Table 5

Endogeneity issue – 2SLS model and propensity score matching

(1) First stage(2) Second stage(3) PSM
VariablesNature_DepWACCWACC
Nature_Dep 0.047***0.002***
 (3.598)(2.671)
Nature_Dep_iv0.185***  
(4.975)  
Size0.016***−0.001***−0.000
(7.239)(−4.000)(−1.025)
ROA0.011−0.016***−0.014***
(0.402)(−6.013)(−4.491)
Lev−0.052**−0.036***−0.039***
(−2.463)(−19.852)(−19.816)
Capex0.252***0.020***0.021***
(3.444)(2.965)(2.621)
Salesgrowth−0.016***0.004***0.003***
(−3.916)(7.510)(5.848)
Tangibility0.259***−0.020***−0.007***
(11.635)(−5.360)(−3.699)
Tobinsq0.0020.001***0.001***
(1.433)(3.857)(4.327)
GDP−0.046*−0.047***−0.037***
(−1.932)(−15.894)(−6.948)
Constant2.536***0.468***0.454***
(8.427)(8.944)(8.519)
Observations56,12356,12345,120
R-squared0.2530.1380.393
Industry FEYesYesYes
Country FEYesYesYes
Year FEYesYesYes
First stage F-statistic of excluded instrument 24.75*** 
Anderson-Rubin 95% confidence interval [0.0214, 0.0728] 
Kleibergen-Paap rk Lagrange Multiplier (LM) statistics 26.905*** 

Note(s): This table reports the fixed-effects regression results addressing potential endogeneity concerns using a two-stage least squares (2SLS) approach and propensity score matching (PSM). In columns (1) and (2), the dependent variable in the first stage is Nature_Dep, while WACC is used in the second-stage regression. The instrument variable (Nature_Dep_iv) is the average nature dependence of industry peers in the same country and year, excluding the focal firm. Column (3) reports the results based on a matched sample obtained from the PSM approach. Industry, country, and time fixed-effects are considered in all regressions. Heteroscedasticity-robust standard errors are clustered at the firm level in parentheses. *, **, *** indicate the significance levels at 10%, 5% and 1%, respectively

Source(s): Authors' calculations
[ViewLarge]
This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal