Industrial characteristics and air emissions: Long-term determinants in the UK manufacturing sector

Paolo Agnolucci,Theodoros Arvanitopoulos

doi:10.1016/j.eneco.2018.12.005

Abstract

Empirical assessments of the relationship between emissions from the industrial sector and the characteristics of the production process are surprisingly scarce in the literature for European countries, despite the industrial sector being one of the major air polluters. In our study, we assess long-term relationship between industrial process and air emissions by building on an existing empirical framework. Our work implies re-estimating published findings for the UK industrial sector on a bigger dataset, incorporating additional observed factors which can plausibly influence the level of emissions and taking into account, for the first time in the empirical literature, unobserved common factors through cross section dependence. In comparison to previous findings we conclude that production inputs, total factor productivity and economies of scale cannot be relied upon to reduce emissions from industrial sector. We provide evidence that reduction in emissions can be reliably delivered by reducing energy consumption, encouraging fuel substitution and by encouraging market competition so that one can counteract the increase in emissions related to higher level of capital investment. We observe considerable similarities in the relationship between market concentration on one side and industrial emissions and innovation on the other side. This is an interesting result in for the energy and environmental economic literature as the relationship between the level of emissions and market structure is a considerably under-researched area.

Highlights

Air pollution affects negatively human health in various ways
Fuel substitution affects more severely SO2 emissions intensity as its elasticity is equal to −1.758, and given that SO2 emissions are highly related to the level of sulphur used in the manufacturing process, this result reflects the fact that coal and oil contain high levels of sulphur while gas contains only negligible quantities (Brown et al, 2017; Wakeling et al, 14 Coefficients in Table 4 are mostly higher in absolute value than those in CES (2005) with the exception of the coefficient on the SO2 and PM10 models
Physical Capital Intensity (PCI) remains significant only for nitrogen oxides (NOx) and its estimated value increases by more than 50%. 16 As mentioned in Section 3, the CCE Pooled (CCEP) estimator is robust to shocks affecting some or all panel units such as the 2008 economic crisis or perhaps the impact of the business cycle, the dynamics of which may very different across industrial subsectors

Summary

Introduction

Air pollution affects negatively human health in various ways. Particular matter exposition, for example, raises the risk of developing cardiovascular diseases and lung cancer (Guerreiro et al, 2016), ecosystems are affected by air pollution through acidification, eutrophication, and ground level ozone (Wald, 2016) while greenhouse gases are the dominant cause of climate change (IPCC, 2014) with related impacts on health (Watts et al, 2015), the economy (Stern, 2008), and ecosystems (Walther et al, 2002). Cole et al (2005), onwards CES (2005), drew attention to the shortage of studies assessing the relationship between industrial activity and air pollution outside of the United States and provided the first empirically investigation of this relationship in the UK manufacturing sector using data from 1990 to 1998. Cole et al (2005), onwards CES (2005), drew attention to the shortage of studies assessing the relationship between industrial activity and air pollution outside of the United States and provided the first empirically investigation of this relationship in the UK manufacturing sector using data from 1990 to 1998. The manufacturing sector is one of the major air polluters in Europe (ONS, 2016a; EEA, 2015), not many studies have responded to the task highlighted by CES (2005) so that this apparent lack of investigative effort is preventing a rigorous understanding of the historical determinants of emissions from the manufacturing sector, at least in Europe. When focusing on the industrial sector, analysis has been often implemented by using decomposition analysis rather than econometric modelling, e.g. Dachraoui and Harchaoui (2006), Kim and Kim (2012), Liaskas et al (2000), Tan and

Results

Discussion

Conclusion