Skip to main content

Table 9 Cross-sectional analysis: heterogeneity of city commercial banks

From: Does the target market affect bank performance? Evidence from the geographic diversification of city commercial banks in China

Dependent variable

Above median

Below median

(1) ROA

(2) ROE

(3) NPL ratio

(4) Cost

(5) Market share

(6) ROA

(7) ROE

(8) NPL ratio

(9) Cost

(10) Market share

Panel A:Bank size

 1–HHI

− 0.003 (0.016)

0.007 (0.039)

1.598b (1.043)

0.002 (0.004)

0.348c (0.091)

− 0.002 (0.026)

− 0.007 (0.071)

1.241 (0.825)

0.008 (0.007)

0.254c (0.045)

 N

447

447

447

447

447

361

361

361

361

361

 R2

0.777

0.780

0.578

0.783

0.979

0.846

0.725

0.702

0.704

0.982

Panel B:Bank age

 1–HHI

−0.002 (0.003)

−0.049 (0.048)

1.425 (1.103)

0.007 (0.005)

0.332c (0.074)

− 0.007b (0.003)

− 0.116b (0.048)

1.377a (0.801)

0.009 (0.005)

0.310c (0.056)

 N

456

456

456

456

456

352

352

352

352

352

 R2

0.648

0.651

0.406

0.625

0.986

0.840

0.741

0.874

0.721

0.981

Panel C:Capital adequacy ratio

 1–HHI

−0.001 (0.003)

−0.018 (0.038)

0.814 (0.670)

0.002 (0.004)

0.187c (0.054)

− 0.007b (0.003)

− 0.121b (0.054)

4.410c (1.226)

0.015b (0.007)

0.519c (0.086)

 N

427

427

427

427

427

381

381

381

381

381

 R2

0.810

0.719

0.532

0.817

0.987

0.652

0.670

0.562

0.595

0.969

Panel D:NPL

 1–HHI

−0.005c (0.002)

−0.087c (0.031)

3.642b (1.502)

0.013b (0.006)

0.411c (0.093)

− 0.004 (0.003)

− 0.069 (0.046)

0.699 (0.507)

0.007 (0.005)

0.161c (0.033)

 N

419

419

419

419

419

389

389

389

389

387

 R2

0.589

0.615

0.611

0.593

0.974

0.832

0.680

0.676

0.797

0.980

Panel E:Non-interest income

 1–HHI

−0.010c (0.003)

−0.115c (0.044)

3.094a (1.613)

0.001 (0.006)

0.262c (0.075)

− 0.003 (0.003)

− 0.036 (0.046)

1.806b (0.823)

0.011b (0.005)

0.373c (0.065)

 N

432

432

432

432

432

376

376

376

376

376

 R2

0.702

0.707

0.510

0.497

0.976

0.832

0.729

0.620

0.865

0.981

  1. This table shows the results of cross-sectional analyses. The sample was divided into two subsamples, Large and Small, based on the sample medians of bank size, bank age, capital adequacy ratio, non-performing loan, and non-interest income. Year and Bank fixed effects are controlled. Standard errors are robust and reported in parentheses. a, b, and c indicate significance at 10%, 5%, and 1%levels, respectively