This paper provides new empirical evidence on delinking trends concerning emission-related indicators in Italy. First, methodological issues regarding the analysis of delinking are discussed and the related Environmental Kuznets Curves literature is critically examined to explore and assess the most value added research lines after more a decade of intense research in the field. The main contribution of the paper is that we provide EKC evidence exploiting environmental-economic merged panel datasets at decentralized level exploiting a long times series and rich cross section heterogeneity both at sectoral and Provincial level. This is a necessary and fruitful research direction, following the unsatisfactory outcomes deriving from cross country analyses, which are less informative for policy purposes given they produce average figures for the environmental-economic relationship. Two panel datasets concerning (i) 1990-2000 emissions at Province level (ii) and sectoral disaggregated NAMEA emissions sources over 1990-2001, are analyzed.
We find mixed evidence in support of the EKC hypothesis. Inverted-U shaped curves for the period here considered arise for some of the pollutants in the NAMEA data, like CO2, CH4 and CO, with coherent within range turning points. Nevertheless, other emission trends show a monotonic relationship, or in some cases an N shaped relationship (SOX, NOX, PM10). Other emissions show relatively less robust results, with mixed evidence arising from different specifications. This partially confirms some of the criticism on the EKC empirical investigation that has recently mounted. All in all, nevertheless, our analysis show that probably the key point is that it does not exist an EKC dynamic, but many EKC dynamics, differing by (i) period of observation; (ii) country/area; (iii) emissions/environmental pressures; (iii) sectors.
In fact, a sectoral disaggregated analysis highlights that aggregate outcome should hide some heterogeneity across different sectors. Services tend to present inverted-N shapes in most cases. Manufacturing industry shows a mix of EKC inverted-U and N shapes, depending on the emission considered. The same is true for industry (all industries, not only manufacturing industry): though a turning point has been experienced, N shapes may lead to increase of emissions with respect to very high levels of the income driver.
The analysis on provincial data shows that inverted-U shapes curves arise for some of the emissions in the SINAnet-APAT database, like CH4, NMVOC, CO and PM10, with coherent within range turning point. Other emission trends show a monotonic relationship (CO2 and N2O), or in some cases an inverted-N shaped relationship (SOX and NOX).
This kind of analysis at level of macro sector and/or specific sector appear to be the most promising and robust field of future research for the assessment of EKC dynamics. National studies grounded on geographical heterogeneity, instead of regional/international analysis, and focused on sectoral trends, are more informative for economic and policy implications. The implementation of such investigations nevertheless needs larger datasets, currently not available. We thus point to the need of spending increasing and constant efforts in the construction of integrated environmental/economic statistical accounts.