Sustainable development and industrial development: manufacturing environmental performance, technology and consumption/production perspectives

ABSTRACT

Industrial development has always been seen as the main engine for economic growth due to its large economic multiplier and technological opportunities. However, manufacturing sectors are directly and indirectly responsible for a large share of overall environmental pressures, raising concerns for the environmental sustainability of manufacturing-based development. In this paper, we evaluate the drivers and decoupling trends of environmental pressures arising (directly or indirectly) from manufacturing production and consumption for a large selection of developed and developing countries. As a first step, we decompose changes in emission intensity of manufacturing sectors into a series of components by means of a shift-share analysis to identify the main drivers of change. A second step will compare direct environmental pressures generated by manufacturing sectors (production perspective) with the amount of emissions generated (domestically and abroad) by the domestic consumption of manufacturing goods (consumption perspective). Finally, we evaluate the possible emergence of an environmental Kuznets curve (EKC) dynamics for production and consumption perspective emissions for the world as a whole and for different continents. Results highlight that first, high income countries are generally more environmental friendly than the average and tend to be specialised in high-tech and greener sectors. Second, emission reduction is driven mainly by unobserved factors such as institutional quality and policy commitment. Finally, while production perspective shows some evidence of EKC dynamics, this result does not hold when shifting to the consumption perspective. Besides, some world area is able to compensate the growth effect exploiting technology dynamics.

KEYWORDS:

• Industrial development
• environmental efficiency
• shift-share analysis
• production–consumption perspective
• environmental Kuznets curve

JEL CLASSIFICATION:

• Q55
• Q56
• O19

Acknowledgments

We thank Nicola Cantore (UNIDO) for very helpful comments. Usual disclaimer applies.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Following EEA (2014b) the production perspective approach considers only direct domestic emissions while final demand consists of total final demand (domestic demand and export) of domestically produced goods. The consumption perspective approach considers also foreign emissions by including in the matrix of inter-industry transactions both domestically produced and imported intermediate inputs, while final demand includes overall demand by resident agents, thus including domestic and imported final consumption but excluding exports.

2. The role of Trade is often complex for what it concerns its dynamic relationship with environmental performances: trade flows embody (increase) emissions but also technological flows.

3. Value added.

4. See Appendix 1 (online supplementary material) for more details on data sources.

5. In doing so, we follow Gilli, Mazzanti, and Nicolli (2013) to which we refer for further details on this approach. Moreover, Appendix 2 presents a mathematical appendix with the derivations of the M, P and A terms.

6. More details about data and methods for this section are reported in Appendix 3.

7. The world-level weighted average of this indicator is not necessarily equal to one as the consumption perspective of manufacturing goods on the one hand does not consider those manufacturing goods that serve to generate final products and services in non-manufacturing industries and on the other hand it does account for emissions occurring to non-manufacturing sectors that serve as suppliers of intermediate goods and services to the manufacturing sector.

8. The authors are aware that splitting the sample in four subsamples drastically reduces the number of observations. However, the results can still offer some interesting descriptive evidences of the relation between economic growth and CO₂ emissions trend, rather than a causality relation between the two.

9. We note that cubic effects are not significant here and for subsamples of countries.

10. In addition to patents, the share of R&D on GDP is also introduced as alternative covariate (results available on request). R&D is similarly not significant across all specifications, and as expected it is positively correlated to patents. There is some similarity with the methodological oriented evidence provided by Eberhardt, Helmers, and Strauss (2012), who highlight the significance of factors which capture unobserved effects over R&D in the estimation of production functions. Technological (and policy) factors are highly related to time events and dynamics. The inclusion of temporal effects often brings about the irrelevance of those factors.

11. Green patent here is referred to the OECD ENV-TECH classification.