Table 9

Entropy balancing estimations

Panel A: Proof of convergence
Without weighting (before-balanced)
TreatedControl
VariablesMeanVarianceSkewnessMeanVarianceSkewness
TANG0.3750.0690.9240.3810.0690.845
SIZE7.6624.0320.4037.8944.0060.295
ROA0.0490.0110.2380.0470.0100.362
LEVE0.3470.0891.3790.3450.0861.315
CASH0.0960.0212.5920.0990.0212.560
SLACK0.6505.5536.8720.6705.7456.698
Ln_OC5.1700.7100.5835.1650.7010.571
ROE0.0920.094−0.8310.0820.086−0.969
CFO0.0610.06414.3400.0523.756−18.830
Ln_Age2.7731.621−1.2242.7661.692−1.214
TOBINSQ0.5601.4423.8330.4861.0344.473
GDP_GRO7.5440.345−0.6045.02217.930−1.652
After weighting variable
TreatedControl
VariablesMeanVarianceSkewnessMeanVarianceSkewness
TANG0.3750.0690.9240.3750.0630.805
SIZE7.6624.0320.4037.6623.7390.304
ROA0.0490.0110.2380.0490.0080.331
LEVE0.3470.0891.3790.3470.0760.944
CASH0.0960.0212.5920.0960.0192.624
SLACK0.6505.5536.8720.6504.9507.015
Ln_OC5.1700.7100.5835.1700.7020.504
ROE0.0920.094−0.8310.0920.053−0.350
CFO0.0610.06414.3400.0615.13979.400
Ln_Age2.7731.621−1.2242.7731.539−1.300
TOBINSQ0.5601.4423.8330.5601.5544.355
GDP_GRO7.5440.345−0.6047.54312.750−3.477
Panel B: Estimations with entropy balanced sample
Investment inefficiencyOverinvestmentUnderinvestment
VariablesModel 1Model 2Model 3Model 4Model 5Model 6Model 7Model 8Model 9
GPR0.0083***  0.0161***  0.0037**  
(0.002)  (0.004)  (0.002)  
GPRT 0.0136***  0.0262***  0.0065*** 
 (0.002)  (0.007)  (0.002) 
GPRA  0.0047***  0.0092***  0.0009
  (0.001)  (0.003)  (0.002)
TANG0.0494***0.0494***0.0494***0.0837***0.0837***0.0837***−0.0187***−0.0187***−0.0187***
(0.010)(0.010)(0.010)(0.013)(0.013)(0.013)(0.003)(0.003)(0.003)
SIZE−0.0057***−0.0057***−0.0057***−0.0157***−0.0157***−0.0157***−0.0021***−0.0021***−0.0021***
(0.001)(0.001)(0.001)(0.002)(0.002)(0.002)(0.000)(0.000)(0.000)
ROA0.0774***0.0774***0.0774***0.1007*0.1007*0.1007*−0.0267***−0.0267***−0.0267***
(0.027)(0.027)(0.027)(0.058)(0.058)(0.058)(0.004)(0.004)(0.004)
LEVE0.0327***0.0327***0.0327***0.0468***0.0468***0.0468***0.0129***0.0129***0.0129***
(0.004)(0.004)(0.004)(0.010)(0.010)(0.010)(0.001)(0.001)(0.001)
CASH0.0339***0.0339***0.0339***0.0752***0.0752***0.0752***−0.0029−0.0029−0.0029
(0.008)(0.008)(0.008)(0.016)(0.016)(0.016)(0.002)(0.002)(0.002)
SLACK0.0016***0.0016***0.0016***0.0071***0.0071***0.0071***0.0014***0.0014***0.0014***
(0.000)(0.000)(0.000)(0.002)(0.002)(0.002)(0.000)(0.000)(0.000)
Ln_OC−0.0005−0.0005−0.0005−0.0071**−0.0071**−0.0071**0.0007*0.0007*0.0007*
(0.001)(0.001)(0.001)(0.003)(0.003)(0.003)(0.000)(0.000)(0.000)
ROE0.00070.00070.00070.00130.00130.0013−0.0026***−0.0026***−0.0026***
(0.003)(0.003)(0.003)(0.011)(0.011)(0.011)(0.001)(0.001)(0.001)
CFO−0.0003−0.0003−0.0003−0.0294**−0.0294**−0.0294**0.00030.00030.0003
(0.000)(0.000)(0.000)(0.011)(0.011)(0.011)(0.000)(0.000)(0.000)
Ln_Age−0.0057***−0.0057***−0.0057***−0.0121***−0.0121***−0.0121***−0.0009***−0.0009***−0.0009***
(0.001)(0.001)(0.001)(0.002)(0.002)(0.002)(0.000)(0.000)(0.000)
TOBINSQ0.00120.00120.00120.00810.00810.0081−0.0017***−0.0017***−0.0017***
(0.002)(0.002)(0.002)(0.006)(0.006)(0.006)(0.000)(0.000)(0.000)
GDP_GRO0.0004***0.0003***0.0004***0.0000−0.00000.00010.0007***0.0007***0.0007***
(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)
Intercept0.0606***0.0345**0.0784***0.1660***0.1154**0.2004***0.1231***0.1562***0.1006***
(0.012)(0.015)(0.010)(0.031)(0.041)(0.027)(0.005)(0.006)(0.004)
Industry FEYESYESYESYESYESYESYESYESYES
Year FEYESYESYESYESYESYESYESYESYES
Observations39,59939,59939,59911,91911,91911,91927,68027,68027,680
Adj. R20.08330.08330.08330.12330.12330.12330.12370.12370.1237

Note(s): This table presents the effects of GPR, GPRT, and GPRA on investment inefficiency, overinvestment, and underinvestment using the entropy balancing method. Firm-year observations with GPR values exceeding the median are categorized into the treatment group, while those with GPR values below the median are assigned to the control group. Panel A demonstrates the proof of convergence after entropy balancing. Panel B reports the regression results post-entropy balancing. All models employ pooled panel OLS estimation. Robust standard errors in brackets are clustered by firm and year to address serial correlation and heteroskedasticity. The measurement of the variables is in  Appendix Table A1. *, **, and *** indicate significance levels of 10%, 5%, and 1%, respectively

Source(s): Authors own work

or Create an Account

Close Modal
Close Modal