Abstract:
The paper assesses the existence of the Environmental Kuznets Curve (EKC) hypothesis, across
48 contiguous states of the US, using recent advances in panel data techniques, given the
existence of cross-sectional dependence, which in turn, makes reliance on time-series evidence
biased. The Common Correlated Effects (CCE) estimation procedure of Pesaran, (2006), allows
us to obtain state-level results, while staying in a panel set-up to accommodate for cross-sectional
dependence, in the presence of cointegration in the relationship between emissions and
a measure of output, and its squared value – a function that captures the inverted u-shaped
relationship postulated by the EKC. Our results show that, the EKC hypothesis holds for only 10
of the 48 states, and hence implies that, the remaining 38 states should reform a number of their
environmental regulatory policies to prevent environmental degradation, since otherwise, lower
levels of emissions would only be possible at the expense of production.
