Table 10.

Heckman selection bias analysis

Variable(s)First stageSecond stage
DEP = GRIDEP = CSR policy
Coefficient
(t-value)
Coefficient
(t-value)
Env0.0155** (2.36) 
Framework 0.1010*** (10.44)
Foreign_sales0.0018 (1.42)0.0002* (1.79)
Leverage0.2385 (0.85)0.0430** (2.13)
Profit−2.1477*** (−3.01)0.1426*** (2.66)
Size0.1035*** (3.13)0.0056** (2.19)
MTB−0.0013 (−0.09)−0.0010 (−0.90)
Analysts0.1006 (1.44)0.0445*** (8.93)
Acnonex0.0017 (0.50)−0.0001 (−0.56)
Shright0.0755 (0.55)0.0066 (0.62)
Csrcommittee0.7918 (8.82)0.0111 (1.23)
Awards0.3495*** (4.85)0.0186*** (3.32)
Diversity0.0088*** (3.97)−0.0012*** (−7.42)
Institution1.9861* (1.91)−0.1180 (−1.58)
GDP−0.8393* (−2.73)0.0257 (1.19)
Civil1.3505*** (11.84)−0.0641*** (−4.24)
Lambda −0.1261 (−8.06)
Constant−40.3968** (−2.33)0.5083** (2.06)
Year_FEYesYes
Industry_FEYesYes
Country_FEYesYes
N24392439
LR chi2 (F)1123.8759.77
Pseudo R2 (Adj.R2)0.36670.5090

Note(s): Table 10 reports the results of the Heckman Selection Bias regarding the relationship between Harmonisation and multiple global CSR pronouncements. The first stage dependent variable is GRI (Indicator variable equal to 1 for firms that uses GRI guidelines/standards, 0 otherwise). The second stage results on the effect of harmonisation and multiple global CSR pronouncements, where dependent variable is CSR_POLICY. All variable definitions appear in  Appendix. ***, ** and * represent statistical significance at the 1, 5 and 10% levels, respectively (two-tailed tests)

Source(s): Table by author(s)

or Create an Account

Close Modal
Close Modal