Table 8 Results for propensity score matching analysis.

From: Does corporate digitalization promote labor investment efficiency? Evidence from Chinese listed companies

Panel A: Balance test

Variables

Sample

Mean

Bias (%)

T-test

Treatment

Control

T-statistics

P-value

Mtb

Unmatched

1.8921

1.9469

−4.5

−3.71

0.000

Matched

1.8921

1.9307

−3.2

−1.51

0.139

Size

Unmatched

22.421

22.031

30.9

25.49

0.000

Matched

22.421

22.383

3.0

2.37

0.018

Quick

Unmatched

1.1961

1.1470

3.8

3.15

0.002

Matched

1.1961

1.2117

−1.2

−0.94

0.348

Lev

Unmatched

0.4643

0.4550

4.7

3.68

0.000

Matched

0.4643

0.4684

−2.1

−1.71

0.087

Divd

Unmatched

0.7117

0.6867

5.5

4.49

0.000

Matched

0.7117

0.7172

−1.2

−0.96

0.336

Sdcfo

Unmatched

0.0467

0.0344

16.5

13.69

0.000

Matched

0.0467

0.0440

3.6

1.70

0.064

Sdsale

Unmatched

0.1966

0.1331

18.8

15.61

0.000

Matched

0.1966

0.1780

5.5

1.78

0.089

Tangib

Unmatched

0.2432

0.2436

−0.3

−0.23

0.621

Matched

0.2432

0.2420

0.7

0.52

0.801

Loss

Unmatched

0.1032

0.1052

−0.6

−0.53

0.293

Matched

0.1032

0.1074

1.3

−1.07

0.586

Inst

Unmatched

0.2413

0.2402

−1.0

0.81

0.001

Matched

0.2413

0.2365

4.3

3.30

0.491

Sdhire

Unmatched

0.5232

0.5418

−1.2

−0.95

0.342

Matched

0.5232

0.5243

−0.1

−0.06

0.953

Labor

Unmatched

0.0775

0.0860

−11.1

−9.09

0.000

Matched

0.0775

0.0784

−1.2

−1.02

0.310

Abinv

Unmatched

0.0304

0.0325

−7.9

−6.50

0.000

Matched

0.0304

0.0304

0.0

−0.02

0.981

Panel B: Univariate tests using the PSM sample

 

Treated

N

Controls

N

Difference

T-Statistics

Ablabor

0.1525

12,631

0.1662

7153

−0.0137***

−3.86

Panel C: Regression results using the PSM sample

 

Dependent variable = Ablabor

 

(1)

(2)

Digital1

−0.0077*** (−4.10)

   

Digital2

   

−0.0503*** (−3.00)

Controls

Yes

Yes

Year

Yes

Yes

Industry

Yes

Yes

N

19,784

19,784

Adj_R2

0.1129

0.1129

  1. *, ** and *** denote significance at the 10%, 5% and 1% level, respectively. T-statistics (in parentheses) are based on standard errors clustered by firm