among middle income States.21 Except Orissa, the performance of the
low income States generally had deteriorated.
Figure 5: PCNSDP Vs. EQI Score: 1990-1996
Figure 6: PCNSDP Vs. EQI Score: 1997-2004
21 While Tamil Nadu needs to control industrial pollution (Mukherjee and Nelliyat,
2006; Appasamy et al., undated); tackling groundwater arsenic contamination in
the rural belt is a major policy challenge for West Bengal (JU, 2006).
34
4.6 Testing the Existence of the Environmental Kuznets Curve (EKC)
For testing the EKC hypothesis, multivariate OLS regression
models are estimated for individual environmental groups. Different
variants of the models are estimated by assuming a non-linearity between
PCNSDP and EQ. Apart from PCNSDP (in Rs. thousand at constant 1993-
94 Prices); various other explanatory variables are introduced to capture
the dynamic aspects of EQ. Tables 6a and 6b present the regression
results which show a mixed picture: while non-linearity exist for a number
of environmental groups like ENERGY, GHGS, LAND, NPS; linear
relationship is observed for other groups like INDOOR, WATER and
FOREST (See Appendix 4 for graphical representation of the obtained
relationships between PCNSDP and various environmental groups).
Similarly, with respect to controlling variables, it is observed that share
of primary sector in GSDP (PRISHARE)22 is negatively related to ENERGY,
LAND, NPS and FOREST. This is because with the fall in share of primary
sector in GSDP; pressure on land, water and forest resources goes down
and EQ improves. With the rise in share of secondary sector in GSDP
(SECSHARE),23 ENRGY score falls and the same for WATER increases
and as the share of tertiary sector improves, the scores of GHGs and
WATER increase. The results imply that composition of income of a
State has substantial impacts on its environmental quality. Increased
share Ofworkers in agriculture (AGRWRK) shows a mixed trend (positive
for GHGS, INDOOR and NPS etc. and negative for LAND and AIRPOL).
Population density (POPD) and level of urbanisation (URB) is generally
showing a negative relationship with EQ.
22 Percentage share of Primary Sector in GSDP (at constant 1993-94 Prices), which
includes Agriculture, Forestry and Logging and Fishing.
23 Secondary sector includes Mining and Quarrying, Manufacturing and Construction.