Table 10 DKSE, FE, Two-step GMM, and Granger causality results

Model 1A using DKSE when the dependent variable is LEXP

 PRUND

-0.1924***

0.042083

-4.57

0.000

-0.28121

-0.10363

F(5,17) = 660.10

Prob > F = 0.0000

within R² = 0.6603

 GDPPC

-0.00023

0.000157

-1.43

0.170

-0.00056

0.000107

 GOVEXP

0.2805

0.264204

1.06

0.303

-0.27686

0.837985

 MNSCHOOL

5.3427***

0.305739

17.47

0.000

4.697669

5.987775

 URBAN

-0.0107

0.010292

-1.04

0.311

-0.03245

0.010973

_ CONS

37.8656***

2.066197

18.33

0.000

33.50632

42.22491

Model 1B using DKSE when the dependent variable is LEXP

 AVRDES

0.1762***

0.041312

4.27

0.001

0.089105

0.263427

F(5,17) = 594.21

Prob > F = 0.0000

within R² = 0.6472

 GDPPC

-0.0002

0.000148

-1.35

0.194

-0.00051

0.000112

 GOVEXP

0.1238

0.304085

0.41

0.689

-0.51776

0.765368

 MNSCHOOL

5.2847***

0.322442

16.39

0.000

4.604474

5.965058

 URBAN

-0.0105

0.009715

-1.08

0.293

-0.03103

0.009964

 _CONS

15.2589***

3.5383

4.31

0.000

7.793746

22.72407

Model 1C using FE when the dependent variable is INFMOR

 PRUND

0.9785***

0.057732

16.95

0.000

0.86511

1.091941

sigma_u = 25.168803

sigma_e = 6.3846649

F test that all u_i = 0: F(30, 522) = 126.57

Prob > F = 0.0000

within R² = 0.6888

 GDPPC

0.0035***

0.000517

6.88

0.000

0.00254

0.004569

 GOVEXP

-0.3123

0.52411

-0.6

0.551

-1.34201

0.717236

 MNSCHOOL

-15.1801***

0.658325

-23.06

0.000

-16.4734

-13.8868

 URBAN

-0.0132

0.036319

-0.37

0.715

-0.08462

0.058075

 _CONS

100.7214***

3.414677

29.5

0.000

94.01316

107.4296

Model 1D using GMM when the dependent variable is INFMOR

 L_INFMOR

22.6035***

5.349301

4.23

0.000

12.11909

33.08797

No of instruments = 21, No of groups = 31

AR(1): z = -1.37 Pr > z = 0.170

AR(2): z = 0.30 Pr > z = 0.768

Sargan test of over-identification: Chi2(15) = 6.80 Prob > chi2 = 0.963

Hansen test of over-identification: Chi2(15) = 17.36 Prob > chi2 = 0.298

 AVRDES

-0.2809

0.175223

-1.6

0.109

-0.6244

0.062461

 GDPPC

-0.0038**

0.001689

-2.27

0.023

-0.00715

-0.00053

 GOVEXP

-3.2419***

1.057435

-3.07

0.002

-5.31445

-1.16938

 MNSCHOOL

6.8813

6.48081

1.06

0.288

-5.82083

19.58348

 URBAN

0.0955*

0.057111

1.67

0.094

-0.01641

0.20746

Dumitrescu and Hurlin [124] Granger causality test

 Null hypothesis

W-bar

Z-bar

Z-bar tilde

p-value

Decision

 PRUND does not Granger-cause LEXP

18.3305

68.2301

50.7996

0.0000

PRUND does Granger-cause LEXP

 AVRDES does not Granger-cause LEXP

14.2891

52.3192

38.8384

0.0000

AVRDES does Granger-cause LEXP

 PRUND does not Granger-cause INFMOR

5.4921

17.6854

12.8020

0.0000

PRUND does Granger-cause INFMOR

 AVRDES does not Granger-cause INFMOR

5.7553

18.7214

13.5808

0.0000

AVRDES does Granger-cause INFMOR