The Implicit Carbon Price of Renewable Energy Incentives in Germany.

• Author: Marcantonini, Claudio
• Position: Report

1. INTRODUCTION

   In adopting the Climate and Energy Package in 2009, the European Union (EU) made the promotion of Renewable Energy (RE) a distinct element of climate policy. As stated at the very beginning of the Renewable Energy Directive (2009/28/EC),

   "the increased use of energy from renewable sources [...] constitutes an important part of the package of measures needed to reduce greenhouse gas emissions and comply with the Kyoto Protocol to the United Nations Framework Convention on Climate Change (UNFCCC), and other further Community and international greenhouse gas emission reduction commitments beyond 2012." Like the Emissions Trading System (ETS), a companion measure in the Climate and Energy Package, the Renewable Energy Directive implies an additional incentive to increase the RE share and thereby reduce greenhouse gas emissions below what they would otherwise be. Unlike the ETS, the additional incentive is not uniform throughout the EU. Instead, each member state is expected to develop a national "support scheme" to ensure achievement of that member states' share of the EU-wide target of a 20% share of gross energy consumption from RE sources by 2020. Those support schemes can take various forms, but all provide some extra incentive that can be seen as comparable to the carbon price created by the ETS. It is only natural then to ask: what is the real price paid by consumers to abate C[O.sub.2] emissions? And what is the implicit C[O.sub.2] abatement cost embodied in these RE support schemes?

   To respond to these questions, we estimate the annual carbon surcharge and implicit carbon price associated with RE incentives (REI). The RE carbon surcharge measures the additional cost to reduce C[O.sub.2] emissions in the power sector over and above the carbon price resulting from the EU ETS. The implicit carbon price is the sum of the annual average EU ETS carbon price paid by conventional generators and the carbon surcharge, thereby providing an estimation of the C[O.sub.2] abatement efficiency of the REI. This paper calculates the RE carbon surcharge and the implicit carbon price for wind and solar energy in Germany for the years 2006-2010. Germany is the member state that has played as large a role as any in the expansion of RE in the EU. The German Renewable Energy Act (EEG), which came into force in 2000, defined a system of feed-in tariff (FIT) for all renewable technologies that triggered an impressive growth of wind and solar capacity. Wind capacity grew more than four-fold from 6 GW in 2000 to 27 GW in 2010, solar capacity more than twenty-fold from 76 MW in 2000 to 17 GW in 2010 (BMU, 2012).

   In this paper we look at RE only as a tool to abate C[O.sub.2] emissions. There are other important reasons to promote RE in addition to reduce C[O.sub.2] emissions such as energy security, innovation, jobs, regional development. They are clearly stated in the Renewable Energy Directive (2009/28/EC) to justify the 2020 renewable target and in the documents of the German government to justify the cost of the EEG (BMU, 2012). (1) The analysis of these other rationales to develop RE however is beyond the scope of this work. We leave it to other researchers to quantify these other benefits and to allocate the costs accordingly

   The RE carbon surcharge is calculated as the ratio of the net cost of RE over the C[O.sub.2] emission reductions due to the RE injections into the electric power system. For the quantity of C[O.sub.2] abated as a result of injections of wind and solar energy for the years 2006-2010, we use the estimates of Weigt et al. (2012) calculated using a deterministic unit commitment model of the German electricity system. Most of the paper is devoted to estimating the net cost of RE, which is the sum of the costs and cost savings associated with the use of RE in generating electricity. Other benefits--whether they are expressed as energy security, innovation, jobs, non-C[O.sub.2] emissions, etc.--are not included, nor are costs associated with transmission and distribution. We define the implicit carbon price as the sum of the RE carbon surcharge plus the average carbon price paid in the EU ETS by conventional generators. This can be seen as the hypothetical carbon price that would make RE economic or, in other words, as an estimation of the equivalent total carbon price being paid when we think of REI as a carbon instrument alone (without EU ETS). Our analysis is restricted to the impact of RE on the power sector. When we refer to emissions abatement, we always mean the reduction of C[O.sub.2] emissions in the German electricity generation system. Actually, because of the EU ETS cap on these emissions, increasing RE in the German electricity sector does not reduce the EU-wide C[O.sub.2] emissions. Instead, additional emissions reduction in the German electricity sector due to the RE incentives displace emissions to other ETS sectors in Germany and to other EU member states. Hence, the RE carbon surcharge should not be considered as the total C[O.sub.2] abatement cost due to the injection of RE but as an estimate of how much German consumers have paid to reduce C[O.sub.2] emissions in the German power sector in addition to what is already paid as a result of the EU ETS. (2)

   There is a number of studies that have analyzed the costs and benefits of renewable generation from an ex-ante point of view (e.g. Denny and O'Malley (2007) for Ireland, Dale et al. (2004) for UK, Holttinen (2004) for Nordic countries, DEWI et al. (2005) for Germany). Some of them, such as Holttinen (2004) and DEWI et al. (2005), have also estimated the cost to reduce C[O.sub.2] emissions resulting from the injection of the RE into the power system. DEWI et al. (2005) estimated the "C[O.sub.2] avoidance cost" of wind energy, which is the equivalent of our RE carbon surcharge. It compares the net cost and C[O.sub.2] emissions the system would have in 2007, 2010 and 2015 between two scenarios: the first one with the future wind capacity remaining the same as in 2003, and second one with a larger wind capacity that is developed thanks to the RE support scheme. Results depend on the assumptions made for the fuel and carbon prices. With a carbon price in the range of [euro]5-10 per tC[O.sub.2], the estimated annual C[O.sub.2] avoidance cost of wind in the years 2007 and 2010 goes from a minimum of [euro]56.6/tC[O.sub.2] to a maximum of [euro]168.0/tC[O.sub.2]. The results from these works are difficult to compare because of the different methodologies, data and scenarios analyzed (Holttinen et al., 2011).

   Regarding ex-post analyses using historical data, the research on the costs and benefits of RE into the power system has mostly focused on the impact of RE on the electricity price (e.g. Sensfu[beta] et al. (2008) for Germany, Saenz de Miera et al. (2008), Gelabert et al. (2011) for Spain, Jonsson et al. (2010) for Denmark). The analyses show that the injection of RE reduces the wholesale price of electricity, often called the merit order effect, and that the savings can be large enough to exceed the total annual expenditure for FIT, as was the case for Germany in 2006 Sensfu[beta] et al. (2008). Others (Gelabert et al. 2011) have found that, although present initially, the merit order effect disappears over time. There is also a substantial amount of literature available--both theoretical and empirical--on renewable incentives. The focus of the empirical studies is mostly on the comparison of the different support schemes and in their effectiveness to promote the deployment of renewable technologies (Lipp, 2007; Fouquet and Johansson, 2008; Steinhilber et al., 2011), but not on the cost to reduce C[O.sub.2] emissions.

   In contrast to these studies, our paper is the first, to our knowledge, to estimate the cost of RE to abate C[O.sub.2] emissions from an ex-post point of view, that is, using actual and not presumed or projected values for RE generation, the fossil fuel displaced, and the prices for those fuels. We do not assess the benefits of RE in terms of impact on the electricity price, but we directly estimate the costs and benefits of RE by analysing the power generation costs. However in Section 4 we explain how the benefits of RE that we take into account relate with the merit order effect.

   In the remainder of the paper, Section 2 provides a categorization and general discussion of the costs and cost savings associated with the use of wind and solar energy. Section 3 describes in detail the methodology used to estimate these costs and cost savings. Section 4 presents and comments on the results. In this section we also present a sensitivity analysis and we discuss the inclusion of the learning rate on our methodology. Section 5 concludes.

2. COSTS AND COST SAVINGS OF RENEWABLE GENERATION

   This section briefly describes the six cost and cost saving components taken into account in calculating the cost of abating C[O.sub.2] emissions by promoting wind and solar energy in the electricity sector. We also discuss why the merit order effect is not one of these components. The included components are associated with the cost of generation behind the busbar, that is, excluding the cost that may be incurred in connecting these generating sources to the grid, as well as any costs or cost savings associated with congestion in the operation of the transmission and distribution system. Finally, as stated earlier, other possible benefits from the use of RE related to energy security, non-C[O.sub.2] related emissions, or jobs are also excluded. The cost components that are included are the remuneration to generators, additional balancing cost and additional cycling costs. The cost savings components are the cost savings from the avoided fossil fuel use and carbon costs, and those from added generating capacity.

   An important aspect of the costs of wind and solar energy is intermittency, which includes two independent aspects...