Embodied emissions in trade (EET): insights from bilateral trade data Misato Sato ESRC Centre for Climate Change Economics and Policy LSE Friday 4th March 2011 Envecon 2011 @ The Royal Society

!"

Presentation outline !  What is EET? !  In which policy context is it relevant? !  How much carbon is embodied in international trade? Some

estimates from the existing literature. !  An alternative approach !  3 key findings !  discussion

“Embodied” emissions in trade

Largest interregional fluxes of emissions embodied in trade (Mt CO2 /y) from dominant net exporting countries to the dominant net importing countries. Davis & Caldeira (2010)

Fig. 1. sions e domin the do Fluxes aggreg France Luxem

Relevant policy contexts !  High level policy questions !  Sustainable development and ecological foot-printing !  Supply chain security and natural resource dependency !  National responsibility: production- vs consumption-based

approach to carbon accounting !  Environmental integrity of Kyoto Protocol etc.

!  Specific and current debates !  Carbon leakage !  Border carbon adjustments !  Role of trade in decarbonising global supply chains

How much carbon is embodied in trade? Findings from the literature so far !  Many studies (over 50) quantifying EET !  Several approaches: !  Single Region Input Output (SRIO) !  Full / Partial Multi-Regional-Input Out (MRIO) !  Others e.g. material balance approach, multiplying trade balance and average carbon intensity of GDP, Hybrid LCA-MRIO, CGE models !  Simplified picture emerging and high uncertainty. !  Broad conclusions: !  Large volumes of EET have been found (for 2005-2006, between 4Gt7Gt annually or roughly 15-30% of global emissions) !  In general find that industrialised countries are net importers of EET..

Industrialised countries are in general net importers of EET. BEET against production by country The Balance of EET against CO2CO2 production by country in 2001(2001) (Peters & Hertwich (2008)) 25 South Africa Venezuela Malaysia Indonesia

20 Australia

Romania

10

10000

India

Canada

10000

15000

0

-15 -20

Italy

-10

Japan

-5

France

5000

Germany

0

UK

BEET (%)

5

China

Russia

15

Rest of Middle East

Czech Republic Belarus/Ukraine

USA

20000

Spain

- Total Global Emissions: 24.8 Gt CO2 - Total Embodied Emissions in Imports/Exports: 5.3Gt CO2 - Annex B Net import 1.34Gt (roughly 10% of Annex B emissions in 2001)

-25 CO2 emissions from production (Mt CO2)

Yet uncertainty around the results are high – a comparison of some estimates for China Study

Model

Year

Productio Consump EEE n (Mt) tion (Mt)

EEI

Weber et al (2008)

SRIO

2005

5030

5560

1679 (33%)

2200 (44%)

Huimin & Ye (2010)

SRIO

2005

5699

5039

1760 (31%)

1100 (19%)

Yan & Yang SRIO (2010)

2005

5429

4699

1180 (22%)

450 (8%)

Lin & Sun (2010)

Partial MRIO

2005

5458

3370

2670 (49%)

583 (11%)

Nakano et al Full (2009) MIRO

2005

4508

3921

794 (18%)

207 (5%)

Bruckner et al (2010)

2005

4449

3459

1357 (31%)

366 (8%)

Full MRIO

See forthcoming Sato (2011) www.cccep.ac.uk/Publications/Working-papers/home.aspx

The existing studies provide some interesting insights, yet….

1. 

Non-trivial uncertainty in country-level EET estimates, as they are sensitive to base-data, model structure and multiple assumptions made

2.  Appears to paint a very simplified picture 3.  Not suitable for gaining more detailed insights about EET flows

and inform the “more specific” policy debates # A more granular approach to quantifying EET !  Forthcoming paper “Embodied carbon flows in global supply chains:

A study drawing on bilateral trade data” www.cccep.ac.uk/ Publications/Working-papers/home.aspx

Approach: “material balance” r,s r,s w EEEj =ΣXj * EFj !  Following Muradian et al. (2002) and others. !  EF from LCA data : around Cradle-to-gate estimates gathered from 9

databases and around 30 papers

!  UN COMTRADE trade data, physical weight values !  Key strengths: !  Weight-based data, not monetary !  Transparent !  Disaggregated: 970 products (aggregated to 60 sectors) and all countries. All countries (195); Standard International Trade Classification (SITC) Rev 3, 4digit (plus some 5digit for important sectors), total some 970 commodities.

Some caveats !  No input-output - double counting !  Trade data: !  Reliability. Need to reconcile EEIr,s and EEEs,r !  No distinction between different production processes !  Problem of attributing exports to production

!  Carbon intensity data !  Lack of harmonization in LCA, comparability of estimates !  world average Efjw !  No product classification – don’t readily match trade data

Despite the focus on China in the literature, most of the global EET occurs within Europe 41'!-% (4-% "3-% .!:% )*"% 8*9% '37% 56(% 1'*% 5(6% 34-% 12(% 0-*% ./)% -")% &*(% +,(% !"#$% ")*% &'(% !"#$% EU-internal ;%

!!5% !!!%

#;;%