The Variation in Capacity Remunerations Requirements in European Electricity Markets.

• Author: Bunn, Derek

1. INTRODUCTION

   Harmonisation of the trading arrangements and regulations between connected electricity markets has produced efficiency gains in many regions and the economic case for further har-monisations has been persuasive (Mansur and White, 2007; Cramton, 2017; Cicala, 2019). Trading across larger markets provides access for the most efficient generating facilities, the sharing of reserves and greater dispersion of weather effects. Perhaps the most notable example has been within the EU where a single energy market for electricity has been legislated (European Union, 2012) and wholesale market coupling, intraday balancing, emission trading and industry code harmonisations have resulted. However, at the same time, support for renewable and other low carbon generating technologies has been selective and this has distorted the fundamental economics of the wholesale market. Subsides for renewables have allowed them to be profitable, despite their low marginal costs causing wholesale market prices to fall (Green and Staffell, 2016). Furthermore, harmonised interregional market coupling has also facilitated the greater penetration of renewable technologies, as intermittent large swings in output are more easily accommodated across the interconnected markets. As a consequence, substantially lower wholesale prices have emerged (Newbery et al., 2018).

   Whilst there are many benefits from this evolution, nevertheless, for the fossil generators, such as gas-fired power plants, these lower wholesale prices, together with their lower load factors, have reduced revenues and caused asset impairments (Tulloch et al., 2017). This is a concern to policy makers, as well as the asset owners, because these facilities remain essential for the security of the system. Therefore policy-makers have increasingly, often reluctantly, accepted that there is a need for capacity remuneration mechanisms (CRMs) to maintain their essential presence in the system and to incentivise adequate new investment (Bublitz et al., 2019). CRMs arise to compensate generators for the "missing money" i.e. insufficient returns from the energy only market to recover capital costs and incentivise investment.

   As expected, EU policy makers, for example, have been seeking to follow their principles of co-ordination and harmonisation in this respect (Haffner et al., 2017). Specifically, EU energy market harmonisation objectives include implementing a process of open, transparent and non-discriminatory practices to allow foreign bidders to gain access to capacity markets. EU policy makers expect that harmonised capacity markets should ensure that overall costs are reduced, and that cross-border investment incentives and short-term merit order operations of the integrated electricity system are not distorted (Tennbakk et al., 2016). Within the EU single energy market, competition is fair across countries because generators generally face the same input costs and state aid decisions have sought to avoid country advantages. However, with the introduction of CRMs, there is additional competition between generators benefitting from selective state aid, even if approved by the EU.

   Different levels of "missing money" to keep incumbent facilities operational could exist between EU member states because of their existing generation portfolio mix, demand, penetration of renewables, levels of interconnection etc, impacting wholesale electricity prices and capacity factors. (1) It is usually assumed by policy makers however that to incentivise new capacity to meet reliability standards, the long-run marginal cost of a gas peaking facility would provide the basis for the required capacity remuneration. Whilst the technology cost of gas-fired power plants may be the same in each member state, sovereign risk (cost of debt), fiscal measures (taxes and allowances on profits), and the impacts of gas network legacy infrastructure (tariffs) will create variations in the amount of costs which will need to be remunerated through CRMs. These factors could create country specific cost advantages in cross border mechanisms and distort cross border investment decisions. For example, if a generator in a low risk country prefers to get higher capacity payments from a high risk country, is the justification of the higher returns based upon compensation for higher counterparty risk? If so, there would be a contradiction with the wholesale energy trading, which is fully harmonised to avoid any country risk premia in the transactions. Plant developers susceptible to higher sovereign risk, network legacy tariffs and taxes with reduced taxable allowances on profits are unlikely to be able to compete with investors facing more favourable cost advantages. This could create challenges in how the fairness of state aid impacts these markets. The open question that follows therefore, and which is analysed in this paper, is whether these CRMs can be introduced into co-ordinated markets in a manner consistent with harmonisation objectives.

   An emerging body of research suggests that the unilateral implementations of CRMs, as in Europe, have negative impacts for welfare. Inefficiency results from under/over capacity procurements in interconnected markets in which the CRMs differ (Meyer and Gore, 2015; Bhagwat et al., 2017; Hawker et al., 2017; Cepeda, 2018; Hoschle et al., 2018). This arises from capacity payments which are awarded to some generators, who can then offer their electricity production more competitively in their own market and in neighbouring bidding zones. Generators which are non-recipient of capacity payments rely fully on the energy market for their revenues, and therefore would not be able to lower their energy market offer prices. Thus, capacity payments implemented in one bidding zone, but not in a neighbouring one, may potentially distort dispatch decisions. Since the differences in capacity payments tends to be higher than differences in generation tariffs, it is likely that these distortions are more significant than any distortions that would be caused by the lack of transmission tariff structure harmonisation (ACER, 2015b). This issue is raised by Bhagwat et al., (2017) in a different context in which they argue that in an interconnected market a capacity market causes crowding out of generators in the adjacent energy only market. Prior research from McInerney and Bunn, (2013) shows that in order to achieve full market coupling and price convergence between neighbouring electricity markets, the price spread has to be greater than the capacity payment when capacity payments are based on actual power flows. This may create an effective "deadband" where the "energy only" price spread has to be greater than the value of the capacity payment to incentivise export.

   The overall objective of this paper is to compute the capacity payments necessary to facilitate investment in new gas assets in each EU country in 2030. CRMs generally evaluate gas peaking facilities as the marginal providers of energy security. We demonstrate that variations in capacity payments required to incentivise resource adequacies arise from different sovereign risks and infrastructure legacies in addition to market operation. We use the results of a European Commission EU Reference Scenario (EC Ref) as a starting point for our analysis. This is a projection of how the EU energy system might evolve in the future assuming all EU and Member State policies and measures implemented by December 2014 are taken into account. Many of these member state polices taken by individual Member States may make sense when agreed at Member State level, but may appear "irrational" when the collective impacts of all Member State policies is viewed through the lens of results from an EU wide energy systems model scenario analysis. Collins et al., (2017) scrutinise the EC Ref in the context of market and operational impacts of renewable energy ambition, and Gaffney et al., (2018) investigated RES-E exports and imports between member states and highlights concerns regarding uncoordinated support mechanisms, price distortions and cost inequality.

   Although we develop our results from extensive and detailed modelling of the capacity remuneration requirements across the various countries in the EU, two research questions feature strongly in our analysis, with general implications beyond the EU. Considering the same technology, combined cycle gas turbines (CCGTs), subject to the same input commodity costs (wholesale gas), we test how the cost of debt, which typically varies by country, influences the costs of capital and thereby becomes a differentiator in the capacity remuneration requirements. Secondly, as load factors for gas generation decrease, and decarbonisation scenarios may project reduced market shares for gas, the legacy costs of gas infrastructure becoming somewhat stranded in different regions could increase the use of system costs for gas generators in substantial and regionally discriminated ways. It is an open question, how material this factor may be in the capacity remuneration requirements. These two locational factors, capital investment risks and stranded gas infrastructures, are considered carefully in our analysis, and contribute to the novelty of our modelling.

   Despite the significant research on capacity remuneration in power markets (Joskow, 2008; Meyer and Gore, 2015; Newbery, 2016; Hogan, 2017; Brown, 2018; Cepeda, 2018; Fabra, 2018; Bublitz et al., 2019; Milstein and Tishler, 2019), the challenges of cross-border solutions in this context are apparently under-researched. Using the official European Commission energy system modelling scenario for 2030 (EC Ref), we assess the capacity remuneration requirements using an investment model of gas-fired power plants in each European member state. To generate endoge-nously a set of inputs for the valuation model, we adapt the approach of Deane et al...