Dynamiken hos ett EU-system för handel med gröna elcertifikat

(1)

DEPARTMENT OF THEMATIC STUDIES

MASTER OF SCIENCE THESIS

Dynamics of an EU System for Tradable Green Certificates

Niklas Knutsson

Linköpings Universitet, Campus Norrköping, Environmental Science Programme, SE-601 74 NORRKÖPING

(2)

Rapporttyp Report category Licentiatavhandling Examensarbete AB-uppsats C-uppsats x D-uppsats Övrig rapport ________________ Språk Language Svenska/Swedish x Engelska/English ________________ Titel

Title Dynamics of an EU System for Tradable Green Certificates Dynamiken hos ett EU-system för handel med gröna elcertifikat

Författare Niklas Knutsson

Author Sammanfattning Abstract ISBN _____________________________________________________ ISRN LIU-ITUF/MV-D--02/04--SE _________________________________________________________________ ISSN _________________________________________________________________

Serietitel och serienummer

Title of series, numbering

Handledare Mattias Hjerpe

Tutor

Nyckelord

Keywords Green certificates, obligation, renewable energy, EU, harmonisation, cost-effectiveness

Datum

Date 7 June 2002

URL för elektronisk version

http://www.ep.liu.se/exjobb/ituf/

Institution, Avdelning

Department, Division

Institutionen för tematisk utbildning och forskning, Miljövetarprogrammet

Department of thematic studies, Environmental Science Programme

In electricity markets evolving towards liberalisation and internationalisation, green certificates are seen as an important tool to promote production of renewable energy. In a green certificate market producers sell certificates received per unit of electricity generated from renewable energy. Trade in renewability is thereby decoupled from the the physical electricity trade. Tradable green certificates in combination with renewables obligation are implemented or planned in several EU member states. Integration at EU level and the creation of a common certificate market seem like a possible next step. Using a simple model, the effects of an EU system for tradable green certificates, compared to national implementation, are tested for the period 2000–2010. The simulations show a clear increase of cost effectiveness and large changes in the geographical balance of renewable energy when reaching the EU 22 per cent target. Less competitive technologies, such as solar power, are not deployed at all. The situation is however similar without international trade. Rather than implementing a certificate market with all EU member states, but with different support schemes, a smaller market, only with countries that combine the certificate market with renewables obligation, is recommended.

(3)

(4)

Abstract

In electricity markets evolving towards liberalisation and internationalisation, green certificates are seen as an important tool to promote production of renewable energy. In a green certificate market producers sell certificates received per unit of electricity generated from renewable energy. Trade in renewability is thereby decoupled from the the physical electricity trade. Tradable green certificates in combination with renew-ables obligation are implemented or planned in several EU member states. Integration at EU level and the creation of a common certificate market seem like a possible next step. Using a simple model, the effects of an EU system for tradable green certifi-cates, compared to national implementation, are tested for the period 2000–2010. The simulations show a clear increase of cost effectiveness and large changes in the geo-graphical balance of renewable energy when reaching the EU 22 per cent target. Less competitive technologies, such as solar power, are not deployed at all. The situation is however similar without international trade. Rather than implementing a certificate market with all EU member states, but with different support schemes, a smaller mar-ket, only with countries that combine the certificate market with renewables obligation, is recommended.

(5)

(6)

List of Figures

3.1 Demand and supply in a green certificate market . . . 15 4.1 Graphical illustration of the core process of the TGC market model . . 23 4.2 Relative cost development of electricity generation technologies, with

2000 as index year . . . 26 4.3 Average accumulated certificate costs of new RES-E since year 2000 . 27 4.4 Highest certificate costs of new RES-E since year 2000 . . . 28 4.5 Highest certificate costs in member states by year 2010 . . . 29 4.6 Increase of RES-E share relative to total domestic electricity

consump-tion in member states in scenario A . . . 29 4.7 Additional cost per capita and year of new RES-E since year 2000 in

member states by year 2010 . . . 30 4.8 Penetration of technologies for new RES-E in scenario A . . . 30 4.9 The difference in geographical balance of RES-E increase between

scenario A and reference scenario 1 . . . 31 4.10 The difference in technology penetration between scenario A, and

ref-erence scenario 1 and 2 . . . 31 4.11 The difference in geographical balance of RES-E increase between

scenario B and scenario A . . . 32 4.12 The difference in technology penetration between scenario B and

sce-nario A . . . 32 4.13 Additional cost per capita and year for alternative support schemes in

scenario B by year 2010 . . . 33

(9)

List of Abbreviations

A Austria

AEO annual energy output

B Belgium

CRF capital recovery factor

D Germany DK Denmark E Spain EL Greece EU European Union F France

FCR fixed charge rate

FIN Finland

I Italy

ICC initial capital cost

IRL Ireland

L Luxembourg

LAE levelized annual operating expenses

LCC levelized cost of capital

LCOE levelized cost of energy

NL The Netherlands

O(&)M operating and maintenance charges

P Portugal

REBUS Renewable Burden Sharing (model/project)

RECerT The European Renewable Electricity Certificate Trading

Project

RECS The Renewable Energy Certificate System

RES-E electricity supply from renewable energy sources

S Sweden

TGC tradable green certificate

TPA third party access

UK United Kingdom

(10)

1

Introduction

1.1 Background

The European energy system is evolving through several processes. Within the context of liberalising markets and widened domains of the EU single market, electricity mar-kets are, more or less rapidly, opening up for competition. In a parallel process in the context of green house gas mitigation and the Kyoto Protocol, countries strive for in-creased deployment of renewable energy. Green certificates and renewables obligation fits within the open market as a tool for promotion of electricity from renewable energy sources. A market for green certificates can be seen as a market for the green of elec-tricity, decoupling the renewable part from the physical electricity trade. Several EU member states have implemented or plan to implement national TGC, tradable green certificate, systems. There are also thoughts of implementing a market for green cer-tificates at international level, within the context of internationalisation, globalisation and increased European integration. Is this an alternative worth consideration? How will an EU TGC system, and its implications, differ from its national counterpart?

1.2 Purpose

The target of this study is to find and estimate possibilities and considerations of an integrated green certificate market at EU level, compared to national implementations. Focus is set on the dynamics of an integrated EU system including possible efficiency gains and technological and geographical balance shifts. Secondary, the effect of de-sign choices, as degree of harmonisation, is considered. For this it is necessary to have a clear picture of how an EU TGC system might be designed and how it theoretically will function. Though main focus is on EU level compared to national level, this can hardly be achieved isolated from adherent questions regarding general effectiveness of green certificates and comparison with other support schemes.

The study is conducted from a socio-economic perspective. That is, it will mainly look at the economic effects of the certificate system, but also try to interpret the results in terms of environmental and political benefits and considerations. The study will look at the whole system with nations, energy technologies and vertical market groups—

(11)

2 CHAPTER1. INTRODUCTION

producers, suppliers, consumers, et cetera—as smallest parts. It is neither part of the purpose to evaluate the benefits of renewable energy and consideration of different renewable technologies, nor to evaluate different environmental implications of the energy system.

The purpose can be broken down to three successive areas, with focus on the sec-ond, and six primary questions:

1. Functionality

(a) How can an EU tradable green certificate system be designed and how will it function?

2. Implications

(a) How large increase of the cost-effectiveness is possible?

(b) How will the distribution of the increased RES-E, electricity supply from renewable energy sources, share amongst participating countries be af-fected?

(c) How will the economic burden of the RES-E capacity increase be dis-tributed?

(d) How will the penetration of different renewable, especially less mature, technologies be affected?

3. Implementation

(a) How will the degree of harmonisation affect above mentioned implica-tions?

The first question/area is mainly treated as a prerequisite and background for the fol-lowing, and is therefor not strictly part of the purpose, but the process. Further, the implications of the answers to these questions is discussed.

1.3 Method

The main effort is concentrated on building a model for simulation of cost efficiency differences, as well as development for renewable technologies and trade patterns, comparing international with domestic approaches. A simple, straight-forward, single-purpose model, which is mainly complicated by its requirement of large amounts of complex data input, will be used. The model is fit and restricted to available data, relying on previous surveys.

Three main scenarios will be simulated. The first represents an EU TGC system with full harmonisation, the second with less harmonisation and additional national support schemes active. A reference scenario represents implementation of national

(12)

1.4. OUTLINE 3

TGC systems in all member states. The reference scenario is actually three scenarios with different assumptions for geographical distribution of renewable energy. The simulations will be run for the years from 2000 up to 2010, although should mainly be seen as proof-of-concept. The EU target for consumption of electricity from renewable energy sources is presupposedly reached in 2010.

The simulation is preceded and complemented by a rather theoretical reasoning about the overall implications and determination of the functionality of and prospects for an integrated certificate market, with information from previous research.

1.4 Outline

The study begins with the provision of a context and background information on con-cepts and general definitions of electricity markets, renewable energy and support schemes in the next chapter. Thereafter it will go deeper into the development of an international certificate market, defining issues and functionality, and describe the process towards integration. The fourth chapter presents the main empirical part of the study, simulating the possible cost advantages using a simple model. It builds upon the former chapters and includes an extension of the above method description, as well as results. Then follows a discussion/analysis of how an integrated certificate market at EU level will affect the production of electricity from renewable energy sources and implications thereof, as well as important implementation consideration.

(13)

2

Concepts and Definitions

2.1 The Internal Electricity Market

The electricity systems in members states of the European Union are evolving towards open, competitive markets. Six, out of fifteen, states have reached a full, or close to complete, market opening, while the others are slowly moving forward from an inclusion of at least 30 per cent of consumers1. The number of customers using this new freedom of switching suppliers is however low, below 5 per cent except in the United Kingdom2.

The electricity directive of 19963 regulates the opening of European electricity markets. It was brought into force in member states in early 1999, with some ex-ceptions. It contains common rules for generation, transmission and distribution of electricity, as well as a gradual time schedule for the opening to consumers.

The directive states that third party access, TPA, to the electricity net shall be fair and implemented as either

1. regulated TPA, meaning public tariffs equal for all; 2. negotiated TPA; or

3. a ‘Single Buyer Model’.

Most member states use the first alternative, while foremost Germany has negotiated TPA.

Further, unbundling shall be achieved through separate accounts, secrecy and des-ignation of a competent dispute settling authority, to avoid discrimination, cross sub-sidisation and distortion of competition. Most member states have chosen a model with full separation of vertical functions, such as generation, transmission and supply. The opening for competition in capacity expansion is mostly implemented through authorisation, but a tendering procedure is also allowed, using objective, transparent

1_{European Commission (2000).} 2_{OXERA et al. (2001).}

3_{Directive 96/92/EC of the European Parliament and of the Council of 19 December 1996}

concern-ing common rules for the internal market in electricity

(14)

2.1. THE INTERNALELECTRICITYMARKET 5

and non-discriminatory criteria. Subsidies for renewable and domestic energy sources, ‘priority production’, are allowed if the sources at most have a 15 per cent share of the domestic electricity production.

The Commission has in a recent directive proposal4 suggested a full market open-ing by 1 January 2005, as well as removal of unused options such as the sopen-ingle buyer model.

The new markets are characterised by a focus on competition, meaning freedom to choose whom you buy electricity from, competitive power generation and capac-ity building, and third party access to the electriccapac-ity net. Although the liberalisation process aims at competitive markets, network activities are excluded as they are con-sidered naturally monopolistic. There is, however, not a single solution for the design of liberalised electricity markets, but a few different models exists. Following is a list of different market types, of which the electricity market might consist of one or a few5:

• Spot market or pool contracts—these are power exchanges where trade usually takes place in periods of hours.

• Bilateral forward contracts—contracts between two parties are sold for a certain time.

• Green market—in addition to the regular spot market, a separate market can be created for electricity generated from renewable energy sources.

• Green certificates—an alternative to the green market, where trade in renewabil-ity is unbundled from the physical electricrenewabil-ity trade.

Pools are organised anonymous markets for electricity, where market players make bids in advance after which the spot price is set. The pool is either voluntary, al-lows bilateral contracts, or mandatory. At present a relatively small share of the total market volume is traded in European power exchanges6. The anonymity of the pool keeps transaction costs low, while bilateral contracts provide higher price stability but costly generator switches. Bilateral prices do to a high degree follow the price in spot markets.

Although there are many small actors, the national markets are characterised by having a few large dominant electricity producers7. The merging and integration of national markets will theoretically change this situation. The nationally dominant gen-erators will be rather small in a European market, although there have already been several mergers in the European electricity industry in recent years.8

4_{Proposal for a Directive of the European Parliament and of the Council amending Directives}

96/92/EC and 98/30/EC concerning common rules for the internal market in electricity and natural gas

5_{Morthorst (1999).} 6_{OXERA et al. (2001).} 7_Ibid.

(15)

6 CHAPTER 2. CONCEPTS AND DEFINITIONS

It is important to note that the electricity market has some specific characteristics, which makes it differ from trade with other goods. This is mainly due to9

1. the balancing problem, and 2. the constraints problem.

Market factors can not respond fast enough to the necessity of immediate supply re-sponse to the constantly changing demand. Grid constraints reduce possible trades. To handle these issues a regulating market is additionally established, regulated by a sys-tem operator to ensure grid security and power quality. The capacity of transmission systems is a crucial factor for increased intra-EU trade. Congestion, lack of capacity, is already often occurring in the cross-border transmission systems10.

2.2 Renewable Energy

2.2.1 Definition

At its core, renewable energy can be defined to include ‘usage of any energy storage reservoir which is being “refilled” at rates comparable to that of extraction’11. Main benefits are perceived to be local production, environmental benefits, and provision of additional diversity of supply12. Increased deployment of renewable energy is a mean for fulfilment of the commitments of the Kyoto Protocol.

Renewable energy does in the definition of the renewable energy directive of 200113 include

• wind power; • solar power; • geothermal power; • wave and tidal power; • hydro power;

• biomass, meaning ‘the biodegradable fraction of products, waste and residues from agriculture (including vegetal and animal substances), forestry and re-lated industries, as well as the biodegradable fraction of industrial and municipal waste’; and

9_{Van Roy et al. (2000).} 10_{Haubrich et al. (2001).} 11_{Sørensen (2000), p. 3.} 12_{Alder (1999).}

13_{Directive 2001/77/EC of the European Parliament and of the Council of 27 September 2001 on the}

(16)

2.2. RENEWABLE ENERGY 7

• landfill gas, sewage treatment plant gas and biogases.

This definition will be used throughout, if nothing else is stated. Note however that other definitions might be used elsewhere, often excluding large hydro power with a capacity of more than 10 MW. The directive also formulates a target, stating that 22 per cent, compared to approximately 16 per cent in year 2000, of the total electricity supply shall come from renewable energy sources by year 2010. Indicative targets are present for all member states. An overwhelmingly large part of the electricity supply from renewable energy sources, RES-E, does at present come from large hydro power. The opening of electricity markets changes the market conditions for renewable energy. The fluctuating character of most renewable technologies and high investment costs will lead to difficulties in a competitive market characterised by profit maximi-sation14. There are similar economic conditions for most energy projects. Capital is obtained through a combination of debt—loan, for low-risk investments—and equity investment—ownership, for high-risk investments. Variables in financing include cap-ital structure, debt-equity ratio; return on equity; debt maturity, referring to length of loan; debt interest rate; debt amortisation; and the debt service coverage ratio, which is the yearly operating income per total debt service. Renewable energy technologies are perceived to have high resource and technology risks. They are often capital intensive, which together with unpredictable policies, and high real and perceived risk leads to more costly financing. Interest rate and equity return are highly important for the final cost of technologies such as wind power and photovoltaics.15

2.2.2 Support Schemes

There are several alternatives for support of electricity from renewable energy sources. These foremost include

1. capital cost subsidies; 2. tax exemptions; 3. feed-in tariffs; 4. tendering/bidding;

5. voluntary schemes, such as green pricing; and

6. tradable green certificates, with renewables obligation.

The main market stimulation instrument for renewable energy in Europe is a com-bination of capital cost subsidies and payment of premium prices for produced energy. Premium prices are either fixed and provide an incremental amount over the value of

14_{Morthorst (1999).} 15_{Wiser and Pickle (1998).}

(17)

8 CHAPTER 2. CONCEPTS AND DEFINITIONS

the electricity alone, or established competitively. Capital cost subsidies are on capac-ity and feed-in tariffs subsidies output.

A feed-in tariff is a price per electricity unit that distribution companies have to pay for local renewable power generation fed into the local distribution grid. There are usually different tariffs for different technologies, as well as time variations.16 Feed-in tariffs give high security for Feed-investments, at least Feed-in the short term, but provides limited incentives to reduce costs17. Possibly very high costs for utilities have led to limitations on maximum required accepted amount, for example 5 per cent of the electricity supply in Germany.

Tendering gives a limited number of investors a premium price on output, in com-petition. In the bidding, potential project developers are invited to submit offers for building new capacity. They bid under different technologies for the amount of finan-cial support to be paid for each produced electricity unit. The lowest bidder is awarded with the bid feed-in tariff for a predefined time. Long-term fixed contracts provide security and tendering, as opposed to feed-in tariffs, tends to drive down costs.18

Feed-in tariffs are presently used as main strategy in Denmark, Germany, Italy, Luxembourg, Portugal and Spain. Tendering is used in France and Ireland, while tax relief and rebates are main support measures in Finland and Sweden.19

The last listed support measure, green certificates, is further investigated below.

2.3 Green Certificates and Renewables Obligation

The green certificate market will practically function solely as a financial market, where the producers sell certificates received per unit of electricity generated from renewable energy. The certificates are physically disconnected, decoupled, from the distribution of electricity. Consumers, or more likely distributors, might buy electric-ity from a local producer, while buying certificates from a completely different seller, possibly in an international market. Thus, the trade in green certificates does not suffer from the restrains of the physical supply in an open electricity market. Both a spot market, for past issued certificates, and a forward market, with hedge for price risks and negotiated long term contracts on future certificates, are expected to develop for green certificates20.

The producers in the green certificate market are generators of electricity using renewable sources, which are allowed to sell an amount of certificates corresponding to the amount of produced green electricity. Producers receive both the electricity wholesale price and a certificate price. Demand for green certificates can originate either from21

16_{Ackermann, Andersson, and Söder (2001).} 17_{Hoogland and Schaeffer (1999).}

18_{Ackermann, Andersson, and Söder (2001).} 19_{Faber et al. (2001).}

(18)

2.3. GREEN CERTIFICATES ANDRENEWABLESOBLIGATION 9

1. voluntary consumer demand; 2. a tendering procedure; or 3. obligation on (a) generation, (b) transmission, distribution, (c) supply, or (d) consumption.

Only obligation on consumers, which might be passed on to suppliers, will be con-sidered here. Electricity consumers are obliged to annually buy a certain share of electricity generated by renewable technologies, or pay a penalty. This is expected to be mainly handled by distributors, except for larger consumers. In a liberalised internal EU market, obligation on the first two groups can only be imposed synchronously, oth-erwise consumers can switch to companies without an obligation, resulting in unfair competition22. Surveys have shown that consumers would consider explicitly buying electricity from renewable energy sources at a higher price23. It is however hard to tell if there would still be a noticeable additional voluntary demand in a market with renewables obligation.

While the supply of electricity from non-renewable energy is determined by the electricity price, for renewables this will change to the electricity price plus a certificate price, and a demand determined by electricity price and the certificate price in relation to the obligation quota.24 A new equilibrium price, p∗, will be present in the market, which can be formulated as25:

p∗= b +αs∗+ q∗ (2.1)

where b is the mark-up excise taxes in distribution,αthe RES-E ratio, s the certificate price and q the wholesale price of electricity. The maximisation problem for RES-E producers will be

MaxΠ(zi) = [q + s] zi (2.2)

where z is the RES-E production of producer i.

A tradable green certificate, TGC, system requires new administrative functions, which will generate administrative costs. Functions in institutionalisation include is-suing of certificates, verification of the isis-suing process, registration of certificates and trade, exchange market, banking of certificates, and withdrawing of certificates from circulation26.

21_{Schaeffer et al. (2000).} 22_{Schaeffer et al. (1999).}

23_{Farhar (1999); Roe et al. (2001).} 24_{Jensen and Skytte (2002).}

25_{Amundsen and Mortensen (2001).} 26_{Schaeffer et al. (1999).}

(19)

3

Green Certificates at EU level

3.1 Towards a European Certificate Market

3.1.1 Design Issues

Within the concept of green certificate there are several alternatives and issues, which must be solved, decided on, before implementation. This is true both at national and international level. An international implementation struggles with the issues that oc-cur at national level, because there must be a consensus on harmonisation of design alternatives. In addition, lifting up the system to international level creates new issues, as well as possibilities.

The TGC system should ensure enough security for investors, a cost effective way of deploying renewables and be non-discriminating and transparent to all actors1. RES-E producers want long-term contracts to minimise risk, but if the TGC policy is unstable, purchasers will only buy short-term contracts. The resulting uncertainty leads to increased financing costs, because of decreased debt leverage and reduced debt term, by up to 25–50 per cent compared to in a stable market2. A clear and consistent government policy creates conditions for a stable certificate system.

The first question that must be answered concerns what will be traded—which producers and energy sources shall qualify for participation in the green certificate market? There are mainly two considerations present—will renewable energy tech-nologies that already are competitive in the regular electricity market be included, and where should the line be drawn for more or less questionable renewable energy sources such as wastes? The former mostly concerns large hydro power, which has a very large and highly competitive share of the present RES-E capacity. Inclusion would create a large cash flow to hydro producers, while the incitement for new capacity from other sources would be lower. Nationally this has in most cases been solved through inclu-sion only of new capacity from large hydro. The latter consideration concerns foremost the inclusion of biogas and municipal solid waste3.

1_{Schaeffer et al. (2000).} 2_{Wiser and Pickle (1998).} 3_{Voogt et al. (2001).}

(20)

3.1. TOWARDS A EUROPEAN CERTIFICATEMARKET 11

To reflect the different competitiveness of technologies, separate obligations for different technologies or technology classes can be implemented, although causing less market homogeneity and transparency4. Alternatively, different technologies can be given different values, for example one unit of photovoltaics might equal five cer-tificates, but this makes the final RES-E share unsure. The obligation can also be combined with a stepped feed-in tariff, which evens out the producer surplus5.

A second design question concerns the structure of the market, how certificates will be traded. As in the regular electricity market, certificates can be traded bilaterally or at an exchange, or both. An exchange may increase market transparency.

Target setting and obligation level are of great importance. The obligation should stimulate building of new RES-E capacity, while keeping consumer costs down. The lag time in building of new capacity can be compensated by transitionally low obli-gation quotas. Well-set long run targets may create a stable demand and investment climate.6

Further design issues concern level of upper and lower price limits, and tools for increased flexibility. The upper limit price, providing security for consumers, is often the sanction price for not meeting the target and the lower limit a guaranteed price for producers. The sanction or upper price limit can be a fixed price per certificate or relative to the average market price of the compliance period. In the first case the penalty is acting as a ceiling for the price.7

Borrowing and banking can help even out price variations caused by RES-E pro-duction fluctuations. Borrowing allows for extent validity of certificates, to fill quota before actual production. Banking means that a certificate can be kept for use in a later compliance period. Interest is possible as an incentive to generate more RES-E than demanded today, or a levy can be introduced to discourage banking. Similarly borrowing can allow demand today to be met by future RES-E production. If borrow-ing is allowed without interest rates, there is an incentive to buy certificates as late as possible, to earn interest on the cost. Banking and borrowing might be limited to a certain share of the obligation.8

The biggest issue at EU level is whether to implement a voluntary or obligatory TGC system; whether the market will be combined with obligations, or not. If a voluntary system is used, which might be more in accordance with the subsidiary principle, there will be a problem with oversupply of certificates. If only some of the member states use obligation, then obligations might be covered by the present EU RES-E production, if the issue is not dealt with.

An international trading system must solve questions regarding how to deal with different levels of subsidies and which country will get the political credit if a cer-tificate is not consumed in the country where produced. The cercer-tificate might be

ac-4_{Schaeffer et al. (2000).} 5_{Huber et al. (2001).}

6_{Hoogland and Schaeffer (1999).} 7_{Schaeffer et al. (2000).}

(21)

12 CHAPTER 3. GREEN CERTIFICATES ATEULEVEL

counted for as production in one country and as consumption in another. Alternatives for handling of subsidies include abolishment of all other incentive schemes, com-pensation for all subsidies at the border, allowance of only certificates that have not profited from subsidies, bilateral agreements, or simply letting it be.9

Another implementation issue concerns the issuing of certificates. Alternatives include issuing at national level, at EU level, or at national level monitored by the EU. Additionally there are other minor issues, concerning for instance length of compliance period, reference period, time of proof and implementation of compliance control.

3.1.2 National Experiences and Plans

National TGC systems are implemented or planned in Belgium, Denmark, Italy, The Netherlands, Austria, Sweden and United Kingdom. These implementations are gen-erally characterised by

• disallowance of foreign certificates and international trade, except if in connec-tion with physical electricity trade as in The Netherlands and Italy;

• exclusion of present large hydro power in definition of renewable energy, except in countries with small hydro capacity;

• usually a smooth transitional period;

• an obligation normally set on suppliers or, formally, on consumers; and • one-year quota periods.

Belgium has a federal TGC system for large industrial consumers connected di-rectly to the grid with an obligation rising from 2 per cent to 6 per cent 2010. For other consumers, there are different systems on local level10. The obligation on suppliers in the Flemish Region increases from 0.96 per cent in 2000 to 3 per cent 2004 and then with a yearly growth factor of more than 1.09 between the years 2005 and 2009, while in the Wallon Region it rises from 2.9 per cent 2001/2002 to 12 per cent 2009/2010. In the Wallon Region banking is allowed for a certificate lifetime of 3 years. The Flemish Region has an upper price limit of 12 eurocent per kWh11.

The Electricity Supply Act of 199912is planned to take over the function of subsi-dies and feed-in tariffs in Denmark. It includes a 20 per cent obligation on consumers 2003 and indicative 50 per cent in 2030. It excludes waste incineration and has a price floor at 1.4 eurocent per kWh and a cap at 3.7 eurocent per kWh13. The system will be complemented by a subsidy to important, costly technologies. The start date has been

9_{Schaeffer et al. (2000).} 10_{Meyer and Nielsen (2001).} 11_{Faber et al. (2001).} 12_{Meyer and Nielsen (2001).} 13_{Faber et al. (2001).}

(22)

3.1. TOWARDS A EUROPEAN CERTIFICATEMARKET 13

postponed and it is unclear—especially after the transfer of power—when the system eventually will come into force.

A TGC system is planned for the year 2003 in Sweden. The proposal for a Swedish certificate system14 contains an obligation increasing from 6.4 per cent in 2003 to 15.3 per cent 2010. Electricity intensive industrial consumers are excluded from the requirement, which makes the actual obligation level lower. New large hydro capacity is proposed to be included, but some restriction will be put in the definition of biofuel and waste. The penalty is set to 150 per cent of the average certificate price in the previous year, but not more than approximately 2.2 eurocent per kWh for the years 2003–2007. A price guarantee is set to decrease from approximately 0.65 eurocent per kWh to zero in 2008. The certificate system is proposed to replace current support schemes, but with a transition period for old plants.

The Renewables Obligation Order 2002 which came into force in April contains an obligation on suppliers in the United Kingdom of 3 per cent gradually increasing to 10.4 per cent in year 2011. The order only concerns England and Wales, but Scotland has plans on following with similar regulation. New large hydro is included and some waste restrictions apply. There is a 4.8 eurocent per kWh alternative buy-out price, upper limit, which follows the retail price index, but no lower price limit. Banking is allowed for 25 per cent of the obligation from the previous period.

Austria has so far a system with an 8 per cent obligation for 2001, which only includes small hydro15. Italy has in the year 2002 an obligation of 2 per cent on producers and an expected certificate price of 5.7–7.2 eurocent per kWh. Finally, in the Netherlands only a voluntary implementation has been implemented so far in the ‘Regerling groencertificaten Electricteitswet 1998’ of 7 May 2001. The price level is free, but the trade volume has been relatively small. It is uncertain whether the voluntary system will survive in the long run.16

Additionally the concept of green certificates has recently been tried in Australia and the US state of Texas.

3.1.3 Progress at EU level

The EU directive on promotion of electricity produced from renewable energy sources states:

This Directive does not require Member States to recognise the purchase of a guarantee from other Member States or the corresponding purchase of electricity as a contribution to the fulfilment of a national obligation quota. However, to facilitate trade in electricity produced from renewable energy sources and to increase transparency for the consumer’s choice between electricity produced from non-renewable and electricity produced from

14_{Elcertifikatutredningen (2001).} 15_{Faber et al. (2001).}

(23)

renewable energy sources, the guarantee of origin of such electricity is necessary.17

It does however state that ‘a legislative framework for the market in renewable energy sources needs to be established’, but that it is ‘too early to decide on a Community-wide framework regarding support schemes’, though being necessary after a transi-tional period. The directive does also provide a framework of definitions and targets appropriate for a green certificates system. A commission report shall be presented before 2006 on experiences of application and coexistence of different support mecha-nisms, and a proposal for a Community-wide framework for RES-E support, if neces-sary. This framework shall foremost be cost-effective and take into account geograph-ical and technologgeograph-ical differences.

There are no concrete legislation proposals yet, but several TGC related research projects funded by the European Commission, including

• The RECerT project, which had as main objective to ‘ensure that TGC mar-ket development was coordinated, and that information and understanding was shared among key stakeholders in the EU, with the aim of minimising barriers to Tradable Green Certificate (TGC) trade between Member States’18and included a cost-benefit analysis and an Internet trade simulation;

• REBUS, the Renewable Energy Burden Sharing project, which was ‘initiated to provide insights of the effects of implementing targets for RES-E generation at EU Member State level and the impact of introducing burden sharing systems within the EU; such as a Tradable Green Certificate (TGC) system’19;

• ElGreen, with the objective to consider ‘how to bring about the enhancement in market penetration of RES-E in the cheapest and most efficient way’20, through analysis of different promotion strategies and recommendation of steps towards harmonisation of these; and

• IntraCert, which ‘aims to explore the possibility of integrating the existing and planned TGC schemes in the European Union’21.

Not all of these projects deal with an EU TGC system explicitly and exclusively, but rather include green certificates in evaluation of different support schemes.

The Renewable Energy Certificate System, RECS, is an international initiative for green certificate trade. It is based on voluntary demand of separate market actors and is therefor easier implemented and in principle possible without government interference. Only certificates that have not profited from subsidies are allowed.22

17_{2001/77/EC, §10} 18_{Energie (2001), p. 4.} 19_{Voogt et al. (2001), p. 15.} 20_{Huber et al. (2001), p. 1.} 21_{Boots et al. (2000), p. 4.} 22_{Schaeffer et al. (2000).}

(24)

3.2. FUNCTIONALITY AND IMPLICATIONS 15

3.2 Functionality and Implications

3.2.1 Demand, Supply and Price

A simple value of certificates can be determined as the cost difference between re-newable and non-rere-newable energy sources. Alternatively the value can be determined through a valuation of societal benefits. This would however most certainly not match the consumer price in a real-world market. The actual price will be determined by several factors.

Demand will almost totally be determined by the required certificate ratio, if it is assumed that consumers’ willingness to pay for additional certificates are small. The demand will then be almost perfectly inelastic and the demand curve straight vertical up to the price limit, if present, as shown in figure 3.1.

Penalty Price Supply Voluntary demand Demand Price guarantee Volume obligation

Figure 3.1: Demand and supply in a green certificate market

Short-term supply will be characterised by the high volatility of most renewable en-ergy source. Variability in climatic conditions—approximately ±20 per cent for wind power23—introduce uncertainty in supply of certificates. This will lead to large price variations over time in a spot market. If the system is inflexible, then the price might vary from the level of the penalty, when installed capacity is lower than the target, to almost zero when capacity is higher than target.24 Flexibility can be reached through allowance of banking and borrowing. Looking at longer time frames the volatility should even out.25 Price volatility can also, at least initially, be lowered using a band-width for prices26. Although, if minimum prices are not harmonised, certificates will be offered in the country with highest minimum price, leading to under-supply in other

23_{Morthorst (1999).} 24_{Schaeffer et al. (1999).}

25_{ECON Senter for økonomisk analyse (2001).} 26_{Hoogland and Schaeffer (1999).}

(25)

countries and a certificate price increasing to the highest minimum price. Alternatively, if in line with EU internal market rules, the minimum price might be allowed only to be received by domestic production.

The capability of RES-E suppliers to reach the obligated target depends on the transparency of the green certificate market and the difficulty for investors to decide whether to increase installed capacity or not; the stochastic climate patterns, which influence renewable energy technologies; and external factors such as granting of li-censes. Additionally the absolute level of obligation is uncertain, as it is dependent on the development of electricity consumption.27

The supply and demand of electricity will most likely increase, but might actually decrease. The additional certificate price should put a downward pressure on the regu-lar electricity price, thus lowering the absolute level of the quota.28 There seems to be a non-linear price relation which should be considered when setting the quota depend-ing on if main interest is in keepdepend-ing down consumer prices or fast development of new renewable capacity.29 The certificate market can be considered asymmetric—if price exceeds penalty payment, there will be no demand, but on the other hand if price is not high enough producers might be carrying over an amount of green certificates to the next year.30

The RECerT trade simulation31 showed that penalties act as a price-cap. Even though the equilibrium price theoretically should not be influenced by the penalty, it will in practice clearly act as a price signalling function. The profits for generators was high because of the lack of demand elasticity, which causes the price to be dependent on the penalty. Well-founded estimates of the equilibrium price is therefor necessary to set an appropriate upper price limit. Long term determination of the obligation quota and penalty level will be of utmost importance. The simulation also showed that without harmonisation of renewables support mechanisms, the market will suffer from low liquidity, poor information and relatively high uncertainty.32

3.2.2 Implications

Cost effectiveness is expected to be the one main argument put forth in favour of an EU TGC system. Theoretically the TGC system will be more efficient at EU level as there is a larger selection of renewable energy sources, from a larger geographical area where energy resources vary. Solar power can be most efficiently used in Southern Europe and the wind resources are as largest along the coasts of Western Europe. Additionally administrative costs are expected to decrease compared to separate implementations in all member states33. To fully exploit the possibility of cost savings, the system must

27_{Schaeffer et al. (1999).}

28_{Amundsen and Mortensen (2001).} 29_{Jensen and Skytte (2002).}

30_{Morthorst (1999).} 31_{Energie (2001).} 32_Ibid.

(26)

3.2. FUNCTIONALITY AND IMPLICATIONS 17

function optimally. General conditions for a market to work competitively apply34. That is, there need to be sufficient suppliers and consumers to ensure that a single participant cannot influence the price much, market transparency and equal access to relevant information, as well as no entry barriers and negligible transaction costs.

Depending on the size of trade between countries in an EU green certificate mar-ket, the geographical RES-E balance will change compared to a scenario with closed national markets. Due to the different availability of different renewable sources, the technical balance should also change. This will naturally have an impact on economic, social and environmental conditions in affected countries. Interestingly the balance might also be changed with separate national TGC systems. For example, a green certificate market in Sweden is expected to lead to a lower RES-E production in Den-mark35.

Further an EU TGC system will likely have an impact on the support measures used nationally. In two cases support might leak abroad, meaning that domestic support is unintentionally given to foreign parties. If subsidies and other support are given to domestic producers and there is a large export of certificates, there will be large costs that mostly gain foreign consumers. If on the other hand certificates are imported, exemptions from energy taxes will similarly support foreign producers.36

These implications will be further explored in the following chapters.

33_{Energie (2001).} 34_{Schaeffer et al. (1999).} 35_{Profu i Göteborg AB (2001).} 36_{Voogt et al. (2001).}

(27)

4

Market Simulation

4.1 Scenarios

Five scenarios have been constructed to represent the differences between national and EU TGC systems, as well as different levels of integration. Common assumptions and choices for all scenarios are listed below:

• The target, set at 22 per cent of total electricity consumption, for the EU-wide RES-E ratio in year 2010 is assumed to be reached.

• The simulated time period is set, fit to available data and target, to the years from 2000 up to 2010.

• Increase of RES-E is assumed linear.

• The certificate trade is not connected with physical electricity trade.

• TGC systems include all technologies considered renewable, including large hy-dro and waste. As only new production is considered, no decision for exclusion of present capacity was necessary

• Market players are simplified as rational groups, who always choose the most cost-effective alternatives and make full use of trade.

• Full transparency of constituents, prices, transaction costs and subsidies is as-sumed.

• The current electricity capacity marginal—capacity normally not used for pro-duction—is maintained.

One primary, one secondary and three reference scenarios are simulated as described below:

A. Full harmonisation. Renewables obligation is used as the only support scheme in a common market for certificates.

(28)

4.2. THE MODEL 19

B. Less harmonisation. Different support schemes are used in a common market for certificates. Renewables obligation is used as the only support measure in member states where a TGC system is planned, otherwise current schemes apply. Only major support schemes are considered. As a work-around for the over-supply issue, it is still presupposed that all member states have obligations. The purpose of the scenario is to estimate any distortion of additional RES-E support measures that may occur.

R. Reference scenarios: National TGC systems. Renewables obligation is used as only RES-E support scheme in closed national markets for green certificates.

1. State targets. National targets are reached, although adjusted to sum up to the EU target for comparability.

2. Equal increase. The RES-E share in each member state is equally in-creased relative to the total domestic electricity consumption.

3. Equal targets. The final domestic RES-E share is equally high in all mem-ber states.

4.2 The Model

4.2.1 Core Functions

To determine the distribution of new production capacity amongst technologies and states, the relative competitiveness of each combination must be known. Three al-ternative approaches for calculation of a simple and reliable comparison index are recognised, assuming the production cost is known:

1. Simply compare the production cost directly. This would only possibly be accu-rate with the assumption of full competition in a completely integaccu-rated internal EU market. It will tell how the highest theoretical cost efficiency is reached, but less how it will actually be.

2. Compare production costs relative to the wholesale electricity price, thus com-paring profit differences of production. The problem is the different structures determining the price in different countries and that it does not compare new renewable capacity with new non-renewable capacity.

3. Compare renewable production costs relative to lowest alternative production cost. This alternative should give a more accurate picture than both above, as it compares the cheapest options for building new capacity.

The optimal solution would naturally be to incorporate a complex model of market functioning and behaviour of actors, but that is beyond the scope of this study. Instead production costs of non-renewable and renewable technologies have been compared.

(29)

20 CHAPTER 4. MARKET SIMULATION

The inelastic demand for green certificates, gd, follows the electricity consumption and is given by1

gd=αx (4.1)

where α is the percentage share of electricity consumption from renewable sources, the obligation, and x is the total consumption of electricity.

The net profit of RES-E producers,Π, can be formulated as

Πi= q + s − ci (4.2)

where q is the wholesale price of electricity, s is the certificate price and c the produc-tion costs of producer i. For producers of electricity from non-renewables the profit is similarly formulated as

Πi= q − ci (4.3)

In the case of equal income requirements for all electricity producers and equal whole-sale price, the required certificate price for a producer, i, can then be derived as

s = ci−Cy (4.4)

where Cyis the production cost per unit electricity from non-renewable energy sources. This required certificate price for separate producers will be referred to as certificate cost for the market as whole, as it represents the cost difference between production of electricity from renewable and non-renewable energy sources.

Incorporating consumer side taxes—Bz for renewable energy and By for the non renewable alternative—the cost of green certificates, Cg, is defined as

Cg= (Bz+Cz) − (By+Cy) (4.5)

where Cz is the cost per unit electricity from renewable energy sources, Cyis the cost of the non-renewable alternative. The cost of green certificates is defined as the differ-ence between the levelized cost of new electricity capacity generation from renewable energy and from non-renewable energy sources. Note that this refers to the cost, not the price of certificates.

The cost of renewable and non-renewable electricity, C, is determined at a specific time using

Ct = C for t = 0

Ct = Ct−1× (1 +βt) for t > 0

(4.6) where βt is a technology dependent cost development factor for year t. For the start

year, t equals zero.

(30)

4.2. THE MODEL 21

Electricity consumption, X , for a certain time period is calculated as

Xt= X0× (1 +γ)t γ= n s  Xn− X0 n + 1 − 1 (4.7)

whereγis a factor for increase of electricity consumption calculated from the differ-ence between the estimated consumption of the final year, n, and present consumption. The present consumption is derived from production and net import of electricity.

If the production increase factor is complemented with a decommissioning factor,

δ, the needed electricity production from new capacity,∆X , each year can be expressed

as

∆Xt= X0× (γ+δ)t (4.8)

The increase of RES-E production, ∆Z, each year towards the target year is

as-sumed linear according to

∆Zt=

(αn× Xn) − Z0

n (4.9)

whereαn is the RES-E share, target, for the final year, n, and Z0 is the present RES-E

production.

The cost minimisation function for green certificates can thereby be defined as

C_tg= min " m

∑

j=1 n

∑

k=1 Cg_jkt×∆Zjkt # (4.10)

where j and k are indexes for states and renewable technologies, respectively, and m and n their numbers. The sum of national RES-E production increases shall equal the target for the EU RES-E increase:

m

∑

j=1 n

∑

k=1 ∆Zjkt =∆Zt (4.11)

and restrictions apply to national increase:

∆Zjt ≤∆Xjt ∆Zjkt ≤ Ojkt, Ojkt= Ojk− t−1

∑

T =0 ∆ZjkT (4.12)

where O jk is the potential increase of production for renewable technology k in coun-try j, and T a secondary time index equivalent to t. Increase of RES-E is limited to the total needed new electricity capacity.

(31)

4.2.2 Energy Cost Analysis and Impact of State Support

Although the energy production costs where precalculated in used data, it is important to recognise the methods used for analysis of energy costs, at least to enable postpro-cessing for inclusion of state support. The levelized unit cost of energy is considered standard in the electricity industry. Methods such as life-cycle payback can alterna-tively be used, but are more tailored to individual financial decisions.2

The levelized cost of energy, LCOE, is calculated as

LCOE = LRR

AEO,

LRR = LCC + LAE + OM

(4.13)

where LCC is the levelized cost of capital—the initial capital cost, ICC, multiplied by the fixed-charge rate, FCR—LAE is the levelized annual operating expenses, O(&)M the fixed operating and maintenance charges, and AEO is the annual energy output. In its simplest form, without the inclusion of taxes, the fixed charge rate is the same as the capital recovery factor, CRF, which can be calculated as

CRF = i(1 + i)

n

(1 + i)n_{− 1} (4.14)

where i is the annual interest rate and n is the depreciation lifetime, in years.

From here the impact of different support measures can be calculated approxi-mately. Capital subsidies can be incorporated in the levelized capital cost using

LCC = (ICC −U ) × FCR (4.15)

where U is the investment subsidy, sometimes implemented as a rate relative to the initial capital cost as

U = ICC × u (4.16)

where u is the subsidy rate. The capital cost with subsidy, LCCu, can thereby be

calculated from the flat capital cost, LCCf, as

LCCu= LCCf+

LCCf

ICC ×U (4.17)

Knowing the regular tax rate, br, and the lowered tax rate, be, the energy production cost with tax exemption, Ce, should simply be

Ce= Cz− (br− be) (4.18)

The cost, Cf, supported by a feed-in tariff is similarly calculated as

Cf = Cz−tf− tr (4.19)

(32)

4.2. THE MODEL 23

where tf is the feed-in tariff and tr is the regular tariff. Strictly this is not the actual production cost, but the relative cost, compared to other production. In some cases there is a limit on the amount of RES-E receiving the premium tariff. In this case a limit has been put on the RES-E potential that has the adjusted cost. The cost in a bid-in system is calculated as the feed-in tariff using a predicted price, with limit on potential.

Combined, these equations will redefine the certificate cost as

Cg= Bz+Cz−tf− tr− (br− be) − LCC ×U ICC × AEO − (By+Cy) (4.20)

Green electricity prices, green tariffs and other voluntary approaches are not con-sidered.

4.2.3 Implementation

The core of the model—implementation of the sequential cost optimisation and output of final results—has been implemented as a small C++ program using POSIX and ANSI/ISO compliant code. Calculation and processing of input data were done using the spreadsheet application gnumeric version 1.0.5.

For each year the relatively most cost effective technology and state combination is selected sequentially until the particular year’s quota is filled. In the third scenario this process runs separately for each member state. The model is run sequentially for every year, t, with the previous output, present state for t = 1, as input. Figure 4.1 shows a graphical presentation of the process.

Select lowest relative price (state/technology) If potential quota unfilled

Allocate If yearlyquota filled Else

Increase time counter

Figure 4.1: Graphical illustration of the core process of the TGC market model

Each renewable technology is categorised in groups after ranges of economic and technological feasibility. These bands have mainly been chosen dependent on preci-sion in available data on production and cost potentials for different energy sources.

4.2.4 Limitations

Every model has some limitations in their representation of reality. Importance lies in awareness of those. Some factors and functions have deliberately been left out

(33)

for maintenance of simplicity and because of limits in available data. Limitations considered important, but not crucial, are summarised below:

• The different financial structures, mostly between non-renewable and renew-able generation but also within the two main categories, are not analysed. This includes implications of high capital costs or fuel costs and different risks, gen-eration fluctuations, et cetera.3

• Parallel systems such as the usual green electricity concept, as well as consumer technology preferences, are ignored.

• Development and progress of the common electricity market are not comprehen-sively analysed.

• Most data are static, without feedback on consumption, price and costs, eco-nomics of scale.

• Time lag in development of renewable capacity is not included in the calcula-tions. For wind turbines it might take 2–3 years, including planning, and for other technologies even longer to build new capacity4.

• Climate patterns, which influence local production from renewable capacity, are not taken into account. Generation of electricity from wind energy might vary at most ±20 per cent from year to year5.

• Premature decommissioning is assumed economically disadvantageous.

• The model does not take into account the progress of renewable energy in other energy systems, such as increased use of biomass for heating.

4.3 Input Data

Potentials and basic cost data were taken from the REBUS model6. The data includes costs and potentials for 16 technologies, each divided in a number of bands after costs as shown in table 4.1.

The cost data consists of harmonised and non-harmonised costs. The non har-monised costs are variable costs of renewable energy projects, defined in REBUS as planning, labour, infrastructure, installation, finance, fuel and land availability. Har-monised costs are costs that in an integrated electricity market are assumed to be equal in all member states. These include international technology costs, international fuel

3_{Awerbuch (2000), pp. see further.} 4_{Morthorst (1999).}

5_Ibid.

(34)

4.3. INPUT DATA 25

Table 4.1: Renewable energy technology bands

Band 1 Band 2 Band 3 Band 4 Wind onshore >7m/s 6-7m/s 5-6m/s 4-5m/s Wind offshore >9m/s 8-9m/s 7-8m/s 6-7m/s Small hydro (<10 MW) Capital Costs1 Capital Costs1 Capital Costs1

Large hydro (>10 MW) Capital Costs1 Capital Costs1 Capital Costs1 Photovoltaics High Radiation2 Medium Rad.2 Low Radiation2 Solar thermal electricity High Radiation2 Medium Rad.2 Low Radiation2 Power stations and CHP

- Solid fuels Forestry Energy Crops - Solid wastes Agricultural Industrial - Liquid wastes Industrial

Farm slurries Farm Slurries

Municipal solid waste centralised decentralised Sewage sludge centralised decentralised Landfill gas centralised decentralised

Geothermal electricity Capital Costs1 Capital Costs1 Capital Costs1 Wave Capital Costs1 Capital Costs1 Capital Costs1 Tidal Capital Costs1 Capital Costs1 Capital Costs1

1_{Based on capital costs. For these technologies the banding is based upon the cost of site development.} 2_{Country specific bands to account for variations in solar radiation.}

costs and a harmonised discount rate. The discount rate are in all projects, regardless of technology and risk, set to be 8 per cent.

Potentials are defined in such a way that they are realistic and realisable, leaving economics as the only final barrier. The potentials are limited theoretically and tech-nically, taking into account energy flow, technical feasibility, land availability, accept-ability and planning. Different key factors determine the final potentials of different technologies. The potentials have been adjusted for the progress since the original date of the data.

For further information about the REBUS model and data used, see the manual7 and the final report8of the project.

Development of RES-E production costs, shown in figure 4.2, was derived from estimates of the ATLAS project9. General energy projections, including projected de-commissioning and electricity consumption in year 2010, are present in the European Union energy outlook to 202010. The input data might be a bit out of synchronisa-tion in a fast evolving market and therefor, rather than as absolute numbers and exact

7_{Voogt (2001).} 8_{Voogt et al. (2001).}

9_{European network of energy agencies (2002).} 10_{European Commission (1999).}

(35)

forecasts/predicts, results should be considered as proof-of-concept.

0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Wind Power Small Hydro Photovoltaics Solar thermal electricity Biomass

Municipal Solid Waste Geothermal Landfill gas Natural Gas (GTCC) Coal (PFBC)

Figure 4.2: Relative cost development of electricity generation technologies, with

2000 as index year

The alternative non-renewable technology does for Denmark, Germany, Italy and Portugal refer to coal-fired fluidised bed combustion, and for remaining states gas tur-bine comtur-bined cycle using natural gas. The cost development for these technologies is determined by increased fuel costs, increased fuel efficiency and decreased capital costs.

Below is a list of data sources and adjustments used for model input:

• RES-E production costs: REBUS. Adjusted in scenario B for state support. • Production costs for non-renewables: European Commission (2001). Operating

at 7000 hours with excise taxes and subsidies included.

• Cost development of renewable technologies: European network of energy agen-cies (2002). The cost development of technologies is calculated as a smooth curve derived from the prospects of year 2000, 2005 and 2010. The cost devel-opment is for simplicity assumed equal for each band within a technology. • Cost development of non-renewable technologies: European Commission (1999);

International Energy Agency (2001b). Assumptions about development in be-tween years present in data.

• Production potentials: REBUS. Adjusted approximately with latest available data on production capacity in European network of energy agencies (2002); In-ternational Energy Agency (2002); Swedish National Energy Administration (2001); International Energy Agency (1998); Meyer and Nielsen (2001).

• Electricity production (renewable and non-renewable sources): International En-ergy Agency (2002); Swedish National EnEn-ergy Administration (2001).

(36)

4.4. RESULTS 27

• Future electricity consumption: European Commission (1999).

• Decommissioning of production capacity: European Commission (1999). Ad-justed to fit a shorter time span than present in data.

• State support: country reports from Faber et al. (2001); Huber et al. (2001); Meyer and Nielsen (2001); Cerveny and Resch (1998), and miscellaneous other sources mentioned throughout this document. Green certificates and renew-ables obligation are assumed to be implemented in Belgium, Denmark, Italy, The Netherlands, Austria, Sweden and the United Kingdom, and are thus not further examined for state support.

• National targets: International Energy Agency (2001a); Elcertifikatutredningen (2001); Hoogland and Schaeffer (1999). Adjusted proportionally to match the EU target. The increase of renewable energy for each year towards the 2010 target is assumed linear. Different variants for the progress have been suggested in different countries—in Sweden at decreasing speed and in UK first slowly, than faster, than slow again.

• General statistics: Eurostat (2001). Includes population data.

4.4 Results

0 1 2 3 4 5 6 7 8 9 10 eurocent / kWh 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 A. Full harmonisation B. Less harmonisation B: All costs included R1. State targets R2. Equal increase R3. Equal targets

Figure 4.3: Average accumulated certificate costs of new RES-E since year 2000

Figure 4.3 shows a comparison between the additional RES-E production costs in different scenarios. The less harmonisation scenario clearly gives the lowest certificate costs, staying below 2 eurocent per kWh, but if costs of additional support are included, the curve ends up near the values of the reference scenarios. Full harmonisation pro-vides the highest overall efficiency with cost developing from 1 to 2.9 eurocent per

(37)

kWh. The reference scenario using national targets gives a certificate cost 1 eurocent higher, while an equally increased RES-E share is a bit more efficient with approxi-mately 3.4 eurocent per kWh. In the scenario without harmonisation and with equal RES-E targets, the costs do, after a few years, run away towards levels that makes the scenario highly inefficient and unrealistic.

To translate the cost output into RES-E benefit, the RES-E target might be switched to a cost target using the outcome of the reference scenarios. The cost differences of less harmonisation and equal increase compared to full harmonisation equals approx-imately an increased RES-E share of 1 per cent and for the state target reference sce-nario 2 per cent. This can also be expressed as a new target of 23 per cent and 24 per cent, respectively, compared to the original 22 per cent target.

0 2 4 6 8 10 12 14 16 18 20 eurocent / kWh 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 A. Full harmonisation B. Less harmonisation R1. State targets R2. Equal increase R3. Equal targets

Figure 4.4: Highest certificate costs of new RES-E since year 2000

If instead the highest certificate costs are looked upon, in figure 4.4, it gives a hint of the equilibrium price of green certificates. This represents a theoretical consumer price of certificates, the required price for the demanded supply to be profitable. These figures are even more favourable for the EU TGC system. The final maximum cost evens out at approximately 4 and 4.6 eurocent per kWh in the harmonisation scenarios, while in the reference scenarios the price do sooner or later jump up to very high levels, when expensive technologies are needed to fill the quotas. With full harmonisation this means an average producer surplus of 1.7 eurocent per kWh.

These high numbers are however a bit deceptive, as the price will be unique in each country with national systems, as shown in figure 4.5. The rapid increase is only true for some countries. If foremost Belgium, with small renewable resources and low al-ternative production costs, is excluded, the highest costs in the reference scenarios will be significantly lower. Even if the average certificate price seem to be higher in the na-tional approaches, this is not true for every state. For Denmark, Ireland, Italy and Fin-land it is considerably lower. Some states, such as IreFin-land and Denmark, where there are large cost-effective renewable energy sources will probably see a large increase in

Dynamiken hos ett EU-system för handel med gröna elcertifikat

DEPARTMENT OF THEMATIC STUDIES

MASTER OF SCIENCE THESIS

Dynamics of an EU System for Tradable Green Certificates

Niklas Knutsson

Table of Contents

List of Figures

List of Abbreviations

1

Introduction

1.1

Background

1.2

Purpose

1.3

Method

1.4

Outline

2

Concepts and Definitions

2.1

The Internal Electricity Market

2.2

Renewable Energy

2.2.1

Definition

2.2.2

Support Schemes

2.3

Green Certificates and Renewables Obligation

3

Green Certificates at EU level

3.1

Towards a European Certificate Market

3.1.1

Design Issues

3.1.2

National Experiences and Plans

3.1.3

Progress at EU level

3.2

Functionality and Implications

3.2.1

Demand, Supply and Price

3.2.2

Implications

4

Market Simulation

4.1

Scenarios

4.2

The Model

4.2.1

Core Functions

∑

∑

∑

∑

∑

4.2.2

Energy Cost Analysis and Impact of State Support

4.2.3

Implementation

4.2.4

Limitations

4.3

Input Data

4.4

Results