Table 6.

Panel regression results for predictors with robust standard errors

VariableModel 1 ESG_perf Coeff. (p-value) No lagModel 2 ESG_perf Coeff. (p-value) One-year lagModel 3 ESG_perf Coeff. (p-value) Two-years lag
B_size0.0769 (0.0354)0.0583** (0.0355)0.0672** (0.0371)
B_gend0.1765 (0.0872)0.2880 (0.1402)0.4466 (0.1832)
B_mgend0.2198* (0.1675)0.3159** (0.1626)0.6945** (0.3638)
B_ind0.0554* (0.0306)0.0593** (0.0299)0.0625** (0.0307)
CSR_com0.0339* (0.0234)0.0386* (0.0223)0.0368* (0.0237)
CEO_dual−0.0125(0.0166)*−0.0137 (0.0154)−0.0125 (0.0162)
Size0.0173*** (0.0056)0.0147*** (0.0066)0.0167*** (0.0046)
Lev−0.3905 (0.7493)−0.4077* (0.7534)−0.3342* (0.7426)
Roe0.0146** (0.0067)0.0148** (0.0059)0.0157** (0.0053)
Gdp0.0155 (0.1613)0.0011 (0.1383)0.0057 (0.1478)
RegressionF18.73***15.88***17.31***
R2 within0.43170.23690.1266
R2 between0.57100.49860.0133
R2 overall0.45870.37050.0019
Wald χ279.23**  
Hausman χ223.5429.95*37.71
Fixed/random effectsFixedFixedFixed
Note(s):

n = 119 (number of utility firms). ∑iTi.n = 595 (number of utility firm-year observations). The robust standard errors of the estimated coefficients reported in parentheses are clustered at the firm level *, ** and ***denotes level of significance at the 0.10, 0.05 and 0.01 levels, respectively

Source(s): Authors’ own creation

or Create an Account

Close subscription notice
Close access options