Assessing the performance of the UK Emissions Trading Scheme

Full Reference:   Stephen Smith and Joseph Swierzbinski (2007)  “Assessing the performance of the UK Emissions Trading Scheme”, Environmental and Resource Economics, 37(1), pp 131-158. ISSN 0924-6460. DOI: 10.1007/s10640-007-9108-5


In 2002 the UK introduced an Emissions Trading Scheme (UK ETS) for greenhouse gases, the world's first large-scale greenhouse gas trading scheme, pre-dating the wider EU Emissions Trading Scheme, which began in 2005. Firms in over 40 industrial sectors which had negotiated “Climate Change Agreements” setting quantitative energy efficiency targets could use the UK ETS to trade over-and under-compliance with these targets. In parallel, a limited number of firms became major participants in the UK ETS as a result of an auction of subsidies for additional abatement commitments. The paper describes the UK arrangements and assesses the economic efficiency and environmental effectiveness of the UK ETS, drawing on evidence of participants’ behaviour both in the incentive auction, and in subsequent trading. For key conclusions, see below

.pdf file at SpringerLink

For a brief description of the UK ETS and its relationship to other instruments in the UK Climate Change Programme click here

The paper formed part of a special issue of Environment and Resource Economics which commemorated the life of Professor David Pearce, who was a colleague of both the authors at UCL. One of the main themes of David Pearce’s academic research was the potential of economic instruments to achieve environmental goals with greater efficiency.


Key conclusions

Given the relatively limited experience with emissions trading outside the US, there are good reasons to look closely at the experience of the UK Emissions Trading Scheme (ETS), the world's first large-scale greenhouse gas trading scheme, which began in 2002. This paper considers the impact of the scheme on the abatement incentives and emissions performance of participants, and the efficiency of the emissions market.

To some extent, the complex institutional context within which the UK ETS is embedded makes it difficult to draw lessons of wholly general applicability about the economic efficiency and environmental effectiveness of emissions trading. The approximately 1400 firms which participated in emissions trading in the first three years of the UK ETS, trading between themselves more than 6 million tonnes of CO2e, are drawn from two very distinct groups. 32 firms entered the scheme as Direct Participants as a result of an auction in 2002 which allocated a budget for abatement subsidies. In return for an abatement subsidy of £17.79 /tCO2e these firms were assigned a fixed ceiling on emissions, and were directly allocated tradeable ETS allowances. But the bulk of the participants entered the scheme as a way of achieving compliance with quantitative commitments for energy efficiency improvements which they had been allocated in sectorally-negotiated “Climate Change Agreements”, and the reason to negotiate such agreements ultimately derives from the 80% discount on the Climate Change Levy, an energy tax introduced in parallel with the ETS. The impact of the ETS on abatement incentives and overall emissions is a joint result of this set of interlocking mechanisms, and identifying the separate contribution of the ETS is not always practical or straightforward.

Nevertheless, there are at least three areas where the experience of the UK ETS suggest lessons of general interest, beyond the specific UK institutional context. These concern the static efficiency case for emissions trading, the extent to which market power issues are of practical relevance to emissions trading, and the importance of long-term market stability and policy commitment.

First, it is clear that the standard textbook analysis of static efficiency in an emissions market has considerable explanatory power in relation to the UK ETS market. In the standard model, efficiency gains arise through emissions trading because firms with low marginal abatement costs sell allowances to firms with high marginal abatement costs. The data on the pattern of trades analysed in the paper show systematic patterns of trading which are consistent with the predictions of this model, with the vast majority of firms trading largely or entirely in a single direction. The three large net sellers in the market, Ineos Fluor, Rhodia and Du Pont (each with negligible or zero purchases) together accounted for over 50% of total sales. Moreover, their willingness to take on additional uncredited abatement obligations in November 2004 strongly suggests that these three firms have low marginal costs of abatement. In the opposite direction, there appear to be many firms – mainly Climate Change Agreement members – which were only buyers in the market, purchasing allowances for compliance purposes in preference to undertaking additional abatement.

However, it is clear that there are also significant trading phenomena in the market that cannot easily be explained by the standard textbook model of static efficiency. At least 20% of sales in the permit market were by net buyers of allowances, or by firms buying and selling equal amounts. This outcome may be due to intertemporal arbitrage, with firms selling when the allowance price was high and buying when it was low, or perhaps other factors. Whether or not these trades have implications for market performance depends on our understanding of why they occurred, a matter which needs further research.

Second, experience of the UK ETS suggests that issues of market concentration and potential market power should not be neglected in analyzing emissions trading and in market design. In the UK ETS market, despite the large number of potential participants, sales are very concentrated. The four-firm concentration ratio (ie the fraction of total sales accounted for by the four largest sellers, a common measure of market concentration) is 65.7%, a value generally regarded as high in analyses of market power. The four largest sellers, moreover, are all Direct Participants that entered the ETS through the auction. One of the possible explanations of the relatively high auction price, compared to the price in subsequent trading, would be that the auction was insufficiently competitive. However, there are other possible explanations for this outcome, and it is also clear that the firms bidding in the auction did not bid as a monopoly would have done (otherwise the auction would have stopped at close to the maximum price).

This aspect of the UK market suggests, at the very least, that the interest in the academic literature concerning the possible effects of market power in permit markets (eg Hahn, 1984, and Misiolek and Elder, 1989) does not seem misplaced. It also suggests that greater policy attention needs to be paid to the possibility of market concentration in market design. The concentration of the seller side of the UK market arises partly as a result of abatement decisions by individual firms, but also reflects concentration in the auction outcome.

Thirdly, it is clear from the UK ETS experience that various aspects of market design matter crucially to the outcome.

The overall constraint on emissions is critical, and determines the permit value. In the UK system, the cap on emissions is derived implicitly from decisions about the targets for abatement which participants are given and the emissions baseline against which these targets are defined. The UK experience indicates that excessive generosity in setting the baseline for individual firms can expose the system to – in effect – excessive allowance allocations, with an aggregate implied emissions cap that may require little additional abatement effort, and a consequently low permit value. Defining the emissions baselines for individual firms can therefore become the Achilles’ heel of emissions trading. The authorities are at an informational disadvantage in negotiating baselines with individual firms or sectors, but the aggregate effect of setting baselines too high for individual sources can undermine the effectiveness of the system in encouraging abatement. The point is not just that there should be a emissions market, but also that there should be a market in which participants face clear incentives for abatement, and in which trading achieves efficiency gains by reallocating abatement between sources facing different abatement costs.

Given the difficulty of defining appropriate baselines, because of the natural information asymmetry at the government’s expense, emissions trading policies need to take account of the risk of initial error, and the need for subsequent adjustment. However, such adjustment is both difficult and potentially costly, and it may be more costly to undo "mistakes" with emissions trading than with alternative market instruments such as emissions taxes. Efficient functioning of the market requires stability, and confidence about current and future property rights in quantities traded. Retrospective adjustment of targets or other parameters of the scheme undermines this stability, and the alternative of repurchasing rights once allocated can be costly.

Stability is important too in defining the future prospects for the market. Many energy use decisions involve investments based on expected costs and benefits over a time horizon of a decade or more, and emissions trading will influence these decisions efficiently only if the market provides clear long-term price signals. The UK ETS has been affected by policy uncertainty in various ways, particularly as its incompatibility with the design of the EU ETS became clear. Making credible commitments to policy stability will always be difficult. But the long-term effectiveness of ETS markets will depend on achieving reasonable clarity about the future of the market, and stability in the policy parameters critical to the value of allowances.