Carbon Leakage and Competitiveness - focus on heavy industry -

1. Carbon Leakage and Competitiveness- focus on heavy industry - Julia Reinaud Energy Efficiency and Environment, IEA ICC - Commission on Environment and Energy 22 October 2008

2. Industrial output growth: 1981-2005 Main products / world regions

3. Outline Definitions of carbon leakage – sector perspective Elements influencing carbon leakage under an ETS Preliminary assessment of EU-ETS Phase 1 Solutions and pitfalls

4. Carbon leakage - channels Competitiveness-driven carbon leakage: Short term : Immediate loss of market share for carbon-constrained industrial products Long term : Changes in investments The fossil fuel price channel Increase in prices of low emitting feedstock (e.g. recycled scrap metal), lowering its consumption in the non-carbon constrained countries. Lower unitary emissions in new vintages outside the region, as the constrained producers’ process innovations may spill over to other regions.

5. Competitiveness- driven CL- a national sector’s perspective -

6. Carbon cost impact:Estimating orders of magnitude Direct costs: allowance purchase (EUAs currently trading at around €22 /tCO2) Indirect costs: effect of CO2 price on electricity prices Ability of a sector to “pass-through” extra costs without inducing increased competition from outside Which activities? Trade-exposed, energy- or GHG-intensive Aluminium:76%, of global output is traded, both GHG and electricity intensive Iron and steel: 32%, high CO2 content Cement : 6% but very high carbon cost per value added

7. Unit investment estimates in Europe and allocation (@20/tCO2)

8. Which elements influence carbon leakage?

9. Summary of EU-ETS Phase 1 (2005-2007) Preliminary assessment No statistical evidence of a change coinciding with the introduction of the EU ETS Great differences btw sectors … Trade intensity EU-ETS costs: emissions intensive vs. electricity intensive sectors Allocation … but some common features across these activities High price environment for industrial commodities Recent slow-down in these activities Yet, Phase 1 is a poor indicator of what may come End of long-term electricity contracts concluded pre-liberalisation More stringent targets (i.e. higher CO 2 prices ) Not enough time to see investment decisions change Using 2005-2006 trade data (imports to and exports from the EU), we find no statistical evidence of a change coinciding with the introduction of the EU ETS. Had the Scheme had an impact, the EU would import more, cheaper products from unconstrained regions, and export less to the rest of the world. While this has been a general trend for some of these sectors (aluminium, steel), the additional impact of the EU cap-and-trade system has not been felt until now. These results are in stark contrast with so-called ex-ante analyses that project the effect of a unilateral constraint on emissions from these industries. The analysis exposes great differences, and some common features across these activities. Cement has a low value per tonne, and is not transported easily on land; coastal production may be at risk, provided there is excess capacity in the rest of the world. Iron and steel is much more traded and therefore exposed to international competition – but it was obviously over-allocated in Phase 1. The refining sector, not unlike steel, has enjoyed growing margins, likely to dwarf any carbon cost introduced by a free allocation of CO2 allowances. Aluminium, not covered by the EU ETS, ought to have felt the pinch of higher electricity prices, yet most smelters still enjoy long-term contracts with their suppliers and when they do not, they have also surfed on the wave of high international aluminium prices. On the other hand, Phase 1 is a poor indicator of what may come: CO2 prices were very volatile and on average relatively low; many installations in these sectors were granted allowances over and above their actual emissions (with, whether this is economically rational or not, limited incentive to pass on an opportunity cost to product prices); and two years of observations say little, if anything, about where companies will take their next investments. Using 2005-2006 trade data (imports to and exports from the EU), we find no statistical evidence of a change coinciding with the introduction of the EU ETS. Had the Scheme had an impact, the EU would import more, cheaper products from unconstrained regions, and export less to the rest of the world. While this has been a general trend for some of these sectors (aluminium, steel), the additional impact of the EU cap-and-trade system has not been felt until now. These results are in stark contrast with so-called ex-ante analyses that project the effect of a unilateral constraint on emissions from these industries. The analysis exposes great differences, and some common features across these activities. Cement has a low value per tonne, and is not transported easily on land; coastal production may be at risk, provided there is excess capacity in the rest of the world. Iron and steel is much more traded and therefore exposed to international competition – but it was obviously over-allocated in Phase 1. The refining sector, not unlike steel, has enjoyed growing margins, likely to dwarf any carbon cost introduced by a free allocation of CO2 allowances. Aluminium, not covered by the EU ETS, ought to have felt the pinch of higher electricity prices, yet most smelters still enjoy long-term contracts with their suppliers and when they do not, they have also surfed on the wave of high international aluminium prices. On the other hand, Phase 1 is a poor indicator of what may come: CO2 prices were very volatile and on average relatively low; many installations in these sectors were granted allowances over and above their actual emissions (with, whether this is economically rational or not, limited incentive to pass on an opportunity cost to product prices); and two years of observations say little, if anything, about where companies will take their next investments.

10. Solving carbon leakage? Allocation modes (EU, US proposals, Canada, Aus, NZ, Sw) Free allocation + closure rule under an absolute cap Needs to address indirect cost (electricity) to be effective Encourages installations to stay Allocation follows output volume Removes incentive to pass CO2 cost in commodity prices Border adjustments (US and EU) Levelling the carbon costs for imports and exports (rebates) Include imports in the ETS On the basis of which goods? Which CO2 content for goods? Think carefully about CO2 price effects and indirect effects (electricity) WTO compatibility uncertain “Sectoral approaches”: to be determined

11. What is meant by “sectoral approaches” (SA)? Bottom-up sectoral analysis mitigation potentials of AI Parties Complementary sector-specific goals for AI Parties Cooperative sectoral approaches supported and enabled by finance and technology Sectoral crediting in non-AI Parties ? What effects on Competitiveness and CL?

12. Concluding remarks How significant could the carbon leakage problem be? Do not speculate: simulate and monitor expected effects Short term = changes in international trade flows Long run = changes in investments patterns Yet drivers of investment are multiple EU sectors are not operating in a vacuum A slow down of the booming commodity market would certainly accelerate closures – and yet one would surely not attribute them to climate policy Policy / modelling challenge: properly defining the counterfactual scenario is critical for finding evidence of leakage Ambitious climate policy implies changing relative competitiveness of sectors, encouraging low-carbon innovations and preparing for new playing field ? Policy challenge: Balance prime mover advantage (R&D) with risk of carbon leakage

13. Reinaud J. (1st Nov 2008)Issues behind Competitiveness and Carbon Leakage Thank you further questions? julia.reinaud@iea.org The ideas expressed in this presentation are those of the author and do not necessarily represent views of the IEA Secretariat or of the IEA member countries