Table 4

First-stage ID-TPA choice estimates

Dependent variable: IDTPA
(1)(2)(3)(4)(5)
SOE0.5046***   0.4937***
(10.91)   (10.35)
PC0.2264***   0.2289***
(3.41)   (3.43)
POVERTY0.4132***   0.4341***
(8.54)   (8.90)
DECENTRAL−0.1246***   −0.1254***
(−5.88)   (−5.85)
ADV 0.4046  0.4215
 (1.51)  (1.46)
COVER −0.0144  0.0092
 (−0.66)  (0.37)
MAND_CSR  0.3928*** 0.3534***
  (6.82) (5.94)
ADJ_ROA   0.47961.1068**
   (1.03)(2.09)
SIZE0.3478***0.3689***0.2945***0.3575***0.2748***
(17.34)(17.35)(13.84)(18.38)(11.34)
LEV0.00430.3411***0.4201***0.3906***0.2753*
(0.03)(2.71)(3.43)(2.88)(1.91)
SLACK0.0842−0.0322−0.0250−0.08330.0267
(0.31)(−0.12)(−0.10)(−0.32)(0.10)
IDratio4.9605***5.5145***5.4889***5.4717***5.0531***
(5.67)(6.57)(6.54)(6.52)(5.76)
Constant−5.3275***−5.2143***−4.6956***−5.1286***−4.9520***
(−24.94)(−25.29)(−22.48)(−25.60)(−21.69)
Industry dummiesYesYesYesYesYes
Year dummiesYesYesYesYesYes
Pseudo R20.22150.16090.16870.16060.2298
N8,0358,0358,0358,0358,035

Note(s): This table presents the first-stage estimation results of the two-stage Heckman selection model (Eq. (1)). The dependent variable is IDTPA, an indicator variable equal to one if a company engages in ID-TPA projects in the current year. Columns (1)–(4) report the results when the test variables for hypotheses H1–H4 are individually added, respectively; while column (5) reports the results when all test variables are added simultaneously. Numbers in parentheses are z-statistics. ***/**/*: significant at the 1%/5%/10% level (two-tailed), respectively. Please refer to Appendix 2 for variable definitions

Source(s): Table by authors

or Create an Account

Close subscription notice
Close access options