A structural decomposition analysis of the pollution terms of trade

ABSTRACT

The Antweiler Pollution Terms of Trade Index (PTTI) measures environmental gains or losses sustained by a country from international trade. Measuring total exports and imports in value terms distorts the results when the index is used for comparisons and analysis. Structural decomposition analysis (SDA) can provide an estimate of such distortions, due to price effects. This paper proposes a procedure of SDA of PTTI changes to deal with this problem. It measures impacts of changes in prices of traded outputs, changes in the real trade mix and volume, and changes in technology, but can be extended to account for other factors too. The pollution content refers here to pollution generated directly in the production of outputs, actually traded in the international markets. However, additional induced effects due to international trade can be included, if they are estimated. Pollution intensity is expressed per unit of total exports and imports respectively, and not per unit of value added, an issue discussed. A case study for The Netherlands is provided for the period 2007–2010. PTTI values and their annual changes are estimated for emission contents in greenhouse CO₂ equivalent. Results confirm the significant and distortive role of prices in measurements and comparisons.

KEYWORDS:

• Greenhouse emissions
• pollution content
• pollution terms of trade
• structural decomposition analysis

Acknowledgments

The authors wish to thank an anonymous referee for useful comments and suggestions. The authors also thank Avanti Pinto and the participants of the 4th ENVECON Conference and the 2017 EAERE Annual Conference for their constructive comments and suggestions.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Ang and Zhang (2000) discuss developments and survey 117 IDA studies, up from 51 from a previous survey in 1995. A survey is also provided in Xu and Ang (2013). SDA reviews can be found in Rose and Casler (1996) and Miller and Blair (2009) as well. Hoekstra and van den Bergh (2002) discuss 27 SDA studies analyzing physical flows in the economy. Su and Ang (2012) discuss 43 SDA studies of the period 1999–2010, on energy and emissions, and classify them according to their different characteristics including their particular decomposition method.

2. A multiplicative SDA method was proposed by Dietzenbacher and Los (1998), and has been adopted in several studies. In Su and Ang (2014) it is shown that this method is conceptually similar to the generalized Fisher index method in IDA (Ang, Huang, and Mu 2004).

3. In Su and Ang (2015) four alternative IO models are introduced to study an economy's aggregate carbon intensity and an application in the study of China's intensity is provided. The different models are distinguished by different combinations of IO analyses, imports assumptions and GDP calculation. The alternative way of treating imports (competitive versus non-competitive) which is in general particularly significant in the construction of IO tables and analysis, is addressed by the authors for the Leontief and also for the Ghosh models. A formal analysis of the two treatments of imports and their consequences is provided, as is also the case in Su and Ang (2013).

4. In Johnson and Noguera (2012) the value added content of trade is measured to avoid the problem of double counting in the value of trade when there is increased world-wide production sharing.

5. The authors considered SO₂ emissions for 62 developing and developed countries over the 1990–2000 period. They account for different technologies of production in similar sectors of different countries resulting in different emission intensities. They use input output tables for the countries of the study where in final demand, international trade of every country is broken down to bilateral trade with all other countries separately.

6. In Su et al (2010) a formal presentation and analysis of the two approaches is presented together with distorting results due to assumption or lack of accurate data. Moreover, the impact of sector aggregation on estimated emissions embodied in China's exports is investigated.