(−0.021)(−0.010)(0.058)
FINCRED+ 0.0050.0050.005
  (1.324)(1.277)(1.320)
EXPORT+0.139***−0.008−0.008−0.008
 (2.649)(−1.005)(−1.009)(−1.072)
FINCONS–−0.062***−0.001−0.001−0.001
 (−7.391)(−0.587)(−0.572)(−0.581)
BUSRISK–−0.001−0.017***−0.017***−0.017***
 (−0.070)(−10.879)(−10.920)(−10.955)
CORP+−0.022−0.009*−0.009*−0.009*
 (−0.776)(−1.685)(−1.673)(−1.663)
EXTDEP+−0.184***0.0110.0110.011
 (−3.017)(1.371)(1.381)(1.359)
INVOP+0.051*0.004**0.004**0.004*
 (1.790)(1.988)(1.990)(1.981)
INVOP_IND+3.148***–––
 (7.039)   
IMR?–−0.028−0.029−0.028
  (−0.787)(−0.820)(−0.801)
CONSTANT?−2.021***0.283***0.292***0.258***
 (−5.298)(4.395)(4.524)(3.968)
Year fixed effects YesYesYesYes
Industry fixed effects YesYesYesYes
Country fixed effects YesYesYesYes
Observations 36,18536,18536,18536,185
Pseudo R-squared/R-squared 0.0790.0920.0920.092
Test: INVOP + INVOP × FEMOWN = 016.56*** 
Test: INVOP + INVOP × CONTROL = 0 15.56***

Note(s): Superscript ***, **, and * represent statistical significance at the 1, 5, and 10% levels, respectively, two-tailed. Coefficient values (robust t-statistics) are shown with standard errors clustered at the country level. Descriptions of variables are provided in  Appendix

Source(s): Authors’ own work

or Create an Account

Close subscription notice
Close access options