Table 5

Propensity score matching (PSM) analysis

Panel A: first-stage regression using caliper matching
Dependent Variable = FF
Coefficientz-statisticp-value
SIZE−0.051−3.3100.001
LEV1.3177.0100.000
ROA0.7102.7800.005
BM0.0180.2200.827
RET−0.043−1.1600.247
SGROWTH0.0390.6800.497
CFO−0.141−2.0500.040
STD_CFO−0.324−0.8600.389
FAGE0.83118.2300.000
EMPTURN0.2982.1200.034
BLOCK0.0111.7000.090
BSIZE−0.834−10.7000.000
BIND−0.720−5.6400.000
CEO_DUAL−0.022−0.5800.564
CEO_GENDER−0.211−2.5500.011
CEO_OWN−0.661−1.7100.088
Constant0.3651.0100.312
Year fixed effects Yes 
Industry fixed effects Yes 
Observations 15,726 
Pseudo-R2 0.097 
Panel B: comparison of firm characteristics between treatment and control groups
FF = 1FF = 0p-value
SIZE8.0558.1260.011
LEV0.0750.0690.003
ROA0.0330.0360.169
BM0.5070.5050.646
RET0.1220.1240.874
SGROWTH0.0530.05400.900
CFO0.0530.0570.414
STD_CFO0.0720.0730.362
FAGE3.1883.1370.000
EMPTURN0.1420.1390.226
BLOCK0.4950.4650.583
BSIZE1.9431.9600.000
BIND0.2390.2460.017
CEO_DUAL0.3450.3440.921
CEO_GENDER0.9440.9530.036
CEO_OWN0.0070.0070.915
Panel C: second-stage regression results using propensity score matching (PSM) sample
Dependent Variable = CEO_TURN
Model (1)Model (2)
FF−0.272***−0.255***
(–5.413)(–4.982)
FF×TRANS 0.042*
 (1.789)
TRANS −0.059***
 (–3.386)
SIZE0.021−0.011
(0.963)(–0.431)
LEV−0.035−0.088
(–0.134)(–0.329)
ROA−3.465***−3.336***
(–9.900)(–9.228)
BM−0.061−0.059
(–0.532)(–0.506)
RET0.160***0.164***
(3.300)(3.190)
SGROWTH0.0730.075
(0.933)(0.881)
CFO0.164*0.157
(1.695)(1.545)
STD_CFO2.113***2.067***
(4.156)(4.034)
FAGE−0.033−0.027
(–0.504)(–0.405)
EMPTURN1.312***1.309***
(7.104)(7.047)
BLOCK0.0070.006
(0.829)(0.765)
BSIZE0.316***0.319***
(2.882)(2.915)
BIND−0.235−0.272
(–1.199)(–1.388)
CEO_DUAL−0.303***−0.306***
(–5.368)(–5.356)
CEO_GENDER−0.114−0.125
(–1.024)(–1.085)
CEO_OWN−0.273−0.243
(–0.484)(–0.434)
Constant−1.140**−0.863
(–2.296)(–1.645)
Year fixed effectsYesYes
Industry fixed effectsYesYes
Observations10,15210,152
Pseudo-R20.0380.039
Test: TRANS + FF×TRANS 11.47***

Note(s): This table presents the propensity score matching (PSM) analysis. Panel A presents the first-stage logistic regression results. Panel B presents the comparison of firm characteristics between treatment and control groups. Panel C presents the second-stage regression results for the PSM analysis. Numbers in parentheses are t-statistics clustered by firm and year. Superscript ***, ** and * represent significance levels at the 1, 5 and 10% levels, respectively. Variables are defined in  Appendix

Source(s): Authors' own work

or Create an Account

Close Modal
Close Modal