Sectoral agreements and competitive distortions - a Swedish perspective
The objectives of this study are to: 1. Give an overview of the current discussion concerning competition distortion in relation to climate policy, 2. Describe results from some studies estimating the actual competition situation for selected activities, 3. Describe what sector agreement models are suggested/ discussed by EU, 4. Describe what sectors are most interesting to target with a sector agreement from a Swedish point of view, 5. Analyse what parameters are important for reducing competition distortion for Swedish Industry. Two studies, for the United Kingdom (Hourcade et al 2008) and Germany (Graichen et al 2008), have recently assessed the potential cost impact for different industrial sectors of CO2-prices due to the EU ETS. Maximum value at stake was used as metrics. The sectors with high potential impact, with a maximum value at stake larger than 10%, are in the United Kingdom Lime and cement, Basic iron and steel, Starches, Refined petroleum, Fertilizers and Nitrogen compounds and Aluminium. In Germany, the sectors with a maximum value at stake larger than 10% are: Cement and lime, Fertilizers and nitrogen compounds, Basic iron and steel, Aluminium, Paper and board, Other basic inorganic compounds and Coke, refined petroleum and nuclear fuels. Ex-ante studies of the impacts of competitiveness and carbon leakage due to the EU ETS fail to find actual impacts. However, that does not mean that there will be no impact in the future, which hold changes both in the EU ETS (method for allowance allocation, allowance prices etc) and possibly also other important circumstances (global demand for certain products and global product prices). In this study, based on official Swedish statistics, the maximum value a stake has been calculated for 52 Swedish sectors. Seven sectors have a maximum value a stake of more than 4%: Coke and refined petroleum (21%), Pulp and paper (11%), Basic metals (10%), Non-metallic mineral (9%), Metal ore mines (6%), Air transport (5%) and Electricity, gas and heat (4%). If Air transport and Electricity, gas and heat are omitted, the five remaining sectors account for 22% of Sweden's carbon emissions. In the Swedish Non-metallic mineral sector (including Cement and lime) the maximum value at stake is considerably lower than for Cement and lime in the UK and Germany. This is most likely due differences in system boundaries. In the Swedish statistics, the Cement and lime industry is a minor part (in terms of value added) of the Non-metallic mineral sector, a sector that also includes Stone, sand and soil industry. The calculated maximum value at stake for Non-metallic mineral is therefore a poor proxy for the Cement and lime sector since other sub sectors may 'dilute' the maximum value at stake. Differences in system boundaries may also explain the significant difference in maximum value at stake between the Swedish steel industry and UK and German steel industries. Other possible explanations may be a higher value added per unit, differences in how value added is calculated, different years applied for the analysis and lower CO2-intensity for Swedish products. In late 2008, the EU proposed three types of sector approaches to be discussed under the Ad-hoc Working Group on future commitments for Annex I Parties under the Kyoto Protocol (AWG-KP): i) Sector CDM - a CDM crediting mechanism with a previously established baseline ii) Sectoral no-lose mechanism - Sectoral crediting against a previously established no-lose target iii) Sectoral emission trading based on a sector emissions cap Based on these three sectoral models, we have analysed what parameters are important for reducing competition distortion for Swedish industry. We have assumed that these sector agreements are implemented in a developing country (DC). We conclude that if sector agreements are to reduce distortions on competition, it is important that the sector agreements create a real carbon price in the DC, i.e. that emissions of carbon dioxide are associated with a cost for the emitter. All three sector agreement-models suggested by the EU can potentially create a carbon price. The driver for emission reductions are in all three cases the international demand for offsets. As a potentially large buyer of off-sets, the EU demand for off-sets is likely to increase the carbon price in the DC sector. The choice of EU policy with respect to imports of off-set will therefore have great importance. Other buyers, such as other countries, emission trading systems or the voluntary credit market will of course also be important. Moreover, imports of off-sets may reduce the price on EU ETS allowances, thus further narrowing the carbon price gap between the two markets. If an important objective of a sectoral agreement is to reduce competition distortion it should be implemented in sectors where the corresponding Swedish industry has significant carbon related costs and where there is significant trade intensity between Sweden and regions outside the EU. Our preliminary analysis indicates that Swedish sectors with potentially high maximum value at stake (direct carbon and indirect electricity cost) are Refineries; Pulp and Paper; Iron and Steel;Cement and Lime; and Metal ore mining. The sectors Aluminium and Fertilizers may be important, but have not been assessed explicitly in this study. In addition, electricity production can be important to include in a sectoral agreement since the electricity price may be a significant cost for certain sectors exposed to international competition. Pass-through of costs - consumer incentives. If a sectoral agreement is to reduce competition distortion it is important that the sector participating in the sectoral agreement can pass through the additional carbon costs on the commodity so the carbon intensive products become more expensive for the consumer. A full pass through of the carbon cost could be compromised in countries with centrally regulated prices on carbon intensive commodities or other measures that shield the true price of carbon from the consumer. Target setting - producer incentives. The rules for setting the targets in the DC sector are crucial from a producer incentive point of view. There are two main options here: 1) absolute targets and 2) intensity targets. Absolute targets create high incentives for carbon reductions as long as the targets are not re-negotiated. The disadvantage is that they might be difficult to negotiate due to difficulties in finding an appropriate emission level, risk for hot air and the inflexibility to future adjustments. Intensity targets are based on output times an intensity factor (called benchmarking). But benchmarking leads to reduced incentives: i) as a production subsidy it encourages overproduction and ii) dis-incentivises the substitution to carbon efficient products. A third, theoretical, option would be absolute targets that are updated according to historic emissions. This model would, however, seriously undermine the incentives for emission reductions. In this study, we have argued that from a competition point of view, it's important to create a carbon price in the developing country. A different issue relates to how different sector agreement models influence the compliance costs of participating firms. We describe a situation where a DC industry sector is linked to the EU ETS, and where the EU industry pays for allowances (no free allocation). For a Sector emission trading system where the DC industry has to pay for allowances, the compliance costs could be compatible in the two regions. For Sector CDM and Sector no-lose mechanism, if the government implements a domestic carbon tax, the compliance costs may also be compatible in the two regions. However, if allowances are allocated freely to the DC industry and no tax is implemented, the DC industry would have no costs associated with the carbon emissions below the compliance level. There could here be a significant difference in compliance costs between the industries in the two regions. We have, however, not analysed if significant asymmetries in compliance costs can lead to competitive distortions between regions.