Environmental Performance Index and Economic Development: Evidence from Panel Data Analysis in India’s EAG States
Karuna Shanker Kanaujiya
Department of Applied Economics, University of Lucknow, Lucknow, India.
Sonakshi Singh *
Department of Applied Economics, University of Lucknow, Lucknow, India.
Shivam Agarwal
Department of Applied Economics, University of Lucknow, Lucknow, India.
*Author to whom correspondence should be addressed.
Abstract
This study analyzes the link between economic development (NSDP) and environmental degradation (EPI) in India’s eight Empowered Action Group (EAG) states, testing the Environmental Kuznets Curve (EKC) hypothesis. The EKC posits that environmental degradation first worsens with economic growth before improving. Using panel data, our model includes NSDP and its squared term (NSDP²) to test for this non-linear relationship. We employed Pooled Ordinary Least Squares (OLS), Fixed Effects (FE), and Random Effects (RE) estimation techniques. A rigorous set of model selection tests (Breusch-Pagan LM, F-test, and Hausman) decisively identified the Random Effects (RE) model as the most statistically appropriate and efficient specification, as it properly accounts for state-specific heterogeneity. Our primary finding, based on the preferred RE model, is that no statistically significant relationship exists between NSDP and the EPI. This result, also supported by the FE model, indicates that the EKC hypothesis is not supported for the EAG states. Although a simpler Pooled OLS estimation does show a significant inverted U-shaped curve, we demonstrate this finding is spurious. It is an artifact of model misspecification that fails to control for the significant differences between states. We conclude that economic growth alone does not guarantee environmental improvement. This highlights the critical need for direct, state-specific policy interventions to achieve sustainability.
Keywords: Environmental Kuznets Curve (EKC), EAG states, Environmental Performance Index (EPI), Net State Domestic Product (NSDP), panel data analysis