Using a Policy Mix to Combat Climate Change – An Economic Evaluation of Policies in the German Electricity Sector



Department of Economics


Foundation of German Business


1/2006 – 6/2009


Strategies to mitigate climate change are based on a policy mix in many industrialized countries. In order to reduce greenhouse gas emissions, a complex portfolio of policies, such as emissions trading, taxes, subsidies for renewable energy technologies, energy-efficiency standards and voluntary agreements, is employed. Within this mix, policies may complement each other. However, there is also the risk of adverse interactions between policies. Consequently, the question arises whether existing mitigation strategies represent an efficient policy mix – or rather a suboptimal ‘policy mess’. Economic answers to this question are rare and insufficient. This project makes three important contributions to closing this research gap and improving the economic understanding of a climate policy mix:

Firstly, this project clarifies under which general conditions a policy mix for pollution control may be more efficient than a single-policy strategy. It develops a novel framework to organize the diffuse body of existing policy mix studies. The framework introduces transaction costs into the analysis of pollution problems and of public governance structures, i.e. regulation, to solve these problems. It understands the existence of a pollution problem not only as a market failure – as it is common in neo-classical economics – but more generally as the failure of private governance structures, such as the market, the firm or bilateral bargains. Based on this framework, two overarching rationales for using a policy mix for pollution control are derived: (1) A pollution externality may be reinforced by other failures of private governance structures, such as technological spillovers or asymmetric information. In this case, one policy is needed for each failure. (2) Single policies to cope with a pollution externality may bring about prohibitively high transaction costs. A policy mix may then reduce transaction costs while providing for a similar level of pollution control. With the identification of these two general rationales, this project sets a counterpoint to the array of economic studies assuming that a single pollution problem can always be addressed more efficiently by a single policy than by a policy mix.

Secondly, the project overcomes the high degree of abstraction in economic policy mix research. Simplified assumptions about policy design in existing models are substituted by more complex design characteristics as they can be found for actually existing policy mixes. Thereby, the project approaches economic theory of a policy mix to real-world conditions and increases its relevance for policy-making. It is shown that the consideration of complex real policy design may deliver results that are contrary to those derived under a simplistic model. Two selected policy combinations are considered in particular: an emissions trading scheme combined with a feed-in tariff for renewable electricity generation, and with a tax on emissions or output.

Economic theory suggests that the combination of emissions trading and a feed-in tariff is justified when a pollution externality is reinforced by spillovers related to learning-by-doing. Nevertheless, it is demonstrated in this project that the expected dynamic welfare gains of such a policy mix may be impaired by inefficient interactions when two characteristics of real feed-in tariff design are taken into account: the payment of the tariff irrespectively of the prevailing electricity price and the funding of the tariff by an add-on to the electricity price.

Existing economic analyses also point out that the combination of emissions trading and a tax is likely to be inefficient in mitigating climate change when the incentives of both policies overlap, i.e. when firms have to hold an allowance and pay a tax in addition for each unit of emission. The resulting conclusion is usually to abolish the tax. However, it is emphasized in this project that this conclusion may be flawed in the real world where taxes address a variety of policy objectives, such as raising fiscal revenues or promoting equity. Given these restrictions, this project addresses three important issues, which have been neglected so far: (1) it identifies factors that drive the actual extent of inefficient abatement under a policy mix of emissions trading and a tax, (2) it shows that the policy mix is in fact likely to reduce this inefficiency compared to a single suboptimal tax, and (3) it proposes modifications of emissions trading design which may bring down welfare losses under the policy mix even further. These new insights are necessary for the evaluation and design of a policy mix in the presence of multiple policy objectives and criteria.

The third important contribution of this project is the evaluation of the climate policy mix in the German electricity sector – a case study that is representative for the industrialized world. This evaluation can claim to be the most extensive analysis which is available for an existing policy mix. It applies the theoretical findings made in this project but also goes beyond them by considering further details of policy design and institutional environment. Particular attention is paid to the question which policies should complement emissions trading. It is shown that the additional implementation of energy efficiency labelling and low-interest loans promoting technological innovation and diffusion clearly raises efficiency. Support schemes for renewable energies and combined heat and power plants as well as energy-efficiency standards are likely to also produce welfare gains when combined with emissions trading. However, certain design features of these policies – such as funding the support schemes by an add-on to the electricity price – may impair the overall efficiency of the policy mix. Finally, it is argued that only electricity taxation and voluntary agreements cannot be justified as components of a policy mix. In summary, the German case study thus confirms that there may be many instances in which a policy mix is needed – but also that the details of policy design matter for the overall performance of the policy mix.